#linuxcnc | Logs for 2013-04-28

Back
[00:13:57] <c0ldfront> anyone using cncPlanet usb driver?
[00:24:25] <ssi> anyone know how I can invert my axis enable signals in hal?
[00:28:28] <ssi> not component oughta do it I suppose
[00:29:01] <KimK> Yes indeed, "man not".
[00:30:30] <ssi> I'm still trying to get to a point where i can make the motor work
[00:30:36] <ssi> drive enable seemed to be important somehow :D
[00:32:15] <KimK> You could also just tie enable wherever it needs to go, and then do your testing.
[00:32:24] <KimK> Just for now.
[00:32:31] <ssi> well what I mean is
[00:32:34] <ssi> the motor's running constantly
[00:32:41] <ssi> and when I try to enable axis, it just instantly faults
[00:32:51] <ssi> although I guess that's more an issue of needing to set a very high ferror or sometching
[00:33:03] <ssi> anyway I should have valid enables here in just a second
[00:34:10] <KimK> The running "away" (sort of) makes sense if you're still in torque mode and there's no torque mode in any of the sample configs (and I'm assuming not).
[00:34:40] <ssi> I don't have the foggiest idea what is different about the configs for torque mode
[00:35:43] <ssi> oh wow I'm going to all this not trouble
[00:35:50] <ssi> but there's an analogena-invert parameter
[00:37:06] <ssi> there we go! drive enables work :D
[00:39:10] <ssi> now then
[00:39:12] <ssi> torque mode configs
[00:39:26] <ssi> let's see what google has to say
[00:39:48] <KimK> Hey great, congrats! And the difference is that normally the drives are in velocity mode and the tach-gens close the velocity loop to the drives, Linuxcnc only closes the outer (position) loop. You might choose to have Linuxcnc close the middle (velocity) loop also.
[00:40:04] <ssi> yeah, I mostly get that in theory
[00:40:13] <ssi> but I don't really understand it from a practical perspective
[00:40:48] <ssi> so, in a velocity mode drive, the drive attempts to maintain the motor at a given velocity
[00:40:56] <ssi> so a 0v ref signal would attempt to maintain 0rpm
[00:41:07] <ssi> and a +10v ref signal would attempt to maintain some max rpm
[00:41:31] <ssi> and it uses the tachgen as feedback to increase or decrease motor voltage (or maybe current?) to attempt to maintain the speed
[00:42:12] <KimK> Yes, 0v = 0 rpm, which means the drive could be moved, but only slowly, against velocity resistance.
[00:42:15] <ssi> then linuxcnc closes the position loop by varying commanded velocity to try to keep the axis in position
[00:42:50] <KimK> Yes, exactly, you've got it.
[00:42:51] <ssi> so if I have this drive in current mode, does that mean my ref signal is a commanded current?
[00:43:29] <KimK> Yes, the reference/command signal is then a current command.
[00:43:54] <ssi> so how does that change the configuration?
[00:44:04] <ssi> I'm guessing the signal wiring is the same, but the PID tuning is different
[00:44:17] <KimK> And the feedback is the sense resistor. And the +/- 10V is proportional to whatever the current gain is scaled to.
[00:45:04] <KimK> So if your drives have a current resistor sense voltage test point, that would be useful during setup.
[00:45:18] <KimK> Or wou could use an ammeter of some kind.
[00:45:25] <KimK> s/wou/you/
[00:45:39] <ssi> Current Monitor. Analog output signal proportional to the actual current output. Scaling is 2.3 A/V. Measure relative to signal ground.
[00:45:42] <ssi> how's that?
[00:46:10] <KimK> Hey, it's just all falling into place for you.
[00:46:12] <ssi> there's also this one:
[00:46:13] <ssi> Measures the command signal to the internal current-loop. This pin has a maximum output of ±7.25 V when the drive outputs maximum peak current. Measure relative to signal ground.
[00:46:27] <ssi> it's *trickling* into place :)
[00:47:57] <KimK> You'll want to set the max current (apparently at +/- 7.5 V ?) to some maximum value according to your motor data sheets.
[00:48:29] <ssi> is that max continuous? or max peak
[00:49:15] <KimK> Ha, probably depends on whether it was written by engineering or marketing. The data sheet might give a clue.
[00:50:41] <ssi> ok this is the datasheet I have:
[00:50:42] <ssi> http://cncpd.com/pdf/SEM_MTS30_Chart.pdf
[00:50:46] <ssi> and it's an F4-52
[00:50:54] <ssi> so 52V/krpm
[00:51:09] <ssi> (which means i'll only be able to get it up just shy of 1000rpm with my paltry 48v supply that I have on hand
[00:51:24] <ssi> stall current continuous 3.4A
[00:51:48] <ssi> I'm assuming that's the number I'm shooting for
[00:52:47] <KimK> Sounds good, you can adjust it more later if there's a need. Also, you might want to set it "quickly" by watching the current sense voltage on a scope and pulsing the command voltage to take a reading. Rather than just letting everything heat up unnecessarily.
[00:53:20] <ssi> I may be able to do that
[00:53:35] <KimK> And you'll have to block the motor to get a stall.
[00:53:40] <ssi> hm
[00:53:54] <ssi> I wonder if there's a hole in the armature to allow for that
[00:54:11] <ssi> http://www.a-m-c.com/download/support/an-015.pdf
[00:54:13] <ssi> I might oughta read that
[00:56:57] <ssi> maybe just a machinists clamp on the shaft would do the trick
[01:01:05] <ssi> working on setting the scope up now
[01:08:08] <ssi> ok servo is blocked
[01:08:12] <ssi> scope is running
[01:08:30] <ssi> I gather what I need to do is send a low-freq square wave that's 0 to 10V
[01:08:41] <ssi> and trace both the command signal and the current sense
[01:08:57] <Gamma-X> kinda wackjy to just turn a driver on
[01:09:16] <KimK> If you're unable to get a 3.4A reading due to the lower than normal power supply voltage, you'll just have to command less current so you don't run out of headroom.
[01:10:16] <KimK> You don't have to use a signal generator or anything, you can just "tap" the connection when you're ready to read.
[01:10:48] <KimK> s/the connection/the selected command voltage/
[01:11:30] <ssi> I have a signal gen, fwiw
[01:11:35] <ssi> not sure if that's easier
[01:12:38] <KimK> OK, that's good, if it can generate "occasional" pulses (low duty cycle) that would be nice. Anything faster than visual would be OK.
[01:13:17] <KimK> And then if you can set the upper voltage, you've got it made.
[01:14:13] <ssi> there's a LOT of noise
[01:14:39] <ssi> it's the 48v switcher I'm using I'm pretty sure
[01:15:36] <KimK> It may be the drives themselves, they're probably switching too. I'll be off for 20-30 minutes or so, then back on. Anything before I go?
[01:16:53] <ssi> nothing I can think of
[01:17:00] <ssi> just trying to figure out if this is even feasible
[01:17:06] <ssi> the noise is blanking almost any reading I might could get
[01:17:43] <ssi> maybe not, maybe this will work
[01:17:44] <ssi> hmmmm
[01:17:45] <ssi> :D
[01:18:43] <Gamma-X> I think that means I have 2 instances of iirc runnin on mylaptop!
[01:19:05] <ssi> I wish the frontpanel of my scope wasn't dying :/
[01:19:14] <Gamma-X> anyone have a mesa card setup they wanna sell for servos?
[01:19:44] <KimK> What kind are you looking for?
[01:20:09] <KimK> Or which models, I mean.
[01:20:38] <Gamma-X> Id love the 6i25 plug and go kit...
[01:20:44] <Gamma-X> just short on cash...
[01:21:51] <KimK> Yes, lots of DIY folks seem to like the "25" series.
[01:22:03] <ssi> this is my first time using it, and I like it a lot
[01:22:09] <ssi> I got the 5i25/7i77 kit
[01:22:20] <ssi> my lathes both use the older style hardware
[01:22:27] <ssi> one's a 5i23, the other a 7i43
[01:22:39] <Gamma-X> KimK you have any ones for sale? :D
[01:22:43] <ssi> I also have a spare 7i43 and 7i33 lying around ;)
[01:24:18] <Gamma-X> ssi.... I dont suppose youd be willing to part with them?! :D
[01:24:35] <KimK> No, not me, in fact, I was wondering if there was any interest in a group buy on the 7i22.
[01:25:49] <KimK> OK, now I'm going, back on in a bit.
[01:26:39] <ssi> you're probably better off just buying the 6i25 kit you want
[01:27:17] <Gamma-X> I keep spending money.... i need to bring some in before I can buy anything...
[01:35:50] <Gamma-X> I have a 3" northfield air chuck that I have no use for but I dont have anything else on the lathe to hold the work piece...
[01:36:44] <ssi> on what kind of lathe?
[01:36:50] <Gamma-X> hardinge hc
[01:37:00] <ssi> threaded nose spindle or taper?
[01:37:11] <Gamma-X> I need like a hc collet closer...
[01:37:15] <Gamma-X> threadednose
[01:37:25] <Gamma-X> machine didnt come with one.
[01:37:36] <ssi> what's the chuck worth to you?
[01:37:59] <Gamma-X> nothing
[01:38:12] <Gamma-X> in reality in terms of making parts right now lol
[01:38:17] <ssi> well shoot, I'd buy it for that price ;)
[01:39:29] <Gamma-X> I need some kind of work holder.... id love to get a pneumatic 5c closer of some type where I can set it up to auto pull the working rod through the spindle and make more parts over and over.
