#emc | Logs for 2010-09-11

[00:00:50] <Connor1> interesting..... 1st parallel port has issues with X, 2nd doesn't.
[00:01:36] <MattyMatt> there were some nice double stack 5 phase recently, around 30 quid . I coveted
[00:02:15] <MattyMatt> is LPT1 on a header?
[00:02:23] <Connor1> No.
[00:02:31] <Connor1> LPT2 Is.
[00:02:39] <MattyMatt> strange
[00:03:04] <MattyMatt> maybe LPT1 is not EPP in the bios?
[00:03:22] <Connor1> LPT1 is working fine now.. LPT2 isn't.. both are on the same PCI Card.
[00:03:35] <Connor1> main LPT on mobo isn't recognized.
[00:04:17] <MattyMatt> can you config pci LPT1 as LPT3 and renable the mobo one?
[00:04:19] <theorb> theorb is now known as theorbtwo
[00:04:41] <Connor1> mobo is enabled.. it's just not seen by linux.
[00:04:45] <MattyMatt> old IO cards had a jumper for that
[00:05:08] <Endeavour> Hello
[00:05:54] <MattyMatt> ah if linux sees it as LPT1 then it may be clashing with the mobo one. disable that in bios
[00:08:30] <MattyMatt> " 1st parallel port has issues with X" I presume he meant the X axis, not X server?
[00:08:49] <Connor1> Yes, X axis.. and it was the 2nd LPT
[00:09:57] <MattyMatt> ah if it's the header one that's missing a pin, it'll be the header, either end is vulnerable
[00:10:24] <MattyMatt> good solid 1980s tech :) IDC and Dsub
[00:11:27] <andypugh> Could be a duff IDC connector. My CNC machine has one 7i43 channel missing as they simply forgot to put a contact in that hole of the IDC (Not Mesa, the actual connector manufacturer)
[00:12:03] <MattyMatt> I usually ended up soldering all my IDCs back then, but they seem to be generally more reliable these days
[00:13:10] <andypugh> Does soldering make them more reliable? I would have thought there was a danger of melted insulation and shorts.
[00:13:31] <andypugh> Or whatever sort of trousers you were wearing at the time.
[00:14:44] <MattyMatt> you strip all wires to same length, then solder all one side, insert cardboard insulator, solder other side, encapsulate in Araldite
[00:15:44] <MattyMatt> I suppose you'd use hot glue these days for flexibility
[00:17:24] <MattyMatt> nobody uses ribbon cables externally these days, that was the main cause of unreliability probably
[00:18:10] <andypugh> I managed to crimp one of my Mesa IDCs half a pitch out, every pin shorted to earth. It didn't work, but on the impressive side, nor did it break.
[00:18:36] <MattyMatt> :)
[00:20:39] <andypugh> Right, need to sleep. Tomorrow is a day of changing fork oil and brake pads prior to hooning round Cadwell Park next week.
[00:21:21] <MattyMatt> I've topped up my Sturmey Archer. I'm good to go
[00:22:12] <andypugh> I used 4 Sturmey Archers as hub gearboxes in SMIDSY, they survived 6 series of Robotwars without a single failure.
[00:22:44] <MattyMatt> that's a lie. I broke a workmate trying to bend these mountain bike calipers in a vice to fit a road wheel
[00:23:08] <andypugh> (There were some mods, the drive went in through the axle, the hub was welded to the chassis, and the wheels were mounted on the sprocket flange)
[00:23:35] <MattyMatt> sweet. you had the gears changable?
[00:24:33] <andypugh> No. We planned to then found that doing so was incompatible with being able to go backwards..
[00:25:42] <andypugh> The gearing relies on engaging and disengaging various ratchets (which is why they "tick" in second, the third gear ratchet is over-running)
[00:27:05] <andypugh> And next time, get the calipers hot before you bend them, then you can burn the workmate too.
[00:27:26] <MattyMatt> :) nothing hot enough to get it red here
[00:27:32] <andypugh> (If aluminium, apply the force then heat till they give)
[00:27:45] <MattyMatt> steel
[00:28:30] <MattyMatt> yeah I need a forge, or at least a gas torch
[00:29:21] <MattyMatt> hmm, I have a tiny butane one
[00:29:52] <andypugh> Get a MAPP gas one. Seriously hot.
[00:30:09] <moopy> dont bother with gas
[00:30:17] <andypugh> I have been using mine to harden and temper Silver steel (O1 for the US folks) parts
[00:30:17] <MattyMatt> that's just propane isn't it?
[00:30:23] <moopy> get in induction heater
[00:30:42] <andypugh> I bought the parts for an induction heater.
[00:30:44] <MattyMatt> I'm saving the dead psu to make induction heater :)
[00:30:50] <MattyMatt> and the dead fidges
[00:30:51] <andypugh> There are plans on 'tinternet.
[00:30:59] <MattyMatt> yep
[00:31:23] <andypugh> But I reckon that is old-hat, an Arduino can do all the sequencing and tuning in software.
[00:31:24] <MattyMatt> his kits aren't far short of a small commercial one tho
[00:32:29] <andypugh> One of the Screwfix MAPP torches is 3000W. (And is one of the cheaper ones too)
[00:33:17] <andypugh> But an Arduino as the frequency generator could auto-tune and handle the interlocks too.
