#emc-devel | Logs for 2007-09-07

Back
[01:16:00] <jmkasunich> I'd be dissapointed if it was sift
[01:16:02] <jmkasunich> soft
[01:16:10] <jmkasunich> although in this case that might be better
[01:33:15] <skunkworks> take a look when anyone gets a chance. http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=320155952581
[01:34:10] <jepler> found one typo so far: limit swithes
[01:34:13] <jepler> 2nd para
[01:35:04] <jepler> didn't notice any others
[01:35:07] <jepler> good luck with your auction
[01:35:25] <SWPadnos> "descided" should be "decided" - third sentence of paragraph 2
[01:35:39] <skunkworks> you guys are great :)
[01:35:45] <SWPadnos> also the word "to" should be "do" in that sentence
[01:35:54] <SWPadnos> "what it could do hooking ...
[01:35:57] <SWPadnos> "
[01:36:31] <SWPadnos> pluged should be plugged
[01:36:52] <jepler> I clearly didn't look hard enough for the typos
[01:36:58] <SWPadnos> "if it where me" should be "if it were me"
[01:37:16] <skunkworks> heh
[01:37:24] <skunkworks> * skunkworks takes notes
[01:37:45] <SWPadnos> "approximately" is usually abbreviated as "approx", not "aprox"
[01:38:54] <SWPadnos> ok - that's it (plus the one jepler mentioned)
[01:40:10] <skunkworks> Thanks guys
[01:40:14] <SWPadnos> any time
[02:01:58] <jtr> skunkworks: I saw a couple more typos: "where burned" should be "were burned", and "by 3 gecko" should be "buy 3 gecko"
[02:03:43] <skunkworks> jtr: thank you
[02:10:02] <jtr> you're welcome. Good luck on the auction.
[02:14:45] <skunkworks> Thank you
[03:38:38] <SWPadnos> jmkasunich, is bfload working for the 5i20?
[03:38:58] <SWPadnos> (I assume the answer is yes, but wanted to make sure)
[04:03:05] <jmkasunich> yes
[04:03:42] <SWPadnos> ok cool. I fiddled with it a little today but couldn't spend much time due to parts ordering stuff and audits and things
[04:03:59] <SWPadnos> so I'll be getting that going for the 5i22 tomorrow, I think
[04:04:04] <jmkasunich> * jmkasunich is chillin' (finally) - I can turn on my air conditioner
[04:04:08] <SWPadnos> cool
[04:04:18] <SWPadnos> :)
[04:04:38] <jmkasunich> I had to re-do the condensate drain, since the old one emptied into the shower drain of the former basement bathroom
[04:04:53] <jmkasunich> its about 23 feet longer now
[04:05:22] <jmkasunich> 17 feet of which ran behind cupboards, washer/dryer, and a computer rack
[04:05:41] <jmkasunich> s/ran/now runs/
[04:07:45] <cradek> I think skunkworks wants to keep the engraver, starting the bidding at $1k
[04:07:51] <cradek> (I don't blame him)
[04:08:08] <fenn> nah that was the plan originally
[04:08:19] <fenn> he says they usually go for $2k
[04:08:40] <cradek> sure but you've got to start the auction at $2 to get the morons to bid it up
[04:08:52] <cradek> we'll see :-)
[11:33:06] <skunkworks> cradek: http://www.cnczone.com/forums/showthread.php?p=339991#post339991
[11:33:07] <skunkworks> bbl
[12:03:15] <Guest871> Guest871 is now known as skunkworks_
[12:11:37] <cradek> skunkworks: thanks, I'm looking for anything that could explain how/why the fault light comes on. I checked the tach signal and it looks great, and the power is good, trying to figure out what to check next, maybe the manual has suggestions
[12:13:26] <cradek> (I do have a basic wiring diagram already)
[12:14:04] <skunkworks_> http://www.cnczone.com/forums/showthread.php?p=340013#post340013
[12:14:58] <cradek> haha
[12:19:21] <skunkworks_> :)
[12:19:53] <skunkworks_> I talk for those who can't talk for themselves ;)
[12:37:05] <skunkworks_> Heh - I just got an email from someone with a hermes 7000 wondering if he could do the same thing.
[12:37:26] <skunkworks_> (he doesn't have the EP module)
[12:38:37] <skunkworks_> cradek: http://www.cnczone.com/forums/showthread.php?p=340019#post340019
[12:44:50] <skunkworks_> maybe your phase converter isn't puting out a clean enough voltage?
