#emc | Logs for 2011-07-28

[00:00:06] <IG-garage> you mean metric collets will be capable to clamp almost any diameter until the maximal is reached, and imperial will not?
[00:01:43] <andypugh> ER32 have a 1mm range, and come in 1mm increments. So any diameter can be held. The sets don't work out so neatly in imperial.
[00:02:10] -!- cjdavis1 [cjdavis1!~cjdavis@cpe-71-67-99-208.cinci.res.rr.com] has joined #emc
[00:02:12] <andypugh> http://www.ctctools.biz/servlet/the-13/FULL-ER11-COLLET-SET/Detail
[00:02:41] <FinboySlick> ER11 do 0.5mm range and come in 1mm increments if I remember correctly.
[00:03:18] <FinboySlick> That's still not a whole lot of collets given that the biggest is 7mm.
[00:03:19] -!- SWPadnos has quit [Read error: Connection reset by peer]
[00:03:24] -!- SWPadnos_ [SWPadnos_!~Me@74-92-8-214-NewEngland.hfc.comcastbusiness.net] has joined #emc
[00:03:30] SWPadnos_ is now known as SWPadnos
[00:03:59] -!- theorb [theorb!~theorb@] has joined #emc
[00:04:22] -!- theorbtwo has quit [Ping timeout: 264 seconds]
[00:04:37] theorb is now known as theorbtwo
[00:04:49] -!- SWPadnos has quit [Changing host]
[00:04:49] -!- SWPadnos [SWPadnos!~Me@emc/developer/SWPadnos] has joined #emc
[00:05:00] -!- cjdavis has quit [Ping timeout: 255 seconds]
[00:05:20] <andypugh> 13 of them. 0.5-1, 1-1.5, 1.5-2....6.5-7 is 13..
[00:05:43] <Tom_itx> http://www.maritool.com/p28/ER-11-COLLET-13-PIECE-METRIC-SET/product_info.html
[00:06:09] <FinboySlick> Tom_itx: Those are very nice.
[00:06:35] <andypugh> ANd only 10x the price of the Hong Kong set ;-)
[00:06:52] <Tom_itx> better quality
[00:07:02] -!- pjm__ [pjm__!~pjm@uhfsatcom.plus.com] has joined #emc
[00:07:37] -!- MOGLI has quit [Remote host closed the connection]
[00:07:52] <Tom_itx> i got an angle drill from harbor freight once (HK) and was able to use the drill one time before the gears broke in the head
[00:07:56] <andypugh> Aye, if I was buying a set to make money with, I would probably go for the expensive set. When I am spending my wages on toys, it's the cheap set.
[00:08:08] <FinboySlick> Maritools has some pretty awesome tools, but not a whole lot of the kind I need besides those collets.
[00:09:02] <FinboySlick> andypugh: I'll probably go for the maritools. I'm in a dreadful situation of having to buy more equipment to avoid taxes ;) Oh the terrible terrible life of a small corporation.
[00:09:44] <andypugh> You need to invest in your UK manufacturing facility :-)
[00:09:48] <FinboySlick> Besides, given the current economic times, currency won't hold its value as long as the tools.
[00:10:26] -!- pjm has quit [Ping timeout: 260 seconds]
[00:10:32] <FinboySlick> Which says a lot considering that we're talking about cutting tools ;)
[00:12:03] <FinboySlick> Hmm, apparently no SPTM on maritools.
[00:13:22] -!- Valen [Valen!~Valen@2001:44b8:3199:9100:21e:8cff:fe63:55df] has joined #emc
[00:13:41] <FinboySlick> Should get some corner radius for roughing, right? Doesn't wear as quickly.
[00:13:52] <FinboySlick> Or is it not worth it?
[00:14:06] <Tom_itx> if you need radius'd corners
[00:14:19] <Tom_itx> if you want roughing, get roughing cutters
[00:15:09] <PCW> Andy have you tried the new SSLBP bitfile with the 8I20?
[00:15:17] <FinboySlick> Tom_itx: Heh, yanking me back to obviousland, aren't you?
[00:15:27] <andypugh> The V23?
[00:15:37] <andypugh> Yes. Though not with a motor attached.
[00:15:37] <PCW> V24
[00:15:52] <andypugh> Just a moment
[00:15:56] <PCW> Cant get it to work at all
[00:16:01] <tom3p> in the the 'master' branch set of configs, every .ini file says "# motion controller, get name and thread periods from ini file" but i cant find the words base-thread or servo thread in any of them.
[00:16:02] <tom3p> how are thread names specified? i need a very fast thread for a hal based oscillator
[00:16:48] <FinboySlick> Holy crap, these things get expensive fast :P
[00:16:56] <Tom_itx> http://www.travers.com/skulist.asp?RequestData=CA_Search&navPath=All+Products%2F%2F%2F%2FUserSearch1%3D1%2F2%22+roughing+cutter&q=block+id+99858+and+class+level3+id+30168&minPrice=$24.44
[00:17:02] <Tom_itx> just a little
[00:18:32] <FinboySlick> Tom_itx: I can't take bigger than 1/4" so they won't get overly expensive individually. I just scrolled down the page to a 200$ endmill...
[00:19:23] <Tom_itx> you may not find a 1/4" roughing endmill, i'm not sure
[00:19:45] <andypugh> PCW: You are using my latest patch?
[00:20:06] <FinboySlick> With the kind of speed I spin it at, it might be pretty pointless anyway, I don't think my mill can move fast enough to give it enough meat.
[00:20:19] <FinboySlick> I'd have to go two-flute at 1/4.
[00:20:24] <PCW> Im using what Matt is using
[00:20:45] <andypugh> OK, he seems to have my patches
[00:20:54] <Valen> FinboySlick: the mill is the cheap part of owning a mill ;-P
[00:21:14] <andypugh> There is a lot to be said for using a shaper in the home
[00:21:44] <andypugh> Tooling is just HSS lathe tools, you can sharpen yourself. And nobody seems to want them.
[00:22:12] <andypugh> PCW: What is not happening?
[00:22:27] <FinboySlick> Valen: It's just a hobby anyway. Most of my coworkers spend more on golf.
[00:23:09] <FinboySlick> Tom_itx: Thanks for that site, btw. They seem to have lots of cool stuff.
[00:24:02] <PCW> Just flakeyV22 runs for a while but fails in a random few seconds, ver 24 wont even start
[00:24:42] <andypugh> I seem to have v23.
[00:27:52] -!- mikegg has quit [Read error: Connection reset by peer]
[00:28:01] <tom3p> the M1xx commands get P & Q parms as floats like 1.000000, alway 6 decimal places.
[00:28:03] <tom3p> in bash, do PP=${P/.*} and PP will be trunc'd to a dang integer
[00:28:20] <andypugh> What changed? I haven't done much to the driver at all, so it could be a change that I have missed.
[00:29:58] <PCW> is it svstua6_6_6.bit from freeby.mesanet.com?
[00:29:59] <PCW> Only change was minor (fix ID code)
[00:30:40] <PCW> well theres a lot internal that changed (support for parameter discovery on new remote cards)
[00:31:00] <andypugh> Hang on, are you talking about Smart Serial or UART here?
[00:31:17] <PCW> but both work fine and are interchangeble wth our testbench
[00:31:19] <PCW> SSERIAL
[00:31:25] <andypugh> I was using svss8_8
[00:32:04] <andypugh> svstua6_6_6 doesn't detect any SSerial devices at all.
[00:32:13] <PCW> sssvst8_8_8?
[00:32:42] <andypugh> loadrt hm2_pci config="firmware=hm2/5i23/svss8_8.bit num_sserials=8"
[00:33:25] <PCW> freeby.mesanet.com/sssvst8_8_8.bit should be rev 24
[00:33:37] -!- danimal_garage has quit [Ping timeout: 246 seconds]
[00:34:05] -!- danimal_garage [danimal_garage!~kvirc@ip68-111-132-33.sd.sd.cox.net] has joined #emc
[00:35:23] <PCW> btw if you skip writing the command reg if busy, this can be done with data instead of a conditional write
[00:35:25] <PCW> (any command reg write with MSB (bit 31) set is ignored
[00:35:36] <PCW> )
[00:37:48] <andypugh> Ah, yes, good point there, as I am not at all sure that the current driver actually does skip a write..
[00:38:03] <Connor> back from dinner
[00:38:13] <andypugh> I might have forgotten that it is all done by the magic of tram
[00:39:19] <PCW> I added the data dependancy so it could be done compatibly with TRAM
[00:39:20] <Connor> What is M2Al ?
[00:39:21] <PCW> If you ever write to the command register before its cleared, all bets are off
[00:40:36] <PCW> (I need to do this to the encoder control register as well so TRAM/DMA can always have the same address block for read/write)
[00:42:12] <andypugh> I need to move my 7i44 to a different header. Which header is sserial on in that bitfile?
[00:43:14] <PCW> It may be the V22.V24 difference is a race condition during startup if the command register handshaking
[00:43:15] <PCW> SSERIAL is on last conn
[00:46:11] <andypugh> The driver is not finding any sserial devices on P4
[00:46:25] -!- mikegg [mikegg!~mike@c-71-56-61-208.hsd1.ga.comcast.net] has joined #emc
[00:49:23] <PCW> yeah thats the symptom, let me verify that bitfile , just a sec...
[00:50:48] <andypugh> It contains modules, the sserial discovery code is being run.
[00:53:07] -!- vladimirek has quit [Remote host closed the connection]
[00:53:08] <PCW> Works fine here
[00:55:54] <andypugh> Where should I look for devide IDs
[00:56:07] <andypugh> 187505.635409] hm2/hm2_5i23.0: Smart Serial Firmware Version 24
[00:56:07] <andypugh> [187505.688827] hm2/hm2_5i23.0: Device Code 0 at addr 5c00
[00:56:07] <andypugh> [187505.688837] hm2/hm2_5i23.0: Device Code 0 at addr 5c04
[00:56:07] <andypugh> [187505.688845] hm2/hm2_5i23.0: Device Code 0 at addr 5c08
[00:56:08] <andypugh> [187505.688853] hm2/hm2_5i23.0: Device Code 0 at addr 5c0c
[00:56:08] <andypugh> [187505.688861] hm2/hm2_5i23.0: Device Code 0 at addr 5c10
[00:56:08] <andypugh> [187505.688869] hm2/hm2_5i23.0: Device Code 0 at addr 5c14
[00:56:09] <andypugh> [187505.688877] hm2/hm2_5i23.0: Device Code 0 at addr 5c18
[00:56:09] <andypugh> [187505.688885] hm2/hm2_5i23.0: Device Code 0 at addr 5c1c
[00:56:24] <PCW> I suspect that the command register protocol is being violated somewhere ,and the v22,23,24 difference is just a race condition
[00:56:26] <PCW> we couldn't get V22 to run more than a few seconds under EMC and V24 wouldn't even start, but on matts machine v22 sort-of worked
[00:57:19] <andypugh> No do-it-not-cleared here. So I think it isn't that
[00:57:52] <PCW> Device IDs are deprecated (look for string in user 1) but are still present in V24 V22/23 had a bug and show them as 0
[00:57:54] <PCW>
[00:58:31] <andypugh> I should look in 5e00 then?
[00:59:08] <PCW> for the device string yes
[00:59:34] <PCW> ID is still in MSByte of CSR (5CXX)
[01:00:38] <PCW> If you read a 0 at 5C00 MSByte you have a handshaking error
[01:03:54] <andypugh> I am seeing something in the user 1 now.
[01:07:33] <PCW> You should also see the device ID
[01:07:37] <andypugh> I have no idea how to cast a u32 into char[4]
[01:09:17] <PCW> Oh for the string
[01:09:55] <PCW> does the device ID show up now? (it should)
[01:10:54] <andypugh> I need to recompile to see
[01:13:42] <PCW> V24 does have the single byte device ID (just checked)
[01:17:27] <andypugh> I don't see them.
[01:17:28] <PCW> Its not important in the sense that the single byte device ID is deprecated,
[01:17:30] <PCW> but its important for debugging that you can read it
[01:17:47] <andypugh> [188708.072272] hm2/hm2_5i23.0: Device Code 1E188 at addr 5c00
[01:17:48] <andypugh> [188708.072282] hm2/hm2_5i23.0: Device Code 1E188 at addr 5c04
[01:17:48] <andypugh> [188708.072290] hm2/hm2_5i23.0: Device Code 1E188 at addr 5c08
[01:17:48] <andypugh> [188708.072298] hm2/hm2_5i23.0: Device Code 3E0000 at addr 5c0c
[01:17:49] <andypugh> [188708.072305] hm2/hm2_5i23.0: Device Code 1E188 at addr 5c10
[01:17:49] <andypugh> [188708.072313] hm2/hm2_5i23.0: Device Code 1E188 at addr 5c14
[01:17:49] <andypugh> [188708.072321] hm2/hm2_5i23.0: Device Code 2E188 at addr 5c18
[01:17:49] <andypugh> [188708.072329] hm2/hm2_5i23.0: Device Code 3E0000 at addr 5c1c
[01:17:57] <PCW> yeah works fine here for 8I20 and 7I64
[01:18:33] <andypugh> (that is with the mask and bitshift commented out, we should see 80nnnnnn and 74nnnnnn
[01:19:06] <PCW> looks like all errors
[01:19:08] <andypugh> Are you using EmC2 or your own software?
[01:19:21] <PCW> Our own
[01:19:22] -!- scanf has quit [Ping timeout: 264 seconds]
[01:20:27] <PCW> heres the sequence
[01:20:29] <PCW> start all (write 9ff to command register)
[01:20:30] <PCW> wait for command register to be clear
[01:20:32] <PCW> read all CSRs
[01:21:15] <andypugh> No offence, but at 0220 the sequence I am most interested in starts with cleaning teeth.
[01:21:35] -!- Guest719 [Guest719!EMCIRCUser@CPE0060673b2e91-CM00169242b37e.cpe.net.cable.rogers.com] has joined #emc
[01:22:07] <Guest719> any way to change the .ngc extention to just .nc by default?