[01:39:58] <ssi> got any pics of the chuck?
[01:40:01] <ssi> I wanta chuck for my hnc
[01:40:04] <Gamma-X> uhhhh
[01:40:16] <Gamma-X> not on me right now.
[01:40:36] <Gamma-X> its got everything with it...
[01:40:39] <ssi> I think I might have blown a fuse in my 48v supply :/
[01:42:02] <ssi> yeah, I definitely did :/
[01:47:23] <ssi> KimK: so I don't really understand what I'm looking for, or what I should be tuning
[01:47:27] <Rob__> ssi, sorry had some trouble
[01:47:29] <ssi> but I did manage to blow the fuse in my supply
[01:47:43] <Rob__> ssi, out of curiosity what is it worth to you?
[01:47:58] <ssi> well I'd want to see some pics
[01:48:23] <ssi> 3" sounds pretty small, but if it's useful to me, I'd send you this mesa card and servo daughtecard
[01:48:53] <ssi> it's a 7i43, which is a parport card, dunno if t hat's useful to you
[01:49:00] <ssi> and a 7i33 servo daughtercard
[01:49:13] <ssi> lemme verify that's what I have
[01:49:13] <ssi> heh
[01:49:29] <Rob__> cant do that... I need some kinda of fixture to hold the work like this, http://www.ebay.com/itm/Royal-5C-Rotary-Station-Pneumatic-Collet-Closer-62020-/280641992382?pt=BI_Tool_Work_Holding&hash=item415790f6be
[01:49:32] <Rob__> prefferably at least...
[01:50:00] <Rob__> or some sort of 5c chuck
[01:50:08] <ssi> I don't have anything for you
[01:50:19] <ssi> I have the pneumatic 5c closer built into the hnc, but you certainly can't have mine ;)
[01:50:35] <GammaX> I wish! lol
[01:51:22] <ssi> blah
[01:51:31] <ssi> not having tachs in these servos is really ruining my day :(
[01:52:23] <GammaX> not having a pneumatic 5c closer is ruining myne!
[01:53:38] <KimK_3> OK, I'm back (sort of). How's it going?
[01:54:08] <ssi> KimK: so I was able to pulse the drive input, and observe a signal on the current monitor pin
[01:54:21] <ssi> I turned the frequency down pretty low,
[01:54:26] <ssi> and popped the fuse on my supply
[01:54:33] <ssi> I really mostly have no idea what I'm doing hahah
[01:54:50] <ssi> I don't know how to interpret what I'm seeing, or what I should be tuning
[01:55:56] <KimK_3> That (...able to...observe a signal...) sounds good, too bad about the fuse.
[01:56:03] <ssi> I just replaced it
[01:56:08] <ssi> I may be able to try again
[01:56:15] <ssi> but I want to have a better idea of what I'm doing
[01:57:38] <KimK_3> You can also just tap or touch a connection when you're ready to drive the current to whatever max value you've arranged. What did you end up with for a maximum with your reduced voltage, anyway?
[01:58:01] <ssi> I don't really know what the max should be
[01:58:14] <ssi> I was driving the ref signal with a square wave of 9.8v peak to peak
[01:58:23] <KimK_3> s/maximum with/maximum current with/
[02:01:01] <KimK_3> OK, that (50% "on") was probably too much. I was thinking more like a 0.1% or less on "square" wave.
[02:01:23] <KimK_3> Although I didn't ask what size fuse?
[02:01:41] <DJ9DJ> moin
[02:02:17] <KimK_3> Hi back
[02:02:22] <ssi> I can reduce the duty
[02:02:26] <ssi> here's the weird thing
[02:02:33] <ssi> I just did it again,
[02:02:40] <ssi> if I reduce the amplitude of the square wave,
[02:02:48] <ssi> the amplitude of the current monitor waveform doesn't change
[02:02:55] <ssi> I would expect it to, yes?
[02:04:58] <GammaX> would this fit on a lathe or is it for just 4th and 5th axis on a mill? http://www.use-enco.com/1/1/34102-a1-212-5c-eagle-rock-5c-air-collet-5c-non-indexing-collet-fixtures.html
[02:05:43] <ssi> Gamma-X: I dunno how you'd spin those air fittings :P
[02:06:49] <GammaX> this sucks
[02:06:58] <KimK_3> Since you had planned to have 140v (call it 150V) pushing the 3.4a, and since you had to use the 48v (call it 50v) instead, I'd expect your max current to be about 1/3 of 3.4a, unless I've overlooked something. Yes, if you reduce the amplitude of the command signal, the armature current should decrease.
[02:07:09] <KimK_3> Since you had planned to have 140v (call it 150V) pushing the 3.4a, and since you had to use the 48v (call it 50v) instead, I'd expect your max current to be about 1/3 of 3.4a, unless I've overlooked something. Yes, if you reduce the amplitude of the command signal, the armature current should decrease.
[02:07:27] <KimK_3> Sorry, I don't know what happened there.
[02:07:38] <GammaX> ssi, ill trade that air chuck for a hc closer and 5c collet set?
[02:09:25] <ssi> GammaX: I don't have anything for you, sorry
[02:11:39] <KimK_3> They do seem to make a distinction between lathe collets and collet fixtures: http://www.use-enco.com/CGI/INSRHI?PMSECT=0000001009 plus ssi's point about the air fittings, so I'd guess those are just for holding round parts on a mill.
[02:11:42] <ssi> KimK: I'm seeing like 3v on the current monitor pin, which according to the datasheet means 7A or so
[02:11:47] <ssi> which seems very high
[02:12:25] <ssi> I'm varying the amplitude of the input signal smoothly, but the current monitor stays about the same, until i get all the way down to the minimum voltage on my signal gen
[02:12:31] <ssi> and then the current monitor drops to 0.5v
[02:12:39] <ssi> I don't understand that at all
[02:12:47] <KimK_3> Well, you could try again while letting the motor spin, that should reduce the current demand.
[02:13:06] <GammaX> is there a 3 or 4 jaw chuck that will close all 4 jaws when turning 1 screw?
[02:13:26] <KimK_3> Yes, there are both. For a lathe, you mean?
[02:13:34] <GammaX> yes
[02:14:23] <GammaX> KimK_3, can I ask for a reference on per say... ebay?
[02:14:28] <KimK_3> But four jaws are usually independent, for rework. But (rarely) they are available for producing new parts from square stock.
[02:15:16] <KimK_3> There are also 6-jaw (scroll) chucks, great on hex stock.
[02:15:36] <KimK_3> Sure, but what would you like to see from eBay?
[02:15:46] <GammaX> hopefully ill be doing only round work in the begining
[02:16:07] <GammaX> a chuck that would close all 3 jaws with the turn of one bolt
[02:16:11] <KimK_3> OK, then the classic 3 jaw should hold you for awhile.
[02:16:22] <GammaX> hmmm ok
[02:16:32] <KimK_3> What kind of spindle nose does your lathe have?
[02:17:11] <KimK_3> The back of the chuck has to fit your lathe. Or at least an adapter or faceplate that fits your lathe.
[02:17:26] <ssi> ok so I think it's doing what I expect, it's just too sensitive to input voltage
[02:18:03] <KimK_3> If there's a current scaling, you could try tuning it down.
[02:18:12] <KimK_3> s/tuning/turning/
[02:18:18] <GammaX> KimK_3, I have a backing plate for it. Its at the shop. ...
[02:20:04] <KimK_3> Gamma-X: OK, so you can get some kind of universal flat-back chuck and bolt it to the plate, or (better) you can get a chuck that fits your lathe like the backing plate does.
[02:20:26] <ssi> you could get the CDCO 5C chuck, it's relatively cheap
[02:20:34] <Tecan> these things are awesome for linux http://www.ebay.ca/itm/USB-6-Channel-5-1-External-Audio-Sound-Card-Adapter-S-PDIF-US-Laptop-Netbook-PC-/200833972608?pt=US_Sound_Card_External&hash=item2ec2a33980
[02:20:34] <ssi> I bought one and machined a backplate for my G0602
[02:21:33] <KimK_3> Gamma-X: What kind of spindle nose does your lathe have?
[02:21:56] <GammaX> KimK_3, oh god.... uhhh
[02:22:01] <GammaX> threaded nose? lol
[02:25:17] <KimK_3> OK, well, what kind of lathe is it?
[02:26:56] <GammaX> hardinge hc
[02:28:52] <ssi> KimK_3: ok I think I'm where I expect to be now
[02:29:06] <ssi> and his hardinge is something like 1-3/8x10
[02:29:09] <KimK_3> Gamma-X: Like this? http://www.mscdirect.com/product/85029908
[02:29:24] <ssi> 2-3/16x10 maybe
[02:30:29] <ssi> more like this:
[02:30:30] <ssi> http://www.ebay.com/itm/6-3-142-MM-B-C-Flat-Back-Backplate-Hardinge-2-3-16-10-Thread-/190748406309?ViewItem=&item=190748406309&nma=true&si=nm892M4vISdMOjzR8S5C7ZPUnuI%253D&orig_cvip=true&rt=nc&_trksid=p2047675.l2557
[02:30:58] <ssi> GammaX: http://www.cdcotools.com/index.php
[02:30:58] <GammaX> ssi, correct
[02:31:00] <ssi> ergh
[02:31:11] <ssi> top right
[02:31:15] <ssi> $140 5c chuck
[02:31:21] <ssi> mount it on a backplate and you're good to go
[02:31:40] <ssi> KimK_3: so now I'm in a place where varying the input voltage varies the current
[02:32:16] <GammaX> hmmmm thats enticing
[02:34:18] <KimK_3> OK, sounds good. Now you need to reconfigure hal to get what you want.