[00:33:18] <MattyMatt> yeah that's all I need. I don't need oxy/acy bottles or anything just yet
[00:33:58] <MattyMatt> I could just buy a new bike from asda for 50 quid :)
[00:34:49] <MattyMatt> or make brackets to hold the shoes
[00:34:53] <andypugh> My dad has Oxy-Acetlylene, he is considering getting rid. The bottle hire is expensive, and it doesn't do anything that a combination of mIg/arc/blowtorch/plama cutter can't do cheaper.
[00:35:42] <andypugh> A £50 Asda bike is a great plan if you are wanting excercise :-)
[00:36:01] <MattyMatt> mainly transport. I need panniers etc
[00:36:06] <moopy> i have some oxy bottles that dont have hire charge
[00:36:43] <andypugh> You would not be able to get them filled in the UK.
[00:37:11] <MattyMatt> all your oxy is belong to BOC
[00:37:18] <moopy> the company went bust and never asked for them back
[00:37:30] <andypugh> Or Air Liquide (beastly frogs)
[00:37:44] <moopy> but the oxy is never used so no need to get them filled
[00:38:10] <MattyMatt> I'd be huffing it every morning
[00:38:14] <andypugh> Helium, that's a better investment
[00:39:00] <MattyMatt> wouldn't argon be better for MIGing in?
[00:39:14] <MattyMatt> closer in weight to air
[00:39:24] <moopy> i imagine industrial oxy may have other nasty gas in it, so bad idea to try breathing it
[00:39:29] <andypugh> Prices are likely to rise hugely soon. For some bizarre reason the US gov owns half the world stocks, and for another odd reason is legally obliged to get rid of it all. When they have, the price will rocket.
[00:40:07] <moopy> its a double dip just round the corner
[00:40:26] <MattyMatt> argon is the cheapest and most plentiful too
[00:40:37] <moopy> best get rid of your savings before the intrest rates go negative
[00:41:07] <andypugh> There is more Argon than CO2 in the atmosphere, so I suspect it will be the last thing we ever run out of.
[00:42:54] <andypugh> BOC and AL have a great business model. They pull argon out the air and sell it to us. We put it back, and they sell it to us again.
[00:43:23] <MattyMatt> it's pretty good. is it illegal to freeze your own air?
[00:43:39] <andypugh> I hope not.
[00:44:06] <MattyMatt> I wonder how small that kit could be made
[00:44:07] <andypugh> I have done it.
[00:44:29] <andypugh> (only using already-cryogenic gas though)
[00:46:01] <MattyMatt> I'd expect you'd have to prime the inner cooling loop with LN
[00:46:33] <MattyMatt> but otherwise it should be as simple as firidge in a fridge in a fridge
[00:46:41] <andypugh> http://en.wikipedia.org/wiki/Siemens_cycle
[00:48:05] <andypugh> (Or the Linde Cycle)
[00:48:22] <andypugh> It's conceptually pretty simple
[00:49:09] <MattyMatt> yep
[00:49:38] <moop> are you into laser physics?
[00:49:55] <andypugh> http://en.wikipedia.org/wiki/Pulse_tube_refrigerator looks even better
[00:50:25] <andypugh> Who? I am into all physics :-)
[00:50:27] <MattyMatt> laser physics is beyond me. I can do plumbing
[00:51:09] <andypugh> I would cheerfully do my physics degree all over again, paying a bit more attention.
[00:51:40] <moop> yeah, education is wasted on the young
[00:54:27] <andypugh> Right, I mean it this time. Goodnight.
[00:54:29] <MattyMatt> I'm 'attending' madras uni on YouTube
[00:56:07] <MattyMatt> http://www.youtube.com/watch?v=Q6Gw08pwhws&feature=channel
[00:56:41] <MattyMatt> Indian Institute of Tech, Madras
[01:13:39] <Connor1> okay, bad header cable for the lpt2
[01:13:55] <Connor1> pin 1 (wire) must be broke.. I'll get a new header cable if I need lpt2.
[01:14:40] <Connor1> so, question, microstepping... My screws for X and Y are 1 turn per 10mm, and Z is 1 per 5mm.. I want to do CNC and Printing..
[01:14:51] <Connor1> recommended setting for microstepping ?
[01:16:48] <MattyMatt> hmm, I have all mine at 1/2 stepping including my 10mm pitch ballscrew, but I haven't really experimented
[01:18:46] <MattyMatt> for fine milling I guess finer steps would be better, but mine's close enough for woodwork so far
[01:19:57] <Connor1> Hmm... I try to go faster, and it stops spinning, it just chatters
[01:20:31] <MattyMatt> I guess you should try 1/16 and see if the PC can keep up full speed like that
[01:21:06] <MattyMatt> chattering when you try and go too fast is lack of torque
[01:21:19] <MattyMatt> what voltage is your PSU?
[01:21:36] <Connor1> 24v
[01:22:25] <Connor1> 80mm/s is fast as I can go without it chattering or stalling.
[01:22:30] <MattyMatt> better than mine, I can do 700mm/min on my ballscrew on 12V, with 2A nema 23s on 50% current
[01:22:32] <_AR_> connor1, stepper motors cannot go very fast
[01:23:40] <Connor1> OKay, what is the deal with the Step Time, Step Space, Direction Hold and Direction Setup?