[01:22:21] <andypugh> Yes, in the ini file
[01:22:27] <PCW> Sure, 'nite (or 'morning)
[01:25:08] <andypugh> Guest719: http://linuxcnc.org/docs/html/config_ini_config.html#sub:%5BFILTER%5D-Section
[01:29:37] -!- danimal_garage has quit [Ping timeout: 246 seconds]
[01:30:12] <Guest719> OPEN_FILE = /full/path/to/file.ngc
[01:30:27] <Guest719> dont see where the extention default is
[01:31:54] <andypugh> Add an entry to PROGRAM_EXTENSION
[01:32:56] <Guest719> ok still new to this so bare with me
[01:33:46] -!- Gensor has quit [Ping timeout: 252 seconds]
[01:34:30] <PCW> andypugh is it possible you have issued a setup start rather than normal start? the 3E state code indicates a setup start
[01:34:50] -!- scanf [scanf!~x32@unaffiliated/scanf] has joined #emc
[01:36:04] <andypugh> Yes. It is definitely a temporary setup start at that point, while we query the hardware. There was a reason for that which I have forgotten
[01:36:31] <andypugh> But is is working with the strings now.
[01:36:41] <andypugh> I am about to send you a patch.
[01:38:04] <PCW> OK you probably should not use a setup start at all
[01:38:57] <PCW> Are you using the file from the website? the setup start will also give you the id byte
[01:39:10] <PCW> (with V24)
[01:39:10] <andypugh> Yes.
[01:39:17] <andypugh> Odd, isn't it?
[01:39:43] <PCW> OK so theres still some timing violation or something
[01:39:57] <andypugh> Do you not need setup start to read the ID string? Presumably that register contains actual data in normal running
[01:40:48] <PCW> No you use the normal start, setup start is only for control communication
[01:41:26] <andypugh> Was that always the case? I have a very clear comment that I am starting in setup mode in order to read ID bytes.
[01:42:23] <PCW> like for EEPROM parameter setup etc
[01:42:25] <PCW> In any case you need to run normal start till the command register is clear before normal real time mode comm starts (the first DOIT)
[01:42:56] <andypugh> Once the board ID is known the driver shuts it straight back down again.
[01:43:44] <PCW> It was always the case, setup mode is only for setup of remote device
[01:43:47] <andypugh> I think I used setup start so that the watchdog doesn't bite.
[01:44:24] <andypugh> Whether that was wise and/or necessary I am not sure.
[01:44:29] <PCW> The watchdog will never bite until you issue the first doit
[01:44:47] <andypugh> OK, lets try it
[01:44:50] <PCW> (well about 50 mS after the first doit)
[01:46:53] <PCW> Still not convinced you have the right bitfile can you verify the ver?
[01:47:01] <andypugh> v24
[01:47:48] <PCW> Maybe you dont print all the data.
[01:48:18] <PCW> ID is MSB of CSR
[01:48:39] -!- Guest719 has quit [Quit: Visitor from www.linuxcnc.org]
[01:49:07] -!- tom3p [tom3p!~tomp@74-93-88-241-Illinois.hfc.comcastbusiness.net] has parted #emc
[01:49:10] -!- mrnuke_ [mrnuke_!~mrnuke@2001:55c:4c1e:55f3:f5f6:c989:36b2:3678] has joined #emc
[01:49:16] -!- Guest719 [Guest719!EMCIRCUser@CPE0060673b2e91-CM00169242b37e.cpe.net.cable.rogers.com] has joined #emc
[01:49:18] -!- Guest719 has quit [Client Quit]
[01:49:25] -!- mrnuke_ [mrnuke_!~mrnuke@2001:55c:4c1e:55f3:f5f6:c989:36b2:3678] has parted #emc
[01:49:43] <andypugh> OK, starting up in normal mode is somewhat disastarous.
[01:50:38] <Tom_itx> what board is this for?
[01:51:17] <andypugh> PCW: [190881.061122] hm2_5i23.0: firmware: requesting hm2/5i23/sssvst8_8_8.bit
[01:51:18] <andypugh> [190881.274903] hm2/hm2_5i23.0: Smart Serial Firmware Version 24
[01:52:31] <andypugh> I am not sure where the problem lies, bit if the initial discover startup is in normal mode then dmesg fills with millions of "Current request bigger than EEProm max" errors.
[01:52:51] <PCW> EPP?
[01:53:29] <PCW> Probably means it bed time...
[01:54:07] <andypugh> EEPROM
[01:54:38] <PCW> what does that have to do with the price of onions in china?
[01:54:49] <Tom_itx> they eat onions?
[01:55:48] <andypugh> It's actually a bit odd, as the 8i20 part of the driver triggers both over-current and under-current (setting) errors every thread if I initially discover the board in normal mode.
[01:57:47] <andypugh> Let me send you what I have now, which at least seems to work. I won't be coding again until Monday now.
[01:58:38] <PCW> over/under current? how would it know?
[02:00:07] <andypugh> Setting. It seems to think that the max-current pin is both <0 and >nvram_max_current
[02:00:08] <PCW> Or does the 8I20 part read the 8I20s current setting? (which requires setup mode)
[02:00:22] <andypugh> That will be it, yes.
[02:00:40] <andypugh> That is probably why I am using setup mode, in fact.
[02:01:31] <andypugh> Yes, it reads out max current and serial number during device discovery.
[02:03:28] <PCW> Serial is in User0
[02:05:05] <andypugh> Yes, I just spotted that. However, it also fetches parameter 0x8e8 so that it can do software current limiting effectively (it needs to know the max to calculate the proportion to use)
[02:05:43] <andypugh> I imagine other boards might have parameters that need to be queried at startup?
[02:08:13] <andypugh> PCW: Please let me know if the second of those patches works better. But I really do need to log off now
[02:08:31] -!- andypugh has quit [Quit: andypugh]
[02:11:03] -!- mendelbuild [mendelbuild!~MendelBui@77-95-78-56.bb.cust.hknett.no] has joined #emc
[02:11:05] <PCW> Maybe theres some error in the transition (we have no problem reading the device IDs in setup or normal mode)
[02:11:07] <PCW> If you need setup mode I would do this
[02:11:08] <PCW> Setup start all
[02:11:10] <PCW> wait for command complete
[02:11:11] <PCW> find device strings, serial numbers in user0 and 1
[02:11:13] <PCW> do parameter access (per channel doit, wait for command complete)
[02:11:14] <PCW> stop
[02:11:16] <PCW> wait for command complete
[02:11:17] <PCW> normal start
[02:11:19] <PCW> wait for command complete
[02:11:21] <PCW> start real time when ready (doits to active channels)
[02:11:22] <PCW> once doits are started in normal mode, remote watchdogs can timeout so dont issue a doit other than in the normal servo thread
[02:11:24] <PCW> (should you read the log)
[02:13:52] -!- ries has quit [Quit: ries]
[02:32:01] -!- yk2 [yk2!~IceChat77@] has joined #emc
[02:53:10] <yk2> anyone here have experience with arduino + emc2?
[02:55:06] -!- Gondorian [Gondorian!~IceChat77@] has joined #emc
[03:01:57] * Gondorian slaps yk2 with a rainbow trout
[03:03:04] -!- yk2 has quit [Quit: Download IceChat at www.icechat.net]
[03:03:17] -!- Gondorian has quit [Quit: When the chips are down, well, the buffalo is empty]
[03:04:25] -!- pulsar has quit [Read error: Operation timed out]
[03:13:23] -!- factor [factor!~factor@] has joined #emc
[03:14:41] -!- pulsar [pulsar!~pulsar@memoryleaks.org] has joined #emc
[03:29:39] -!- cevad has quit [Ping timeout: 252 seconds]
[03:30:09] -!- cevad [cevad!~davec@host-174-45-229-40.bln-mt.client.bresnan.net] has joined #emc
[04:08:40] -!- FinboySlick has quit [Quit: Leaving.]
[04:11:14] -!- Calyp has quit [Remote host closed the connection]
[04:24:30] -!- sumpfralle has quit [Ping timeout: 258 seconds]
[04:26:20] -!- theos has quit [Ping timeout: 260 seconds]
[04:38:58] -!- theos [theos!~theos@unaffiliated/theos] has joined #emc
[04:39:18] -!- mhaberler [mhaberler!~mhaberler@extern-181.stiwoll.mah.priv.at] has joined #emc
[04:52:33] -!- cncbasher has quit [Ping timeout: 255 seconds]
[04:53:05] -!- cncbasher [cncbasher!~david@cpc15-hart9-2-0-cust101.11-3.cable.virginmedia.com] has joined #emc
[05:17:51] -!- ve7it has quit [Remote host closed the connection]
[05:26:10] -!- mhaberler has quit [Quit: mhaberler]
[05:29:07] <L84Supper> yk2: I don't recall anyone getting EMC2 to work on an Arduino
[05:29:28] -!- ve7it [ve7it!~LawrenceG@S0106009027972e37.pk.shawcable.net] has joined #emc
[05:29:52] <L84Supper> oh, i see he's been slapped out of the channel already.... with a big fish no less
[05:34:32] -!- ve7it has quit [Remote host closed the connection]
[05:45:51] -!- The_Ball has quit [Ping timeout: 276 seconds]
[05:45:59] -!- vladimirek [vladimirek!~vladimire@bband-dyn21.178-41-212.t-com.sk] has joined #emc
[05:46:09] <Loetmichel> mornin'
[05:46:42] -!- The_Ball [The_Ball!~The_Ball@123-2-12-83.static.dsl.dodo.com.au] has joined #emc
[05:52:05] -!- nullie [nullie!~nullie@dhcp249-232.yandex.net] has joined #emc
[05:56:36] <mikegg> woa, time for bed
[05:57:12] <mikegg> so I can pull the source for 2.5 and start mucking around with that, huh?
[05:57:14] -!- ve7it [ve7it!~LawrenceG@S0106009027972e37.pk.shawcable.net] has joined #emc
[05:58:39] <mikegg> how do you list the different branches?
[06:04:21] -!- ve7it has quit [Remote host closed the connection]
[06:04:29] <mikegg> night all..
[06:09:03] -!- mikegg has quit [Ping timeout: 255 seconds]
[06:22:48] -!- mendelbuild has quit [Ping timeout: 246 seconds]
[06:49:04] -!- mendelbuild [mendelbuild!~MendelBui@77-95-78-56.bb.cust.hknett.no] has joined #emc
[06:53:23] -!- e-ndy [e-ndy!~jkastner@nat/redhat/x-tngvlcjhymtvxcpo] has joined #emc
[07:17:04] -!- mhaberler [mhaberler!~mhaberler@] has joined #emc
[07:32:14] -!- mhaberler has quit [Quit: mhaberler]
[07:37:50] -!- mhaberler [mhaberler!~mhaberler@] has joined #emc
[07:50:53] -!- robh__ [robh__!~robert@5ace7086.bb.sky.com] has joined #emc
[07:50:57] -!- psha[work] [psha[work]!~psha@] has joined #emc
[08:44:52] <scanf> hi, for a db25 parport cable for emc on a simple xyz step/dir + estop config, is it okay if ive tied pins 21-25 together and grounded them?
[08:45:47] <mrsunshine> isnt 21-25 ground pins? :)
[08:45:53] <scanf> yes
[08:45:58] <scanf> i think
[08:46:01] <scanf> thats why im asking :)
[08:46:14] <mrsunshine> hmm, dont know
[08:46:16] <mrsunshine> seems to be alot different here and there
[08:46:16] <psha[work]> look at db-25 pinout description and see if they are gnd
[08:46:23] <mrsunshine> saw one now where 22- 24 is some data thingies also
[08:46:47] <scanf> if they arent in use as per emc2 config sholdnt it not matter if they are gnd'd or not
[08:46:56] <psha[work]> no
[08:46:58] <psha[work]> it matters
[08:52:04] -!- mhaberler has quit [Quit: mhaberler]
[09:19:12] -!- mendelbuild has quit [Ping timeout: 246 seconds]
[09:21:17] -!- nicko [nicko!~nicko@125-237-76-242.jetstream.xtra.co.nz] has joined #emc
[09:21:28] <nicko> ello all!
[09:21:30] -!- mhaberler [mhaberler!~mhaberler@] has joined #emc
[09:21:40] <nicko> question:
[09:21:55] <nicko> what is 'NML' ?
[09:22:31] <nicko> from the documentation: "Halui is a HAL based user interface for EMC, it connects HAL pins to NML commands."
[09:30:01] <psha[work]> it's messaging layer
[09:30:09] <psha[work]> used for communication between components
[09:30:27] <psha[work]> other layer is using hal objects in shared memory (pins, singals)
[09:34:05] <nicko> messaging layer as in taking input from the GUI to hal ?
[09:34:22] <nicko> what does the 'N' stand for ?
[09:34:46] <psha[work]> network
[09:34:52] <nicko> thanks!
[09:35:02] <psha[work]> no, taking input from gui to emc internals
[09:35:15] <psha[work]> not hal - it's using shared memory communication
[09:35:17] <psha[work]> mostly
[09:35:21] <nicko> hmmm
[09:36:08] <nicko> if I wanted to make my own GUI for non-CNC applications would I replace 'EMC' or hack into the NML commands ?
[09:36:24] <nicko> or of course - something else ...
[09:37:51] <nicko> imagine I had a interface that eventually after some user input had a either a list of velocity or positional values over time - oooor, some kind of representation of a motion path for an axis in the form of a function or parametric equation
[09:38:17] <nicko> so... no G-code
[09:38:47] <psha[work]> and what do you want to do next?
[09:38:49] <nicko> could something like that be easily slotted somewhere into the framework of EMC ?
[09:38:55] <psha[work]> with this parametric equation
[09:39:27] <psha[work]> sure, but if you don't need motion, realtime and other stuff it has no point in fitting in
[09:39:36] <nicko> I want to make an object move along the path defined by that equation - maybe custim kinematics, maybe not
[09:40:10] <archivist> you can put equations in gcode
[09:40:13] <psha[work]> if you want movements and EMC's modules responsible for them - then yes, you'd generally want to fit into emc
[09:40:31] <psha[work]> however if you don't need Gcode you don't need interp
[09:40:37] <psha[work]> but need to replace it with something similar
[09:41:09] <nicko> I was hoping to avoid making my own G-code conversion process
[09:41:59] -!- cevad has quit [Quit: Leaving]
[09:42:34] <nicko> I thought that somewhere within EMC - G-code would be stripped down to some representation of the motion paths that was pretty much a raw info - here now, here now, here now - or this fast now, this fast now, this fast now etc...