[02:34:32] <ssi> I set the current limit on the drive so at max input I get right about 1.3V
[02:34:41] <ssi> which is ~3A
[02:34:55] <ssi> ok so how do I go about configuring hal for this?
[02:38:23] <KimK_3> Take a look at the diagrams on page 1 (what you have now) and page 2 (what you want): http://www.motiontech.com.au/assets/pdf/Kollmorgen%20Tuning%20Position%20and%20Velocity%20Loops%20260712.pdf
[02:39:23] <KimK_3> You're taking the velocity loop which was external to HAL (and HAL was "unaware" of it) to internal to HAL, so HAL must be very aware of it.
[02:40:57] <GammaX> ssi, Im debating on trying to get this to work... http://www.ebay.com/itm/5c-collet-closer-tube-/261205868270?pt=LH_DefaultDomain_0&hash=item3cd1153eee
[02:40:59] <ssi> I don't really know how to translate that into signal netting
[02:41:17] <KimK_3> Instead of a velocity output to some external I/O, you'll have a torque/current output. The velocity output/input will be in the middle of HAL.
[02:41:19] <ssi> GammaX: might not be a bad place to start
[02:41:51] <ssi> KimK_3: ok I think i'm starting to fathom a bit
[02:42:00] <ssi> I already have a pid component that turns position into velocity
[02:42:09] <ssi> I'm going to need another pid component that turns that velocity signal into a current signal
[02:42:13] <GammaX> just wish I had the push pull setup on all other kinds of lathes...
[02:44:39] <KimK_3> ssi: yes, that's right
[02:46:37] <ssi> hooboy
[02:46:43] <KimK_3> Gamma-X: Are you working small jobs on small parts? Maybe you want a lever-operated collet closer? (I think they're actually collet openers, but nevermind).
[02:46:45] <ssi> this is gonna stretch my knowledge of hal a bit
[02:47:07] <ssi> also I wish I could get this machine on the network so I could install a proper vi
[02:48:18] <KimK_3> Anytime you add a component, you add it to the functions list, then add it to the desired thread, then connect it as needed.
[02:48:32] <ssi> yea I have that much at least :D
[02:48:40] <ssi> PID isn't my strong suit tho
[02:48:56] <KimK_3> You can download debs from another machine and move by flash drive maybe?
[02:49:05] <ssi> yea that'd
[02:49:06] <ssi> work
[02:49:17] <ssi> ok so do I actually want two pid components per axis?
[02:49:23] <ssi> right now I have a pid.x for instance
[02:49:33] <ssi> am I gonna want a pid.vel.x, pid.tq.x?
[02:49:52] <KimK_3> Let me look...
[02:50:39] <Loetmichel> moin
[02:50:48] <Loetmichel> aehm... mornin_'
[02:54:36] <KimK_3> I think these are the other two articles in that series: http://www.motiontech.com.au/assets/pdf/Kollmorgen%20Tuning%20PID%20Position%20Loops%20260712.pdf http://www.motiontech.com.au/assets/pdf/Kollmorgen%20Tune-up%20Feed-forward%20in%20Position%20Loops%20260712.pdf
[02:56:11] <GammaX> KimK_3, I plan on producing many of 1...
[02:57:30] <GammaX> I would love to buy a pneumatic 5c closer but there 2 expensive...
[02:58:02] <ssi> http://permalink.gmane.org/gmane.linux.distributions.emc.user/34837
[02:58:09] <ssi> has some info from PCW about torque mode
[02:59:41] <ssi> unfortunately a lot of it is lost on me
[03:04:30] <KimK_3> It looks like two PIDs are needed, as far as I can tell. And the other point he was making is that the velocity/current output loop may have to run at a higher speed than the usual 1ms servo loop. So it's easy enough to create a second thread running a little faster. Some experimentation and latency testing may be needed.
[03:05:29] <KimK_3> s/may be needed/may be needed to learn exactly how much faster the second loop has to run/
[03:08:24] <KimK_3> I'd guess if it was too slow that it would have to increase more than twice, so maybe 4 times faster(?) that might be enough. But if it was me, I'd let that go and just do it all in the same servo thread until I had it working and was on to "fine-tuning" (nitpicking?).
[03:10:45] <ssi> right now I'm trying to just trace through the hal and see if i can understand how the existing single pid is wired
[03:11:37] <ssi> so axis commands a position in axis.0.motor-pos-cmd
[03:11:43] <ssi> and that's wired to the pid command pin
[03:13:51] <ssi> what I don't get is, right now the x-output signal is showing a value of 4.7
[03:14:06] <ssi> I'd expect that to show up as a voltage on the Aout pin of axis 0 on my 7i7
[03:14:07] <ssi> 7i77
[03:14:10] <ssi> and I get zero voltage
[03:15:05] <ssi> hm2_5i25.0.7i77.0.1.analogout0 shows a value 4.732272
[03:15:46] <ssi> I wonder if I need field voltage for that
[03:34:20] <KimK_3> You might try the "names" feature on your PIDs, it would be a lot easier to follow "pid.Xvel.output, pid.Ypos.output, etc." as compared to the usual "pid.0.output, pid.1.output, etc."
[03:38:24] <ssi> yeah UI'm already using names
[03:38:44] <ssi> right now I'm just trying to figure out why a value on the analog out signal of the 7i77 doesn't actually manifest a voltage
[03:38:53] <ssi> I'm pretty sure I need to apply field voltage, but I did that
[03:38:55] <ssi> and still no good
[03:39:00] <ssi> anymay I think now it's time for bed
[03:40:21] <KimK_3> Me too, we can chat more tomorrow maybe. Goodnight.
[03:41:12] <ssi> thanks for the help :)
[03:42:44] <KimK_3> Sure thing.
[05:00:40] <Bojangles> Hey any .au cncer's want to cut bootleg frames for me?
[05:01:04] <Bojangles> 0.06" g10 or there abouts
[05:06:36] <chatch> omg this is the bestest channel
[05:07:03] <Bojangles> hay
[05:07:13] <Bojangles> chatch: can you cnc something for me?
[05:07:17] <Bojangles> do you live in aus?
[05:07:51] <chatch> i r florida
[05:08:06] <chatch> almost aus but without the llamas
[05:08:13] <Bojangles> useless
[05:28:33] <chatch> cack
[05:59:01] <Tuipveus> less is more
[05:59:18] <Tuipveus> useless -> use less
[06:22:25] <DJ9DJ> lol
[06:22:35] <DJ9DJ> thats more or less true ;)
[06:34:07] <AITEK203> hello
[06:35:42] <Tuipveus> but with "less", true is endless, see it yourself writing: /bin/true |less
[06:35:46] <Tuipveus> and you will see end
[06:36:32] <Tuipveus> but with "more", true is nothing more, it is even less if you use "more"
[06:36:47] <Tuipveus> just try it your self, write "/bin/true |more"
[06:38:40] <Tuipveus> so with less, you will see "true" to the bitter end, but with more, you will see even less true
[06:39:44] <DJ9DJ> hehe
[06:40:27] <Tuipveus> and therefore also less is more, since you will see "end" with less, but nothing with more, even more says it is more or less more than less
[07:21:36] <ktchk> Hi which 5axis configuration is good for head model?
[07:40:13] <jthornton> what is a head model?
[07:54:27] <Tom_itx> 3d model of a 5 axis mill head?
[07:59:03] <jthornton> or a bust of someones head?
[07:59:21] <Tom_itx> that came to mind as well
[07:59:37] <Tom_itx> i got all the buttons etc working
[07:59:54] <jthornton> cool
[07:59:59] <Tom_itx> i'm not happy how the timer works though
[08:00:23] <jthornton> why
[08:00:50] <Tom_itx> i'd like to capture actual spindle time and it either resets at M0 or runs thru it
[08:01:54] <jthornton> it is a cycle timer, what were you expecting?
[08:02:09] <jthornton> it resets each time you start it
[08:02:44] <Tom_itx> i can add up the times
[08:03:01] <Tom_itx> but i wanted actual spindle time
[08:03:25] <Tom_itx> without adding them up between M0
[08:04:06] <jthornton> change the component to do what you want, rename it and install it
[08:04:07] <Tom_itx> i'm displaying both right now
[08:05:34] <Tom_itx> it would help to tweak the actual run time displayed in lcnc
[08:05:53] <jthornton> I don't follow
[08:06:21] <Tom_itx> i've never looked but i thought lcnc gave an estimated run time for code
[08:06:24] <Tom_itx> my cad cam does
[08:06:29] <Tom_itx> but it's not that accurate
[08:06:48] <Tom_itx> i can tweak it with tool delays etc to get it accurate
[08:07:20] <Tom_itx> so i can look at the code output and know it's gonna take x hrs to do this job
[08:07:31] <jthornton> no, the number is a total of the rapids and feed moves only
[08:07:53] <Tom_itx> ok
[08:07:55] <jthornton> it does not take into account acceleration or trajectory planning
[08:08:17] <jthornton> the only way to get that number is to run a sim with the exact same settings as your machine
[08:08:27] <Tom_itx> true but some software had fudge factors you can get pretty close with
[08:08:28] <jthornton> and use the cycle timer...