[12:53:49] <cradek> skunkworks_: only Z does it, and only sometimes (always when it's sitting still, not under load)
[12:54:29] <cradek> when I swapped two amps, the problem stayed with Z, so it's not a bad amp, and probably not a power problem (the DC bus is very clean 110 V)
[12:55:34] <cradek> hmm I wonder if the 110 is too high, I found something that said the transformer output should be 70, and 70 sqrt 2 is 99
[12:56:03] <skunkworks_> hmm - could be.
[12:56:26] <cradek> I should check that
[12:56:51] <cradek> when I went through the power check section of troubleshooting, it did not mention that voltage - I will have to dig deeper :-)
[14:42:46] <jmkasunich> are there taps on the transformer to let you adjust the voltage?
[14:42:52] <jmkasunich> (usually on the primary side)
[14:52:03] <skunkworks_> overvoltage error would make sense sitting still - mostlikely the supply is at its highest voltage.
[14:53:07] <cradek> yes there are some taps - it's wired right
[14:53:49] <cradek> maybe my power is a little high or something
[14:54:03] <cradek> (it would be nice to know what the acceptable input range for the amp is)
[15:05:01] <SWPadnos> cradek, you have a rotary phase converter?
[15:05:20] <cradek> yes
[15:05:38] <SWPadnos> have you noticed if you get more faults when the spindle is off?
[15:05:55] <cradek> I've had them both ways
[15:06:06] <cradek> often it's right at startup, I can't get the Z amp to enable
[15:06:16] <cradek> but, it has done it a few times while cutting (but not moving Z)
[15:06:33] <SWPadnos> ok. you may want to check the wild leg of the converter. they're often pretty high - as much as 20% or so
[15:06:34] <SWPadnos> or more
[15:07:09] <cradek> one phase is hooked right to the line and reads 230 or so, the other two phases are low, reading 200-210
[15:07:22] <cradek> I moved all the electronics onto the real phase
[15:07:29] <SWPadnos> ok, that's a good thing
[15:07:41] <cradek> only things on the fake phases are the main DC transformer and spindle motor
[15:08:25] <SWPadnos> does that main DC transformer feed the servo amps?
[15:08:31] <cradek> yes that's all it does
[15:09:05] <SWPadnos> and the wild leg is lower than line? interesting
[15:09:11] <cradek> everything else has its own power supplies running on the one phase - there are about 35 of them sprinkled all over the place
[15:10:21] <cradek> I'm sorely tempted to wire all the primaries on the main transfomer together and try it single-phase
[15:10:34] <cradek> might be a bit more ripple?
[15:10:45] <SWPadnos> seems likely
[15:11:01] <cradek> could it have any other problem (saturation?)
[15:11:12] <SWPadnos> plus, if it's a 3-phase transofrmer, you can't do that (3 separate 1-phase you can obviously)
[15:11:53] <cradek> it has three primaries on the same iron, one on each phase
[15:12:17] <SWPadnos> well, I'm sure it's different, but I'm not sure how :)
[15:12:31] <cradek> http://timeguy.com/cradek-files/emc/ECNC4.tif
[15:14:42] <SWPadnos> err - I don't understand that coil at first glance, and I can't think about it right now ;)
[15:15:10] <cradek> the pairs (X,Y,Z) are strapped parallel
[15:15:42] <cradek> then X(2,4) to Y1, etc etc
[15:18:24] <cradek> hm I don't understand the secondary wiring though
[15:23:43] <skunkworks_> 3 phase bridge rectifier in a delta config
[15:26:25] <cradek> so could I tie together all the (1,3) (2,4) for primary and parallel all the 5 and 7 as secondary (and then use only 4 diodes)?
[15:28:21] <skunkworks_> umm - I think that is a question for jmk.. I don't know how well paralleling transfomers work. you would have to make damn sure they where all in phase.
[15:28:49] <SWPadnos> or out of phase ...