[09:43:36] <psha[work]> yes, that's what done in interp - translate gcode into motion queue
[09:43:40] <nicko> interp - is that a component behind the scenes in EMC - or are you refering to g-code in a general sence
[09:44:16] <nicko> well, I think I may have most of interps functionality already here ...
[09:44:16] <psha[work]> however if you can provide motion queue yourself you don't need interp
[09:44:38] <nicko> I just need to learn how to get the info I have into a form that EMC expects (from interp?)
[09:45:16] -!- mhaberler has quit [Quit: mhaberler]
[09:46:16] <nicko> 'motion queue' - what exactly is that ? a defined way of translating motion data/info ?
[09:46:36] <psha[work]> no, just a list of expected motions
[09:46:48] <nicko> at some time period ?
[09:46:49] <psha[work]> living somewhere in 'motion' module
[09:47:00] <nicko> does it talk in position or velocities ?
[09:47:11] <psha[work]> a sort of prepared list which may be processed in hard realtime environment
[09:47:26] <nicko> ok - this sounds promising !
[09:47:29] <psha[work]> i guess positions
[09:47:49] <nicko> you can switch between the two easily enough I guess
[09:47:57] <psha[work]> yes, better say 'coordinates'
[09:48:14] <nicko> hmmmm
[09:48:23] <psha[work]> since there is really lot of conversion involved between X0 and real joint jogs
[09:48:26] <archivist> similar to how a list of g1.... is just a list of movements and required speeds
[09:48:38] <nicko> where can I find info on this - in the sources ?
[09:49:12] <psha[work]> yes
[09:49:24] <archivist> it may be simpler to provide gcode sequences
[09:50:21] <nicko> but then I have to make them - which is hard
[09:50:48] <psha[work]> it's more simple then to make your own motion queue
[09:50:56] <psha[work]> remember that you need to work in realtime env
[09:50:59] <nicko> as I dont know it for starters
[09:51:04] <psha[work]> which impose lot of limitations
[09:51:30] <nicko> so if I had a profile - for instance, keeping things simple
[09:51:30] <psha[work]> for example in hard realtime you even don't have floating point operations
[09:51:41] <nicko> 2 axis -
[09:51:44] <nicko> a circle
[09:52:06] <psha[work]> consider gcode as a language to talk with interp
[09:52:15] <nicko> each axis can be defined by sin(t) and a phase difference
[09:52:21] <psha[work]> you'll get lot of goodies like backplot
[09:53:11] <nicko> you then go into Matlab or whatever and calculate a heap of units for each axis -
[09:53:15] <psha[work]> with your own motion you wont' be able to preview what you are doing (using standard stuff)
[09:53:39] <psha[work]> Heh, or either tell interp to draw circle :)
[09:54:00] <archivist> yes gcode knows about arcs
[09:54:08] <psha[work]> even about splines
[09:54:17] <nicko> yeh sure - but I'm keeping things simpel to illustate the, um, er, point (which I'm yet to think up)
[09:54:28] <nicko> ok, so you have a table of values
[09:54:38] <archivist> gcode is simple, you seem to be making it hard
[09:54:46] <nicko> positions for each axis - they are 't' apart
[09:55:10] <archivist> see inverse time mode
[09:55:11] <psha[work]> awk '{print "G1 X$1 Y$2"}'
[09:55:31] <nicko> how can I 'shove' those numbers into the EMC system to make my pretty machine make a pretty circle ?
[09:55:49] <archivist> a sequence of g1
[09:56:03] <nicko> at what frequency could I do that ?
[09:56:31] <archivist> a singly file of all, emc will use at the required speed
[09:56:52] <nicko> my required speed ?
[09:57:06] <nicko> t = um, I dunno - 1000Hz ?
[09:57:10] <psha[work]> yes, your required speed (if it's in allowed range)
[09:57:25] -!- cevad [cevad!~davec@host-174-45-229-40.bln-mt.client.bresnan.net] has joined #emc
[09:57:26] <nicko> yeh, that was the question - what is the allowed range ?
[09:57:29] <psha[work]> or what speed do you mean?
[09:57:32] -!- El_Matarife has quit [Quit: Nettalk6 - www.ntalk.de]
[09:57:35] <psha[work]> movement speed?
[09:57:38] <archivist> depends on your hardware
[09:57:40] <psha[work]> or feeding emc with commands?
[09:57:47] <nicko> feeding EMC with commands
[09:57:59] <nicko> which effectively becomes the resolution
[09:58:01] <psha[work]> your equation is changing 1000 times per second?
[09:58:33] <nicko> if I wanted to make a table of values like that - yeh easy ...
[09:58:50] <archivist> what is your real application?
[09:58:57] <psha[work]> hm, maybe you'll shed some light about your task?
[09:58:57] <nicko> I could program the 't' setting as big as my harddrive woudl allow all the output
[09:59:12] <nicko> real application:
[09:59:25] <nicko> a boom arm on a track
[09:59:39] <nicko> it has rotate, tilt, boom extension and track
[09:59:40] <archivist> eg camera?
[09:59:43] <nicko> yes
[09:59:53] <nicko> the camera has pan roll tilt also
[10:00:05] <nicko> the lens has zoom, focus, aperture etc...
[10:00:27] -!- mendelbuild [mendelbuild!~MendelBui@77-95-78-56.bb.cust.hknett.no] has joined #emc
[10:00:35] <archivist> ok then just program positions at a time and the trajectory planner will blend and make smooth
[10:00:37] <nicko> custom kinematics will come into play if I decided to use inverse kine from the camera ...
[10:00:39] <psha[work]> and what type of control do you want? kind of programming? 'follow this part, arm!'
[10:00:58] <psha[work]> s/part/path
[10:01:04] <nicko> I will model the machine in a 3D program
[10:01:19] <nicko> where all this carry on is done for me ...
[10:01:45] <nicko> then output its positions and/or velocities at time period 't'
[10:01:59] <archivist> you can also model the machine in emc see vismach models
[10:02:05] <nicko> yeh ?
[10:03:05] <nicko> thing is - EMC wont give me a through the lens view that maya will allow -
[10:03:15] <nicko> with all the pre-vis CGI build in
[10:03:26] <archivist> it cannot do that I know
[10:03:37] <nicko> (this is where the film/cine world kicks in)
[10:04:08] <nicko> Was thinking It would be neat to have a Maya plugin that was basically the EMC UI
[10:04:21] <archivist> btw my toy in the garden is an outside broadcast truck
[10:04:28] <nicko> and feed it the numbers it needs as its 'playing' in the Maya
[10:04:54] <nicko> and also encoder wheels going the other direction ... programming maya
[10:05:05] <nicko> oooh - cool!!
[10:05:11] -!- Connor has quit [Ping timeout: 240 seconds]
[10:05:17] -!- mhaberler [mhaberler!~mhaberler@] has joined #emc
[10:05:22] <psha[work]> so what you get from maya?
[10:05:23] <archivist> I think the positions you get will be a simple translation to gcode
[10:05:29] <psha[work]> planned trajectory?
[10:05:34] <nicko> I work with one of the Oscar winners of the Cablecam system !
[10:05:45] <nicko> we are building a version of it here in NZ
[10:05:48] <psha[work]> for what time period? 1s 10s? full trajectory?
[10:05:55] <nicko> will be used for sports and other OB
[10:06:10] <archivist> you can just tie focus zoom etc to other axes
[10:06:38] <nicko> psha >> tell me what it needs and I can (maybe) tell you if it can do it
[10:06:49] -!- ssi has quit [Ping timeout: 258 seconds]
[10:06:54] <nicko> Maya is a bit of a swiss army knife
[10:07:19] <psha[work]> you need trajectory points which you may feed into interp (using g0/g1)
[10:07:24] <archivist> just output a list and then awk that list to gcode
[10:07:35] -!- ssi [ssi!~ssi@app2.prototechnical.com] has joined #emc
[10:07:35] <psha[work]> this you'll get smooth movement and all goodies
[10:07:40] <nicko> 'tie focus to other axes' ? as in proprotional to keep a plane ? hmmm, yep - prob do that in maya
[10:07:50] -!- Connor [Connor!~Connor@] has joined #emc
[10:08:01] <archivist> emc can control 9 at the same time
[10:08:17] <psha[work]> archivist: ja3 has same limitation or not?
[10:08:34] <mhaberler> NML means 'neutral messaging language'. See tutorial on: http://www.isd.mel.nist.gov/projects/rcslib/
[10:08:55] <archivist> psha[work], dunno what you mean by that question
[10:09:25] <nicko> track, boompan, boomtilt, camtilt, campan, roll, focus, aperture, zoom
[10:09:40] <nicko> theres 9 - of course, you hardly ever use all of them
[10:09:57] <mhaberler> nicko: if you want to do 'raw trajectory progamming' see src/emc/task/emccanon.cc
[10:10:05] <nicko> roll ... awkward to set up
[10:10:25] <mhaberler> I think you're better of staying in gcode or like the python canon wrapper:
[10:10:28] <anonimasu> nicko: also you can tie other functions in
[10:10:51] <anonimasu> like write custom comps or m codes for extra functions
[10:10:53] <mhaberler> lib/python/rs274/author.py
[10:10:55] <archivist> well emc can respond to digital inputs too for you roll
[10:10:55] <anonimasu> if you run out of axes
[10:10:57] <psha[work]> btw mhaberler's python cannon is nice idea
[10:10:57] <nicko> hello all
[10:11:28] <nicko> yeh ?
[10:12:26] <elmo40> if you guys could incorporate what I have been reading, it would make EMC a killer piece of software!
[10:12:38] <nicko> anonimasu>> thats good to know - there are other things, like digital switches and so on
[10:12:40] <archivist> it is already :)
[10:13:17] <nicko> The Terminator ?
[10:13:37] <nicko> Hal got kinda nasty eventually also huh
[10:13:45] <mhaberler> elmo40: so what you've been reading?
[10:14:25] -!- mhaberler has quit [Quit: mhaberler]
[10:14:34] <nicko> thanks for the links guys - a good chat - will read up
[10:14:38] <nicko> g'night!
[10:15:04] <archivist> nicko eg look at spindle speed output, that can be a pwm value that could be your aperture or
[10:16:22] <archivist> read the integrators manual
[10:20:57] <nicko> archivist> you get what I mean by multiple pass moco for vfx ?
[10:21:12] <nicko> imagine one pass at 24fps
[10:21:35] <nicko> another at 96fps with the mocon rig going 4x faster
[10:21:40] <nicko> for instance
[10:22:34] <nicko> the positions would need to be interpreted 4 x faster in EMC than the 24fps pass to ensure the same interpolations were made
[10:24:36] <nicko> vismach could be helpful yes -
[10:30:26] <psha[work]> nicko: that's a problem if you don't have full trajectory before move
[10:30:32] <psha[work]> if you have - it's not a problem
[10:32:02] <nicko> I need to know what you define a 'full trajectory' as
[10:33:11] <nicko> The only ways I can see that a trajectory could be defined is as either a function (spline whatever) - or a table of values over a given period
[10:33:12] <psha[work]> a part of trajectory you want to move on in next 1-5-10 seconds
[10:33:25] <psha[work]> points or splines - not important
[10:33:29] <nicko> yeh - sure I can have that any time its needed
[10:33:41] <nicko> at any resolution that you want
[10:33:48] <psha[work]> then you may load it into emc and travel through it with speed you want
[10:33:53] <nicko> the limit is the size of the harddisk !
[10:34:35] <nicko> but that infers that EMC has no limitation on the speed it can interp between these points I give it
[10:35:24] <nicko> say I had a nice motion profile (over time) I wanted an axis to follow
[10:35:33] <nicko> over one second
[10:35:49] <nicko> I have two points - beginning and end
[10:36:12] <nicko> great ! it made it ! but the profile was destroyed
[10:36:26] <nicko> so I give it say 100 middle points
[10:36:36] <awallin> normally the PID loops run at 1kHz but I think I've tried 4 kHz also, probably the mechanical bandwidth of your actuator will be the limiting factor?
[10:36:45] <nicko> but now the interp function is working 50 times faster
[10:36:58] <nicko> ah ha!
[10:37:10] <nicko> a number!
[10:37:13] <nicko> 1000Hz
[10:38:10] <nicko> which is without the hindsight of testing - sounding ok ;)
[10:39:28] <awallin> I think by default traj-calculations run at 1kHz also, not sure if that can be made to run faster. traj calculations is the input (set-point) for the pid-loop.
[10:40:18] <nicko> that gives me 41 interpolated points per standard cinema frame
[10:41:13] <nicko> which I think is going to work - there wil be other mechanical factors that will smudge the system before that did
[10:42:19] -!- mendelbuild has quit [Ping timeout: 250 seconds]
[10:44:10] <nicko> thats the PID - but does that mean you could throw "G1 X$1 Y$2" moves in the same *timely* fashion ?
[10:44:19] <nicko> this is where I have no idea ...
[10:45:20] <awallin> there is a limit of roughly one g-code block per traj-period. this isn't easy to work around. If you need faster than that you could stream position data from disk/RAM and not use g-code
[10:45:26] <psha[work]> i guess you don't need to throw 1khz G1 commands, just give some splines to cover trajectory and you'll get smooth movement without overcomplicating path description
[10:46:14] -!- jthornton has quit [Remote host closed the connection]
[10:46:34] <nicko> awallin >> thats exactly what I was thinking
[10:46:47] <nicko> stream the data into ?
[10:46:53] <nicko> and what format woudl it be in ?
[10:47:28] <nicko> I know I have a bit to read and I;m aksing the same questions again - but I'd like to soak this stuff up while you're all talking ! :)
[10:47:54] <awallin> probably one joint-position per period in some format (ascii?) there is a hal streamer component
[10:48:06] <nicko> brilliant
[10:48:21] <awallin> http://www.linuxcnc.org/docs/2.4/html/man/man1/halstreamer.1.html
[10:48:34] <awallin> not sure what happens if/when your disk decides to have a hickup...
[10:48:46] <nicko> RAM ?