[08:08:56] <Tom_itx> i realize every machine will run differently
[08:09:33] <Tom_itx> it's not something i really need but i thought it would be cool to have
[08:10:08] <Tom_itx> i suppose with the 2 timers you can see how lazy the operator is too
[08:10:17] <Tom_itx> how much time is spent in the M0 stop
[08:12:26] <jthornton> or just count the number of parts made
[08:12:59] <Tom_itx> i just thought it would be a nice tool for a production shop to have
[08:14:04] <jthornton> once they know the cycle time the rest is a couple of calculator clicks away
[08:14:35] <Tom_itx> i was at a place where engineering would come down every so often and do time studies. everybody would slow down when they saw em coming
[08:14:49] <Tom_itx> that's true
[08:15:27] <jthornton> yep, we did the same at the shipyard... nobody move the safety guy is here so lets have a safety meeting
[08:15:56] <Tom_itx> don't get me wrong, i like this. i just always look for ways to do things different
[08:16:25] <jthornton> I understand
[08:17:07] <Tom_itx> i'm not going for high production on my sherline anyway
[08:17:09] <jthornton> you could modify the timer to record run time and stop time and have a reset button if you want as it is a dead simple comp
[08:17:23] <Tom_itx> i do have a few parts i'd like to keep track of time on though
[08:19:58] <Tom_itx> i have been getting an error once in a while when i start it up
[08:20:15] <Tom_itx> i'll have to capture it next time to see just what it is
[08:21:30] <Tom_itx> thanks again for all your help
[08:22:24] <jthornton> a latency error?
[08:24:05] <jthornton> the cycle timer should keep track of the parts but M0 pause would have to be dealt with in a different manner
[08:25:41] <jthornton> the man page includes examples that keep the timer running during a pause
[08:26:46] <jthornton> http://linuxcnc.org/docs/html/man/man9/time.9.html
[08:32:30] <Tom_itx> i set it up that way as well
[08:44:57] <jthornton> I just ran a file with M0 in the middle and the cycle timer keeps timing
[08:51:31] <Tom_itx> same here
[08:51:36] <Tom_itx> i set one up both ways
[09:08:02] <Tom_itx> can sim be modified to test pyvcp add ons?
[09:08:17] <Tom_itx> without breaking it
[10:16:21] <JT-Shop> anyone see where I put the slitting saw arbor?
[10:16:53] <Tom_itx> right where you could easily find it
[10:16:58] <Tom_itx> so you thought
[10:18:05] <rahul> HELLO FRNDS...... :)
[10:18:26] <Guest86291> HELLO FRNDS...... :)
[10:19:08] <Guest86291> can any one help me with linuxcnc HAL settings thx u.. :)
[10:20:08] <Guest86291> i am new to linuxcnc servo system
[10:20:51] <Tom_itx> there are dozens here to help. ask specific questions
[10:21:24] <Guest86291> i just need some who can guide me first step for servos... thx.. :)
[10:21:45] <Tom_itx> not all are awake yet either
[10:21:56] <Guest86291> thx TOM
[10:22:21] <Guest86291> my question is how to i setup hal settings for servo
[10:22:32] <Guest86291> do*
[10:23:28] <Guest86291> i am confused can i setup servo settings without any internal motion cards or cards like mesa5i20 etc.
[10:24:08] <Guest86291> just need some help if you guide me
[10:24:09] <Guest86291> can*
[10:26:56] <Guest86291> ???
[10:27:07] <JT-Shop> do you have a question?
[10:27:47] <Guest86291> can i setup servo settings without any internal motion cards or cards like mesa5i20 etc. in linuxcnc
[10:28:42] <Guest86291> may be my question sound strange but thats why i am here just need someone who can guide me :)
[10:28:44] <JT-Shop> sure but why would you try and run a servo machine on a parallel port?
[10:29:06] <JT-Shop> if you can give us more information on what you want to do it will help
[10:31:42] <Guest86291> thx jt-shop i have just setup 3 -axis stepper motor now i want to use 4th axis to i want my 4th axis as servo spindle with control of spindle rpm so read user manual but its too confusing for me try to setup hal settings but got stuck
[10:32:45] <JT-Shop> what does your spindle use for control inputs?
[10:38:20] <JT-Shop> the user manual is for running LinuxCNC for setting up you refer to the Integrators manual if your looking at the PDF manuals
[10:42:37] <guest12345687645> hello jt-shop i am sorry i have internet prblm 2day i asked you about servo motors
[10:43:01] <JT-Shop> <JT-Shop> what does your spindle use for control inputs?
[10:43:19] <guest12345687645> i was telling you about my setup i
[10:45:52] <guest12345687645> i am sorry my english is not good
[10:45:55] <guest12345687645> i have simple induction motor i give simple power from plug
[10:46:44] <JT-Shop> I thought you had a servo motor...
[10:46:46] <guest12345687645> but i will try my best to tell you evrything
[10:50:43] <JT-Shop> ok we will wait
[10:56:29] <JT-Shop> why does everyone think private chat is the place to ask questions?
[11:00:38] <GammaX-Laptop> mornin all
[11:03:47] <JT-Shop> hi
[11:04:25] <Tuipveus> evening
[11:17:30] <L84Supper> JT-Shop: often it's to not appear ignorant on the topic
[11:17:59] <L84Supper> to shy to ask in public
[11:18:40] <JT-Shop> yea, that makes some sense but your here to ask questions because you don't know the answer yet and to get the best answer you have to ask it in public
[11:19:36] <archivist> ircs noobs do not realise that :)
[11:19:47] <JT-Shop> so how do you steer them back to here after they PM you and not piss them off?
[11:19:51] <L84Supper> the crowd in here tends to be mature, in many other channels you are ridiculed for asking too simple a question
[11:20:04] <Roguish_> Hello again. back after a long absence. I need to control 3 stepper motors (size 11). i have external encoders for position monitoring/control. QUESTION: what is the best MesaNet USB controller system/board(s)??
[11:20:05] <JT-Shop> that makes sense too
[11:20:30] <JT-Shop> none that I know of work over USB
[11:20:50] <Roguish_> does not have to be fast. just accurate
[11:22:10] <Roguish_> Serial?
[11:23:19] <L84Supper> PCIe, PCI or EPP
[11:23:52] <JT-Shop> 5i25 7i76 combo is the best bang for the buck
[11:24:43] <L84Supper> USB has too many issues to be used for real time control
[11:27:05] <Roguish_> I don't need real time control. it is running motors to position optics in an test set up. I would like to run on a laptop. How about a usb/serial converter?
[11:27:33] <L84Supper> USB could be made to work with step rates to maybe 1K/sec, but nobody has bothered to write the driver
[11:28:21] <JT-Shop> how about a usb to parallel port adapter if there is such a thing
[11:29:18] <Roguish_> like: http://www.belkin.com/us/p/P-F5U002
[11:29:44] <JT-Shop> yep it would only cost you $40 to find out if it worked
[11:30:06] <Roguish_> Hee, hee....
[11:30:21] <JT-Shop> but I doubt it will work unless it can pretend to be a parallel port
[11:31:07] <L84Supper> ask PCW about the 7i43 and using it with SoftDMC
[11:31:28] <Roguish_> ok, assume that adapter works and i can get a parallel port. what mesanet then? of course with driver in linuxcnc
[11:31:30] <L84Supper> http://www.mesanet.com/pdf/parallel/7i43ds.pdf is has EPP and USB
[11:32:22] <L84Supper> http://www.mesanet.com/pdf/parallel/7i43man.pdf
[11:32:38] <JT-Shop> if you have a parallel port you just need a breakout board perhaps...
[11:33:10] <Roguish_> 7i43 says it has hostmod2. can this be confirmed?
[11:33:36] <L84Supper> Roguish_: why are you looking for USB? Is there some distance between the PC and the optical table?
[11:34:25] <L84Supper> http://linuxcnc.org/docs/html/drivers/hostmot2.html
[11:34:31] <Roguish_> no. this will be a test setup, probably in a cmm. i need to run on a laptop,
[11:35:00] <L84Supper> it would have to run really slow on a laptop to not miss steps
[11:35:25] <L84Supper> laptops have ACPI and SMI that will freeze the OS during events
[11:36:20] <Roguish_> i have an actuator from harmonicdrive with a 100:1 gearing and a .5mm pitch screw. it will position a mirror to the microradian level.
[11:36:28] <L84Supper> say the battery needs a change in recharge state, the OS gets frozen in time for may 0.5 sec while the EC changes some register in the battery charger
[11:36:47] <Roguish_> high dollar renishaw encoder for nm positioning.
[11:37:30] <L84Supper> Roguish_: I'm doing something similar with a 5-axis stage and a zoom microscope and 1um dia spot laser
[11:37:58] <Roguish_> this is a photon chicane in an FEL.
[11:38:12] <L84Supper> <1 arc second repeatability
[11:39:13] <Roguish_> i can't think in degrees anymore. everything is in radians, i mean mrad or urad or smaller....
[11:39:40] <Roguish_> little numbers with lots of zeroes.
[11:39:46] <L84Supper> ~4 urads
[11:40:07] <Roguish_> yeah, same ballpark.
[11:40:52] <Roguish_> if not mesanet, any other worthy alternatives?
