#linuxcnc | Logs for 2012-02-01

[00:00:35] -!- CIA-96 has quit [Ping timeout: 252 seconds]
[00:03:59] -!- theorb [theorb!~theorb@] has joined #linuxcnc
[00:04:05] -!- theorbtwo has quit [Ping timeout: 248 seconds]
[00:04:12] theorb is now known as theorbtwo
[00:16:26] -!- factor [factor!~factor@] has joined #linuxcnc
[00:19:47] -!- Valen has quit [Quit: Leaving.]
[00:32:19] -!- factor_ [factor_!~factor@] has joined #linuxcnc
[00:32:39] -!- factor_ has quit [Read error: Connection reset by peer]
[00:39:59] -!- GoSebGo has quit [Quit: Bye]
[00:40:14] -!- GoSebGo [GoSebGo!~Seb@] has joined #linuxcnc
[00:41:28] -!- jarray52 has quit [Quit: Leaving.]
[00:42:39] -!- jarray52 [jarray52!~solargrin@unaffiliated/jarray52] has joined #linuxcnc
[00:46:15] -!- Loetmichel [Loetmichel!cylly@p54B157A7.dip.t-dialin.net] has joined #linuxcnc
[00:48:29] -!- Loetmichel_ has quit [Ping timeout: 252 seconds]
[00:57:03] -!- Nick001-Shop has quit [Remote host closed the connection]
[00:58:20] <JT-Shop> say goodnight Gracie
[01:02:08] <Jymmm> G'Night Groucho
[01:04:08] -!- ries has quit [Quit: ries]
[01:05:05] <JT-Shop> Mr. Grinch to you my good man
[01:05:48] <JT-Shop> http://www.youtube.com/watch?v=9LjbMVXj0F8
[01:14:54] <Jymmm> JT-Shop: ping
[01:32:53] -!- rob_h has quit [Ping timeout: 247 seconds]
[02:07:50] <SWPadnos> Jymmm, for a buyer, you'd just use "up to 1000", "up to 5000", "up to 10000" ...
[02:15:55] -!- Lasivian [Lasivian!Lasivian@c-50-132-66-96.hsd1.wa.comcast.net] has joined #linuxcnc
[02:15:58] <Lasivian> hiya
[02:19:28] <Jymmm> SWPadnos: Thanks, I'm still thinking all the possibilities =)
[02:19:34] <Jymmm> SWPadnos: How ya doin?
[02:28:33] <SWPadnos> ok
[02:56:21] -!- Gast608 [Gast608!~Gast608@c-68-84-74-152.hsd1.pa.comcast.net] has joined #linuxcnc
[02:57:13] <Gast608> Good evening folks - I have a question on gcode subroutines
[02:57:53] <Gast608> I am using the oword subroutine that Jeff Epler mashed up for me about 2 years ago
[02:58:39] <Gast608> I normally cut and paste code into one of the subroutines to create duplicate parts
[02:59:11] <Gast608> recently I have been playing with calling a file from in the subroutine
[02:59:26] <Gast608> it has work - mostly-
[02:59:55] <Gast608> it got cranky when I called a file with lots of gcode
[03:00:10] <Gast608> I am mostly calling small files and they will work
[03:00:25] -!- seb_kuzminsky has quit [Ping timeout: 276 seconds]
[03:01:07] <Gast608> Question: instead of contantly cutting and pasting into one of the suboutines, is it possible to pass the file name with the subroutine call?
[03:01:42] <Gast608> I did try to do this carefully adjusting the sub
[03:02:23] <Gast608> when I run it, my call line fails with "NEAR LINE XXX - Unknown Word Where Unary operation could be
[03:02:38] <Gast608> any suggestions?
[03:02:46] -!- mhaberler has quit [Quit: mhaberler]
[03:20:57] -!- Gast608 has quit [Quit: Bye Bye]
[03:37:04] -!- West0n has quit [Quit: Bye]
[03:37:21] -!- West0n [West0n!~west0n@unaffiliated/west0n] has joined #linuxcnc
[03:38:24] -!- jarray52 [jarray52!~solargrin@unaffiliated/jarray52] has parted #linuxcnc
[03:49:34] -!- cstop [cstop!~cal@c-50-133-194-221.hsd1.nh.comcast.net] has joined #linuxcnc
[04:02:31] -!- Jim_ has quit [Quit: Leaving]
[04:11:52] -!- Kirk_Wallace_ [Kirk_Wallace_!~kwallace@smb-191.sonnet.com] has joined #linuxcnc
[04:41:39] -!- automata [automata!~amit@] has joined #linuxcnc
[04:50:54] -!- [n0b0dy] [[n0b0dy]!~asdf@99-109-227-209.lightspeed.hstntx.sbcglobal.net] has joined #linuxcnc
[04:58:01] -!- factor has quit [Read error: Connection reset by peer]
[05:03:02] -!- Kirk_Wallace_ has quit [Quit: Ex-Chat]
[05:04:46] -!- cstop has quit [Quit: Leaving]
[05:09:49] -!- factor [factor!~factor@] has joined #linuxcnc
[05:13:41] -!- seb_kuzminsky [seb_kuzminsky!~seb@] has joined #linuxcnc
[05:55:23] -!- maximilian_h has quit [Quit: Leaving.]
[05:58:54] -!- seb_kuzminsky has quit [Ping timeout: 245 seconds]
[06:09:36] -!- ve7it has quit [Remote host closed the connection]
[06:17:24] -!- automata has quit [Ping timeout: 252 seconds]
[06:38:46] -!- vladimirek [vladimirek!~vladimire@bband-dyn226.178-40-36.t-com.sk] has joined #linuxcnc
[06:45:36] -!- West0n has quit [Ping timeout: 272 seconds]
[06:50:09] -!- psha [psha!~psha@] has joined #linuxcnc
[06:58:07] -!- vladimirek has quit [Remote host closed the connection]
[07:21:52] -!- packrat [packrat!~packrator@c-98-209-146-133.hsd1.mi.comcast.net] has joined #linuxcnc
[07:54:45] -!- jonnyATroot [jonnyATroot!~jonny@] has joined #linuxcnc
[08:19:12] -!- jonnyATroot has quit [Quit: Leaving]
[08:19:30] -!- jonnyATroot [jonnyATroot!~jonny@] has joined #linuxcnc
[08:29:44] -!- mhaberler [mhaberler!~mhaberler@] has joined #linuxcnc
[08:32:23] -!- mk0 [mk0!~x-ray@fiztech.basnet.by] has joined #linuxcnc
[08:48:47] -!- Valen [Valen!~Valen@2001:44b8:3199:9100:21e:8cff:fe63:55df] has joined #linuxcnc
[09:09:37] -!- rob_h [rob_h!~rob_h@5ace704a.bb.sky.com] has joined #linuxcnc
[09:15:01] -!- jonnyATroot has quit [Ping timeout: 248 seconds]
[09:31:56] -!- automata [automata!~amit@] has joined #linuxcnc
[09:35:50] -!- automata has quit [Client Quit]
[10:07:41] -!- maximilian_h [maximilian_h!~bonsai@] has joined #linuxcnc
[10:07:51] -!- maximilian_h [maximilian_h!~bonsai@] has parted #linuxcnc
[10:24:16] <Loetmichel> mornin'
[10:27:09] -!- Loetmichel has quit [Read error: Connection reset by peer]
[10:27:24] -!- Loetmichel [Loetmichel!cylly@p54B157A7.dip.t-dialin.net] has joined #linuxcnc
[10:33:56] <Loetmichel> anyone knows a Zen Tooolworks DIY CNC? some ppl i have met have one and i want to make them a new mount for the dremel/proxxon
[10:34:13] <Loetmichel> can someone gice me the distance of the mointing screws?