[10:48:49] <nicko> http://www.linuxcnc.org/docview/html/man/man9/streamer.9.html
[10:49:00] <nicko> heh heh
[10:49:08] <nicko> you beat me to it
[10:50:02] <psha[work]> nicko: you'd really try gcode way first - it's much simplier than using streamer
[10:50:07] <psha[work]> since you have feedback
[10:50:22] <psha[work]> with streamer you are just dumping coordinates and hope that everything is going right
[10:50:42] <psha[work]> you'll end up with reimplmenetation of interp and friends
[10:51:48] <nicko> 240Hz G1 commands could work
[10:51:59] <awallin> improving the trajectory blending or "lookaehead" comes up now and then, it's not an easy task :) I think it requires some assumptions on the kinematics of the machine (i.e. 3-axis trivkins mill) before progress can be made
[10:53:14] <nicko> well, the kinematics fo the mahine should be worked out in Maya
[10:53:33] <nicko> it would push out data for the individual axes ...
[10:53:45] <nicko> not the camera position in world space
[10:53:54] <nicko> so yeh trivkins
[10:54:44] <nicko> oh hang on -0 yes I see what you're saying now
[10:55:14] <nicko> G-code assumes some kind of machine with orthoganal axes and so on
[10:55:59] <nicko> anyway - actual bedtime this time
[10:56:07] <nicko> goodnight and thank you all very much
[11:01:13] -!- jthornton [jthornton!~chatzilla@216-41-156-49.semo.net] has joined #emc
[11:08:04] -!- sumpfralle [sumpfralle!~lars@g226019085.adsl.alicedsl.de] has joined #emc
[11:14:32] -!- Calyp [Calyp!~Caly@unaffiliated/calyptratus-tzm] has joined #emc
[11:24:14] -!- mendelbuild [mendelbuild!~MendelBui@77-95-78-56.bb.cust.hknett.no] has joined #emc
[11:45:17] -!- servos4ever [servos4ever!~chatzilla@74-45-200-106.dr01.hnvr.mi.frontiernet.net] has joined #emc
[11:46:58] -!- ries [ries!~ries@] has joined #emc
[12:10:22] -!- sumpfralle has quit [Ping timeout: 264 seconds]
[12:29:25] -!- mhaberler [mhaberler!~mhaberler@] has joined #emc
[12:30:24] -!- mhaberler has quit [Client Quit]
[12:38:12] -!- theos has quit [Disconnected by services]
[12:38:25] -!- theos [theos!~theos@unaffiliated/theos] has joined #emc
[12:38:54] -!- factor has quit [Quit: Leaving]
[12:45:09] -!- i_tarzan has quit [Ping timeout: 252 seconds]
[12:47:25] -!- theos has quit [Ping timeout: 260 seconds]
[12:48:37] -!- FinboySlick [FinboySlick!~shark@] has joined #emc
[13:00:29] -!- theos [theos!~theos@unaffiliated/theos] has joined #emc
[13:01:01] -!- mendelbuild has quit [Ping timeout: 260 seconds]
[13:04:21] -!- mhaberler [mhaberler!~mhaberler@] has joined #emc
[13:11:08] -!- skunkworks [skunkworks!447329d2@gateway/web/freenode/ip.] has joined #emc
[13:11:37] <JT-Shop> the capacitors are here but I forgot to press order for the flux pen :/
[13:12:15] <skunkworks> flux pen?
[13:13:40] <JT-Shop> for the solder wick
[13:14:44] <JT-Shop> I'm removing the old capacitors from the power supply for my VMC drive
[13:17:52] <SWPadnos> solder wick should have quite a bit of flux in it already
[13:17:58] <SWPadnos> if you got the right kind
[13:18:49] <SWPadnos> it's a dry resin, but it will get all sticky gooey when you heat it up, and will leave lots of nice brown goo on the PCB :)
[13:19:15] <skunkworks> These are big caps - right?
[13:20:04] <JT-Shop> yea 30mm x 35mm or there abouts
[13:20:23] <skunkworks> I would use a suction end on your air hose. - get it liquid - quick spirt of air.
[13:21:00] <skunkworks> siphen - whatever they are called
[13:21:13] <skunkworks> or spelled ;)
[13:22:06] <JT-Shop> I actually have a rather large vacuum generator at the other shop...
[13:22:42] <JT-Shop> I'll bring it over here and test it out
[13:24:19] <JT-Shop> in the mean time I have to make a funny looking part on the lathe :)
[13:26:06] -!- Valen has quit [Quit: Leaving.]
[13:37:43] cjdavis1 is now known as cjdavis
[13:46:43] -!- theos has quit [Disconnected by services]
[13:46:57] -!- theos [theos!~theos@unaffiliated/theos] has joined #emc
[13:50:23] <skunkworks> SWPadnos: what have you been up to?
[13:50:37] <SWPadnos> mostly work
[13:51:09] <SWPadnos> heading to London next week though, for fun
[13:54:31] <skunkworks> awesome!
[13:55:41] -!- mhaberler has quit [Quit: mhaberler]
[13:56:02] <SWPadnos> gotta get to work. see you later
[14:23:28] -!- ktchk [ktchk!~eddie6929@n219079181189.netvigator.com] has joined #emc
[14:23:35] -!- ktchk [ktchk!~eddie6929@n219079181189.netvigator.com] has parted #emc
[14:27:51] -!- psha[work] has quit [Quit: Lost terminal]
[14:55:49] -!- cncbasher [cncbasher!~david@cpc15-hart9-2-0-cust101.11-3.cable.virginmedia.com] has parted #emc
[14:55:55] -!- ries_ [ries_!~ries@] has joined #emc
[14:57:21] -!- ries has quit [Ping timeout: 255 seconds]
[14:57:21] ries_ is now known as ries
[14:58:18] -!- sumpfralle [sumpfralle!~lars@] has joined #emc
[15:06:13] -!- cncbasher [cncbasher!~david@cpc15-hart9-2-0-cust101.11-3.cable.virginmedia.com] has joined #emc
[15:14:29] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #emc
[15:21:03] -!- syyl [syyl!~syyl@p4FD144BF.dip.t-dialin.net] has joined #emc
[15:26:02] -!- mendelbuild [mendelbuild!~MendelBui@77-95-78-56.bb.cust.hknett.no] has joined #emc
[15:30:32] -!- nullie has quit [Quit: Ex-Chat]
[15:31:33] -!- theos has quit [Ping timeout: 255 seconds]
[15:35:55] -!- tom3p [tom3p!~tomp@74-93-88-241-Illinois.hfc.comcastbusiness.net] has joined #emc
[15:44:20] -!- theos [theos!~theos@unaffiliated/theos] has joined #emc
[15:49:22] <Loetmichel> re @ home
[15:53:31] -!- psha [psha!~psha@] has joined #emc
[15:59:18] -!- JT-Shop has quit [Read error: Connection reset by peer]
[15:59:50] -!- JT-Shop [JT-Shop!~chatzilla@216-41-156-49.semo.net] has joined #emc
[16:18:20] -!- psha has quit [Quit: Lost terminal]
[16:18:25] -!- Birdman3131 has quit [Read error: Connection reset by peer]
[16:29:32] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[16:32:24] <anonimasu> skunkworks: your resolver to quadrature converter how does that work?
[16:32:28] <skunkworks> hmm - how come extruded aluminum (6061) is cheaper than even hotrolled the same size?
[16:33:14] <Connor> Higher Demand or Cheaper manufacture process ?
[16:33:18] <skunkworks> anonimasu: I have only played with it a little. It seems that it needs some work. - I would go with pico systems or I think mesa has them also - but I don't know if they are supported)
[16:34:08] <skunkworks> A few on here have used the pico systems resolver->quadautre converter.
[16:36:35] <anonimasu> my ardim trying to do similiar stuff
[16:36:42] <anonimasu> convert a sine/cos signal to quadrature
[16:36:49] <anonimasu> but it seems the arduino if pretty slow
[16:41:56] <anonimasu> any idea how to solve that?
[16:42:03] <anonimasu> prehaps setting the prescaler to a bigger number...
[16:42:03] <skunkworks> have you looked at the code that andy wrote?
[16:43:18] <L84Supper> skunkworks: you're probably getting pricing on low volumes, steel is still lower cost per Kg than Aluminum
[16:44:12] <L84Supper> I get all kinds of crazy pricing on small pieces
[16:45:46] <anonimasu> no, i'm trying it now
[16:50:49] -!- kbarry [kbarry!~chatzilla@173-13-173-249-sfba.hfc.comcastbusiness.net] has joined #emc
[16:54:31] <skunkworks> anonimasu: the code also only does 1000 counts per rev. Which for some things might be too small.
[16:54:57] <skunkworks> while the comercial ones do 4000+ counts/
[16:55:09] -!- IchGuckLive [IchGuckLive!~chatzilla@95-89-104-26-dynip.superkabel.de] has joined #emc
[16:55:41] <anonimasu> skunkworks: my commercial one ws supposed to handle like 64000counts/rev
[16:55:51] <anonimasu> but they failed to get me a adapter for setting it up and programming it
[16:55:57] <anonimasu> and the manufacturer cant sell them separate
[16:56:00] <anonimasu> BISS
[17:03:57] -!- ve7it [ve7it!~LawrenceG@S0106009027972e37.pk.shawcable.net] has joined #emc
[17:04:04] <anonimasu> I'm supposed to have 100khz sine signal at maximum speed from my motors..
[17:04:13] <anonimasu> err max allowed by the encoder..
[17:05:20] -!- tom3p has quit [Quit: Ex-Chat]
[17:09:57] -!- ktchk [ktchk!~eddie6929@n219079181189.netvigator.com] has joined #emc
[17:13:47] kbarry is now known as kb18951452
[17:14:06] kb18951452 is now known as kbarry
[17:20:41] -!- ktchk [ktchk!~eddie6929@n219079181189.netvigator.com] has parted #emc
[17:28:13] -!- nullie [nullie!~nullie@nullie.telenet.ru] has joined #emc
[17:30:37] <JT-Shop> YEA! new energy efficient AC unit is in and running in the garage
[17:31:35] <pcw_home> I guess a smart interpolator could just square up the signals for high speeds and interpolate accurately for low
[17:31:37] <pcw_home> my guess is you need at least a 1-2 M sample per second A-D.
[17:31:39] <pcw_home> Where you run into trouble with uProc based interpolators is outputing high speed quadrature
[17:31:41] <pcw_home> This is easier in a FPGA where the output integrator/counter is in hardware (this is what our resolver interface does)
[17:32:46] -!- mendelbuild has quit [Ping timeout: 250 seconds]
[17:37:46] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #emc
[17:41:27] -!- e-ndy has quit [Quit: Ex-Chat]
[17:47:34] -!- scanf has quit [Ping timeout: 264 seconds]
[17:53:50] -!- danimal_garage [danimal_garage!~kvirc@ip68-111-138-74.sd.sd.cox.net] has joined #emc
[17:54:26] <danimal_garage> hi
[17:55:53] <danimal_garage> so i'm trying to install ubuntu 10.04 on my pc from the live cd, but the screen is just text, like terminal. it doesn't show the standard install or run from live cd screen
[17:56:04] <danimal_garage> is that a sign my pc is too crappy?
[17:56:48] <anonimasu> pcw_home: i cont program that kind of things fpga's..
[17:56:50] <anonimasu> cant..
[17:56:55] <danimal_garage> it's an old celleron 1.1khz with i think 128mb of pc133 ram
[17:57:20] <archivist> that is a bit ram starved
[17:57:49] <danimal_garage> yea
[17:57:55] <danimal_garage> i dont know if i have any more
[17:57:57] -!- Loetmichel has quit [Ping timeout: 258 seconds]
[17:57:57] <archivist> and not chip has clocked that slow since the 1950's
[17:58:03] <archivist> not/no
[17:58:06] <danimal_garage> ghz
[17:58:08] <danimal_garage> sorry
[17:58:23] <archivist> slight difference :)
[17:58:49] <danimal_garage> i told you it was old!
[17:59:03] <danimal_garage> should i dump that one?
[17:59:21] <archivist> just add ram and a sensible video maybe
[18:00:28] <danimal_garage> ah yea it's onboard video, probably shared memory
[18:00:43] <danimal_garage> unfortunately no agp slot
[18:00:56] <danimal_garage> so video would cost as much as a new used pc
[18:01:08] <archivist> pci slot videos are ok like an old matrox card
[18:01:26] <danimal_garage> trying to find one isnt easy
[18:02:02] <archivist> its easy when one has a few lying around...but Im wrong side of the pond for you
[18:02:06] <JT-Shop> Dan try 8.04 on your antique computers :)
[18:02:33] <danimal_garage> JT-Shop: na, not worth it
[18:02:33] -!- Loetmichel [Loetmichel!Cylly@p54B1361F.dip.t-dialin.net] has joined #emc
[18:02:47] <danimal_garage> i have a better suitable pc, but i was trying to save that for something else
[18:04:01] <anonimasu> pcw_home: I Cant even handle normal hand turning on the Ác
[18:07:48] -!- mozmck has quit [Remote host closed the connection]
[18:07:53] <danimal_garage> i got the mini machine ready for my mesa boards when they get here today
[18:08:07] <danimal_garage> just gotta drill and tap some holes for the standoffs
[18:08:07] <anonimasu> pcw_home: what im basically doing is im reading the input and quantizing it
[18:08:54] <danimal_garage> it's pretty nice, all the home and limit switches are already wired up and marked
[18:09:10] -!- mozmck [mozmck!~moses@client-] has joined #emc
[18:09:36] * anonimasu sighs
[18:09:41] <anonimasu> is 100khz so damn fast?
[18:09:51] <danimal_garage> it has 3 opto22's in there already, i think i can snag one for the spindle maybe
[18:10:04] <archivist> anonimasu, for software yes
[18:10:42] <anonimasu> archivist: so to get stuff working, I need a schmittrigger :S
[18:10:50] -!- IchGuckLive has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.18/20110628230241]]
[18:11:07] <archivist> no you need a faster loop in the software or a real hardware counter
[18:11:26] <archivist> think loop time
[18:11:37] <anonimasu> archivist: did you miss that i have proper hardware but im unable to configure it
[18:11:39] <anonimasu> :/
[18:11:59] <anonimasu> since my supplier screwed up and they cant produce a separate adapter without refunding me a board ordering a new kit from the manufacturer
[18:12:02] <anonimasu> and..