[11:41:06] <L84Supper> laptops aren't a good fit for Linuxcnc
[11:41:29] <JT-Shop> any desktop computer is a worthy alternative
[11:41:31] <L84Supper> wrong model for where the processing has to be with USB
[11:41:44] <Roguish_> I know, but like i mentioned, it doesn't have to be real 'real time'
[11:41:46] <JT-Shop> the micros are not much bigger than a laptop
[11:41:49] <L84Supper> $80 used PC with EPP or PCI/e
[11:42:18] <L84Supper> run the latency test on the laptop
[11:42:45] <L84Supper> it might be 0.1 seconds
[11:42:45] <JT-Shop> didn't someone on the forum cobble up some kind of USB thing after months of hard work?
[11:43:28] <L84Supper> he trying to use the worst possible combination, a laptop and USB
[11:44:03] <JT-Shop> aye
[11:44:35] <JT-Shop> about as useful as this box of slitting saws with 7/16" bore
[11:44:37] <L84Supper> there a bunch or Arduino boards and applications to control steppers
[11:45:18] <Tom_itx> the reprap software might work
[11:45:29] <L84Supper> or use the 7i43 with SoftDMC
[11:46:03] <L84Supper> not sure where he posts any howto's for softDMC
[11:46:31] <JT-Shop> I've not seen anything on softDMC anywhere except the Mesa stie
[11:46:33] <JT-Shop> site
[11:47:01] <L84Supper> http://www.mesanet.com/software/parallel/7i43.zip
[11:47:38] * JT-Shop tires and wanders off to take a nap
[11:49:09] <L84Supper> https://github.com/araisrobo/linuxcnc
[11:49:17] <L84Supper> http://en.araisrobo.com/
[11:51:04] <L84Supper> http://en.araisrobo.com/linuxcnc
[11:51:30] <L84Supper> for servos
[11:51:46] <L84Supper> it's more difficult with steppers
[11:57:24] <Roguish_> hey, you guys have been very helpful. i think i'll contact PCW tomorrow and see what he thinks. (he's just a few miles away.) gotta run. ciao.
[12:10:25] <r00t4rd3d> http://www.amazon.com/Intel-1333MHz-LGA771-Dual-Core-Processor/dp/B000I1OFK2
[12:10:27] <r00t4rd3d> wow
[12:10:38] <r00t4rd3d> You Save: $772.03
[12:15:10] <mhaberler> L84supper: probably you want to investigate first what arais actually have done before coming to the conclusion this is the "wrong model for where the processing has to be with USB" - it is not
[12:17:31] <IchGuckLive> hi all
[12:26:37] <GammaX-Laptop> is there a post for linuxcnc? I think I will be using Cam works.
[12:26:58] <IchGuckLive> standard iso
[12:26:59] <GammaX-Laptop> hey IchGuckLive
[12:27:20] <IchGuckLive> yoiu can take the most posts for linuxcncn
[12:31:51] <IchGuckLive> GammaX-Laptop: cam works has a UPG
[12:32:09] <IchGuckLive> so up to you to define your needs
[12:32:15] <GammaX-Laptop> upg?
[12:32:25] <IchGuckLive> Universal Post Generator
[12:32:35] <GammaX-Laptop> oh nice
[12:32:51] <IchGuckLive> http://www.de.camworks.com/utility/images/add_softwares_02.jpg
[12:33:53] <IchGuckLive> but fanuc11 will fit your need
[12:35:05] <GammaX-Laptop> awsome.
[12:35:41] <IchGuckLive> you have seen dekproto ?
[12:35:47] <GammaX-Laptop> no sir
[12:37:02] <IchGuckLive> what on staying open for a while
[12:37:14] <GammaX-Laptop> ?
[12:37:31] <IchGuckLive> or in 30day option mode on 12 CADCAM systems gives you a year
[12:38:03] <IchGuckLive> GammaX-Laptop: you have seen ma youtube channel
[12:38:38] <GammaX-Laptop> i have no idea wat your talkin about lol
[12:38:52] <IchGuckLive> heekscnc
[12:39:37] <GammaX-Laptop> still lost lol
[12:39:42] <IchGuckLive> http://www.youtube.com/user/magic33de/videos?flow=grid&view=0
[12:39:56] <IchGuckLive> what is your OS
[12:39:57] <rob__H> GammaX-Laptop, i made a camworks post for linuxcnc its on the forum and wiki
[12:40:37] <GammaX-Laptop> rob__H, awsome ill go look for it now
[12:40:44] <GammaX-Laptop> rob__H, how do u like camworks?
[12:40:56] <ReadError> is anyone here strong with composites ?
[12:41:00] <ReadError> (CF)
[12:41:05] <rob__H> based on a fanuc post and modded for linuxcnc, it also suppors subcalls and dataum shift
[12:41:32] <GammaX-Laptop> ReadError, what do you wanna know/do?
[12:41:36] <rob__H> 4 axis i know works but 5axis dont know should do but nowt to try it on
[12:41:37] <ReadError> well
[12:41:44] <ReadError> i was wondering about torsion
[12:41:52] <ReadError> specifically, the direction of the weave
[12:42:09] <rob__H> GammaX-Laptop, http://linuxcnc.org/index.php/english/forum/31-cad-cam/26124-camworks-post-processor-for-linuxcnc
[12:42:10] <ReadError> like the fibers now are # pattern (long and lat)
[12:42:20] <ReadError> if I where to rotate it so the fibers ran at 45 and cut it
[12:42:21] <GammaX-Laptop> ReadError, yeah im deff not that strong lol
[12:42:27] <ReadError> would that provide better torsion resistance ?
[12:43:06] <GammaX-Laptop> although most companies leave te weave it its regular rate unless make some kind of tube...
[12:43:28] <rob__H> GammaX-Laptop, we use to use there tecksoft before camworks/solidworks was born so it was a upgrade path for us , we did look at other cam at the time but didt like things they did or did not do. but now all are farly even in cam terms
[12:44:00] <rob__H> some cam i find is not very production friendly but grate for a1 off quick program.. while others are the other way around
[12:44:14] <rob__H> only you realy know what you want from CAM and what works for you
[12:44:35] <GammaX-Laptop> they make cf light poles with a twist in the lay out of the weave, How its made had an episode on one.
[12:44:53] <GammaX-Laptop> rob__H, I like how its built in to solidworks... just need to get more comfortable wit it.
[12:45:13] <rob__H> does turning too :) at the time cam people said, turning & cam what u want that for lol
[12:45:49] <rob__H> yea why we had it no exporting etc all there and finds model changes etc etc.... HSM was a nice one but now its own by Autocad
[12:46:43] <GammaX-Laptop> rob__H, know of any good training/tutorials for camworks?
[12:46:48] <rob__H> yea camworks has a lot of options when doing the cam.. which i like you have full control of it not locked into its way of doing things so much
[12:46:55] <rob__H> solidprofessor
[12:47:05] <GammaX-Laptop> AHHHH yes
[12:47:13] <ReadError> hsmxpress is nice for solidworks (and free)
[12:47:18] <rob__H> youtube, alot of realy grate vids on there now from users
[12:47:23] <ReadError> very simple to learn
[12:47:45] <IchGuckLive> un the USDA Rhino is a good option 2
[12:49:26] <ReadError> Rhino3d?
[12:49:32] <IchGuckLive> GammaX-Laptop: no student license around to get a professional one for less in 2year trial
[12:49:52] <IchGuckLive> yes rhino3d
[12:51:49] <rob__H> solidcam is other one that plugs in solidworks nicly and new one does look good... but no idea what its like to use
[12:52:07] <rob__H> but all these ones do come with a price tag
[12:52:20] <ReadError> Gamma-X, grab hsmxpress for solidworks
[12:52:24] <ReadError> its *awesome*
[12:52:31] <rob__H> id love to checkout topsolid
[12:53:06] <IchGuckLive> from moldtech
[12:53:41] <IchGuckLive> awaysome on our C40 Hermle moldmaker
[12:54:30] <rob__H> looks like it has some nice mold function for cam etc which we do alot of
[12:54:50] <IchGuckLive> and cheeper then hypermill
[12:57:51] <IchGuckLive> im of by
[13:04:17] <JT-Shop> Roguish_: why not just put a motherboard in your machine and use the telnet remote thing to control your thing
[13:09:30] <GammaX-Laptop> I realy like the looks and functionality of cam works
[13:32:14] <rob__H> GammaX-Laptop, what we like for production we can have many different parts and maximize our tools accross them so not changing tools every part etc.. and not having to edit the posted code todo so
[13:54:14] <GammaX-Laptop> rob__H, you know if a way to easily create a jig from within solidworks for a certain part?
[13:54:49] <r00t4rd3d> http://i.imgur.com/qFGfzlg.jpg
[14:14:36] <toner> nice r00t4rd3d :)
[14:14:39] <toner> very nice
[14:26:59] <ReadError> osh
[14:27:02] <ReadError> psh
[14:27:06] <ReadError> he got that off reddit ;)
[14:33:26] <GammaX-Laptop> where do ou guys commonly shop for cheap tools such as taps, end mills etc...