[01:23:47] <_AR_> just guess
[01:23:52] <_AR_> depends on your driver
[01:24:17] <MattyMatt> I left those as default on mine, seems to work
[01:24:26] <Connor1> and my driver has a setting for current, 100, 75, 50 and 25 %
[01:24:34] <Connor1> Looks like the board came default to 75%
[01:24:34] <_AR_> what driver are you using, connor
[01:24:40] <MattyMatt> that depends on your motors
[01:24:48] <_AR_> tb6560?
[01:24:52] <_AR_> chinese?
[01:24:53] <MattyMatt> yeah
[01:24:54] <Connor1> TB6560AHQ
[01:24:55] <_AR_> yeah
[01:24:59] <_AR_> i just got one too
[01:25:15] <_AR_> i already burnt up a chip on it lol
[01:25:26] <_AR_> but that's totally my fault i think
[01:25:26] <Connor1> how did you do that?
[01:25:42] <_AR_> well, my 3 good motors worked at first
[01:25:48] <_AR_> but one was getting way too hot and chattery
[01:25:54] <_AR_> so i figured it was a goner
[01:26:02] <_AR_> so i tried to hook up another motor i had
[01:26:23] <_AR_> an 8 wire 1.75A
[01:26:28] <_AR_> which should be fine
[01:26:34] <_AR_> considering my other ones are 2A
[01:26:35] <_AR_> but
[01:26:44] <_AR_> i dont know if my soldering was bad or what
[01:26:58] <_AR_> because i had to solder wires onto the pins on the motor, it doesnt have wires
[01:27:10] <_AR_> i think i had a short and there was just infinite current running through
[01:27:45] <_AR_> also, i wanted to turn the current setting down to 25% at first to be safe, but left it at 75%
[01:27:50] <_AR_> probably wouldnt have mattered
[01:27:57] <_AR_> the chip just popped and smoked
[01:28:04] <_AR_> ordered a replacement
[01:28:19] <_AR_> the chips look easy to replace on this board at least
[01:28:50] <MattyMatt> yeah no through holes to wreck
[01:29:11] <_AR_> lol
[01:29:18] <_AR_> id be fine with through holes too
[01:29:39] <_AR_> just as long as it isnt like a 50 pin surface mounted chip
[01:30:14] <_AR_> suck you gotta take the heatsink off all 3 chips though just to fix 1
[01:30:18] <_AR_> sucks*
[01:30:33] <Connor1> table travel. Is that in mm ?
[01:30:44] <MattyMatt> yeah
[01:31:06] <Connor1> back later.
[01:31:12] <Connor1> oh. How do you set direction of motor?
[01:31:20] <Connor1> reverse the coils?
[01:31:27] <_AR_> yes or invert signal
[01:31:33] <Connor1> ok.
[01:31:36] <Connor1> thanks.
[01:32:02] <MattyMatt> iirc stepconf asks you whether you want to spec in in or mm at the start
[01:32:23] <_AR_> yeah
[01:32:48] <_AR_> what type of computers are you guys running emc on?
[01:33:03] <MattyMatt> 1.4 athlon I found at the tip
[01:33:45] <MattyMatt> with 512mb iirc
[01:33:50] <_AR_> lol
[01:34:00] <_AR_> im using an HP Vectra VL
[01:34:00] <_AR_> rofl
[01:34:02] <_AR_> http://genisu.com/IMG/vectrap2.jpg
[01:34:05] <_AR_> looks like that
[01:34:21] <_AR_> i think it has a 4 gig harddrive
[01:34:24] <_AR_> pentium 2
[01:34:29] <_AR_> 256 ram
[01:34:35] <_AR_> MAYBE 512
[01:34:35] <MattyMatt> mine's HP as it happens :)
[01:34:37] <_AR_> i forget
[01:34:49] <_AR_> lol, this thing is slow but it works
[01:34:55] <_AR_> sometimes better than other times
[01:34:56] <_AR_> it's weird
[01:35:15] <_AR_> it took me about 12 tries until i finally got the emc2 livecd to boot to the point where i could install it
[01:35:46] <_AR_> im actually using the XUbuntu version
[01:36:03] <_AR_> actually
[01:36:05] <MattyMatt> I had that problem with an old 500Mhz machine. it was bad ram
[01:36:09] <_AR_> i dont even remember
[01:36:24] <_AR_> i think i just installed xubuntu and then installed emc2
[01:36:27] <_AR_> same thing
[01:37:07] <_AR_> ive installed so many different OSes at different times on that machine
[01:37:10] <_AR_> lol
[01:37:14] <_AR_> it's probably just dead
[01:37:19] <_AR_> from exhaustion
[01:37:34] <_AR_> i used to actually have it running good enough to host a small website
[01:39:11] <MattyMatt> bleh, I've done that on Win98se :) not a very long lived website
[01:39:55] <MattyMatt> and I never noticed the bad chunk of ram all the years I was running windows :D
[01:40:15] <_AR_> lol
[01:40:16] <MattyMatt> maybe I did, every 36 hours when I needed to reboot
[01:41:04] <_AR_> when i have an older computer, i prefer linux on it simply because it makes doing the simple things like internet and what not a lot faster
[01:41:17] <_AR_> but on my good computers for my main stuff i use windows
[01:41:32] <MattyMatt> I'm the other way around
[01:41:45] <MattyMatt> only win box now is XP on lappy
[01:41:46] <_AR_> lol
[01:42:24] <MattyMatt> mainly because it's a thinkpad, so has licenced XP pro
[01:42:36] <_AR_> i really dont do anything that would require one OS over another
[01:42:46] <_AR_> but i got windows 7 for free from my university
[01:42:55] <_AR_> so i use that on my laptop for my regular stuff
[01:43:22] <MattyMatt> I deliberately got an OLD thinkpad, so I got an XP licence not a Vista one :)
[01:43:24] <_AR_> i mainly keep windows as my regular os for ease of compatibility
[01:43:31] <_AR_> lol
[01:43:36] <_AR_> win7 is good
[01:43:52] <_AR_> vista was OK for me
[01:44:04] <MattyMatt> I wouldn't know, nor probably ever will. XP is the sweet spot for NT
[01:44:10] <_AR_> lol
[01:44:34] <_AR_> it's just a matter of getting used to whatever you are using
[01:44:42] <MattyMatt> like 98SE was the sweet spot for 9x, they jumped the shark with ME
[01:44:59] <_AR_> xD
[01:46:23] <MattyMatt> actually I've heard there is a 98 3rd and 4th edition, with even better USB support than SE, but I haven't seen them
[02:16:13] <Jymmm> Win98 A, B, C, and D
[02:16:39] <Jymmm> iirc C was the USB addon, and D had USB included
[02:18:16] <L84Supper2> is anybody going to IMTS next week? http://www.imts.com/
[02:32:25] <_AR_> L84Supper2, my dad is with his company
[02:32:26] <_AR_> lol
[02:32:41] <_AR_> you going?