[18:12:19] <archivist> no, I also didnmiss that you are using a microcontroller
[18:12:42] <anonimasu> this is a emergency try, i want to spin my motors somehow
[18:13:10] <anonimasu> and waiting 4 weeks for a damn adapter to see why my signal is too high/low/too much offset/wrong settings... is kindof shitty
[18:23:07] <anonimasu> archivist: btw, the figure is at something like 9k rpm or something
[18:23:45] <Tom_itx> anonimasu are those the servos you got cheap?
[18:23:57] -!- Tech_Talk [Tech_Talk!~quassel@ip70-178-238-227.ks.ks.cox.net] has joined #emc
[18:24:00] <anonimasu> yeah
[18:24:05] <Tom_itx> worth it?
[18:24:14] <anonimasu> yes
[18:24:26] <Tom_itx> kinda frustrating it seems
[18:24:37] <anonimasu> right now yes, but it'll get better
[18:24:48] <Tom_itx> how many did you get?
[18:24:50] <anonimasu> 9
[18:24:59] <Tom_itx> send me 3 once you get them figured out :D
[18:24:59] <anonimasu> for 500eur
[18:25:02] <anonimasu> haha
[18:25:13] <Tom_itx> or 4?
[18:25:13] <anonimasu> I'll happily send them to you for 2000eur each :)
[18:25:25] <anonimasu> it's still cheaper then retail ;)
[18:25:25] <Tom_itx> i bet you would
[18:25:58] <anonimasu> im mostly annoyed because the stupid distributor here in sweden cant produce a adapter for setting up the decoder chips
[18:26:13] <Tom_itx> oh, they're programmable?
[18:26:17] Tech_Talk is now known as Birdman3131
[18:26:17] <anonimasu> yeah
[18:26:44] <Tom_itx> you can't find a data sheet and make an adapter?
[18:27:30] <anonimasu> no data about that
[18:27:44] <Tom_itx> figgrs
[18:28:23] <danimal_garage> how big are they?
[18:28:28] <danimal_garage> the servos
[18:28:35] <anonimasu> 1.9kw..
[18:28:59] <Tom_itx> what were they off of?
[18:29:03] <anonimasu> robots
[18:29:13] <danimal_garage> nice
[18:29:23] <danimal_garage> i'll take a few
[18:29:26] <anonimasu> joking aside they arent for sale.
[18:29:29] <anonimasu> :)
[18:29:35] <Tom_itx> danimal_garage, take a number
[18:29:39] <danimal_garage> making a transformer?
[18:29:41] <anonimasu> got machines for them already
[18:29:53] <anonimasu> already did, i expect to pull a whopping 500watts out of them
[18:29:57] <anonimasu> for my small lathe..
[18:32:03] <danimal_garage> someone's supposed to come by and look at my small lathe
[18:32:03] <anonimasu> found some docs on the protocol
[18:32:07] <anonimasu> http://www.lna.br/~det/Projetos/Encoder/doc/BiSS_b2es.pdf
[18:32:07] <danimal_garage> hopefully he buys it
[18:32:48] <Tom_itx> why you gettin rid of it?
[18:33:01] <anonimasu> they use the paralell port for interfacing it but you still need a level adapter and some extra stuff
[18:33:12] <danimal_garage> dont need it
[18:33:22] <danimal_garage> i just got it cheap enough to not pass up
[18:33:35] <danimal_garage> i got 2 lathes already
[18:33:37] <Tom_itx> how big is it?
[18:34:26] <danimal_garage> 12x36
[18:34:39] <Tom_itx> anonimasu, do they have software to communicate with them?
[18:34:43] <anonimasu> yes
[18:34:52] <Tom_itx> that's a plus
[18:35:01] <anonimasu> they were supposed to include a adapter in the dev kit i bought but they didnt send the adapter
[18:35:11] <Tom_itx> why not?
[18:35:28] <anonimasu> because their supplier(that makes the chips) failed to send them with the kits
[18:35:41] <Tom_itx> that sucks
[18:36:06] <anonimasu> and they sent me a new board(I bought 2) hoping they would get me the adapter since it fits them both..
[18:36:12] <anonimasu> and with neither one arrived
[18:37:20] <Tom_itx> is that BiSS protocol a robotics protocol?
[18:38:02] <anonimasu> it's a special one that talks to the decoder chips
[18:38:17] <anonimasu> I think allen bradley themselves uses a fpga and a fast adc to handle it with counting internally on their drives
[18:38:52] <anonimasu> the motors just throw a sin/cos output
[18:40:30] <anonimasu> 1024per turn
[18:42:24] <anonimasu> but I dont know what do with them
[18:44:00] -!- cncbasher_ [cncbasher_!~quassel@cpc15-hart9-2-0-cust101.11-3.cable.virginmedia.com] has joined #emc
[18:44:16] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[18:44:33] <danimal_garage> resolvers?
[18:44:49] <anonimasu> no
[18:44:57] <anonimasu> it's the same stuff heidenhain uses for output on their encoders
[18:45:20] -!- micges [micges!~ddd@bxp40.neoplus.adsl.tpnet.pl] has joined #emc
[18:45:23] <danimal_garage> 1024 is the typical count for resolvers, isnt it?
[18:46:01] <anonimasu> you would think a avr can sample the analog inputs at 18hz..
[18:46:07] -!- micges [micges!~ddd@bxp40.neoplus.adsl.tpnet.pl] has parted #emc
[18:46:08] <anonimasu> err 17khz..
[18:46:18] <anonimasu> and throw me a squarewave..
[18:46:35] <anonimasu> :)
[18:46:37] <danimal_garage> ah
[18:47:08] <anonimasu> im hand spinning and it cant keep up now..
[18:48:39] <danimal_garage> that blows
[18:48:48] <danimal_garage> are you sure there isnt noise somewhere?
[18:48:57] -!- sumpfralle1 [sumpfralle1!~lars@g226016237.adsl.alicedsl.de] has joined #emc
[18:49:05] <anonimasu> no
[18:49:23] <anonimasu> but I dont have my scope handy so I can test that
[18:49:36] <danimal_garage> ah
[18:49:57] <anonimasu> they have a receiver circuit descibed in their datasheet, but i dont see the need when all im doing is hand spinning
[18:50:04] <anonimasu> and mesuring voltages over 3.2v
[18:50:10] <anonimasu> and giving a on/off output on higher ones
[18:50:25] -!- sumpfralle has quit [Ping timeout: 240 seconds]
[18:50:27] <anonimasu> but it seems like the avr loses counts when I spin fast by hand..
[18:50:43] -!- mendelbuild [mendelbuild!~MendelBui@77-95-78-56.bb.cust.hknett.no] has joined #emc
[18:50:49] <danimal_garage> no ac power running nearby?
[18:51:00] <anonimasu> well, the computer with the mesa
[18:51:22] <anonimasu> I dont have very extremely short wired tho.
[18:51:40] <anonimasu> it's like the avr cant toggle the output fast enough
[18:52:36] <anonimasu> or maybe it's the arduino overhead stuff.
[18:52:39] <danimal_garage> do you have a regular encoder or something that you can hook up to it? that way you can at least narrow it down to the board
[18:53:13] -!- JT-Shop has quit [Read error: Connection reset by peer]
[18:53:31] <anonimasu> you mean to the arduino?
[18:53:36] <anonimasu> or to the motor?
[18:53:38] <danimal_garage> yes
[18:53:49] <danimal_garage> arduino
[18:54:04] -!- JT-Shop [JT-Shop!~chatzilla@216-41-156-49.semo.net] has joined #emc
[18:54:21] <anonimasu> well, im gonna try after dinner to replace the digitalwrite stuff with a bitwise op
[18:54:54] <anonimasu> since i read that digitalwrite is dead slow
[18:55:29] <anonimasu> haha
[18:55:41] <anonimasu> arduino default libs are 20 times slower then a normal bitwise op
[18:55:41] <anonimasu> :)
[18:56:21] <anonimasu> 100khz should be possible with writedigital and 2.4mhz with the bitwise ops
[18:56:36] <Tom_itx> what if you fed the output to a schmidt trigger and used that on the int0 pin of the avr?
[18:57:00] <anonimasu> then i wouldnt need the avr
[18:57:03] <anonimasu> i could feed the mesa
[18:57:04] <anonimasu> :)
[18:57:09] <Tom_itx> well why not then?
[18:57:53] <Tom_itx> you get sin/cos output at what levels?
[18:59:28] <danimal_garage> 11. everything i own goes to 11
[19:05:08] -!- andypugh [andypugh!~andy2@cpc2-basl1-0-0-cust1037.basl.cable.virginmedia.com] has joined #emc
[19:06:10] <andypugh> It's G-code Jim, but not as we know it: http://pastebin.com/Dtqriqpt
[19:06:11] <anonimasu> 2.9-3.4
[19:08:51] <danimal_garage> ah crap! i got the wrong board
[19:09:07] <danimal_garage> :(
[19:09:15] <danimal_garage> no machine this week
[19:09:29] <cradek> andypugh: it looks angry
[19:09:41] <cradek> M04S00FA!
[19:10:07] <andypugh> I am guessing it is G-code in absolute steps.
[19:10:44] <anonimasu> it looks like someone if feeding the interpreter ulong's
[19:11:09] <andypugh> slongs, I think.
[19:11:20] <anonimasu> err sorry. yes :)
[19:11:39] -!- nullie has quit [Quit: Ex-Chat]
[19:11:44] <andypugh> I think those numbers probably go straight to the step accumulator.
[19:13:12] <cradek> andypugh: it can't be too simple, as it does arcs
[19:13:36] <danimal_garage> did not realize the 7i37 came without terminal blocks
[19:13:44] <cradek> and line 36 is surprising
[19:13:49] <andypugh> cradek: Ah, yes. Good spot.
[19:14:34] -!- mendelbuild has quit [Ping timeout: 264 seconds]
[19:15:17] <cradek> I think G0A is also a gcode
[19:15:22] <cradek> line 104
[19:15:29] <andypugh> G10 maybe?
[19:15:47] <andypugh> Though it doesn't make sense as an EMC2 G10
[19:16:04] <andypugh> Off that the byte-length seems variable in line 36
[19:16:09] <cradek> where did you find this? it might just be bogus.
[19:16:17] <cradek> %x where someone meant %g
[19:16:43] <andypugh> It's snooped from the interface between my friends Boxford lathe and control computer.
[19:17:03] <andypugh> So is raw serial bytes.
[19:18:02] <JT-Shop> danimal_garage: you didn't order the 7i37TA?
[19:18:14] <cradek> andypugh: cool
[19:18:34] <cradek> andypugh: looks easy enough to decipher and generate
[19:19:24] <andypugh> Indeed. I am trying to persuade him that an EMC2 conversion makes sense, but converting G-code to that doesn't look hard.
[19:20:04] <anonimasu> yeah but emc is defenetively better
[19:20:12] <anonimasu> if the controller isnt a really good one.
[19:20:32] <anonimasu> i wouldnt convert my heidenhain for anything because it's too efficient for hand coding
[19:20:34] <andypugh> It's a few eproms and chips on a board.
[19:20:49] <anonimasu> and for making one offs quicker then you can by hand
[19:20:49] <anonimasu> :)
[19:20:57] <danimal_garage> JT-Shop: apparently not
[19:21:05] <andypugh> There is a decent CAD/CAM system on it, but it has a very restrictive education license.
[19:21:08] <JT-Shop> ouch
[19:21:09] <anonimasu> but, if it were a bit older i'd throw it away quickly
[19:21:47] <JT-Shop> I'd never convert my CHNC to anything else as it is too efficient for hand coding
[19:22:09] <anonimasu> anything else then emc?
[19:22:10] <JT-Shop> EMC2 + ngcgui :)
[19:22:17] <anonimasu> hehe
[19:22:27] <danimal_garage> i don't hand code anything
[19:22:39] <anonimasu> i'd bet that heidenhain is better at that.. and more expensive $$$$$$$
[19:22:46] <anonimasu> as tomp said they are the mac's of cnc's :D
[19:22:47] <danimal_garage> mastercam or manual machines for me
[19:23:51] <Jymmm> Dictated by Steve Jobs???
[19:23:58] <danimal_garage> guess i'm a bit of a diva
[19:24:11] -!- mhaberler [mhaberler!~mhaberler@62-47-139-255.adsl.highway.telekom.at] has joined #emc
[19:25:36] <anonimasu> Jymmm: nah, just in how easy they are to use
[19:26:06] <Jymmm> anonimasu: ah
[19:26:57] <archivist> andypugh, have you got the ascii right is it 7bit by any chance
[19:27:01] <Jymmm> Lion has more poofy (useless) things, and they took out a lot of things too.
[19:27:56] <andypugh> My computer won't run Lion <sad>
[19:28:10] <Jymmm> andypugh: You're not missing anything.
[19:28:15] <anonimasu> with the heidenhain I have about 6 lines to make a progam to loop and make a helically interpolated hole
[19:28:20] <andypugh> Xcode 4?
[19:28:39] <cradek> emc 2.5 can make a helical hole with one line, no looping needed
[19:28:52] <andypugh> archivist: Not my logging, but I think it looks right.
[19:29:06] <archivist> I do a helical hole on 2.4 with one line too
[19:29:38] <cradek> archivist: you can't do multiturn in one line in 2.4
[19:29:51] <archivist> but then I cheat with a rotary axis :)
[19:30:08] <archivist> so I can :)
[19:31:30] <Jymmm> Ok, the washer is slow draining and now overflowing if on regular (delicate so-so). I tossed in some sulfuric acid drain opener a few days ago (followed by baking soda after 30m) and no luck this time. Someone mentioned there might be a lint trap, so I tried a 4ft snake no dice, and I can't see any trap on it (goes into the wall). Any suggestions or do I need to get it snaked out?
[19:32:26] <danimal_garage> maybe your snake is too short
[19:32:43] <Jymmm> it's a hand one, and all I got =)
[19:32:43] <archivist> said the actress to the bishop
[19:33:23] <danimal_garage> zing
[19:33:52] <Tom_itx> use a depth charge
[19:33:55] <danimal_garage> sulfuric acid drain opener?