[14:34:28] <Tom_itx> local
[14:34:43] <roycroft> i buy most of my tooling from enco
[14:35:09] <roycroft> local = 3x as expensive as enco, and the only two local shops won't take returns, even on defective product
[14:35:20] <Tom_itx> if you get good ones and take care of them the cost will average out though
[14:35:26] <roycroft> i am willing to pay more to shop locally, but not that much more, and not with such onerous terms
[14:35:42] <Tom_itx> i figure shipping is part of the tool cost too
[14:35:49] <roycroft> yes, of course
[14:35:50] <ssi> I can't seem to get any voltage on the analog outs of my 7i77 :/
[14:35:54] <ssi> not sure what I'm missing
[14:36:01] <roycroft> although enco almost always have a free shipping offer
[14:37:51] <pcw_home> Probably the analog enable pin is not set
[14:40:21] <ssi> pcw_home: I do have enables working
[14:40:23] <ssi> but I have them inverted
[14:40:25] <ssi> using the parameter
[14:40:36] <ssi> because my drives have a /INHIBIT pin
[14:40:41] <pcw_home> inverted will not work
[14:40:59] <ssi> dang
[14:41:10] <ssi> I guess I can send analogena through a not component
[14:41:16] <ssi> that was my original plan before I found the invert parameter
[14:41:20] <pcw_home> nope
[14:41:30] <ssi> no?
[14:41:31] <ssi> hm
[14:41:32] <pcw_home> you need a relay
[14:42:02] <ssi> why's that?
[14:42:05] <pcw_home> its actually a really bad design to have to make a connection to disable a drive
[14:42:10] <ssi> yeah I agree
[14:42:14] <ssi> but that's the way the drive is set up
[14:42:33] <pcw_home> and OPTOs cannot be normally on
[14:43:02] <pcw_home> so the drive would be enabled if you have a logic power failure
[14:43:18] <pcw_home> or even broke a wire
[14:43:47] <ssi> this is the drive datasheet:
[14:43:47] <ssi> http://www.a-m-c.com/download/datasheet/bx15a20.pdf
[14:43:54] <ssi> maybe I'm missing something about it?
[14:44:49] <ssi> /INHIBIT: TTL level (+5 V) inhibit/enable input. Leave open to enable drive. Pull to ground to inhibit drive. Inhibit turns off all power devices.
[14:45:32] <ssi> does the 7i77 need field power in order for the analog outs to work?
[14:45:52] <pcw_home> Looks like its just TTL level with a 10K pullup so you can do it with a stiff pulldown
[14:46:07] <ssi> 1k stiff enough?
[14:46:16] <ssi> yea should be
[14:46:19] <pcw_home> yeah
[14:46:33] <pcw_home> 470 even
[14:46:51] <ssi> ok
[14:47:03] <pcw_home> then pullup to 5V with enable OPTO
[14:47:11] <ssi> right
[14:47:22] <pcw_home> ENA+ to 5V ENA- to /INH
[14:47:56] <pcw_home> analog out needs 5V, not field power
[14:48:01] <ssi> ok
[14:48:07] <ssi> the drives don't provide a 5v line
[14:48:12] <ssi> can I steal it from an encoder connector?
[14:48:18] <pcw_home> from cable or external
[14:48:27] <ssi> right now it's from cable
[14:48:33] <pcw_home> Probably not
[14:48:35] <ssi> ok
[14:48:55] <ssi> the field supply that I'm planning to use is a +12 +5 switcher that came out of the old centroid drive
[14:48:58] <ssi> I can use its 5v
[14:49:06] <ssi> I'll probably drive the external 5v with that as well actually
[14:50:00] <ssi> ok so right now, the drive isn't connected to the 7i77 at all
[14:50:02] <GammaX-Laptop> roycroft, where do you live?
[14:50:04] <pcw_home> Yeah that should be fine (check polarity carefully)
[14:50:04] <ssi> I have machine power on in axis
[14:50:08] <ssi> analogena is high
[14:50:17] <ssi> analogout0 is 0.3837 according to hal config
[14:50:25] <GammaX-Laptop> ssi, still havin issues/
[14:50:25] <ssi> but I still have 0v on the analog pins on the connector
[14:50:26] <GammaX-Laptop> ?
[14:50:32] <ssi> TB5
[14:51:05] <ssi> GammaX-Laptop: clearly
[14:51:41] <ssi> measuring pins 3 and 4 on tb5, and 0v across them
[14:51:47] <pcw_home> analogout 0? (pin 4 from bottom)
[14:51:51] <ssi> yep
[14:52:44] <pcw_home> what is the scalemax parameter?
[14:53:18] <ssi> 10
[14:53:22] <ssi> maxlim 10, minlim -10
[14:53:52] <pcw_home> linuxcnc running?
[14:53:56] <ssi> yep
[14:54:04] <pcw_home> no idea then
[14:55:03] <ssi> that's not what I like to hear! :(
[14:57:02] <pcw_home> can you verify that enable is on? (ohmmeter across enable pins will do)
[14:58:12] <ssi> I get open circuit when analogena is high,
[14:58:17] <ssi> and 13k when it's low
[14:59:00] <pcw_home> so you still have an inversion you need to remove there
[14:59:18] <ssi> hm yea
[14:59:20] <ssi> I thought i'd taken that out
[14:59:37] <ssi> oh you know
[14:59:39] <ssi> I commented it out
[14:59:44] <ssi> but I guess nothig actually SET it to false
[15:02:57] <ssi> ok looking better
[15:03:01] <ssi> now to sort out enables
[15:14:43] <ssi> pcw_home: so this machine I'm working on, the servos don't have tachs is them like I expected
[15:15:14] <ReadError> ssi: know of any good finds in ATL right now?
[15:15:15] <ssi> pcw_home: the drives will run in current mode, but I don't fully understand current mode. KimK and I were talking about it last night, and I'm not 100% clear from our discussion if I'm going to need one PID component per axis or two
[15:15:23] <ReadError> looking to grab a bigger mill
[15:15:29] <ssi> ReadError: not aware of anything, no
[15:15:33] <ssi> I'd like to get a mill myself
[15:16:49] <ReadError> what are you working on now?
[15:17:09] <ssi> I'm doing a mill conversion for a company
[15:17:11] <ssi> it's not my machine
[15:17:32] <ReadError> ah cool
[15:17:43] <ReadError> you do this full time? or just a side project
[15:17:48] <ssi> just on the side
[15:18:02] <ssi> if I did it full time, I'd likely be better at it ;)
[15:18:06] <ReadError> sounds fun though
[15:19:15] <ssi> I enjoy it
[15:19:52] <ssi> https://pbs.twimg.com/media/BI3ljNzCAAAHPvJ.jpg:large
[15:19:59] <ssi> that's as of friday night
[15:20:03] <ssi> it's a bit more of a mess righ tnow
[15:21:39] <mrsun> hmm when cutting timing belt pulleys ... does it have to be done like gears ? with many different cutters for a range of number of teeth ?
[15:21:53] <ReadError> ssi: friend of yours?
[15:22:03] <ssi> some folks that were referred by a friend
[15:22:28] <ssi> mrsun: they're trapezoidal, so I think you could do it with a flat bottom endmill
[15:22:32] <Gamma-X> I need a 5c closer.... are they all the same dimensions?
[15:22:44] <mrsun> ssi, yes but they do have angles
[15:22:47] <mrsun> on the sides =)
[15:22:55] <mrsun> so was thinking fly cutter :P
[15:23:07] <WalterN> hmm
[15:23:19] <ssi> they have angles, but if you have a 4th axis, you can use rotation of the axis to make that cut I believe
[15:23:48] <mrsun> ssi, but i dont :/(
[15:23:58] <ssi> aha!
[15:23:59] <ssi> heheh
[15:23:59] <mrsun> so was thinking of mocking up an indexer =)
[15:24:12] <ssi> well even with an indexer
[15:24:18] <ssi> as long as you can index fine enough
[15:24:23] <mrsun> using lathe gears to index :P
[15:28:08] <r00t4rd3d> my new gantry is wide as my old tables x is long :/
[15:28:23] <r00t4rd3d> if that makes sense
[15:28:49] <r00t4rd3d> 36" :D
[15:29:59] <ssi> pcw_home: if I use this dual voltage switching supply for both the external 5v power and the field 12v power, they'll share a common. Is that a problem?
[15:30:27] <Gamma-X> ssi what was that link u sent me yesterday about the 5c chucker?
[15:30:57] <ssi> http://www.cdcotools.com/index.php
[15:31:08] <Gamma-X> and youve used that before?
[15:31:14] <ssi> yea I have that 5c chuck
[15:31:23] <ssi> it's good for the money
[15:31:40] <Gamma-X> accurate? Also how fast is it to put in or pull through some new stock?
[15:31:40] <dgarr> ssi: some amc drives have a jumper to reverse sende of inhibit pin see J1 note on http://www.a-m-c.com/support/glossary.html
[15:31:49] <dgarr> sende==sense
[15:32:33] <ssi> dgarr: oh man that would be amazing
[15:33:27] <ssi> Gamma-X: yeah if you mount it true, it's got low runout
[15:33:35] <ssi> it's easy enough to use, but it's a chuck key closer
[15:33:40] <ssi> so it's not as fast as a lever or pneumatic closer
[15:34:00] <dgarr> i have some amc 30a8t brush drives with that feature but not sure how i found out about it , it is a well-marked smt jumper on my drives
[15:34:12] <Gamma-X> thanks!