[03:26:20] <L84Supper2> http://chicago.craigslist.org/chc/tls/1938090725.html 2 head version in case you break one
[03:46:36] <Connor1> question, why can't I control the machine in manual mode? If the power button is depressed.. it turns everything off. if it's not.. the steppers are active, but, I can't use the controls..
[03:46:53] <Connor1> It's almost as if the power button is inverted.
[03:47:14] <atmega> perhaps you are looking at it wrong.
[03:47:32] <cradek> sounds like your amp-enable is inverted
[03:47:43] <Connor1> works correctly in stepconf
[03:50:28] <Connor1> okay, that's waked.. it's inverted in EMC, but not in stepconf!
[04:27:43] <Connor1> Dang, stepper motors get HOT!!!
[05:21:49] <Alex7bin> any uC news for EMC2 ?
[06:18:05] <JuniperJaxx> does anyone know of a mini manual mill that does not fail? :P
[06:30:10] <archivist> "does not fail" rather broad spec, define closer
[06:31:57] <WalterN> archivist: accuracy mostly
[06:33:44] <WalterN> archivist: table top manual mill that is decently built and accurate
[06:33:57] <archivist> you can take into account some errors, both in software and with effort fixing hardware
[06:34:45] <archivist> emc has error correction if you need it
[06:35:36] <WalterN> wait, what? nothing electronic in a manual mill... lol
[06:35:48] <archivist> I have mechanical erros in mine and write the gcode to machine unidirectionally as you would for a worn manual mill
[06:35:50] <WalterN> unless you count the DRO maybe
[06:36:23] <archivist> accuracy is in the hands of the user
[06:37:17] <WalterN> mostly, not completely though
[06:39:30] <Jymmm> * Jymmm only uses the best dremel and mill fresh pine for the highest precision
[08:52:48] <roh>
[12:51:51] <morficmobile> Good morning
[12:53:06] <kgartner> mornin
[13:19:52] <jthornton> morning
[13:20:25] <morficmobile> good morning mr thornton
[13:20:47] <jthornton> don't be calling me names LOL
[13:23:49] <morficmobile> ha
[13:24:23] <jthornton> more rain here today... got 7 inches thursday night
[13:25:40] <morficmobile> we got some good rain + wind + lightning mid week
[13:53:03] <morficmobile> i wonder why our CAM app uses ~14% cpu when idle, if the window is active or inactive the same, it shouldn't have to do anything when noone works with it.
[14:00:53] <jthornton> you should see what SolidWorks uses
[14:03:53] <morficmobile> another broken app? :)
[14:33:03] <mendel|pancakes> mendel|pancakes is now known as mndlbld
[14:44:07] <Guest743> hi there
[14:44:54] <Guest743> I have a problem understanding some things can you help me please?
[14:45:25] <micges> ask
[14:46:10] <Jymmm> Dont ask to ask, just ask. If someone knows, they'll answer, though it's early here, so many may not be around for a few hours.
[14:46:42] <Guest743> ok
[14:47:27] <Guest743> what is the max jitter number ?
[14:47:31] <micges> Jymmm: 'don't ask to ask' should be added to channel topic ;)
[14:47:56] <Jymmm> micges: Eh, it's not that common in here.
[14:49:12] <Guest743> anybody knows?
[14:50:02] <Jymmm> If someone knows, they'll answer
[14:50:22] <micges> Guest743: be patience
[14:50:56] <Guest743> jymm would you stop talking please this doesn't concern you
[14:54:06] <morficmobile> He just pointed out something that make IRC work better, and there will be many people talking at once, some talking to you, some talking to others regarding something they might think of because of what you said.
[14:56:52] <cradek> Guest743: he's trying to help - "don't ask to ask, just ask" is part of the culture on IRC because it avoids noise but also because it's presumptive of the asker that someone will promise to help you without knowing first what you want. "just asking" lets all the readers see the question and decide if they are qualified to help.