[10:34:17] <Loetmichel> give
[10:34:36] -!- Fox_Muldr [Fox_Muldr!quakeman@frnk-4d01f7e7.pool.mediaWays.net] has joined #linuxcnc
[10:35:29] <Loetmichel> 'cause THAT is a PITA but no moount
[10:35:34] <Loetmichel> http://www.hackerspace-ffm.de/wiki/index.php?title=Datei:Foto1519.jpg
[12:44:06] -!- Valen has quit [Quit: Leaving.]
[12:55:04] -!- Fox_Muldr has quit [Ping timeout: 245 seconds]
[13:11:22] -!- ries [ries!~ries@] has joined #linuxcnc
[13:13:54] <mrsun> think a quick change holder for the lathe would work if it was made in alu ? :)
[13:15:29] -!- ries has quit [Ping timeout: 240 seconds]
[13:15:55] <jdhNC> why AL?
[13:16:53] <mrsun> cause thats what i know my mill can do :P
[13:16:59] <mrsun> havent tried steel tbh =)
[13:21:33] <archivist> I dont think ally would be strong enough
[13:25:07] <mrsun> problem is that the toolpost on the lathe is to high for my tools :(
[13:25:14] <mrsun> cant get them to centerline with the work
[13:25:19] <mrsun> stupid lathe
[13:37:48] <jdhNC> get smaller tools
[13:38:00] <jdhNC> shave the toolpost?
[13:48:23] -!- shade88 [shade88!~shade88@] has joined #linuxcnc
[13:50:42] -!- ries [ries!~ries@] has joined #linuxcnc
[13:51:37] <shade88> Hello. How i can to increase steplenght and stephold above 500 000 ns?
[13:52:14] <shade88> in stepconf wizard
[13:54:10] <cradek> why do you want that?
[13:58:39] <shade88> because we plan to use our own driver stepper motor. on which we will send signals more than a 1 ms
[14:01:06] <cradek> do you realize that even if you're only half-stepping, you'll only be able to make 1 about rotation per second of the motor with that step length?
[14:02:54] -!- Gromits [Gromits!~chatzilla@] has joined #linuxcnc
[14:05:07] <jdhNC> he's young, has a lot of time left.
[14:06:11] <cradek> aren't we all?
[14:07:19] <shade88> 1 mm per step. acceptable to us
[14:08:04] <cradek> must be a very unusual setup!
[14:08:17] <cradek> if it's just a limit in stepconf, you can post-edit the hal it generates
[14:09:25] <cradek> the stepgen manpage does not mention any limits, so I bet it's just stepconf that's limiting you to reasonable values
[14:09:47] <shade88> well. Thank you. Try do it
[14:09:54] <cradek> welcome
[14:09:58] <skunkworks> are the steppers run with relays?
[14:12:05] -!- FinboySlick [FinboySlick!~shark@] has joined #linuxcnc
[14:18:28] <JT-Shop> Jymmm: pong
[14:23:00] <jdhNC> opto-muscular couplers
[14:23:40] <Gromits> Has anyone else followed the thread on isolcpus on the mailing list?
[14:24:36] <JT-Shop> got too confusing for me to follow it
[14:25:03] <Gromits> The crux of it is pretty simple.
[14:25:54] <Gromits> I was just running linuxcnc on my second core and Gene Haskett is right that it really speeds up the machine
[14:26:23] <Gromits> The question is, are you doing something bad but running a process on the core that you have dedicated to the rtai kernel...?
[14:26:31] <Gromits> but==by
[14:30:11] <skunkworks> The few times I have tried it - I didn't see that much of an improvement.
[14:30:20] -!- psha has quit [Quit: Lost terminal]
[14:32:19] <pcw_home> Or does isolcpus simply disable one CPU requiring taskset so regain its use?
[14:32:55] -!- mk0 has quit [Quit: поздравляю всех с кредитом. халява продолжается.]
[14:33:00] <Gromits> Well, that is a good question, but the info on it claims it pegs the rtai kernel to the cpu designated in isolcpus.
[14:33:46] <Gromits> skunkworks: you have tried isolcpus? or taskset?
[14:34:08] <skunkworks> isolcpus
[14:34:14] <Gromits> isolcpus makes my machine much slower, but the latency test is consistently lower
[14:34:50] <skunkworks> it may have only seen a few percent maybe improvement in latency...
[14:35:01] <pcw_home> So something in LinuxCNC automatically uses the isolated CPU? isolcpus by it self just removes a core from the scheduler
[14:35:02] <skunkworks> it seemed to settle faster.
[14:36:10] <Gromits> by the way, Gene claims that latency is low when using taskset to put it on the idle core. That is true for me too, but if I run another process like glxgears (using taskset) the latency is terrible. Worse than just running on the stock machine.
[14:37:28] <pcw_home> glxgears on the LinuxCNC core> you expect that to be terrible
[14:37:48] <Gromits> why?
[14:38:00] <Gromits> nothing else is running there but the latency test...
[14:38:23] <Gromits> and the latency test only using 2% of the core...
[14:38:32] <pcw_home> latency test with taskset glxgears without?
[14:39:16] <Gromits> no, both with - putting them both on the "idle" core
[14:39:28] <pcw_home> sure that would be a mistake
[14:39:37] <Gromits> why?
[14:40:01] <Gromits> When running the latency test you are suppose to tax the machine...
[14:40:05] <pcw_home> Because you want one core free for LinuxCNC
[14:40:21] <pcw_home> and only LinuxCNC
[14:40:57] <Gromits> Yes, in operation mode... but I am talking about running the latency test...
[14:41:33] <Gromits> Not WHILE linuxcnc is running - but before you run it, to understand the latency of the hardware
[14:42:33] <pcw_home> So if you run the latency test with taskset (to use the isolated CPU core) and everything else on the other core whats the latency?
[14:44:27] <Gromits> I don't know. I would have to use taskset for ALL processes. That would be extremely difficult to do...
[14:45:08] <Gromits> I have used it to put glxgears on the core with the latency test which was telling enough...
[14:46:17] <pcw_home> But thats exactly what you are trying to avoid
[14:47:19] <pcw_home> and what gene avoids with his taskset script AFAIK
[14:47:24] <Gromits> in operation, yes. I would never do that.
[14:47:49] <Gromits> In operation, I would just use taskset to run linuxcnc on the idle core, and that is all.
[14:48:39] <Gromits> The issue I am raising, is that typically with a new system you run the latency-test to understand how fast the base thread can be set to.
[14:49:02] <Gromits> If I do the same thing, but using taskset to put the latency-test on the idle core, then the numbers look great.