[15:28:55] <SWPadnos> it's a 3-phase core, after all
[15:29:16] <SWPadnos> paralleling all 3 coils off one phase could cause saturation problems
[15:29:34] <SWPadnos> or not. I never did like that transformers and power transmission class
[15:29:40] <cradek> yeah I'm not about to do it without being a lot smarter about it
[15:29:52] <cradek> seems like it sure could melt
[15:30:05] <SWPadnos> it didn't help that the professor reminded me of Alfred E. Newman
[15:31:14] <SWPadnos> or Neuman
[15:33:58] <tomp> that 3 phase 'X' style primary is described on this page http://www.federalpacific.com/university/transbasics/chapter2.html
[15:36:45] <SWPadnos> I think you meant http://www.federalpacific.com/university/transbasics/chapter3.html
[15:37:03] <SWPadnos> but maybe not
[15:40:54] <tomp> i think the tif wants primary as ... each #4 to next #1 (needs to flow) and (for 460) each 3,4 'X' pair shorted together (making longer coils), and.. for 230 it seems to want the 3 sets of primary coils (1-2) (3-4) paralleled, which i wouldnt try cuz i'm a sceerdycat
[15:42:27] <SWPadnos> what doesn't make sense to me is that 2 has no connections for X Y or Z
[15:42:32] <SWPadnos> on that diagram
[15:43:21] <SWPadnos> oh - I was looking at the smaller coil on the right. I guess if you look at T2 in the top center, it's a little different :)
[15:43:32] <SWPadnos> wait, they're both T2
[15:43:51] <cradek> T2-left is the 230/460 version, T2-right is the 208
[15:44:06] <cradek> I have T2-left (I hope)
[15:44:21] <SWPadnos> yeah. t2-right requires more caffeine
[15:45:08] <cradek> http://timeguy.com/cradek-files/emc/ECNC3.tif
[15:45:17] <cradek> here are the jumpering instructions
[15:45:34] <cradek> but this is beside the point, I know how to wire it for 3 phase :-)
[15:45:38] <SWPadnos> heh
[15:46:40] <cradek> if I actually have T2-right, it would explain the slightly high DC bus
[15:47:16] <cradek> I will look for taps 6,7,8 and see if it can be adjusted to 70V on all three like it should be
[15:47:34] <SWPadnos> so it would
[15:48:24] <cradek> % units '70*230*sqrt(2)/208'
[15:48:24] <cradek> Definition: 109.46557
[15:48:27] <cradek> hmmmmmmmmm
[15:48:37] <SWPadnos> heh
[15:49:05] <SWPadnos> or 240 less some dipde drop
[15:49:07] <SWPadnos> diode
[15:49:31] <cradek> the caps are rated 200V, so it's safe, but it might be a little high for the amps.
[15:49:56] <cradek> (if the manuals say the acceptable input range, I haven't found it yet)
[15:52:55] <cradek> rayh would know this in a heartbeat
[15:53:05] <cradek> the xformer wiring I mean
[15:53:07] <SWPadnos> he
[15:53:07] <SWPadnos> h
[15:53:09] <SWPadnos> you could email him
[15:54:11] <cradek> I may bother him later... converting to 1ph is not a priority but it would be nice eventually. I could do it when I do the rest of the conversion (have to get a VFD too)
[15:54:45] <SWPadnos> how many horse is the spindle?
[15:55:13] <cradek> 2
[15:55:26] <SWPadnos> oh, that's not so bad fro VFD buying on eBay
[15:55:28] <SWPadnos> firo
[15:55:30] <SWPadnos> for
[15:55:53] <cradek> automationdirect has them for ~ $150 iirc
[15:56:02] <SWPadnos> oh, that's not bad
[15:56:13] <cradek> rayh said he's happy with them
[15:56:31] <SWPadnos> I think that was the model Matt S was fiddling with last year at CNC workshop
[15:57:39] <cradek> interesting, they have an analog input, I wonder why he wants the serial thing
[15:57:50] <SWPadnos> because he has no analog output?
[15:57:55] <cradek> sure
[15:58:32] <SWPadnos> it's that magic 4 number - if you want XYZA, you need another 7i33 (or whatever)
[15:58:43] <SWPadnos> for the spindle
[15:58:56] <cradek> right
[15:59:34] <cradek> if they can be configured for 0-5, parport pwm is plenty fast
[15:59:47] <cradek> but 0-10 or +-10 is harder to come by
[15:59:54] <SWPadnos> man - it's amazing how expensive a pair of 2-56 screws is. $1.80 for 2 screws and hopefully two nuts, to bolt a 50-pin header to the baord
[16:00:14] <SWPadnos> I imagine you can get feedback as well if you do it serially
[16:00:38] <SWPadnos> so no encoder needed - just an index pulse for rotational sync
[16:00:47] <cradek> neat
[20:49:58] <jmkasunich> hi cradek
[20:50:12] <cradek> hi jmk
[20:50:24] <jmkasunich> just lookin at your transformer drawing
[20:50:30] <cradek> ah
[20:50:40] <jmkasunich> unfortunately you probably can't run it on single phase
[20:51:14] <jmkasunich> you need to look at the physical construction - if it has a three leg core with three sets of windings, one set per leg, it is three phase only
[20:51:34] <jmkasunich> thanks to a neat trick that lets them make cheaper transformers
[20:51:41] <cradek> less iron required?