[14:57:36] <cradek> the answer to your question is it depends - what is acceptable latency depends on what you are doing.
[14:57:47] <Guest743> this is your own opinion I respect what you said but don't impose your opinions too late cradek you're so out of the point
[14:58:01] <cradek> software step pulse generation requires the best latency numbers
[14:58:18] <cradek> ok, I'll ignore that and continue with my answer
[14:58:27] <cradek> servo systems have a much looser latency requirement
[14:58:33] <Guest743> please do
[14:58:46] <cradek> as does step generation on an external fpga card
[14:59:25] <cradek> so, can you describe what kind of interfacing with EMC you're doing?
[15:08:27] <cradek> I need to run - if you can describe your setup and tell us the latency numbers you have, maybe someone can help.
[15:14:06] <kgartner> did mesa write their drivers for EMC, or was that all work done by EMC people?
[15:17:14] <xyzee> Hi. Can someone please tell me how to start a program from a specific line?
[15:22:13] <Guest743> I'm runing a hobby cnc
[15:26:21] <morficmobile> The people that can answer the question (not me) will want to know how you drive the CNC, through a interface on your parallel port or via a FPGA on a PCI card for example, the more info about your system you can provide, the better the answer will be
[15:29:37] <micges> xyzee: select that line in preview, then click right mouse button on that gcode line and select hit 'run from here'
[15:51:32] <Guest743> when my mechine finishes the task, how can I make it go back to point 0.0.0
[15:53:43] <Guest743> ??
[15:55:30] <morficmobile> look up if G28 works the way you expect
[16:26:40] <skunkworks> I fail at index enable...
[16:27:21] <skunkworks> x axis works - I didn't notice until today that z is only homing to the switch
[16:28:55] <skunkworks> I think I have everything correct - I have home_use_index = yes but it homes to the switch and the index inable never goes high like it does for x when it homes
[16:29:29] <skunkworks> * axis.2.index-enable
[16:34:36] <Jymmm> skunkworks: are all the axis the same in the ini?
[16:35:13] <Jymmm> skunkworks: is there something unique about Z ?
[16:35:25] <Jymmm> skunkworks: compared to XY
[16:36:08] <skunkworks> not that I can see
[16:36:26] <L84Supper2> anyone here going to IMTS next week? http://www.imts.com/
[16:37:06] <Jymmm> skunkworks: does XY have switches too?
[16:37:41] <skunkworks> I only have x and z working right now.
[16:37:47] <skunkworks> both have switches
[16:38:30] <Jymmm> skunkworks: Ok, is the index position BEFORE or AFTER the switch on both X and Z?
[16:39:37] <Jymmm> skunkworks: (if you only have XZ working, how in the hell did you make the circle video?)
[16:40:37] <skunkworks> Jymmm: that is x and z
[16:40:48] <skunkworks> (horizontal machine)
[16:41:01] <Jymmm> skunkworks: well, make it XY damnit!
[16:41:08] <Jymmm> =)
[16:41:18] <Jymmm> skunkworks: Ok, is the index position BEFORE or AFTER the switch on both X and Z?
[16:42:17] <Jymmm> skunkworks: Are both X & Z the EXACT same in ini ?
[16:43:07] <Jymmm> skunkworks: does one have a longer travel than the other as example?
[16:43:16] <skunkworks> huh - now it works
[16:43:19] <skunkworks> never mind.
[16:43:25] <skunkworks> jeeze
[16:44:31] <Jymmm> skunkworks: heh
[16:45:02] <morficmobile> skunkworks: what was it?
[16:48:22] <skunkworks> I *think it was always working but I wasn't seeing it on halscope. (wrong pin)
[16:48:25] <skunkworks> duh
[16:59:37] <elmo40> L84Supper2: i will attend this instead: http://www.sme.org/cmw
[17:05:22] <IchGuckLive> Hi all
[17:05:35] <Jymmm> What was the name of the "free" cad program from alibre ?
[17:05:41] <Jymmm> x something
[17:05:58] <IchGuckLive> Quest is there a aplication that can produce programms for WING Styro cuting on EMC
[17:06:46] <IchGuckLive> i got a self made Double XY with wire between like EDM
[17:08:21] <morficmobile> Jymmm: Design Xpress?
[17:09:44] <IchGuckLive> Jymmm: Did you check out the free open source heekscad CAD/CAM
[17:12:41] <Jymmm> ah it was x-cad.net
[17:13:44] <Jymmm> IchGuckLive: have you tried it?
[17:13:59] <IchGuckLive> yes i work with it every day
[17:14:17] <Jymmm> IchGuckLive: have you done assemblies with it?
[17:14:32] <IchGuckLive> no only parts
[17:15:22] <IchGuckLive> i only need parts to mill
[17:26:43] <L84Supper2> elmo40: I wasn't aware of that show until now
[17:37:28] <Jymmm> These tarp clips are pretty cool (set of 4), they screw tight http://www.harborfreight.com/pack-of-4-tarp-clips-91858.html
[17:39:54] <Jymmm> They have jaws, but since the tension is adjustable a nice way to clamp something securely for painting or hanging
[17:48:26] <Endeavour> Hello
[19:10:48] <Jymmm> I was looking at a diaphragm air pump yesterday, and the air seemed "pulsed". Is there a way to get rid of the pulse? I thought about just feeding the pulsed air into a coffee can type container. Would something like that work?