[15:34:12] <ssi> I have 30a8ts as well
[15:34:15] <ssi> but I needed hight voltage for this project
[15:34:18] <ssi> so they're bx15a20s
[15:35:37] <ssi> theres' a JS1 and a JS2
[15:35:42] <ssi> JS1 is populated, JS2 isn't
[15:36:25] <ssi> I may try pulling JS1
[15:38:05] <mrsun> ssi, but question still stands .. will it be 40 degrees on on all sizes or will it change over pulley sizes :P
[15:38:21] <ssi> mrsun: I dunno... cad it up and see :)
[15:40:31] <mrsun> archivist, !
[15:41:01] <ssi> dgarr: got the JS1 shunt removed
[15:41:14] <ssi> dgarr: I tore my power wiring apart, so it'll be a bit before I can check it
[15:41:20] <ssi> dgarr: but if that works, I owe you a beer ;)
[15:41:46] <dgarr> good luck -- mine is labeled j1 so i don't know about js1
[15:41:56] <ssi> yeah it's a gamble
[15:42:30] <ssi> there's another set of them
[15:42:33] <ssi> SC1, SC2, SC3
[15:42:38] <ssi> SC1 is populated witha 0R shunt
[15:42:42] <ssi> the others are unpopulated
[15:45:47] <mrsun> hmm it should always be 40 degrees or else the actual tooth will be bent also i guess
[15:50:54] <WalterN> is there a good place I should get servo motors and controls for making a small CNC machine?
[15:54:41] <Gamma-X> ssi would you go fo a 3 jaw... 5c.... or 4 jaw if you only had one choice
[15:57:02] <Gamma-X> WalterN check out http://www.automationtechnologiesinc.com/
[15:57:47] <WalterN> oh, I should probably decide on torque too
[15:57:49] <DJ9DJ> gn8
[15:58:11] <WalterN> for a table top CNC machine, what kind of servo motor torque do I want? :-x
[15:58:14] <ssi> one chuck? definitely 4 jaw
[15:58:48] <Gamma-X> WalterN are you make it yourself or buying and converting?
[15:58:56] <WalterN> making everything myself
[15:59:25] <Gamma-X> depends on the specs you want.... especially if your makin it yourself.
[16:00:04] <WalterN> ok...
[16:00:11] <WalterN> so what specs do I want? (lol)
[16:00:25] <WalterN> I guess I need a reference point...
[16:00:31] <ssi> what kind of machine are you trying to make?
[16:00:31] <Gamma-X> walter, y not buy something even used and convert it?
[16:00:47] <WalterN> making a laser sintering rapid prototyping machine
[16:00:58] <ssi> ok then you don't need high torque
[16:01:03] <ssi> have you built cnc machines before?
[16:01:07] <WalterN> no
[16:01:11] <ssi> start with steppers :)
[16:01:17] <Gamma-X> your in for a treat! lol
[16:01:18] <WalterN> no :-/
[16:01:18] <ssi> much more newbie friendly
[16:01:22] <ssi> no?
[16:01:35] <WalterN> meh, I dont want stupid steppers :P
[16:01:41] <ssi> hahaha ok then
[16:01:46] <ssi> small servo systems aren't as common
[16:01:54] <Gamma-X> you might have better accuracy with steppers lol
[16:02:14] <WalterN> how is that possible?
[16:02:24] <ssi> servos are only as accurate as your feedback
[16:02:36] <WalterN> with a good encoder it should be really accurate
[16:02:41] <ssi> yep
[16:02:43] <WalterN> and ballscrews
[16:03:02] <ssi> anyway
[16:03:08] <Gamma-X> and ALL tolerances are tite
[16:03:30] <ssi> you can estimate the torque you need if you know the mass of the axis you're trying to move and the acceleration you want to be able to achieve
[16:03:33] <Gamma-X> tight*? idk how to spell lol
[16:03:44] <ssi> but you need to know everything about your drive system
[16:03:46] <WalterN> lite tite
[16:03:49] <WalterN> heh
[16:04:01] <ssi> but the answer is "small"
[16:04:06] <WalterN> ssi: drive system?
[16:04:09] <ssi> you don't need much torque to run a laser
[16:04:16] <ssi> because there's no cutting forces
[16:04:17] <Gamma-X> http://www.automationtechnologiesinc.com/products-page/servo-nema23-kits/3-axis-nema23-350ozin-60v20a-psu-g230x-gecko-driver
[16:04:21] <WalterN> oh yeah
[16:04:31] <ssi> so really your torque just limits your max acceleration
[16:04:45] <WalterN> torque is only necessary for a decent acceleration for rapid movements I guess
[16:04:48] <ssi> sorry by drive system I meant how are your motors coupled to the load
[16:05:08] <ssi> you're gonna trade torque for speed in different ratios at each step
[16:06:00] <Gamma-X> what brand ball screws you going with?
[16:06:06] <Gamma-X> how big of a table you going to have?
[16:06:10] <WalterN> the good kind that dont suck
[16:06:14] <WalterN> lol
[16:06:27] <ssi> I hope you have deep pockets
[16:06:35] <Gamma-X> what kind of price range do you have for the project if you dont mind me askin?
[16:07:04] <WalterN> *shrug* whatever till its finished...
[16:07:16] <ssi> if I were in your shoes
[16:07:22] <WalterN> $5,000?
[16:07:26] <ssi> I'd build the machine with some decent roton screws
[16:07:31] <ssi> not the best in the world, but they're inexpensive
[16:07:36] <Gamma-X> agreed ^^^
[16:07:41] <ssi> and I'd use steppers in the first iteration
[16:07:47] <ssi> the drives are much cheaper, much simpler
[16:07:51] <Gamma-X> also agreed ^^^
[16:07:52] <ssi> motors are much cheaper
[16:07:58] <ssi> and they're perfectly adequate
[16:08:12] <WalterN> meh
[16:08:16] <WalterN> (lol)
[16:08:34] <ssi> servo motors and drives are going to easily cost three times as much as equivalent steppers
[16:08:50] <ssi> but if that's what you want to do, who am I to stop you ;)
[16:09:09] <Gamma-X> getting into servos and drives it is much more complex.
[16:09:17] <WalterN> well... I just dont like screwing around..
[16:09:22] <mrsun> good thing about servo is that you wont lose position :P
[16:09:40] <Gamma-X> thats not neccasarily true
[16:10:07] <WalterN> could put an encoder on a stepper... but why not just use a servo in the first place?
[16:10:48] <Gamma-X> differant types of motors from within both devices.
[16:11:17] <ssi> how big do you want this machine to be?
[16:11:31] <Gamma-X> you beat me too it.
[16:11:42] <ssi> I have a 50x50" plasma table that uses 381oz-in steppers
[16:11:44] <ssi> http://www.automationtechnologiesinc.com/products-page/nema23-m-dc-servo-motor/skewed-servo-dual-shaft-3
[16:11:55] <ssi> http://www.automationtechnologiesinc.com/products-page/nema23-m-dc-servo-motor/skewed-rotor-design-nema23-dual-shaft
[16:12:03] <ssi> something like that would run a big table pretty well
[16:12:22] <WalterN> well
[16:12:37] * WalterN grabs tape measure
[16:13:18] <WalterN> maybe 20"-24" of movement
[16:13:34] <ssi> 24x48 is a good size for a laser
[16:13:52] <ssi> you might be able to use something small like a 24v maxon servo
[16:14:00] <WalterN> well... laser sintering rapid prototyping takes a while
[16:14:02] <Gamma-X> how fast do you plan on the gantry moving with the laser?
[16:14:16] <ssi> http://www.ebay.com/itm/MAXON-8-WATT-PRECISION-EC-BRUSHLESS-DC-SERVO-MOTOR-/380349594221
[16:14:22] <WalterN> I'm not cutting sheets of metal or anything
[16:14:26] <ssi> oh right you said sintering
[16:15:07] <ssi> how do you intend to build the linear motion?
[16:15:07] <WalterN> 18" cube of volume is probably enough for such a machine
[16:15:46] <WalterN> I was going to have a fiber optic cable coming in over top of the bed with some optics (that I have yet to buy)
[16:15:59] <WalterN> and just move the optics around overtop of the work space
[16:17:14] <ssi> ok, but what mechanism is going to move the fiber head?
[16:17:36] <WalterN> I recently figured out the laser and optics part of the machine (its really fun to burn stuff with a 30 watt laser)
[16:17:56] <WalterN> uh
[16:18:46] <WalterN> like a router? instead of a spindle the fiber cable comes in overtop of everything
[16:18:54] <ssi> yeah I get that
[16:19:05] <WalterN> so...?
[16:19:07] <ssi> linear bearings?
[16:19:11] <WalterN> oh
[16:19:18] <ssi> I assume you're not going to stand over it and point the laser at stuff by hand
[16:19:26] <Gamma-X> lol
[16:19:42] <Gamma-X> Actually it does sound kinda fun to have a 30 watt laser in your hand...
[16:19:59] <Gamma-X> bye bye gf's kitty!
[16:20:24] <WalterN> Gamma-X: it kinda is... and a diode laser is small enough that you actually could... minus the power supply
[16:20:48] <Gamma-X> I almost bought the one they sell online.. the 1watt but its too much money
[16:20:54] <WalterN> ssi: yeah, linear bearings (that I have yet to select), with ball screws and stuff
[16:21:08] <WalterN> heh "and stuff"
[16:21:37] <WalterN> Gamma-X: I got mine on ebay for $322 :3
[16:21:48] <WalterN> need another $350 for a power supply though
[16:21:50] <Gamma-X> a 30 watt laser on ebay for 322?