[20:51:46] <jmkasunich> yeah
[20:52:01] <cradek> I was afraid of something like that
[20:52:04] <cradek> oh well, no problem
[20:52:25] <cradek> it does look like III but I don't know the exact configuration of the laminates
[20:52:41] <jmkasunich> if there are only three legs, thats it
[20:53:01] <jmkasunich> the lams would look like uppercase E, laying on their side
[20:53:40] <cradek> ok, I just need a different transformer then, no problem
[20:54:05] <cradek> I'll just keep an eye out for something, the converter works fine the way it is
[20:54:11] <jmkasunich> and maybe different or bigger filter caps
[20:54:22] <jmkasunich> single phase probably triples the ripple current in the caps
[20:54:48] <jmkasunich> until you get a VFD for the spindle, I'd leave it along
[20:54:50] <jmkasunich> e
[20:55:03] <cradek> right that was my plan too
[20:55:15] <cradek> little point in doing just one or the other
[20:56:05] <cradek> thanks for the advice
[20:56:15] <cradek> "don't screw with it" heh
[20:56:23] <jmkasunich> how far apart are the taps? 10% or so?
[20:56:47] <jmkasunich> it might be a good idea to switch to the next lower set of taps and see what that does to your faults
[20:56:48] <cradek> on the secondary? I haven't even looked for them
[20:57:17] <cradek> the way it is set, the three output windings measured 70, 70, 80? and they're supposed to be 70
[20:57:23] <jmkasunich> once you find them ;-) put a voltmeter from 5-6, 5-7, and 5-8, and see what you get
[20:57:30] <cradek> so if I can knock that high one down a bit I will
[20:57:35] <jmkasunich> 80 eh? maybe you are getting high bus trips
[20:57:46] <jmkasunich> I would NOT recommend changing just one tap
[20:58:08] <cradek> well the bus is 110-111 while sitting still and the amps are supposed to take 122
[20:58:19] <skunkworks_> cradek: looks like you can error on the low side.
[20:59:28] <cradek> jmkasunich: why would that be bad? they just go to diodes...?
[20:59:58] <cradek> I figured it would be worse (at least hard on two diodes) to have the output unbalanced
[21:00:07] <jmkasunich> in general line unbalance is a bad thing, and I'd never to anything to intentionally unbalance something
[21:00:20] <jmkasunich> that may be a rule-of-thumb that doesn't apply here
[21:00:39] <jmkasunich> because you are starting unbalanced, you may be right
[21:00:44] <cradek> well the imbalance is in the input power - the question is how far that should be propogated
[21:01:21] <jmkasunich> it all depends on how far apart the taps are
[21:01:32] <cradek> yeah I will try to figure that out
[21:01:40] <jmkasunich> if you can change 70 70 80 to 70 70 74, thats good
[21:01:50] <jmkasunich> but you don't want to make it 70 70 68
[21:02:08] <jmkasunich> as it is, the high phase will get the highest load, which will tend to drag it down and balance things out
[21:02:47] <jmkasunich> if you change it too far, the load will be on the phases that are already low, and they'll just get lower under load
[21:03:33] <cradek> if it's not obvious what to do, I'll come back and ask for more advice!
[21:03:50] <jmkasunich> do you know if your phase converter has balancing caps?
[21:04:15] <cradek> I don't think it does
[21:04:50] <jmkasunich> it is possible to balance things out nicely with some caps
[21:06:32] <cradek> gotta run, bbl
[21:06:34] <cradek> and thanks again
[23:32:31] <cradek> jmkasunich: hmm, it seemed like a good idea...
[23:32:42] <jmkasunich> changing taps?
[23:32:45] <cradek> yep
[23:32:56] <cradek> turns out I have the 208V transformer
[23:33:06] <jmkasunich> hmm
[23:33:12] <cradek> but I thought that's cool, because two of my phases are 208 anyway
[23:33:33] <cradek> I moved one of the secondary taps down, and I got a nice balanced 70/70/70 output
[23:33:41] <jmkasunich> cool
[23:33:45] <cradek> DC bus goes down to 98 with the amps on (it was 111)
[23:33:52] <cradek> I found in the book it's supposed to be 100
[23:34:04] <cradek> so I cut some stuff for a while, but ... what's that shellac smell??