[14:49:47] <pcw_home> right but unless you run the latency test in the same CPU environment as LinuxCNC its results dont apply
[14:49:48] <Gromits> BUT - if I run another process on that core as well (which is something you should do while running the latency-test normally), the latency numbers are bad
[14:50:15] <Gromits> But that is exactly what I am doing by using taskset to put it on that core.
[14:51:25] <pcw_home> which seems silly since that's not the final running environment so not representative of LinuxCNCs performance
[14:52:04] <Gromits> You mean running glxgears (or anything) as well? I agree.
[14:53:27] <Gromits> But if the latency-test is ever valid, then why wouldn't it be valid using taskset?
[14:53:38] <pcw_home> running glxgears is fine, deliberately running it on your carefully isolated CPU makes no sense to me since that would never happen in a running system
[14:54:41] <Gromits> I agree - since we have an isolated core - there should be nothing else there beside linux cnc...
[14:55:32] <Gromits> I was just asking the question as to why latency on the second core is worse than the latency on the first - or both?
[14:58:12] <pcw_home> Dont know. maybe if LinuxCNC "owns" one CPU, its core threads/processes never get flushed from the CPU cache
[14:59:21] <Gromits> hmm, might be possible...
[14:59:52] <Gromits> I wish someone who understood the rtai kernel and the scheduler would weigh in. There is just a bunch of speculation and random trials.
[14:59:58] <pcw_home> skunkworks did you try Genes taskset trick with isolcpus?
[15:00:12] <skunkworks> no - not yet
[15:01:11] <pcw_home> I would but our test machine is just single core
[15:01:36] <pcw_home> I should upgrade to a better dumpster machine
[15:02:28] <cradek> I think our code asks rtai to put all the *realtime* threads on the highest cpu
[15:02:40] <cradek> my understanding is that if you isolate that one, you'll get better latency numbers
[15:03:21] <cradek> I am not sure if this is the setup gene is trying to get. There's confusion/disagreement floating around about whether the nonrealtime userland parts should also be on that cpu
[15:04:25] <Gromits> that makes sense. Isolate it so nothing runs on it by the default scheduler, then tell the kernel to puts specific threads there...
[15:04:48] <cradek> yes I think that is exactly what happens
[15:05:06] <pcw_home> I wonder if that has been tested, seems like Gene got a huge improvement
[15:05:07] <pcw_home> (but maybe its because having the non RTAI parts of LinuxCNC running on the same core helps somehow)
[15:05:10] -!- Nick001 has quit [Ping timeout: 276 seconds]
[15:05:12] <cradek> someone even posted the chunk of code
[15:05:43] <cradek> pcw_home: I've done it on my dual P3-1000, and got much-improved latency numbers, but that's been ages ago
[15:06:09] <Gromits> I just tried it as well, and I can concur with Gene - there is a huge improvement of the responsiveness of my machine
[15:06:09] <cradek> I'm not even sure it was the current lucid kernel
[15:06:16] <pcw_home> isolcpus or isolcpus+taskset?
[15:06:23] <cradek> Gromits: define "it"
[15:06:24] <Gromits> isolcpus and taskset
[15:06:39] <Gromits> I have had isolcpus on for a long time.
[15:07:11] <Gromits> When I did it, the latency-test numbers improved quite a bit
[15:07:13] <Gromits> However, my machine has been quite slow
[15:07:43] <Gromits> I have just lived with it. Several people in the shop have commented on how sluggish the machine is.
[15:08:10] <pcw_home> what Motherboard?
[15:08:30] <Gromits> Now, if I use taskset to move linuxcnc to the idle core, everything - both linux cnc AND all other random things I am doing are much much snappier.
[15:08:32] <cradek> what's your base thread? for any speed of machine there's a base thread setting that'll make it suck.
[15:08:37] <Gromits> D510MO
[15:08:56] <Gromits> I have no base thread - only a servo thread as I am using 7i43
[15:09:15] <cradek> huh that shouldn't be too sucky then (although it's a very low power machine)
[15:09:57] <Gromits> Before doing isolcpus the machine was fine - not the fastest - but not noticably poky either
[15:10:12] <Gromits> taking away a core had an obvious impact
[15:11:17] <Gromits> we were doing step gen on the MB (at the very beginning) which is why we cared about latency
[15:12:01] <Gromits> Obviously I care less (not in the snide way :-) now that the Mesa card is doing step gen...
[15:14:05] -!- seb_kuzminsky [seb_kuzminsky!~seb@174-16-121-79.hlrn.qwest.net] has joined #linuxcnc
[15:14:44] <cradek> what's your servo thread rate then? do you have a huge ladder or complex hal or anything like that?
[15:15:39] <Gromits> servo thread is default, I have tried to set it down below 750,000 in the past and always gotten realtime errors. i will try it again the on the other core and see if that changes
[15:16:25] <Gromits> We have a little bit of hal and a bunch of python, but not an outrageously large amount. no ladder.
[15:17:42] <skunkworks> I have also not been able to set the base thread much less than 1ms. I think we figured it was the ladder - but I have not investigated it too much.
[15:19:46] <pcw_home> I current have a 6 KHz servo thread for our testing. pretty sure CL is a time hog
[15:19:47] <skunkworks> the latency test on the k&t I think is between 10000 and 15000
[15:21:27] <Gromits> well, perhaps a small improvement in the servo thread. Looks like I can set it to 600000, but not lower without getting the realtime error message...
[15:21:39] -!- cncbasher [cncbasher!~quassel@cpc15-hart9-2-0-cust101.11-3.cable.virginmedia.com] has joined #linuxcnc
[15:22:45] <skunkworks> I wonder if you could put the ladder in a 1ms thread and made the 'base' thread faster...
[15:24:58] <cradek> sure, you could put ladder in a slower thread.
[15:25:01] -!- pingufan [pingufan!~rainer@goliath.hantsch.co.at] has joined #linuxcnc
[15:25:08] <pcw_home> doesn't that mean the the CL thread would be interrupted by the servo thread?
[15:25:10] <pcw_home> (not sure if that a problem or not)
[15:25:21] <cradek> yes I think it means that
[15:25:30] <cradek> I agree with not being sure if that's a problem
[15:26:38] <cradek> if you could separate reading/writing encoders/dacs (for servo thread) and for reading/writing gpio (for slower io thread) you'd surely be fine
[15:26:46] <cradek> not sure if that's possible, though
[15:29:47] <shade88> i want to change file hal_parport.c. and i added #include<stdio.h> into then. but when i execute make command, i have exeption "no such file or directory". what should i do, to make it work?
[15:29:55] <skunkworks> what is odd - the 'scan time' within the cl editor - show to be pretty fast.. like <50us if I recall correctly
[15:30:53] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #linuxcnc
[15:33:24] bootnecklad1 is now known as bootnecklad
[15:39:37] -!- Fox_Muldr [Fox_Muldr!quakeman@frnk-5f746c80.pool.mediaWays.net] has joined #linuxcnc
[15:41:05] <cradek> shade88: realtime modules run in kernel space, where you can't use stdio or other library calls.
[15:43:12] -!- isssy has quit [Quit: Bye Bye]
[15:44:41] <shade88> <cradek> is there a way to write to a file the values are sent to the parallel port?