[19:34:06] <Jymmm> yep
[19:34:11] <danimal_garage> isn't usually sodium hydroxide?
[19:34:33] <danimal_garage> i guess the acid will disolve the clothes
[19:34:46] <danimal_garage> but it'll probably take longer than 30 minutes
[19:34:53] <Jymmm> Even comes in a sealed plastic bottle sealed in a plastic bag
[19:34:59] <danimal_garage> and it'll probably disolve your washer too
[19:35:29] <Tom_itx> shove a garden hose down the drain
[19:35:33] <Jymmm> Yes, becasue I alwasy pour sulfuric acid in the washing machine instead of, um, the drin
[19:35:36] <Tom_itx> and plug it off
[19:35:37] <danimal_garage> is it a black bottle that said instant power hair and grease?
[19:36:03] <Jymmm> danimal_garage: No, it says "Sulfuric Acid"
[19:36:13] <Jymmm> maybe nitric acid
[19:36:50] -!- The_Ball has quit [Read error: Operation timed out]
[19:37:08] <Jymmm> http://www.idealtruevalue.com/servlet/the-60649/Detail
[19:37:43] <Jymmm> Hmmm, safe for PVC/ABS. I didn't realize that.
[19:38:40] <Tom_itx> call a plumber
[19:39:16] <Jymmm> first one said $150 to snake
[19:39:29] <Tom_itx> tell him there's no snake caught in it
[19:39:47] <Jymmm> I don't know that for sure =)
[19:39:54] <Jymmm> might be alligator
[19:39:54] <Tom_itx> you can rent one for $50
[19:39:57] * anonimasu slaps the arduino
[19:40:05] <Jymmm> anonimasu: kinky
[19:40:08] <Tom_itx> or try a garden hose first
[19:40:37] <Jymmm> I wonder if those expando things work?
[19:41:00] -!- vladimirek has quit [Remote host closed the connection]
[19:41:00] <Jymmm> http://lh4.googleusercontent.com/public/M7W9Jt2a4Z5-9JZpC9iOVUDSgkUcGzNYCmiPr5hbz0qUZs8NFtoNKFC-Nb2A2So-AuN4DQqfDP9tqgnS_4LvJSj7m7A95wjPEbAKEMIES3XYoIhiIqnYKTFpZqBgRvowtOK-NpH1-RTLGrnX_SdPpSIy1n6tecnQWdjXtMkv-cEiz1SwTfm_ua6HVnKjSfI7v-vLmhUIbPw
[19:41:01] <Tom_itx> seal off the drain with a rag and pump some water thru it
[19:41:55] <Tom_itx> it probably has a check ball on a spring to build enough pressure to expand the bladder
[19:42:01] <Tom_itx> then lets the water out
[19:42:21] <Tom_itx> just use a rag
[19:43:04] <Jymmm> 16" from 220V outlet, I'll spend the $8 AND a rag =)
[19:43:21] <Tom_itx> the 220v is on a breaker
[19:43:51] <Tom_itx> put some tape over it if you're worried
[19:44:07] <Jymmm> LOL, like tape has ever sealed water =)
[19:44:26] <Tom_itx> over the outlet
[19:45:05] <Jymmm> I can sorta see in the wall.... The washer drain goes in, then a T comes in from looks like the kitchen sink, then into the ground.
[19:46:04] <Jymmm> I think if I use that bladder thingy, I can slip past the kitchen T, and get to the main pipe. But I suspect that the sewer pipe is broke due to tree roots.
[19:46:35] <Jymmm> that same damn tree is breaking concrete
[19:46:52] <Jymmm> and that's 15ft away from the trunk too
[19:47:07] <Jymmm> I HATE sycamore trees
[19:47:15] <Tom_itx> good shade
[19:47:30] <Jymmm> and those needle balls that it drops
[19:48:43] <Tom_itx> it isn't plastic sewer pipe?
[19:49:16] <Jymmm> cast iron, except for the 12" PVC riser where the washer drain hose fits into
[19:49:30] <Tom_itx> probably clay out in the yard
[19:49:47] <Jymmm> Yeah, I'd suspect.
[19:49:55] <Tom_itx> i replaced all mine
[19:50:06] <Jymmm> I'm renting, so...
[19:50:28] <Tom_itx> tell the landlord the drain is plugged
[19:50:50] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/house/drain1.jpg
[19:50:54] -!- mendelbuild [mendelbuild!~MendelBui@77-95-78-56.bb.cust.hknett.no] has joined #emc
[19:51:00] <Jymmm> Ha! Haven't heard from him in years. Once he took a YEAR to cash a rent check.
[19:51:25] <danimal_garage> sweet, bidding wars on my lathe lol
[19:51:31] <danimal_garage> i should have listed it higher
[19:51:43] <JT-Shop> quit sending them and see how long it takes him to notice
[19:51:50] <danimal_garage> ha
[19:52:02] <Tom_itx> JT-Shop, you done anything on your probe yet?
[19:52:04] <JT-Shop> how much did you list it for Dan?
[19:52:07] <danimal_garage> 600
[19:52:12] <Tom_itx> i finally added adjuster screws to mine
[19:52:21] <danimal_garage> i'm not going to take more than that, i hate those kind of people
[19:52:28] <JT-Shop> Tom_itx: painting the walls in my shop... probe is on hold for now
[19:52:28] <danimal_garage> but at least it'll sell
[19:52:39] <Jymmm> JT-Shop: Tough situation... we have birds and about 95% of the places around here are NO PETS. the other 5% are one small dog/cat.
[19:52:45] <Tom_itx> i swear i'm gonna hide your paint bucket
[19:52:54] <danimal_garage> i thought 600 was fair for it
[19:53:05] <JT-Shop> I swear I'm going to give it to you shortly
[19:53:11] <danimal_garage> lol
[19:53:15] <skunkworks> our house was move to a new lot - new basement. when the did the plumbing - they tied into the old drain tile. after a year - they had to dig up the yard again and replace it all. (thank goodness it was before we bought it ;O
[19:53:17] <danimal_garage> you're not done painting yet?
[19:53:30] <Tom_itx> he's just stalling now
[19:53:46] <danimal_garage> yea, crap, i could paint my whole house in that amount of time
[19:53:49] <Tom_itx> he needs to epoxy the floor before he moves all his stuff in
[19:53:55] <danimal_garage> inside and out
[19:54:05] <danimal_garage> ugh, epoxy sucks
[19:54:05] -!- cncbasher_ has quit [Read error: Connection reset by peer]
[19:54:07] <JT-Shop> I just primed this last bit of wall http://gnipsel.com/shop/inside/inside-17.xhtml
[19:54:13] <Jymmm> skunkworks: Yeah, short cutting bastards. Cast all the way to the 12" main fsckers!
[19:54:24] <danimal_garage> i epoxied my floor..... not much left to it
[19:54:36] <danimal_garage> wore through the traffic areas
[19:54:44] <danimal_garage> and the coolant eats it up
[19:54:51] <JT-Shop> DIy epoxy?
[19:55:00] <danimal_garage> baehr
[19:55:03] <Tom_itx> did you acid etch the cement first?
[19:55:03] <danimal_garage> or whatever
[19:55:09] <danimal_garage> yes tom, it looked like new
[19:55:23] <Tom_itx> time to do it again then :D
[19:55:27] <danimal_garage> never
[19:55:31] <danimal_garage> until i sell the house
[19:55:50] <danimal_garage> then i'll paint it to cover up the evidence of a machine shop lol
[19:55:56] <JT-Shop> I've not seen many good reviews on consumer epoxy coatings for concrete
[19:56:40] <JT-Shop> I might go by the concrete store to see what they have
[19:56:45] <danimal_garage> where do you get yours John?
[19:56:48] <danimal_garage> ah
[19:56:55] <Tom_itx> i just think it would lighten up the room more
[19:57:02] <Tom_itx> if you went with a light grey
[19:57:25] <danimal_garage> the whole room is white, with 10000000000 lumens of light in there
[19:57:32] <danimal_garage> he needs sun glasses as it is
[19:57:34] <Tom_itx> heh true
[19:57:38] <JT-Shop> after reading the reviews if I do it will be commercial grade stuff
[19:57:50] <danimal_garage> slippery when wet
[19:57:52] <JT-Shop> I only have 1/2 of the lights up :)
[19:58:14] <danimal_garage> John, the best stuff i've worked on is the stick on tiles in industrial buildings
[19:58:24] <danimal_garage> seemed to hold up better than the paint
[19:58:44] <JT-Shop> this is how the lights will be in each quadrant http://gnipsel.com/shop/inside/inside-15.xhtml
[19:58:46] <danimal_garage> the ones that look like the old asbestos ones
[19:58:58] <JT-Shop> yea, those are tough ones
[19:59:24] <Tom_itx> you need a party ball in the middle of each one
[20:00:02] <danimal_garage> but i guess opinions are like assholes.... everyone has one
[20:00:29] <danimal_garage> guess i need to clean the drawers of this lathe out
[20:00:48] <Tom_itx> well, i think i would seal it with something to keep the oils from seeping into the cement
[20:00:49] <danimal_garage> scavange anything i want from it
[20:01:12] <danimal_garage> tom, sulferic acid will make the floor look like new in like 2 minutes
[20:02:04] <danimal_garage> so will murietic acid
[20:02:13] <danimal_garage> that's what i used on my driveway, it looked like new
[20:02:22] <danimal_garage> after 2 leaky fords
[20:03:10] * Jymmm counts danimal_garage's fingers... 1, 2, 3, 5...
[20:04:01] <L84Supper> concrete sealers are generally sodium silicate (water glass) + polymer or epoxy
[20:04:03] <Jymmm> danimal_garage: you're not suppose to stir the acid with your finger
[20:04:36] <JT-Shop> you should hear the OSB suck up the first coat of primer
[20:05:07] <L84Supper> the industrial floor coatings are usually epoxy + an aggregate
[20:05:27] <Jymmm> JT-Shop: Jsut wait till the 12th coat of primer
[20:05:51] <JT-Shop> Jymmm: it only takes 2 coats of primer on the OSB and one on the sheetrock
[20:06:15] <Jymmm> JT-Shop: Ah, heh.
[20:06:44] <PCW> Andy: We found that the ID byte is not written in setup mode, we were fooled by the fact we had run normal start previously
[20:07:56] <PCW> will fix in rev 25
[20:07:58] <PCW> tried to build from source but configure is missing and I dont have more time to mess about with it today
[20:08:14] <cradek> ./autogen.sh
[20:08:22] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #emc
[20:08:56] -!- adb [adb!~Moldovean@178-211-230-159.dhcp.voenergies.net] has joined #emc
[20:09:48] -!- The_Ball [The_Ball!~The_Ball@123-2-12-83.static.dsl.dodo.com.au] has joined #emc
[20:09:50] <PCW> Thanks!
[20:10:07] * anonimasu slaps the stupid-duino again
[20:10:13] <PCW> maybe the README should say this
[20:11:08] <cradek> for better or worse, our build instructions are pretty much just on the wiki: http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Installing_EMC2#Building_emc2_realtime
[20:11:38] <danimal_garage> PCW: apparently i ordered the wrong board
[20:11:53] <danimal_garage> didn't realize there was a 7i37 and 7i37ta
[20:12:25] <PCW> Yes we have many options to dazzle and confuse
[20:12:27] <danimal_garage> got the 7i37 :(
[20:12:50] <danimal_garage> can i send it back?
[20:13:15] <PCW> did you look crosswise at it?
[20:13:39] <danimal_garage> i may or may not have released gas in it's general direction
[20:13:49] <danimal_garage> i'll wash it first
[20:14:02] <danimal_garage> and throw it in the dryer
[20:14:06] <JT-Shop> lol
[20:14:18] <JT-Shop> works on my thumb drives
[20:14:21] <danimal_garage> haha
[20:15:05] <PCW> We can send you a 7I37TA
[20:15:26] <danimal_garage> PCW: do i send it to the addy on the top of the packing slip?
[20:15:36] <danimal_garage> cool, you got those in stock?
[20:15:44] <PCW> Yeah
[20:15:49] <danimal_garage> sweet
[20:16:02] <danimal_garage> i'll get it out today
[20:16:24] <danimal_garage> what's the difference in $?
[20:16:49] <danimal_garage> for the shipping too
[20:17:14] <danimal_garage> well i can just wait for her to clal me back for that stuff i guess
[20:17:18] <danimal_garage> call*
[20:17:28] <danimal_garage> i left a message but i assume it was lunch time
[20:18:07] <PCW> Just send the 7I37 back. I think theres a $10 diif
[20:18:46] <danimal_garage> ok, i'll throw some $$ in the envelope
[20:19:06] <danimal_garage> save her from running a card again
[20:19:22] <PCW> So how do you apply a git patch?
[20:20:01] <cradek> git am 0001-the-patch-name.diff
[20:20:09] -!- motioncontrol [motioncontrol!~i@host26-74-dynamic.11-87-r.retail.telecomitalia.it] has joined #emc
[20:22:31] <anonimasu> it works ^___________^
[20:22:47] <anonimasu> the arduino is fast enough it seems :)
[20:22:52] <anonimasu> atleast at 1rpm
[20:23:19] <skunkworks> anonimasu: the issue I was having was non linearity on a resolver (have a turn was not half a count)
[20:23:29] -!- mhaberler has quit [Quit: mhaberler]
[20:23:36] <anonimasu> my issue was that the arduino routines suck very very bad..
[20:23:45] <skunkworks> heh
[20:23:45] <anonimasu> so instead of writing a port taking one cycle it took 13
[20:24:17] <anonimasu> it works perfectly well now :)
[20:26:16] -!- motioncontrol has quit [Quit: Sto andando via]
[20:26:31] <anonimasu> we'll see when I get commutation going how well it'll work
[20:27:45] -!- mhaberler [mhaberler!~mhaberler@62-47-139-255.adsl.highway.telekom.at] has joined #emc
[20:29:21] <PCW> Andy, patch fails for me on mesa_other_sserial, index error
[20:30:57] -!- mendelbuild has quit []
[20:34:31] -!- mhaberler has quit [Quit: mhaberler]
[20:36:46] -!- The_Ball has quit [Ping timeout: 240 seconds]
[20:38:00] -!- The_Ball [The_Ball!~The_Ball@123-2-12-83.static.dsl.dodo.com.au] has joined #emc
[20:38:48] -!- nutbeam [nutbeam!~nutbeam@smtp.electroconnectinc.com] has joined #emc
[20:40:11] <PCW> probably faster to just send me the files screw this patch nonsense
[20:41:01] <JT-Shop> danimal_garage: http://gnipsel.com/shop/inside/inside-18.xhtml
[20:42:05] <anonimasu> nice!