[23:34:18] <jmkasunich> you are probably overfluxing the transformer just a bit, but unless you notice significant heating with no load you might be ok
[23:34:34] <cradek> it was significant...
[23:34:42] <jmkasunich> ouch
[23:34:45] <cradek> I hope I didn't melt it
[23:34:53] <cradek> I've got a fan blowing on it now
[23:34:59] <jmkasunich> the only thing you changed was one secondary tap?
[23:35:03] <cradek> yes
[23:35:12] <jmkasunich> and you've run it at least this long before?
[23:35:17] <cradek> oh yeah
[23:35:29] <jmkasunich> I'm surprised about that
[23:35:49] <jmkasunich> in spite of my comments earlier, I expected that a 70/70/70 arrangement would be good
[23:36:06] <cradek> it sure seemed promising
[23:36:15] <jmkasunich> I'm more concerned about putting 240V into a 208 primary, but you've done that before the change with no probs
[23:37:09] <cradek> I'm not entirely sure it's the 208 transformer, but it sure all makes sense then
[23:37:08] <jmkasunich> how physically accessible is the transformer? can you feel it? it would be interesting to know if the hottest part is the core, or all three coils, or just one or two coils
[23:37:25] <cradek> one coil seemed hotter than the others, but all 3 were hot
[23:38:07] <cradek> I think it's kind of hard to say - I let it heat way too long to tell the source
[23:38:34] <cradek> maybe I should try moving all the taps down so it's the same as before but with lower output
[23:39:13] <jmkasunich> 240 into a 208 xfmr is going to cause excessive flux and thus excessive core loss regardless of where the secondary side taps are
[23:40:14] <jmkasunich> how sure are you of the 208 vs. 240 thing? looks like the 240/480V transformer has 4 leads per primary winding, and the 208 one only has 2
[23:40:44] <cradek> it matches the 208, even the number labels on the secondary (the numbers on the taps are different)
[23:41:11] <jmkasunich> in that case, I'd let it run _unloaded_ for a while
[23:41:15] <cradek> and the measured ratio (voltage) is 208/70, not 240/70
[23:41:24] <jmkasunich> if you get excessive heating in that case, you have a problem
[23:41:35] <jmkasunich> if not, then dropping the secondary taps is probably the way to go
[23:41:40] <cradek> wired the old way you mean?
[23:41:53] <jmkasunich> the primary has never been changed, right?
[23:41:56] <cradek> right
[23:42:40] <jmkasunich> you could disconnect the secondary - leave terminal 3 of each winding connected, but remove disconnect terminal 5
[23:42:47] <jmkasunich> (or whatever they are called on yours)
[23:43:02] <cradek> they do have those markings
[23:43:36] <jmkasunich> "even the number labels on the secondary (the numbers on the taps are different)"
[23:43:57] <jmkasunich> the taps are on the secondary, so I'm a bit confused
[23:44:00] <cradek> different on the 240 vs 208
[23:44:08] <jmkasunich> oh, I see
[23:44:17] <jmkasunich> 3456 vs 5678
[23:44:20] <cradek> I meant that's one of the reasons I think it's the 208
[23:44:20] <cradek> right
[23:44:35] <cradek> these say 3456
[23:44:41] <jmkasunich> the absence of the 2 and 3 pins on the primary is even more significant
[23:44:58] <cradek> let me go look again
[23:46:18] <cradek> pri is 111222
[23:46:34] <jmkasunich> sounds like the 208 then
[23:46:55] <cradek> hmm, all the more reason to get a new transformer
[23:47:04] <cradek> * cradek grumbles about the IPO
[23:47:19] <cradek> someone went through and strapped everything else for 230
[23:47:38] <jmkasunich> IPO?
[23:47:41] <cradek> except for the big transformer the size of his head, at which he just shrugged and got another beer
[23:47:45] <cradek> idiot previous owner
[23:48:11] <jmkasunich> he probably had a balanced line
[23:48:21] <jmkasunich> and if it was a bit low, the 208 transformer would work OK
[23:48:57] <jmkasunich> lots of equipment does work unchanged on both 208 and 240 V
[23:49:25] <cradek> I guess this must have
[23:50:00] <cradek> 114v instead of 100v on the drives, still below the 122 spec