[15:45:17] <cradek> file io is also not possible
[15:45:29] <cradek> use halscope instead
[15:47:25] <shade88> i go read about halscope. thank you
[15:51:12] -!- Nick001-Shop [Nick001-Shop!~chatzilla@clsm-208-111-237-28-pppoe.dsl.clsm.epix.net] has joined #linuxcnc
[15:53:31] <skunkworks> what is odd - the 'scan time' within the cl editor - show to be pretty fast.. like <50us if I recall correctly
[15:53:34] <skunkworks> heh
[15:53:36] <skunkworks> oops
[15:56:23] -!- psha [psha!~psha@] has joined #linuxcnc
[15:57:45] -!- sumpfralle [sumpfralle!~lars@31-17-29-52-dynip.superkabel.de] has joined #linuxcnc
[16:02:06] -!- FinboySlick [FinboySlick!~shark@] has parted #linuxcnc
[16:02:34] -!- FinboySlick [FinboySlick!~shark@] has joined #linuxcnc
[16:02:36] -!- FinboySlick [FinboySlick!~shark@] has parted #linuxcnc
[16:04:46] <Gromits> psha: hi
[16:08:38] <psha> hi
[16:08:49] <Gromits> i have a glade question: should the "Layout" Container in glade work?
[16:09:07] <psha> what's layout container? :)
[16:09:43] <psha> don't know )
[16:09:44] <Gromits> you know how there are vbox and hbox?
[16:09:46] <psha> i've not used it
[16:09:50] <Gromits> really?
[16:10:01] <psha> i've used *box, fixed and table
[16:10:04] <psha> probably others too
[16:10:26] <Gromits> hmm, one of the example I came across used it...
[16:11:01] <psha> it looks like 'fixed'
[16:11:15] <psha> probably it's mhaberler's example
[16:11:23] <Gromits> yes, may be.
[16:11:36] <Gromits> it looks like fixed but has vert and horiz arrows next to it
[16:11:45] <psha> scrollable fixed?
[16:12:17] <Gromits> perhaps
[16:12:28] <Gromits> whenever I use it, no panel appears in axis
[16:13:38] <Gromits> when I hover mouse over it, it claims to be "Layout"...
[16:22:17] <psha> probably it shrinks to 1x1 size
[16:22:30] <Gromits> ah
[16:29:52] -!- factor has quit [Read error: Connection reset by peer]
[16:31:40] <Gromits> yep, that was it. thanks!
[16:35:36] -!- Nick001-Shop has quit [Remote host closed the connection]
[16:35:58] <Gromits> i think it was shrinking to 0x0, btw :-)
[16:36:19] -!- Nick001-Shop [Nick001-Shop!~chatzilla@clsm-208-111-237-28-pppoe.dsl.clsm.epix.net] has joined #linuxcnc
[16:56:00] -!- Gromits has quit [Quit: ChatZilla 0.9.88 [Firefox 9.0.1/20120117091749]]
[16:56:07] -!- sumpfralle1 [sumpfralle1!~lars@] has joined #linuxcnc
[16:56:53] -!- sumpfralle has quit [Ping timeout: 248 seconds]
[17:14:36] -!- sumpfralle [sumpfralle!~lars@31-17-29-52-dynip.superkabel.de] has joined #linuxcnc
[17:14:53] -!- ve7it [ve7it!~LawrenceG@S0106001c10b7770f.pk.shawcable.net] has joined #linuxcnc
[17:15:20] -!- pingufan has quit [Quit: Konversation terminated!]
[17:15:33] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #linuxcnc
[17:16:17] -!- sumpfralle1 has quit [Ping timeout: 252 seconds]
[17:16:56] -!- isssy has quit [Client Quit]
[17:22:44] -!- acemi [acemi!~acemi@unaffiliated/acemi] has joined #linuxcnc
[17:25:30] -!- mhaberler has quit [Quit: mhaberler]
[17:32:01] -!- kanzure [kanzure!~kanzure@] has joined #linuxcnc
[17:32:27] -!- mhaberler [mhaberler!~mhaberler@] has joined #linuxcnc
[17:33:48] -!- shade88 has quit [Read error: Connection reset by peer]
[17:34:03] -!- shade88 [shade88!~shade88@] has joined #linuxcnc
[17:40:15] -!- jonnyATroot [jonnyATroot!~jonny@] has joined #linuxcnc
[17:49:54] -!- shade88 has quit [Quit: Leaving]
[17:53:33] <Jymmm> JT-Shop: Nitrogen regulators, are there two kinds; pressure and flow? Are they interchangable at all, like from flow to pressure?
[17:55:18] <jdhNC> regulators regulate pressure. flow is determined by pressure and orifice size
[17:55:57] -!- morfic has quit [Changing host]
[17:55:57] -!- morfic [morfic!morfic@unaffiliated/morfic] has joined #linuxcnc
[17:57:02] -!- The_Ball has quit [Ping timeout: 252 seconds]
[17:57:18] <Jymmm> Well, this is CF/m not PSI... http://www.harborfreight.com/regulator-gauge-94841.html
[17:59:08] <Jymmm> err CF/hr
[17:59:30] <jdhNC> there is a restricted orifice of known size inside the output barb. The low side gauge is just labeled in flow units for that orifice size
[17:59:53] <jdhNC> obviously, the gauge is only sensing pressure, not flow.
[18:00:24] <Jymmm> So, if I remove the orifice and replace the gauge with a PSI, it'll work?
[18:00:25] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #linuxcnc
[18:00:41] <jdhNC> yes
[18:01:09] <Jymmm> will it be accurate?
[18:01:26] <jdhNC> how accurate do you need it to be.
[18:01:44] <archivist> and accurate at what pressure or flow
[18:01:49] <jdhNC> a cheap gauge is a cheap gauge. It will be 'accurate' up util the flow is restricted at the barb
[18:02:06] <Jymmm> +- 2psi
[18:02:13] <Jymmm> gove or take
[18:02:17] <Jymmm> give or take
[18:02:26] <jdhNC> at what pressure?
[18:02:41] <Jymmm> 10 - 60 psi
[18:03:49] <jdhNC> still depends on flow rate, but probably
[18:05:05] <jdhNC> that's just a single stage reg though, it will change based on input pressure
[18:05:07] -!- fragalot [fragalot!~thomas@gentoo/user/FamousToaster] has joined #linuxcnc
[18:05:16] <fragalot> how long ago did the #emc channel die again? lol
[18:05:28] <fragalot> only just noticed >.>
[18:05:30] <Jymmm> and a dual does what exactly?
[18:05:44] <Jymmm> fragalot: It's still alive, sorta.
[18:05:58] <jdhNC> first one regulates HP down to something reasonable, the second only has to deal with mostly regulated
[18:06:19] <fragalot> Jymmm: meh :P
[18:06:20] <Jymmm> Horse Power?
[18:06:29] <jdhNC> high pressure
[18:06:32] <Jymmm> high press
[18:06:38] <Jymmm> took me a while =)
[18:07:15] <Jymmm> jdhNC: Are there dual reg that will still allow 0-250 psi ?