[20:42:07] <anonimasu> im jealous.
[20:42:08] <anonimasu> :)
[20:42:11] -!- nutbeam [nutbeam!~nutbeam@smtp.electroconnectinc.com] has parted #emc
[20:42:24] <danimal_garage> i'm also jealous!
[20:42:40] <anonimasu> I could use that kind of space
[20:42:48] <danimal_garage> i could use half that space
[20:43:14] <JT-Shop> for some reason it is almost full and nothing is in there :/
[20:43:21] <anonimasu> tho, I have about 400m^2 at work :)
[20:43:26] <anonimasu> but it's not 5 meters from inside
[20:44:24] * anonimasu looks at your chnc
[20:44:35] <anonimasu> my lathe needs paint. and clearcoat.
[20:46:13] <anonimasu> :)
[20:46:34] <JT-Shop> Dan can you see how much difference the second coat makes in the middle where I overlapped just a bit
[20:46:48] <JT-Shop> my lathe needs some wheels
[20:47:22] <danimal_garage> yea
[20:47:29] -!- isssy has quit [Quit: Visitor from www.linuxcnc.org]
[20:47:33] <anonimasu> turn them on it ^_^
[20:47:35] <danimal_garage> my lathe needs wheels too....
[20:47:41] <danimal_garage> and leveling screws
[20:47:54] <anonimasu> did anyone see the dmg machines?
[20:48:00] <anonimasu> 3 feet instead of 4
[20:48:18] <JT-Shop> funny thing the CHNC does not have leveling screws only one to make it a solid mate with the floor
[20:48:49] <JT-Shop> they whole inside floats on the frame so it can't warp the ways
[20:48:57] <danimal_garage> same with mine John
[20:48:58] <JT-Shop> dmg?
[20:49:06] <anonimasu> deckel maho glidemeister
[20:49:16] <JT-Shop> no, I've not seen any
[20:49:53] -!- syyl has quit [Quit: Leaving]
[20:51:35] <JT-Shop> anonimasu: I have about 93m^2 in the shop and the same in the garage both about 12m from the house
[20:52:10] -!- ries has quit [Quit: ries]
[20:52:24] -!- skunkworks has quit [Ping timeout: 252 seconds]
[20:52:32] -!- mrsunshine has quit [Ping timeout: 246 seconds]
[20:52:58] <anonimasu> I cant find a good picture of the dmg machines
[20:55:27] <anonimasu> http://www.dmg.com/query/internet/v3/pdl.nsf/b18b67a883fa6b04c1256bab00757fd6/$file/pm0uk08_dmc60tdmu406080100monoblock.pdf
[20:57:37] <anonimasu> PAGE 8
[20:57:38] <anonimasu> :)
[21:00:00] -!- mrsunshine [mrsunshine!~mrsunshin@c-0a3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #emc
[21:00:34] -!- nutbeam [nutbeam!~nutbeam@smtp.dh-global.com] has joined #emc
[21:01:52] -!- mhaberler [mhaberler!~mhaberler@62-47-139-255.adsl.highway.telekom.at] has joined #emc
[21:02:36] -!- FinboySlick has quit [Quit: Leaving.]
[21:10:11] <JT-Shop> impressive machine
[21:12:39] <anonimasu> http://www.youtube.com/watch?v=nZs4399AEAU&feature=related
[21:12:42] <anonimasu> sounds very stable :)
[21:13:01] <L84Supper> anonimasu: what's the price range of that machine?
[21:13:49] <anonimasu> like 80k eur
[21:14:06] -!- dgarr [dgarr!~dgarrett@adsl-75-61-65-77.dsl.pltn13.sbcglobal.net] has joined #emc
[21:14:40] -!- mhaberler has quit [Quit: mhaberler]
[21:14:43] <anonimasu> over.. i think
[21:15:42] <anonimasu> a dmc64v is like 131000eur
[21:15:52] <anonimasu> err usd..
[21:16:02] <anonimasu> old model..
[21:17:25] <anonimasu> but they are for a 2004 one
[21:17:26] <anonimasu> :)
[21:18:29] -!- mhaberler [mhaberler!~mhaberler@62-47-139-255.adsl.highway.telekom.at] has joined #emc
[21:18:31] <anonimasu> L84Supper: http://cgi.ebay.com/DMU-50-DECKEL-MAHO-5X-Vertical-Machining-Center-2009-/350476927759?pt=LH_DefaultDomain_0&hash=item519a0d830f
[21:19:18] -!- nutbeam has quit [Remote host closed the connection]
[21:22:55] <L84Supper> CNC machines are so much lower cost than inkjet printers with a fraction of the parts
[21:23:04] -!- dgarr [dgarr!~dgarrett@adsl-75-61-65-77.dsl.pltn13.sbcglobal.net] has parted #emc
[21:23:14] -!- dgarr [dgarr!~dgarrett@adsl-75-61-65-77.dsl.pltn13.sbcglobal.net] has joined #emc
[21:27:04] <JT-Shop> one cap done, two more to go on this board... the piab makes it easy to desolder
[21:28:48] <Tom_itx> anonimasu, ditch that arduino crap and write it in c or asm
[21:31:11] -!- mhaberler has quit [Quit: mhaberler]
[21:31:16] <Tom_itx> piab?
[21:37:32] <JT-Shop> they make vacuum pumps, I usually have a sample or two laying about
[21:38:10] <JT-Shop> get the solder molten then hit the vacuum hose to it... sucks it out fast
[21:38:19] <Tom_itx> oh
[21:38:29] <JT-Shop> one more touch on each lead with the solder iron and out it pops
[21:38:30] <Tom_itx> i never cared for vacuum suckers
[21:38:38] <Tom_itx> good deal
[21:38:56] <JT-Shop> I have one that uses a spring but it doesn't suck well
[21:39:16] <Tom_itx> that's what i have and i never use it
[21:39:23] <JT-Shop> one board done, now to put it back before I forget where it goes
[21:40:36] <anonimasu> Tom_itx: it's in c now
[21:41:46] <Tom_itx> adc to pins out?
[21:41:51] <anonimasu> yeah
[21:41:58] <anonimasu> only thing I didnt fix is the adc read yet
[21:42:01] <Tom_itx> a t10 could do that
[21:42:09] <anonimasu> t10?
[21:42:13] <Tom_itx> yes
[21:42:21] <anonimasu> i dont know what that is
[21:42:36] <anonimasu> tiny 10?
[21:42:45] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/temp/tiny/tinyTPI1.jpg
[21:42:47] <Tom_itx> yes
[21:42:51] <Tom_itx> sot23
[21:43:04] <anonimasu> but the question is how fast it is
[21:43:09] <Tom_itx> 8mhz
[21:43:16] <anonimasu> I have a mega232 at 16mhz
[21:43:23] <anonimasu> err 322
[21:43:25] <anonimasu> at 16mhz
[21:43:37] <Tom_itx> it's asm only too
[21:43:43] <Tom_itx> too small for c
[21:44:00] <Tom_itx> 16bit timer
[21:44:00] <anonimasu> I paid 3 eur + crystal and caps
[21:44:01] <anonimasu> :)
[21:44:05] <Tom_itx> adc, pwm
[21:44:12] <Tom_itx> pretty robust for such a tiny chip
[21:44:21] <anonimasu> yeah I have some like it here in dil capsule
[21:44:36] <anonimasu> but my big issue is this can the avr perform it fast enough
[21:44:40] <anonimasu> I have no means of measuring that
[21:45:16] <Tom_itx> the bottleneck would be the adc
[21:45:28] <Tom_itx> figure how fast it can read the samples
[21:48:49] -!- FinboySlick [FinboySlick!~shark@] has joined #emc
[21:49:23] <anonimasu> looks like max adc prescaler is 16
[21:49:24] <L84Supper> anyone know a good source for vacuum tables?
[21:49:39] <L84Supper> heated would be a ++
[21:50:09] <Tom_itx> heated??
[21:50:11] -!- scanf [scanf!~x32@unaffiliated/scanf] has joined #emc
[21:50:27] <anonimasu> that gives a frequency of the adc at 1mhz
[21:50:32] -!- tlab has quit [Read error: Connection reset by peer]
[21:50:59] <Tom_itx> it's 1024 per rev?
[21:51:02] <L84Supper> vacuum table meets electric skillet.... well to 60 deg C or so only
[21:51:04] <anonimasu> yeah
[21:51:14] <Tom_itx> i doubt it
[21:51:27] <anonimasu> http://www.sick-automation.ru/images/File/pdf/Div01/srs_srm_50_60.pdf
[21:51:53] <anonimasu> 1,024 sin/cosine periods per revolution
[21:52:08] <Tom_itx> what's your max rev?
[21:52:12] <Tom_itx> rpm
[21:52:30] <anonimasu> 3000
[21:52:48] -!- tlab [tlab!~tlab@c-98-223-20-100.hsd1.in.comcast.net] has joined #emc
[21:53:01] <Tom_itx> 3000 x 1024 = 3072000 samples per min
[21:53:07] <anonimasu> that should give (3000*1024)/60 counts/sec
[21:53:15] -!- Fox_Muldr has quit [Ping timeout: 276 seconds]
[21:53:19] <Tom_itx> 51200
[21:53:24] <anonimasu> 51khz
[21:53:44] <Tom_itx> it might work
[21:53:54] <Tom_itx> figure the overhead to toggle the pins
[21:53:55] <anonimasu> it's a temporary hack to get stuff working for now
[21:54:55] -!- Fox_Muldr [Fox_Muldr!quakeman@frnk-4d01c039.pool.mediaWays.net] has joined #emc
[21:55:19] <Tom_itx> how many samples do you need to read to get the threshhold point of the signal?
[21:55:22] -!- JT-Shop has quit [Read error: Connection reset by peer]
[21:55:55] -!- JT-Shop [JT-Shop!~chatzilla@216-41-156-49.semo.net] has joined #emc
[21:56:07] <anonimasu> hm, just when it crosses a specific value
[21:56:09] <anonimasu> 3.2v
[21:56:14] <Tom_itx> maybe a comparitor would be better
[21:56:17] <anonimasu> one
[21:56:31] <Tom_itx> but you won't know that unless you sample it alot
[21:56:46] <anonimasu> i need two samples
[21:56:58] <anonimasu> so it can toggle between low/high in one signal period
[21:57:14] <anonimasu> to get a quadrature pulse out _-
[21:57:16] <Tom_itx> right but you won't know where the phase is unless you sample alot
[21:57:54] <anonimasu> im just quantizing the signal on the input
[21:58:00] <anonimasu> over a certain level output is high
[21:58:03] <anonimasu> under input is low
[21:58:06] <Tom_itx> a comparator would trip at a setpoint if you did it that way
[21:58:08] <anonimasu> err output
[21:58:30] <Tom_itx> say 70-80% of the peak
[21:58:45] <anonimasu> that's what my value is
[21:59:16] <anonimasu> I figured it'd give me some extra space for the avr to not do anything except for jmp
[21:59:23] <Tom_itx> but you need a low value too
[21:59:33] <Tom_itx> then you need to compare the results of both channels
[21:59:41] <Tom_itx> to know what direction it's moving
[21:59:44] <anonimasu> no, i dont
[21:59:51] <anonimasu> i have the mesa board doing the counting
[21:59:54] <Tom_itx> ok
[22:00:08] <anonimasu> im just using the avr as a schmittrigger
[22:00:10] <Tom_itx> does that chip have a comparitor on the adc pins?
[22:00:19] <Tom_itx> i'd use the comparitor
[22:00:42] <Tom_itx> it'll trip and interrupt when it matches
[22:00:44] <anonimasu> I think so, or i can use the atmega88's i have around
[22:00:53] <anonimasu> is that faster?
[22:01:03] <Tom_itx> then all you gotta do is spit out a pin change to the output
[22:01:07] <anonimasu> then a free running loop
[22:01:12] <Tom_itx> try it
[22:01:49] <Tom_itx> you don't really need any code in the main loop
[22:01:54] -!- sumpfralle [sumpfralle!~lars@] has joined #emc
[22:02:00] <Tom_itx> maybe a tiny delay or something
[22:02:04] <anonimasu> it has a comparator
[22:02:14] <Tom_itx> it should have 2
[22:02:30] <Tom_itx> use them for each channel and filp the pins inside the interrupt
[22:02:35] <Tom_itx> flip*
[22:03:01] <Tom_itx> problem is, you won't detect the low side with a comparitor
[22:03:06] <Tom_itx> i don't think
[22:03:19] -!- sumpfralle1 has quit [Quit: Leaving.]
[22:03:22] <Tom_itx> hmm
[22:03:33] <anonimasu> cant I toggle the comparator value inside the interupt
[22:03:37] <anonimasu> ie, toggle on low instead
[22:03:41] <anonimasu> that's a single bitwise op
[22:03:44] -!- Birdman3131 has quit [Read error: Connection reset by peer]
[22:03:48] <Tom_itx> will that work?
[22:04:27] <Tom_itx> it may fall before you can read it again
[22:04:29] <Tom_itx> i'm not sure
[22:04:46] <Tom_itx> i haven't tried reading comparitor values on the falling edge
[22:05:02] -!- Birdman3131 [Birdman3131!~quassel@ip70-178-238-227.ks.ks.cox.net] has joined #emc
[22:05:16] <anonimasu> you can do toggle
[22:05:18] <anonimasu> so it flips on both
[22:07:32] <anonimasu> but there is only one comparator.