[19:12:43] <KimK> Feed pulsed air thru a small orifice into a large reservoir? (Lather, rinse, repeat?)
[19:12:49] <morficmobile> a storage tank as buffer, and a air regulator would be my guess
[19:13:08] <morficmobile> not sure what it's for if a coffee can seems sufficient though
[19:13:31] <Jymmm> morficmobile: it's an air pump, like fish tank type, not compressor
[19:13:36] <Jymmm> KimK: say what?
[19:14:44] <cradek> kgartner: Mesa does the FPGA part (and it is fully available under multiple free licenses), and the EMC driver part was done by EMC people, especially Seb K. There is a very effective ongoing relationship between Mesa and the EMC project.
[19:15:02] <KimK> If the air out of the reservoir is still too "pulsy", then do the same thing again. (Lather, rinse, repeat, lol)
[19:15:45] <Jymmm> KimK: Oh, so the coffee can will work, but might need multiple. Would the size of the contianer make a difference?
[19:15:49] <cradek> kgartner: I think the firmware images are actually built by an EMC project person now, using the source from Mesa
[19:16:50] <Jymmm> KimK: I could use 24" long 4" PVC pipe with bib fittings too
[19:17:17] <Jymmm> KimK: I just didn't know if it would work at all
[19:17:25] <KimK> Sure, the larger the reservoir (compared to the pulse volume), the better. Up to a point of diminishing returns, anyway. But a coffee can isn't good for much pressure, I'd guess, is that OK for you?
[19:18:08] <Jymmm> KimK: It's just an aquarium air pump, need flow more than pressure.
[19:18:14] <KimK> You could also use a regulator instead of an orifice, as morficmobile suggests.
[19:18:28] <KimK> Ah, so not a lot of pressure or volume.
[19:18:40] <Jymmm> KimK: Would the reservoir alter the flow?
[19:19:04] <KimK> No, but the orifice/regulator/other might.
[19:19:21] <Jymmm> KimK: I need the volume more than the pressure.
[19:19:25] <KimK> duckbill valve, maybe?
[19:19:58] <Jymmm> KimK: http://people.whitman.edu/~yancey/platypus1.jpg ???
[19:20:56] <KimK> It's like a long-style balloon with the tip cut off, you can blow out easily, but if you try to suck back, it collapses and blocks.
[19:21:31] <Jymmm> KimK: This is the pump http://www.ehydroponics.com/?action=product&id=241&cid=452
[19:25:42] <morficmobile> hm, for fluids, a reduction would not change volume, it should change the speed, with the pressure being lower after the reduction (think thumb on open garden hose), wonder if this applies at all for air....
[19:30:03] <Jymmm> ok 70 lt per minute ~= 2.5 CFM. that should work for what I need.
[19:32:48] <skunkworks> woo hoo - the vfd still powers up.
[19:33:26] <KimK> skunkworks: Just one of the many benefits of joining ubuntu -- it's like AAA!
[19:34:42] <Jymmm> KimK: NOT!
[19:35:51] <KimK> Haha, I'm pulling skunkworks's leg in the wrong channel anyway, maybe I should join something. Joke's on me!
[19:36:19] <Jymmm> KimK: Go sit in #ubuntu for 10 minutes
[19:36:55] <KimK> No thanks, been there, done that, got the t-shirt.
[19:38:07] <KimK> "I sat in #ubuntu for 10 minutes". Size L, please.
[19:38:27] <tlab> I updated to 10.04 and emc rtai modules will not install
[19:40:28] <tlab> maybe I should read... duh
[19:40:55] <jthornton> I like this comment someone wrote in the DX32 Programmers Manual "If tapping first in the part program you must make a G1 move (in outer space is ok) before you rigid tap...
[19:41:09] <KimK> tlab: If you had previously installed an older... Ah, you're way ahead of me.
[19:41:33] <tlab> ya previous version was installed
[19:41:51] <jthornton> skunkworks: any chips yet?
[19:41:53] <KimK> I like tapping in outer space, the chips just float away.
[19:42:46] <jthornton> and your taps never break
[19:43:00] <KimK> Congrats on your new machine, JT.
[19:43:38] <jthornton> ah, thanks... still have a crippled spindle with a 1500 rpm speed limit atm all else is ok
[19:44:21] <jthornton> I'm sure at some point to put EMC on it as EMC2 is light years ahead of the DX32 control
[19:44:36] <KimK> Oh, I missed that. What's the spindle trouble? (Briefly, if you've been over it a million times already.)
[19:45:34] <jthornton> it alarms out if I go 0-2000 or if I step it up past 2800
[19:46:01] <jthornton> I can step it up in 200 rpm increments to 2800
[19:46:15] <jthornton> 0-1500 in one step works too
[19:47:26] <KimK> What's the alarm, overcurrent/overload?
[19:49:07] <jthornton> just says axis alarm... not very specific
[19:50:00] <KimK> Does one power supply feed all the drives, axis and spindle?
[19:50:17] <jthornton> yes, it is a simodrive 611
[19:51:20] <KimK> Are you on 208/120 wye or 240 delta out there?
[19:51:45] <jthornton> 240 rotary phase converter
[19:52:04] <jthornton> less than 1 volt from phase to phase :)
[19:52:40] <KimK> OK. And you're probably getting a solid 240? Not 220 or 210?