[18:07:56] <jdhNC> should be, don't recall seeing one that went that high though
[18:08:31] <jdhNC> you can use two single stage regs in series. Second one doesn't need to be high pressure
[18:08:35] <Jymmm> You might have to reload this page twice, but is $80 a good price? http://www.grainger.com/Grainger/HARRIS-Medium-Duty-Nitrogen-Regulator-5KZ42
[18:08:59] <jdhNC> or if you don't care about pressure increase over (longish) time, it wont' matter
[18:09:53] <Jymmm> jdhNC: Well, I kinda do, but if it's cheaper just to refill the tank than pay $500 for a dual reg, I'll refill the tank =)
[18:10:16] <jdhNC> I have quite a few regs, never paid more than $30ish
[18:10:46] <Jymmm> Got a spare you wanna sell =)
[18:11:18] <Jymmm> I can only find one used sadly.
[18:11:40] <jdhNC> there should be hundreds on ebay
[18:12:06] <jdhNC> that grainger picture isn't the real gauge, it has the wrong connector
[18:12:34] <Jymmm> Yeah, I'm going by the desc CGA-580
[18:13:01] <Jymmm> Though, I'm not sure what "Purging" means in this context.
[18:13:03] <jdhNC> that looks like a cga540/o2 reg
[18:13:42] <jdhNC> it self-purges when the pressure increases on the downstream side
[18:14:04] <Jymmm> Oh, that sucks
[18:14:16] <jdhNC> it's generally a good thing
[18:14:34] <Jymmm> Well if it's a safety thing, ok. If it's a SOP thing, then no.
[18:14:41] -!- GoSebGo has quit [Quit: Bye]
[18:14:50] <jdhNC> it just purges until the pressure reaches the pressure the reg is set to
[18:14:58] -!- GoSebGo [GoSebGo!~Seb@] has joined #linuxcnc
[18:15:24] <jdhNC> otherwise you couldn't set the pressure with the downstream side closed off
[18:15:35] <Jymmm> ah
[18:15:40] <jdhNC> without turning off the supply side on pressure decreases
[18:16:50] -!- factor [factor!~factor@] has joined #linuxcnc
[18:17:23] <jdhNC> that same reg is on ebay for $90 BIN, even the same pic
[18:17:32] <Jymmm> heh
[18:19:08] <Jymmm> Hmmm, I just realized I also need 12CF/s, fudge.
[18:19:18] <jdhNC> damn
[18:19:38] <Jymmm> 42K CF/hr, fsck me
[18:20:04] <jdhNC> you can't afford that :)
[18:20:34] <Jymmm> I can't afford the TANK, much less the regulator =)
[18:21:11] <jdhNC> I lease 4/each Helium & O2, that's painful enough.
[18:21:49] <Jymmm> Well, if I can get 5s @ 12CF/s, that's a 80lb tank.
[18:22:10] <jdhNC> lb?
[18:22:21] <Jymmm> Nitrogen
[18:22:32] <Jymmm> I have a 40lb tank
[18:22:59] <jdhNC> gas by the lb?
[18:23:00] <Jymmm> err cf
[18:23:12] <Jymmm> I have a 40cf tank
[18:23:15] <Jymmm> sorry
[18:23:36] <jdhNC> I get gas in 300ish cf tanks, normal sized gas bottles
[18:24:12] <Jymmm> I just didn't have the room to storage that size tank
[18:24:33] <jdhNC> it's only a 8.5" tank?
[18:25:33] <Jymmm> Heh, hell who needs a regulator, just crack open the valve @ 2000PSI, I should be able to get the 42K Cf/hr no problem =)
[18:26:46] <Jymmm> jdhNC: And it doesn't fit under my bench.
[18:28:29] <jdhNC> wtf do you need with 700cfm?
[18:28:57] <Jymmm> jdhNC: Fire suppression in my laser engraver
[18:29:19] <jdhNC> I would dispute that number.
[18:29:37] <Jymmm> the laser cabinet is 12CF
[18:30:00] <Jymmm> I doubled that for fudge factor
[18:31:22] <Jymmm> I'm guessing 5 seconds if flooding the compartment would extinguish a flame, though there still may be heat/ambers.
[18:31:59] <Jymmm> That should give enough time to shut down the laser, and stop the blower
[18:32:28] <Jymmm> or at least turn off the blower and close a gate valve
[18:33:20] <Jymmm> jdhNC: make sense?
[18:33:37] <jdhNC> what is going to trigger all this?
[18:34:09] <Jymmm> heat sensor and/or BRB (Big Red Button)
[18:34:36] -!- GoSebGo has quit [Quit: Bye]
[18:34:49] <jdhNC> get a fire extinguiser, plumb it in to the enclosure
[18:34:58] <Jymmm> Why?
[18:35:05] <jdhNC> it's cheap and easy
[18:35:29] <jdhNC> http://www.fireboy-xintex.com/portable-fire-extinguisher.htm#M
[18:36:03] <Jymmm> Dry extinguishers are caustic, CO2 will promote condensation, and Nitrogen is cheap and I have an article on Nitrogen as a replacement for Halon for fire supression from NIST.
[18:36:44] <jdhNC> those aren't co2
[18:37:04] <Jymmm> No, those are dry, which is caustic
[18:37:19] <jdhNC> Unlike dry chemical systems, these portable extinguishers leave no residue whatsoever.
[18:37:19] <Jymmm> And EXPENSIVE
[18:37:30] <Jymmm> I looked at them from te marine supply
[18:38:24] <Jymmm> http://www.westmarine.com/webapp/wcs/stores/servlet/ProductDisplay?productId=96210&catalogId=10001&langId=-1&storeId=11151&storeNum=50469&subdeptNum=50502&classNum=50504
[18:38:41] <Jymmm> That has a auto / manual release
[18:38:53] <Jymmm> so you can plumb into engine compartment.
[18:39:06] <Jymmm> my bad on the dry / clean part.
[18:40:10] <Jymmm> jdhNC: This is what happens when you go pee while your laser is running.. http://i37.tinypic.com/sqh6r7.jpg
[18:40:23] <jdhNC> cool
[18:40:56] <Jymmm> Yeah, $30K pile of ashes and not covered by her insurance company.
[18:41:46] <Jymmm> She had ran 12 jobs that morning so far, no issues. On the 13th one, POOF!
[18:43:01] <Jymmm> I'm not sure if she used the bathroom or signed for a package, but if was just within a couple of minutes
[18:45:22] -!- GoSebGo [GoSebGo!~Seb@] has joined #linuxcnc
[18:46:49] <Jymmm> jdhNC: http://www.bfrl.nist.gov/866/HOTWC/HOTWC2006/pubs/R0000296.pdf
[18:55:22] <jdhNC> Extinguishment times vary from 16 to 80
[18:55:27] <jdhNC> seconds
[18:55:44] <Jymmm> for a full room, right?
[18:56:36] <jdhNC> yep, huge room (relatively)
[18:57:14] <Jymmm> He also says he accunts for 32% O2, instead of flooding 100% Nitrogen
[18:57:54] <Jymmm> So I'm hoping I'd get a little better performance even if overkill
[18:58:36] <Jymmm> at least in theory =)
[19:01:53] -!- Anon9845 [Anon9845!~isssy@78-83-51-185.spectrumnet.bg] has joined #linuxcnc
[19:02:14] -!- isssy has quit [Ping timeout: 252 seconds]
[19:04:35] -!- Anon9845 has quit [Client Quit]
[19:04:58] -!- isssy [isssy!~isssy@78-83-51-185.spectrumnet.bg] has joined #linuxcnc
[19:13:38] -!- vladimirek [vladimirek!~vladimire@bband-dyn226.178-40-36.t-com.sk] has joined #linuxcnc
[19:21:16] <A2Sheds> http://reprap.org/wiki/ConductiveMaterials I updated some of the info. Maybe this will inspire someone.