[22:07:40] <Tom_itx> well crap
[22:07:46] <Tom_itx> find one that has 2
[22:08:13] <Tom_itx> or try to get the adc working
[22:13:08] <anonimasu> it is now
[22:13:49] <anonimasu> perferably I'd like to sample x2 so i can have double the resolution
[22:14:50] <Tom_itx> sample as much as you like but only flip the pin inside the interrupt when the values are >= a set value
[22:15:01] <Tom_itx> or <= on the falling side
[22:15:09] <anonimasu> that's how I do it now
[22:15:10] <anonimasu> :)
[22:15:52] -!- Birdman3131 has quit [Read error: Connection reset by peer]
[22:16:22] <anonimasu> looks like most avr's only have one comparator
[22:16:35] <Tom_itx> i've never looked really
[22:16:47] <Tom_itx> there's a pdf on my site that will show
[22:17:11] <anonimasu> looking at the atmega128
[22:17:57] <anonimasu> it can be multiplexed somehow
[22:17:59] <Tom_itx> http://tom-itx.dyndns.org:81/~webpage/pdf/atmel/PARAMETRIC_TABLE.pdf
[22:18:02] -!- Birdman3131 [Birdman3131!~quassel@ip70-178-238-227.ks.ks.cox.net] has joined #emc
[22:19:42] <anonimasu> 76.9ksps
[22:19:44] <anonimasu> maximum
[22:20:55] <anonimasu> crapduino.
[22:20:56] <anonimasu> :)
[22:21:10] <anonimasu> 13Ás conversion time
[22:21:28] <L84Supper> ARM M3's are more bang for the buck now
[22:22:09] <Jymmm> Well, arduino isn't intended for speed, it's for simplisity. If you cont like it, dont use and use avr's instead.
[22:22:21] <Tom_itx> anonimasu, if it's just to test, what does it matter?
[22:22:49] <Tom_itx> you'll have a solution once you get your cable etc
[22:23:31] <anonimasu> Jymmm: it's the same chips im just swearing at the arduino since it's where the chip sits
[22:23:57] <Tom_itx> the atmega is just an innocent bystander
[22:24:18] * Tom_itx randomly smacks Jymmm
[22:24:20] <anonimasu> the adc is slow that's all
[22:24:31] <Jymmm> anonimasu: Well, you're trying to pull a 40K lb trailer with a yugo
[22:24:47] <Tom_itx> it works fine downhill
[22:24:55] <Jymmm> in a hurricane
[22:25:18] <anonimasu> how fast can a cortex m3 do it?
[22:26:02] <L84Supper> what res (bits) ?
[22:26:08] * Jymmm replaces Tom_itx's AVR's with PIC's!
[22:26:23] <anonimasu> L84Supper: more the better
[22:26:24] <Tom_itx> i have a few already
[22:26:42] <Tom_itx> L84Supper probably 8 or 10
[22:27:11] <Tom_itx> much more than that on an integrated adc would be noise anyway
[22:27:15] <anonimasu> L84Supper: i'd like to sample alot faster so I could manage to x4 the input
[22:27:24] <anonimasu> 10 bits for the adc
[22:27:49] <L84Supper> 200KHz 12 bit
[22:28:08] <anonimasu> that would do very fine
[22:28:24] <L84Supper> LPC1768 or any LPC17xx with ADC
[22:28:39] <L84Supper> http://ics.nxp.com/products/lpc1000/all/~LPC1768/
[22:28:48] <L84Supper> 120MHz cpu core
[22:29:06] <anonimasu> I can buy dev boards with them
[22:29:07] <anonimasu> :)
[22:29:18] <anonimasu> http://www.lawicel-shop.se/shop/assets/prod_images/pDEV-09931l.jpg
[22:29:28] <Jymmm> Tom_itx: Just curious... Can you take a arduino "sketch" and convert it to be used on an avr directly?
[22:29:49] <Jymmm> Tom_itx: w/o using the loader
[22:29:53] <L84Supper> anonimasu: yes, low cost
[22:30:35] <anonimasu> Jymmm: yeah pretty much the arduino is just a wrapper lib for the native c functions
[22:30:59] <anonimasu> http://www.lawicel-shop.se/shop/assets/prod_images/pLM-C-M3l.jpg
[22:31:00] <anonimasu> mine.
[22:31:02] <Jymmm> anonimasu: k
[22:31:10] <Tom_itx> Jymmm, i don't see why not
[22:31:16] <anonimasu> but we'll see how much the supplier will fail with the other boards
[22:31:28] <Tom_itx> Jymmm, what avr is it?
[22:31:34] <anonimasu> if they stall me for another month i'll go that route
[22:31:49] <anonimasu> 100khz is the max rate of the sine/cos period
[22:31:53] <Tom_itx> just save the bootloader and reset the fuses so they're not looking for it
[22:32:03] <anonimasu> (that's at the max encoder rpm )
[22:32:17] <anonimasu> so at 200khz I'd have time to sample and trigger x2 at max rpm
[22:32:32] <anonimasu> and at 72mhz, it'd be doable.
[22:33:28] <anonimasu> night.. it's too late here :)
[22:35:07] <Tom_itx> gnite anonimasu
[22:38:19] <PCW> DSPICs can manage 2 Ms/sec at 10 bits
[22:39:54] <L84Supper> yeah, there are faster devices, I wasn't sure what he needs for the CPU, ADC or software support
[22:40:13] <PCW> Oversampling is good for an interpolator
[22:40:18] <L84Supper> I just never got on the Arudino bandwagon
[22:41:11] <anonimasu> indeed, but i cant find a dspic devel board without a million pheripherals that I can just plug in and use
[22:41:38] <Jymmm> anonimasu: those are called computers ;)
[22:42:19] <anonimasu> I need a schmittrigger that does 2x toggling over a sin/cos period
[22:42:33] <PCW> We could send you a 7I73 pendant interface proto, its a DSPIC on a board with connectors
[22:44:23] <anonimasu> that would be a very nice option compared to mucking around trying to do it with the wrong stuff
[22:45:20] <Jymmm> PCW: where is that on teh wbesite?
[22:45:36] <anonimasu> probably not yet i guess
[22:45:52] <L84Supper> also STM32 has 12bit ADC at 1uS
[22:49:01] <L84Supper> http://www.st.com/internet/mcu/product/250484.jsp or similar
[22:49:16] <L84Supper> http://www.sparkfun.com/products/8559 typical dev board
[22:49:49] -!- WalterN [WalterN!~walter@tiwake.com] has joined #emc
[22:51:09] <andypugh> Sorry Pete, I was in the Workshop
[22:54:39] <JT-Shop> what temperature should the soldering iron be set at to desolder the caps?
[22:55:58] <JT-Shop> I put it at 850F but it seemed a bit on the hot side...
[22:57:41] <PCW> Thanks for the file Andy, I'll give it a go
[23:00:14] <PCW> BTW got parameter discovery working in SSLBP V25
[23:00:16] <PCW> So newer cards (and updated firmware on 8I20,7I64)
[23:00:17] <PCW> can report name,type, size of all SSLBP transferred parameters
[23:00:19] <PCW> so the driver can pack/unpack and present to HAL without card specific knowledge
[23:00:43] <JT-Shop> neat!
[23:01:14] <PCW> 750 is what I normally use (850 is pretty hot for PCBs)
[23:05:21] -!- Valen [Valen!~Valen@2001:44b8:3199:9100:21e:8cff:fe63:55df] has joined #emc
[23:06:00] <JT-Shop> thanks Peter
[23:10:19] -!- robh__ has quit [Ping timeout: 252 seconds]
[23:17:00] -!- skunkworks [skunkworks!~chatzilla@184-158-63-77.dyn.centurytel.net] has joined #emc
[23:18:08] <Tom_itx> JT-Shop, better to get on and get off, 800ish
[23:18:16] <skunkworks> stupid question. I have a spindle encoder that is 1024 counts. I am using it for rigid tapping - so that scale is set to 4096
[23:18:18] -!- servos4ever has quit [Quit: ChatZilla 0.9.85 [SeaMonkey 2.0.11/20101206162726]]
[23:18:20] <Tom_itx> too hot and you'll pull up the copper
[23:19:21] <skunkworks> why is the velocity reading 1/1000 rpm? when the spindle is running at 300 rpm - the velocity is .300
[23:19:30] <skunkworks> should it not be in rps?
[23:19:47] <skunkworks> I am not seeing it ;)
[23:20:27] -!- mshaver [mshaver!~mshaver@c-68-50-233-206.hsd1.md.comcast.net] has joined #emc
[23:21:32] -!- micges has quit [Quit: Ex-Chat]
[23:22:08] <skunkworks> or 5rps?
[23:22:57] <danimal_garage> sweet, someone's on the way to look at my lathe
[23:23:07] <danimal_garage> hope they buy it
[23:23:13] <danimal_garage> thats beer monet
[23:23:13] <Tom_itx> http://www.reddit.com/r/funny/comments/j26j0/for_engineers/
[23:23:16] <danimal_garage> money*
[23:23:20] <Tom_itx> ot but funny
[23:25:22] <JT-Shop> Tom_itx: yea, I was on the lead mostly but the tip would oxidize or something real fast
[23:26:24] <Tom_itx> cheap tip?
[23:26:37] <JT-Shop> no, a good one
[23:26:38] <Tom_itx> i got an iron from RS that lasted maybe 30 min
[23:27:11] <Tom_itx> i have a weller i've had for years that if i hadn't lost the tip would still be using it
[23:27:13] <JT-Shop> Weller WEDS51
[23:27:30] <Tom_itx> mine has the magnetic tips
[23:27:36] -!- mrsunshine has quit [Ping timeout: 258 seconds]
[23:27:42] <Tom_itx> WTCPT
[23:27:42] <JT-Shop> my first one with temp control so I was doing the wild ass guess thing
[23:28:09] <Tom_itx> ~7-800 depending on how much heating you need done
[23:28:20] <andypugh> Tom_itx: Tips are readily available.
[23:28:30] <Tom_itx> i use 700 on most of my smt stuff
[23:28:40] <Tom_itx> andypugh yes i have a few spares
[23:28:52] <Tom_itx> but haven't needed them yet
[23:29:05] <andypugh> You said you "would still be using it"..
[23:29:09] <Tom_itx> the tip
[23:29:11] <JT-Shop> I wish I would stop bleeding :/
[23:29:13] <Tom_itx> had i not lost it
[23:29:19] <Tom_itx> JT-Shop, cauterize it
[23:29:20] <Tom_itx> :D
[23:29:22] <andypugh> Ah.
[23:29:44] <JT-Shop> it will quit by the time I finish this glass of red
[23:30:05] <Tom_itx> i thought i needed a heating element for it once but as it turned out for some reason one of the twist connectors inside came loose and quit heating
[23:30:15] <Tom_itx> so now i have a spare heating element too
[23:30:20] <JT-Shop> Patsy Cline should make it stop bleeding right
[23:33:09] <JT-Shop> speaking of that my leather riding gloves are getting stiff on top from exposure to the sun... any thoughts on what to put on them to get them limber again?
[23:33:34] <Tom_itx> umm, what do they use on baseball gloves?
[23:33:40] <JT-Shop> dunno
[23:33:41] <Tom_itx> is it linseed or some other oil
[23:33:49] <Tom_itx> i forget
[23:33:55] <JT-Shop> saddle soap?
[23:34:02] <Tom_itx> that would work
[23:34:22] <JT-Shop> shit I feel like Julia Childs now I bleeding from two spots
[23:34:59] -!- mrsunshine [mrsunshine!~mrsunshin@c-0a3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #emc
[23:35:37] <JT-Shop> I've never used saddle soap...
[23:37:33] -!- kljsdfhklj_ has quit [Ping timeout: 250 seconds]
[23:37:58] <skunkworks> hmm - It looks like it is correct compared to the sample configs.
[23:38:33] <skunkworks> #spindle encoder for rigid tapping
[23:38:35] <skunkworks> setp hm2_5i20.1.encoder.02.scale 4096
[23:38:37] <skunkworks> net spindle-pos hm2_5i20.1.encoder.02.position motion.spindle-revs
[23:38:38] <skunkworks> net spindle-index-en motion.spindle-index-enable hm2_5i20.1.encoder.02.index-enable
[23:38:40] <skunkworks> #spindle velocity for fpr and such
[23:38:41] <skunkworks> net spindlevel hm2_5i20.1.encoder.02.velocity motion.spindle-speed-in
[23:39:23] <skunkworks> but the velocity is .3 for 300rpm
[23:40:02] <andypugh> JT Shop. My bike gloves are probably 8 years old, used every day, and showing no signs of sun damage. I wonder why?
[23:40:36] <JT-Shop> mine might be cheap, they are 4 years old and getting white on top
[23:41:01] <JT-Shop> you do have a higher angle of the sun...
[23:42:11] <Valen> if they are leather use some leather stuff on them
[23:42:24] <Valen> i did it with my missus ones and it made a huge difference
[23:42:36] <Valen> started off with lather oil
[23:42:47] <Valen> then another one thats like a paste
[23:42:48] <JT-Shop> I wonder if sweat does it
[23:43:01] <L84Supper> lanolin
[23:44:00] <JT-Shop> yea, these are Tourmaster summer elite 2 gloves paid about $35 for them
[23:44:27] <JT-Shop> I'll look at the leather shop for something
[23:44:48] <JT-Shop> lots of saddles around here
[23:45:29] -!- kljsdfhklj [kljsdfhklj!~ln@] has joined #emc
[23:45:42] * JT-Shop wanders off to fire up the oven for dinner...
[23:47:21] -!- skunkworks_ has quit [Ping timeout: 260 seconds]
[23:50:54] <andypugh> What is this "sun"of which you speak?
[23:52:31] -!- mrsunshine has quit [Ping timeout: 258 seconds]
[23:53:11] -!- mrsunshine [mrsunshine!~mrsunshin@c-0a3de253.4551536--62697410.cust.bredbandsbolaget.se] has joined #emc
[23:53:45] -!- kljsdfhklj has quit [Ping timeout: 255 seconds]
[23:53:48] <IG-garage> hey guys what's adaptive roughing?
[23:54:11] -!- mozmck has quit [Remote host closed the connection]
[23:54:25] <IG-garage> i mean, whatsa fun with it? Wy is it so special?
[23:55:10] -!- mozmck [mozmck!~moses@client-] has joined #emc
[23:57:17] <IG-garage> anonimasu: give me a hint when it will be a convenient time for you to chatting (wanted to ask how much time do you spend to create g-code for the order or task)