[19:52:54] <jthornton> 246-247
[19:53:00] <KimK> Yeah, OK.
[19:53:26] <jthornton> the axis are all happy going 472IPM all the time
[19:53:40] <jthornton> and they are after the spindle drive in the stack
[19:54:09] <KimK> Excellent.
[19:54:43] <jthornton> http://www.gnipsel.com/Bridgeport/eBayISAPI.dll.htm
[19:56:15] <jthornton> third pic is the drives
[19:57:11] <KimK> What's the little uP board SW of the main fuses in pic 6? Tool changer?
[19:58:17] <jthornton> just a sec I'll get the manual it tells me what that does
[20:02:47] <jthornton> that is the AUF board
[20:03:56] <jthornton> says it is a small programmed controller that controls several machine functions
[20:04:12] <Jymmm> What all that black sut on and above the drives?
[20:04:32] <Jymmm> soot
[20:04:49] <jthornton> there is a fan right in front of them blowing in...
[20:05:01] <Jymmm> Motor drivers roasting over an open fire.....
[20:05:03] <jthornton> filter must have been off at one time or another
[20:05:34] <Jymmm> so they still have their magic smoke?
[20:05:40] <Jymmm> INSIDE?
[20:06:55] <Jymmm> oh, you got a manual with it too? cool!
[20:07:18] <jthornton> 3 manuals
[20:07:21] <skunkworks> hey - it runs with 10v also :)
[20:08:03] <jthornton> what does?
[20:08:08] <Jymmm> * Jymmm hands skunkworks 2 9v batteries to run his ENTIRE machine on!
[20:08:21] <tlab> I'm missing kernel mod rtapi, what package is that in?
[20:08:44] <tlab> ohh RTAPI: Locked memory limit is 64KiB, recommended at least 20480KiB.
[20:08:58] <Jymmm> jthornton: Is that a padlock on top of the drives?
[20:09:22] <jthornton> that's the A axis connector
[20:09:38] <Jymmm> oh, heh
[20:10:37] <Jymmm> jthornton: no inside pics? tool changer?
[20:10:53] <jthornton> 8 station tool changer BT30 tooling
[20:11:46] <jthornton> shortest distance tool changer
[20:11:56] <Jymmm> pic?
[20:12:11] <jthornton> I don't have one atm
[20:13:12] <jthornton> wonder if I need to dig into the spindle area and check the tach???\
[20:14:13] <KimK> Good idea, or maybe put a scope (in floating diff mode) on it?
[20:14:32] <jthornton> the rpm reported on the crt seems to be steady though and close to commanded rpm
[20:14:48] <jthornton> ok, what is floating diff mode?
[20:15:11] <KimK> OK. It would be neat if you could find a 4th drive to put in there.
[20:15:51] <jthornton> it is wired up for a rotary, just need the rotary and the drive
[20:17:11] <KimK> Put scope in A minus B mode, connect scope ground to cabinet/earth ground, connect probe grounds to nothing (or remove them), connect one probe to each tach signal lead.
[20:17:25] <skunkworks> the vfd is running from the mesa card
[20:18:01] <jthornton> I have 4 tach leads R, S, T, and REF
[20:18:26] <KimK> Ah, nevermind, I thought we were talking about a DC tach. Still doable though...
[20:18:51] <jthornton> is it an AC tach?
[20:19:24] <KimK> Pick any two (of R,S,T) at a time. Or pick one at a time vs. REF.
[20:19:54] <jthornton> ok
[20:20:07] <KimK> Not sure. Is it an AC or DC motor? Permanent magnet or not?
[20:20:28] <KimK> How many power leads to motor?
[20:21:10] <jthornton> looks like 2
[20:21:49] <Jymmm> When a less than 5yo freezer says "Manual Defrost", what does that mean? Is that frost free?
[20:22:16] <jthornton> no
[20:22:22] <KimK> You unplug it for several hours?
[20:22:41] <jthornton> or take the hose to it
[20:23:03] <Jymmm> Oh
[20:23:12] <Jymmm> damn
[20:23:42] <jthornton> what does a communating reactor do?
[20:23:45] <KimK> Two power leads to motor sounds like DC. But RST sounds like a hall effect? Confusing.
[20:25:45] <KimK> Steadies the current while something commutates? (commutator, SCRs, etc.?)
[20:26:40] <jthornton> specs say AC servo motors
[20:26:44] <KimK> Where is it in the circuit?
[20:27:09] <jthornton> in front of the power section for the drives
[20:29:48] <KimK> Then possibly to make sure the power line impedance is not too low, in case the machine is located very close to the power distribution point?
[20:30:02] <jthornton> here is a drive on flea bay from a guy with only 80% positive feedback LOL
[20:30:43] <KimK> On how many sales?
[20:30:50] <jthornton> 158
[20:31:52] <KimK> Yikes, that's a lot of unhappy people. Worth a read first, I'd say!
[20:32:18] <jthornton> only one bad and a couple of neutral ...
[20:33:55] <jthornton> * jthornton is wandering back to the shop now
[20:34:22] <jthornton> talk to you guys later
[20:45:08] <ichudov_> Guys, I want to look at my axis-1 following error with halscope, by pressing up and down arrows. What should I use for triggering? Thanks
[20:46:06] <cradek> trigger on f-errored
[20:47:46] <ichudov_> ButI get error all the time? it is just smaller than when moving
[20:48:05] <ichudov_> Is there some component that says axis is commanded to move?