[19:30:35] <mrsun> hmm, drilling int he lathe, the hole becomes conical over 20mm depth ... wth =)
[19:31:01] <mrsun> either the chuck is pointing far far away from the gah whats its called, the part where you put the centers and the drills :P
[19:31:09] <mrsun> or something else is very wrong :P
[19:31:20] <mrsun> (first time using the lathe since i got it :P )
[19:31:22] <jdhNC> tailstock
[19:31:35] -!- sumpfralle1 [sumpfralle1!~lars@31-17-29-52-dynip.superkabel.de] has joined #linuxcnc
[19:32:36] <jdhNC> or the bit is wandering
[19:33:38] -!- sumpfralle has quit [Ping timeout: 244 seconds]
[19:33:39] <mrsun> jdhNC, i center drilled first then drilled ... so dont know how much it can wander =)
[19:34:35] -!- The_Ball [The_Ball!~The_Ball@] has joined #linuxcnc
[19:36:39] <jdhNC> are you using a regular drill chuck in the tailstock?
[19:36:47] <mrsun> jdhNC, yeah
[19:37:53] <jdhNC> so, fix it :)
[19:38:17] <mrsun> jdhNC, the drill chuck or what? :P
[19:38:22] <Loetmichel> *eyes watering* *SNEEEZE* ... THAT was too much CA... burning eyes and nose :-(
[19:38:34] -!- sumpfralle1 has quit [Read error: No route to host]
[19:38:59] <jdhNC> the drill chuck, or the headstock to ways alignment or the tailstock to ways alignment
[19:39:45] <mrsun> jdhNC, yeah ... i guess ill have to do some measuring =)
[19:40:17] <Loetmichel> <- is glueing the 12th cable spool ... a few more and my wire stock is sorted and detangled ;-)
[19:42:50] -!- sumpfralle [sumpfralle!~lars@31-17-29-52-dynip.superkabel.de] has joined #linuxcnc
[19:46:37] -!- vladimirek has quit [Remote host closed the connection]
[19:48:30] -!- vladimirek [vladimirek!~vladimire@bband-dyn226.178-40-36.t-com.sk] has joined #linuxcnc
[19:52:14] -!- The_Ball has quit [Ping timeout: 245 seconds]
[19:54:19] -!- mozmck [mozmck!~moses@client-] has joined #linuxcnc
[20:03:30] -!- The_Ball [The_Ball!~The_Ball@] has joined #linuxcnc
[20:11:49] -!- jonnyATroot has quit [Quit: Leaving]
[20:17:00] -!- kbarry [kbarry!~chatzilla@rrcs-24-153-167-50.sw.biz.rr.com] has joined #linuxcnc
[20:30:23] -!- Vq has quit [Ping timeout: 252 seconds]
[20:31:15] -!- Vq [Vq!~vq@90-227-210-39-no122.tbcn.telia.com] has joined #linuxcnc
[20:34:51] <joe9> i like this irc, has a pretty congenial attitude. glad I stumbled on it.
[20:36:23] <cradek> don't tell anyone that, jerk
[20:38:33] -!- PCW has quit [Remote host closed the connection]
[20:44:39] <JT-Shop> lOL
[20:44:42] <JT-Shop> L
[20:45:11] <fragalot> Anyone here have any idea abt the strongest metal glue you can think of?
[20:45:17] <fragalot> currently considering loctite 435
[20:46:06] <fragalot> it's a very small contact surface (4x10mm) & welding isn't an option (it's the tang of a straight razor)
[20:46:38] <jdhNC> I don't think any glue will work well.
[20:47:20] <fragalot> least I can do is try
[20:47:29] <jdhNC> sure
[20:47:39] <mikegg> JB weld!
[20:47:49] <fragalot> mikegg: loctite 435 is stronger than jb weld
[20:47:53] <fragalot> and faster
[20:47:55] <fragalot> and cleaner :P
[20:48:21] <mikegg> blasphemy!
[20:48:27] <fragalot> fact.
[20:48:36] <jdhNC> jb weld isn't so hot for rejoining small pieces, if you could fillet it, it would be good
[20:49:08] <fragalot> i'm trying to preserve the stamp as much as possible
[20:49:12] <jdhNC> in my (completely worthless) opinion, the 435 would be as good as anything else, which is to say, not very good.
[20:49:21] <fragalot> yeah
[20:49:26] <fragalot> well i'll give it a go
[20:49:32] <jdhNC> it will hold
[20:49:38] <jdhNC> right up until it doesn't
[20:49:53] <fragalot> if it fails, i'll spot weld the top & bottom of it with my TIG & then spend ages with a file getting it to look good again
[20:49:58] <jdhNC> heh
[20:50:16] <jdhNC> resistance weld
[20:50:33] <fragalot> trying not to do anything invasive on the text that's stamped into it
[20:50:56] <jdhNC> don't you think it would be good to reduce the need for abortion as much as possible?
[20:51:03] <jdhNC> <wrong channel>
[20:51:08] <bill20r3> so very wrong.
[20:51:08] -!- acemi has quit [Quit: WeeChat 0.3.2]
[20:52:09] <mrsun> fragalot, braze it? :)
[20:52:18] <mrsun> ahh no heat? :PO
[20:52:25] <fragalot> mrsun: yeah tht'd kill the temper
[20:52:39] <fragalot> spot welding with the TIG would be as far as i'm willing to go in terms of heat
[20:53:08] <fragalot> I've considered drilling holes in the 2 pieces lengthwise, but... that's not gonna happen with this temper, lol
[20:55:44] <mrsun> from what i can tell the alignment is off by 0.1mm over like 400mm
[20:56:02] <mrsun> have to get a better bar to measure with i think =)
[20:56:52] <fragalot> what are you measuring?
[21:02:45] <mrsun> headstock to ways
[21:03:04] <mrsun> using the average method =)
[21:03:13] <mrsun> tho, i havent checked if the bed is twisted
[21:03:57] <jdhNC> did you check tailstock at various extensions?
[21:04:10] <mrsun> nop
[21:04:29] <mrsun> jdhNC, just did this test fast while i was throwing some wood into the "furnace?" ...
[21:04:46] -!- psha has quit [Quit: Lost terminal]
[21:05:18] <jdhNC> furnace? it's 22C outside
[21:06:07] -!- The_Ball has quit [Ping timeout: 244 seconds]
[21:06:52] <mrsun> ...