[20:54:16] <ichudov_> cradek: actually itlooks great! thanks!
[21:18:13] <Connor> Should I use shielded wire for my limit/home sensors, and rpm sensor?
[21:19:08] <andypugh> For the rpm sensor, yes.
[21:19:47] <Connor> I'm using a qrd1114 with a encoder disk I made..
[21:20:14] <Connor> super simple. 1/2 white, other half black.
[21:20:39] <andypugh> Limit and home are best shielded (nuisance limit-switch tripping can be very annoying) but there it the alternative of using an anti-bounce component in HAL. (I actually have both)
[21:21:12] <Connor> I'm trying to decide on what to use for limit and home switches.. and how to set them up..
[21:21:31] <andypugh> I have reflective optos and metal flags.
[21:21:32] <Connor> I have 7 more of these qrd1114's
[21:21:44] <Connor> reflective or interupt ?
[21:21:49] <andypugh> Reflective
[21:22:20] <andypugh> Focussed type and a comparator on the same board.
[21:22:22] <Connor> How do you have it setup? 3 for each axis (2 limit's and a home) or just 2?
[21:22:35] <andypugh> Just 2 limits.
[21:22:47] <Connor> using one as a home?
[21:22:49] <andypugh> One sensor and 2 flags on each axis.
[21:23:03] <Connor> oh.
[21:23:35] <Connor> I read somewhere that for failsafe, you should wire all the limits in series including the E-Stop.
[21:23:57] <Connor> then have 1 sensor each for home.
[21:24:00] <andypugh> I never read that, and will try to avoid doing so :-)
[21:24:21] <andypugh> http://picasaweb.google.com/bodgesoc/Gibbs#5403345722696629362
[21:25:10] <andypugh> Shows the lathe X limits, the next few pictures show the mill Z (since changed) and the latheZ (which is Mill X)
[21:26:03] <Connor> photo 25 looks as if that's two sensors, and 1 flag.
[21:30:40] <andypugh> Yes, but I changed all that to a moving sensor and 2 flags. That arrangement was prone to getting dirty.
[21:31:35] <Connor> so, how does EMC work with a moving sensor? and how does it know which way to home ?
[21:31:49] <andypugh> (I also moved the ballscrew inside, put the motor on the top, added a head-tilt bracket, converted to metal gears and belt primary, converted to oil-bath lubrication...) That's an old picture.
[21:32:40] <andypugh> EMC doesn't know if the flag is movibg past the sensor, or the sensor past the flag.
[21:32:56] <Connor> right, but, your using just one sensor..
[21:33:00] <andypugh> You tell EMC which direction to home in in the INI file.
[21:33:34] <andypugh> Yes, EMC sees home +ve limit and -ve limit trip all at the same time. It doesn't matter.
[21:34:10] <andypugh> In practice, once homed, EMC will never hit a limit switch.
[21:34:31] <andypugh> Unless it has seriously lost position.
[21:34:51] <andypugh> And all EMC does when it hits a limit is stop, so it doesn't matter which one it it.
[21:34:56] <Connor> yea, I need to figure that out.. I want full movement.. I should have added a little extra, but, didn't..
[21:35:03] <andypugh> (There is an option to ignore limits when homing)
[21:35:44] <andypugh> I think my hard limits are 0.2mm from the soft limits on one of my axes.
[21:36:07] <andypugh> The focussed opto and comparator is remarkably consistent.
[21:36:29] <Connor> yea, I'm not sure about the QRD1114's
[21:37:21] <Connor> Hmm.. 6-ft shielded stereo...
[21:37:26] <Connor> cable
[21:38:07] <andypugh> Looking at the datasheet, the QRD does seem a little less likely to be accurate.
[21:38:43] <andypugh> The ones I have have the advantage that they have a shallow depth of focus, so the flag only has to be 3mm higher than the surface behind, and it still works.
[21:39:15] <Connor> I can start off using these.. and do some testing.
[21:39:19] <andypugh> I would suggest getting some slot-detectors.
[21:40:19] <Connor> Yea, I could turn them into slot detectors.. they're individual components in a plastic case. You can pull them out..
[21:41:07] <Connor> okay, off to get some shielded cable..
[21:42:42] <Connor> oh, any suggestion on were to get couplers from? I'm looking for the helical kind, side clamp instead of set screw. I got some from ebay.. and they're not very good.. (set screw design, and not drilled very well, they woble)
[21:47:30] <andypugh> Good couplers are expensive. I made all my axes toothed-belt drive so I didn't have to worry about them.
[21:48:41] <archivist> I made some and later bought some from arc euro trade
[21:51:51] <andypugh> I think the general advice for folk in the US is McMaster Carr for couplers, they have pages and pages of them.
[21:52:19] <andypugh> I get my stuff from RS, but arc-eurotrade is better priced.
[21:58:02] <andypugh> As the neigbours are having a party, I can go back out to the workshop with a clear conscience.
[21:58:05] <andypugh> Bak later.
[22:30:29] <Connor> McMaster doesn't have the helical style
[22:47:29] <skunkworks> vfd under emc2 control
[22:53:04] <skunkworks> so... how do you ground the computer / mesa hardware? like the 7i33? I have everthing grounded back to the bus bar - but I really have not run anything to the computer yet. the case is bolted to the pannel..
[23:24:02] <skunkworks> bbl