[21:06:55] <mrsun> -10C here
[21:07:23] <fragalot> all of the machines in the shop at work today were frozen solid
[21:07:39] <fragalot> I found out when I fired up a 7HP electric motor that did nothing but hum at me
[21:07:43] <fragalot> :/
[21:08:21] <mrsun> fragalot, haha nice =)
[21:08:24] <mrsun> no heat? :P
[21:08:53] <fragalot> heater didn't work either lol
[21:09:35] <fragalot> it's a portable diesel burner, someone borrowed it a few years ago & apparently it worked fine then
[21:09:44] <fragalot> tried to fire it up today,... nothing
[21:10:03] <fragalot> opened the filter only to find a very thick slurry... so I think they put parafin or something nasty like that in it
[21:10:34] <fragalot> so that thing isn't going to work until it either warms up again or I clean the entire unit out
[21:10:37] <mrsun> fragalot, home heating oil ? :)
[21:10:49] <mrsun> it goes to a slurry when cold
[21:10:54] <fragalot> possibly
[21:14:09] -!- JustinXJS [JustinXJS!~Justin@c-98-232-177-150.hsd1.or.comcast.net] has joined #linuxcnc
[21:16:22] -!- Valen [Valen!~Valen@2001:44b8:3199:9100:21e:8cff:fe63:55df] has joined #linuxcnc
[21:18:47] <Jymmm> fragalot: what about one of those propane tank mounted heaters?
[21:20:33] <Jymmm> fragalot: Like this, but without the defective part... http://www.cpsc.gov/cpscpub/prerel/prhtml05/05148.html
[21:22:06] -!- The_Ball [The_Ball!~The_Ball@] has joined #linuxcnc
[21:25:45] -!- vladimirek has quit [Remote host closed the connection]
[21:30:29] -!- The_Ball has quit [Ping timeout: 240 seconds]
[21:34:32] <fragalot> Jymmm: http://www.labeltec.be/files/library/GA_100.jpg we've got a few of these too but none that could heat the shop up properly
[21:34:44] <fragalot> Jymmm: the portable diesel one we have now is basically a vivox on wheels :P
[21:35:03] <Jymmm> fragalot: Too bad I dont know what a "vivox" is =)
[21:35:13] <Jymmm> fragalot: How big is this warehouse?
[21:35:19] <Jymmm> err shop
[21:35:41] <fragalot> 400 square meters with plenty of , er.. "vent holes"
[21:36:00] <Jymmm> fragalot: It sounds like you guys need to leave a heater running 24/7 just to prevent freezing.
[21:36:07] <fragalot> pretty much
[21:36:20] <fragalot> it's really a losing battle anyway
[21:36:30] <fragalot> it's basically a glorified shed
[21:36:51] <Jymmm> Yeah, when a 100,000 BTU heater can't heat the place, you're screwed for the most part =)
[21:37:08] <fragalot> :P
[21:37:25] <Jymmm> fragalot: Have you thought of just moving OUTDOORS, where it's warmer
[21:37:32] <fragalot> I actually did
[21:37:39] <fragalot> did some extreme outdoor welding
[21:37:39] <fragalot> lol
[21:37:48] <fragalot> it was nice & sunny
[21:37:55] <Jymmm> I know my garage is colder right now, than it is outside.
[21:38:08] <Jymmm> weird that way.
[21:38:18] <fragalot> it's -5°C outside right now & my room still has the pilot light on
[21:38:23] <fragalot> good insulation++
[21:38:29] <Jymmm> heh
[21:41:00] <fragalot> Jymmm: in either case - I don't like the type of heater you linked
[21:41:20] <fragalot> it's just infrared, doesn't actually HEAT anything, it just FEELS warm, lol
[21:41:31] <fragalot> switch it off & it's ice cold again
[21:41:32] <Jymmm> Well, isn't that the point?
[21:41:38] <fragalot> it is for people
[21:41:45] <fragalot> not so much for people handling steel all day
[21:41:59] <Jymmm> ah
[21:42:54] <fragalot> still pissed that i broke that razor :'(
[21:43:11] <fragalot> it was worth a good 200 euro before I broke it
[21:43:22] <Jymmm> face razor?
[21:43:26] <fragalot> yes
[21:43:29] <fragalot> straight razor
[21:43:36] <fragalot> ex-mint condition
[21:43:42] <Jymmm> That's one hell of a beard you have there
[21:44:10] <fragalot> broke it while attempting to straighten the tang the same way i've done on the others i've restored
[21:44:16] <fragalot> first one to break
[21:44:37] <fragalot> also the most valueable, go figure, heh
[21:45:03] <fragalot> Well the loctite 435 *SEEMS* to hold up to what it needs
[21:45:09] <fragalot> sofar, atleast
[21:48:07] <Jymmm> Well, sorry to hear that.
[21:50:17] <fragalot> shit happens. :)
[21:51:03] <Jymmm> Yeah, but still.
[21:53:12] bootnecklad is now known as suitnecklad
[21:58:11] -!- tom3p [tom3p!~tomp@75-150-195-235-Illinois.hfc.comcastbusiness.net] has joined #linuxcnc
[22:03:15] <Valen> fragalot: weld it up, grind, heat treat polish QED ;->
[22:04:07] <fragalot> Valen: not going to risk botching the heat treat on a blade that's got it done to perfection
[22:04:31] <Valen> well if its broken, its not that perfect anymore ;-P
[22:04:39] <fragalot> esp. with a 6/8" full hollow grind down to a .12mm thickness
[22:04:47] <fragalot> Valen: the BLADE is still fine :P it's the tang that isn't
[22:05:10] <Jymmm> fragalot: duct tape =)
[22:05:16] <fragalot> I also don't want to ruin the engravings
[22:05:33] <fragalot> so if I do weld, it's going to be on the top & bottom of the tang & just spot welds with a TIG
[22:05:49] <Jymmm> fragalot: and heat sink the blade itself?
[22:05:51] <fragalot> all I have to do then is match the pattern of the .. ribbly bits
[22:05:52] <fragalot> Jymmm: yeah
[22:06:00] <fragalot> have it sit in oil or water while i tack it
[22:06:05] <Nick001-Shop> Using apt-get update at the buildbot- will it just get updates and bugfixes or will it change the ini files and such?
[22:06:22] <fragalot> but that's if I break the glue bond when stropping it tomorrow
[22:06:28] <fragalot> if it survives that, i'm golden
[22:06:37] <Jymmm> Nick001-Shop: It is best to BACKUP everything before doing an upgrade.
[22:07:08] <fragalot> anyway off to bed for me
[22:07:09] <fragalot> gnite :)
[22:07:16] <Jymmm> G'Night fragalot
[22:07:34] <Jymmm> fragalot: Pleasant sharp ass razor dreams!
[22:07:50] <Nick001-Shop> I thought it was an update, not upgrade
[22:07:53] <fragalot> 's fine i've got others to keep my cleanly shaven :P
[22:08:05] <fragalot> none as shiny, but they're funtional
[22:08:28] <Jymmm> Nick001-Shop: My bad, you're right. Still good to backup when doing an update too.
[22:09:14] <Nick001-Shop> just the configs or is there other stuff to backup?
[22:09:50] <Jymmm> Nick001-Shop: *I* am not sure.
[22:10:20] <cradek> our apt packages do not edit configs that are in your home directory
[22:10:38] <cradek> you should have backups of course, but not because of the upgrade
[22:12:45] <Jymmm> If only he had a tape robot.... NOT!
[22:14:19] <Nick001-Shop> what else should I backup besides the config section and how do I store that backup on the same drive as EMC2? I guess I'm asking how to have multiple instances of emc2b on the same drive or do I get into trouble with the boot record
[22:14:44] <cradek> make a copy of your configs somewhere other than on that drive
[22:14:59] <cradek> a second copy on the same drive is the poorest type of backup imaginable
[22:15:26] <cradek> disk drives are ephemeral
[22:15:49] <Jymmm> QUESTION.... Do you think there is a legal reason why device come with a power adapter (wal-wort) as a "plug in" instead of being hard wired into the device directly? OR, do you think it's just cheaper for mfg's to buy them in bulk with the connectors already?
[22:16:05] <Nick001-Shop> I know - just trying not to have to screw more stuff into the box
[22:16:27] <cradek> wall warts make the part you can drop in the sink or tub low-voltage
[22:16:27] <jdhNC> my kids think that having the only copy of their stuff on a thumb drive is reasonable.
[22:17:04] <cradek> also they make the part you hold or sit on your lap lighter and cooler
[22:17:35] <jdhNC> they also make teh device region independent
[22:17:36] <Jymmm> cradek: Right, but why not HARD WIRED (soldered to the PCB) instead of a jack? I'm sure the jack adds to the cost when you get into the 100K's quantity.
[22:17:47] <cradek> oh I get your question
[22:18:04] <cradek> well on a laptop that'd suck
[22:18:05] <Nick001-Shop> can the thumb drive be bootable?
[22:18:06] <Jymmm> jdhNC: Most come 110/220 universal, at least voltage wise.
[22:18:24] <cradek> but for a clock radio, sure it could be hardwired, and I think some are
[22:18:32] <jdhNC> imagine a desktop device with a hardwired wall wart
[22:18:46] <cradek> I bet the real reason is they all come from the same factory
[22:18:48] <jdhNC> you could never route the wire with the wart on the end
[22:18:51] <Jymmm> cradek: Sure, but think of things like cordless phone, clock, or even a wireless router.
[22:19:05] <cradek> oh true - you can't put the wart through the hole in the desk
[22:19:32] <Jymmm> Yeah, but that's few and far in between
[22:19:35] <cradek> but is there a legal reason - who knows
[22:19:53] <jdhNC> Third Annual CNC Workshop
[22:20:01] <cradek> you might have an easier time getting something approved/certified if the HV parts are "separate"
[22:20:03] <jdhNC> I don't think you can get to Ann Arbor, MI from here.
[22:20:15] <Jymmm> cradek: Ah, that's true.
[22:20:24] <cradek> like "china already worked all that out for us"
[22:20:53] <cradek> so they can get it approved once, then sell the same products to everyone
[22:21:16] <cradek> I'm talking out my ass, though, I have no idea
[22:21:28] <Jymmm> cradek: Makes sense, I just wasn't sure if it was a liability thing. As onse it's soldered in place it becomes a "whole unit"
[22:21:36] <Jymmm> once
[22:21:50] <cradek> yeah that sure seems like a likely guess to me
[22:22:26] <Jymmm> cradek: I like your idea better though, the certifications just make sense that way.
[22:23:11] <jdhNC> I'd be unhappy if my wall-warted devices were hardwired
[22:23:12] <Jymmm> If one batch of "CE approved" PS are defective, they can just swap them out.
[22:23:34] <jdhNC> I'd be happier if they were all the same size coax with + in inside
[22:27:51] <Valen> modularity would be the reason
[22:28:09] <Valen> manufacturer of psu's makes them by the bazillion
[22:28:45] <Valen> you have 10000 parts your not going to tool up a factory to save you a few cents by not having the jack
[22:35:37] -!- micges has quit [Quit: Ex-Chat]
[22:37:54] -!- FinboySlick [FinboySlick!~shark@squal.net] has joined #linuxcnc
[22:48:39] -!- JustinXJS has quit [Quit: Leaving]
[22:54:36] -!- Fox_Muldr has quit [Ping timeout: 272 seconds]
[23:07:57] -!- tom3p [tom3p!~tomp@75-150-195-235-Illinois.hfc.comcastbusiness.net] has parted #linuxcnc
[23:11:33] -!- isssy has quit [Quit: Bye Bye]
[23:12:01] <Jymmm> Valen: soldering two bare wires is far cheaper than a jack I'd think.
[23:13:51] <Jymmm> But, they could use the same device, then include a wall-wort with the correct plug for the region, US, EU, etc.
[23:14:13] <Jymmm> Standards, you gotta love em!
[23:15:00] <Jymmm> How the computer industry got IEC connectors as standard is beyond me.
[23:15:43] -!- Fox_Muldr [Fox_Muldr!quakeman@frnk-4d01c2f6.pool.mediaWays.net] has joined #linuxcnc
[23:23:15] <Valen> Jymmm: your presuming the jack costs what you pay in a store for it
[23:23:25] <Valen> it probably costs 5 cents or less
[23:23:41] <Valen> and means that the whole PCB can be mechanically stuffed
[23:24:04] <Valen> IEC connectors are pretty handy really, I see them on more and more electronics these days
[23:25:59] -!- WalterN [WalterN!~walter@host-174-45-122-145.bzm-mt.client.bresnan.net] has joined #linuxcnc
[23:27:31] -!- kbarry has quit [Remote host closed the connection]
[23:35:40] <FinboySlick> A2Sheds: You got your hands on a raspberry Pi yet?
[23:38:02] -!- logger[mah]_ has quit [Ping timeout: 252 seconds]
[23:39:07] -!- logger[mah] [logger[mah]!~loggermah@burgi.mah.priv.at] has joined #linuxcnc
[23:40:41] <A2Sheds> FinboySlick: no plans, I'm actually designing these right now http://elinux.org/Embedded_Open_Modular_Architecture/EOMA-68
[23:42:13] <A2Sheds> using the A10 and AMD APU's http://rhombus-tech.net./allwinner_a10/ http://rhombus-tech.net./amd_g_series/
[23:50:40] <Jymmm> Valen: Even at $0.05/each when you have to buy 200K, it adds up.
[23:51:17] <Valen> yeah, now factor that against getting custom psu's made, and paying somebody to do the soldering
[23:51:42] <Jymmm> Valen: they solder in a jack, or two wires, same thing.
[23:51:54] <Valen> jack is done by the PnP machine
[23:52:03] <Valen> goes through the wave solder bath
[23:52:08] <Jymmm> you sure?
[23:52:40] <Valen> lots of jacks are SMT these days too
[23:52:46] <Valen> (thats why they break lol)
[23:53:05] <Jymmm> oh, ok.
[23:54:31] <Jymmm> Well, with localization of the wall plugs, I can see having a jack in the device. I just wanted to make sue it wasn't a legal liability thing/practice.
[23:54:45] <Jymmm> s/sue/sure/
[23:55:50] <Valen> its probably because they can get the wallplug "ticked" as a seperate item then
[23:55:56] <Valen> regulatory approvals arent cheap
[23:56:26] <FinboySlick> A2Sheds: That would actually fit my needs even better than the Pi.
[23:56:31] <Jymmm> Yeah, it's the plug VS hard wired I was questioning.
[23:57:47] <Jymmm> Valen: But if its' wave soldered and then can swap out wall-wort with localized wall plug, then it makes sense. one device + couple localized version of wall-worts.