[16:22:17] <ssi> sure
[16:22:21] <ssi> you can get 40 watt tubes for $100
[16:22:24] <WalterN> used, but the diode laser was made in USA
[16:22:55] <WalterN> oh yeah
[16:22:55] <WalterN> CO2 lasers are cheap
[16:23:07] <Gamma-X> i want one!
[16:23:21] <WalterN> do it!
[16:23:32] <Gamma-X> you have link to the laser?
[16:24:20] <WalterN> cant fiber optic cable CO2 lasers though I dont think
[16:24:20] <ssi> I want to do a laser too
[16:24:20] <WalterN> well maybe
[16:24:20] <cpresser> Co2-lasers only work with mirrors
[16:24:20] <ssi> I'd like to build a 24x48 80W laser cutter
[16:24:20] <Gamma-X> I just wanna burn crap
[16:24:20] <ssi> co2
[16:24:24] <Gamma-X> ssi dont you remember the movie with val kilmer?
[16:24:31] <ssi> which one? :P
[16:24:38] <Gamma-X> with the laser?
[16:24:43] <Gamma-X> Real Genius
[16:24:46] <WalterN> cpresser: yeah, wavelength is a bit too long for fiber optics
[16:27:30] <WalterN> hmm
[16:28:15] <Real-Genius> ssi is there a way to test a stepper test to see if your losing steps?
[16:29:00] <ssi> in setup, with a dial indicator or tape measure
[16:29:09] <ssi> in normal operation, without axis position feedback, no not really
[16:30:02] <ssi> https://pbs.twimg.com/media/BI9845KCIAAI1xX.jpg:large
[16:30:04] <ssi> field power :D
[16:30:32] <Real-Genius> I did see leadshine has some stepper drivers that somehow stop stepp loss?
[16:30:43] <WalterN> hmm
[16:30:54] <WalterN> should I gear the servo down?
[16:31:08] <ssi> if you need to trade speed for torque, yes
[16:31:16] <ssi> if you were using something like those maxon servos I linked
[16:31:20] <ssi> theyr'e high rpm, low torque
[16:31:31] <WalterN> most ballscrews have a pretty high twist rate
[16:31:33] <ssi> honestly, you're nowhere close to being able to make these kicd of choices yet
[16:31:38] <ssi> you need a motion design first :)
[16:32:09] <WalterN> motion design?
[16:32:38] <ssi> didn't we go through this already?
[16:32:51] <ssi> you need metally bits that slide around and hold your LAZERBEEMZ
[16:33:03] <ssi> you need ballscrews or belts or something to move them around
[16:33:10] <Real-Genius> do you have a sketch or dimensions of your ... "table" and laser holder?
[16:33:11] <ssi> (that's another thing, I'd do belt for a machine like that)
[16:33:39] <ssi> you're honestly probably best off scaling up a reprap
[16:34:02] <ssi> I'm sure you have some reason to not like that idea :P
[16:34:33] <WalterN> had to look it up... and it looks like.. garbage
[16:34:42] <L84Supper> http://www.brandnew-china.cn/product/752734740-218400239/40Kpps_Wider_Angle_Speed_Galvo_Scanner_For_Laser_Show_Projector_Good_Price_and_Quality_With_Stocks.html
[16:34:56] <ssi> heh there are a lot of different types of reprap hardware
[16:35:00] <ssi> some of them are garbage
[16:35:02] <L84Supper> Input Voltage: ±5VDC Differential
[16:35:14] <ssi> http://reprap.org/mediawiki/images/thumb/b/b5/MendelMaxPhoto.JPG/535px-MendelMaxPhoto.JPG
[16:35:27] <ssi> something like that, made of extrusion and thompson shaft rails
[16:35:33] <ssi> belt driven
[16:35:53] <WalterN> L84Supper: I decided against galvo because I dont think it would work very well for engraving stuff
[16:35:55] <L84Supper> he can get a galvo setup for $150
[16:36:33] <ssi> hm that'd be an interesting way to do it
[16:36:36] <ssi> then you don't need any motion
[16:37:44] <Real-Genius> WalterN could just buy this... http://www.ebay.com/itm/2X2-CNC-PLASMA-CNC-ROUTER-CUTTING-TABLE-/261203160816?pt=Power_Tools&hash=item3cd0ebeef0
[16:39:36] <L84Supper> WalterN: what would you be doing wrong for a galvo setup to not work well for engraving?
[16:40:30] <WalterN> L84Supper: donno... I figured the angle would screw with the way the engraving looks?
[16:40:45] <ssi> it'd be fine
[16:40:51] <ssi> a 30w laser doesn't "engrave"
[16:40:55] <ssi> it just burns a coating
[16:40:57] <ssi> so there's no depth
[16:41:15] <L84Supper> http://www.youtube.com/watch?v=oez3XLqyRH0
[16:41:16] <Tecan> (oez3XLqyRH0) "Laser Galvo" by "catherineanseat" is "Tech" - Length: 0:05:08
[16:41:56] <WalterN> and I still dont know how beam focusing would work
[16:48:22] <WalterN> could converge and make the beam roughly straight before it goes in to the galvos I guess... not sure what kind of beam density I could get by doing that
[16:48:55] <WalterN> or if mirrors are designed to handle it
[16:50:43] <WalterN> well, mirrors have a certain power density they are rated for
[16:53:33] <L84Supper> https://www.cvimellesgriot.com/products/Documents/Theta_F_Article.pdf
[16:57:30] <WalterN> oh, a special lens that.. interesting
[16:57:57] <L84Supper> the f-theta lens keeps the laser spot circular vs elongated
[16:58:52] <WalterN> I would still need to converge and diverge the beam though
[17:03:15] <L84Supper> the f-theta lens also keeps the beam at focus across the image plane
[17:03:29] <WalterN> yeah
[17:04:00] <L84Supper> spot the spot diameter doesn't change (along with the power density)
[17:05:32] <L84Supper> heh, so the spot diameter .....
[17:07:49] <WalterN> I dont see any of those kinds of lenses here... http://search.newport.com/i/1/q1/products/q2/Optics/x1/pageType/x2/section/q3/Spherical+%2526+Aspherical+Lenses/x2/section/x3/chapter/nav/1/
[17:11:41] <WalterN> oh, maybe in here? http://search.newport.com/i/1/x1/pageType/q1/Products/q2/Optics/q3/Cylindrical%20%2526%20Achromatic%20Lenses/q4/BK%207%20Precision%20Cylindrical%20Lenses/x2/section/x3/chapter/x4/family/nav/1/
[17:12:43] <L84Supper> http://www.thorlabs.com/NewGroupPage9.cfm?objectgroup_id=6430
[17:12:50] <L84Supper> https://www.cvimellesgriot.com/Products/FTheta-Lens.aspx
[17:12:58] <L84Supper> http://www.jenoptik.com/en-f-theta-objective-lenses
[17:13:08] <L84Supper> http://www.edmundoptics.com/lasers/laser-optics/laser-optic-assemblies/f-theta-scanning-lenses/2961
[17:21:46] <ssi> dgarr: good news, that jumper worked :D
[17:30:21] <WalterN> L84Supper: huh... that thor labs one only has ~85% light transmission for 810nm wavelength
[17:35:29] <ssi> weeee
[17:35:32] <ssi> I have proper drive enables now
[17:35:37] <ssi> that makes me inordinately happy :D
[17:37:38] <ssi> holy hell I think my servos are even servoing
[17:39:00] <ssi> ohhh yeahhh
[17:39:01] <Nick001-Shop> <ssi> Always happens when you connect things properly -)
[17:39:13] <ssi> yeah i suppose that's true
[17:39:33] <ssi> I think it's close enough to start working with it in the machine actually
[17:39:50] <ssi> servo is dead still, and holding position with more torque than I can impart by hand
[17:40:12] <ssi> had to bump P to 60 to get it to hold
[17:40:19] <ssi> good start :D
[17:40:33] <Nick001-Shop> just make sure overtravels are working in case
[17:40:45] <ssi> yeah I'll get them wired before I power up the drives
[17:40:54] <ssi> I need to have hands on the machine to wire limits
[17:41:01] <ssi> machine is 50 miles from here :/
[17:41:21] <Nick001-Shop> long wire??
[17:41:27] <ssi> yeah, very ;)
[18:52:12] <kizer> derpype
[18:54:19] <kizer> yeah wrong box. sorry
[20:53:41] <WalterN> L84Supper: of those f-theta lenses you linked, which one would be best for 810nm wavelength?
[21:42:57] <tjtr33> pcw_home hello, what's the intended 'audience' or purpose for the cubieboard fpga shield?
[21:44:15] <pcw_home> Just a shot in the dark (but I think it would run LInuxCNC very well especially the A20)
[21:46:13] <tjtr33> i got one for that purpose. but i see so much talk about BBB, and it isnt FPGA, and I guess that means the BBB would be slower for IO.
[21:46:20] <jef79m> hi guys, trying to find someone who can cut g10 in australia, any recommendations?
[21:46:24] <tjtr33> (no hdwr expert here )
[21:47:39] <tjtr33> garolite? dulls HS steel immediately. some sugestions in history of this ric, but i have only had bad experience with single lip HS cutters.
[21:47:48] <pcw_home> The PRU (200 Mhz 32 bit fixed point coprocessor)makes it capable of a lot of fairly fast I/O b
[21:47:55] <tjtr33> ric/irc
[21:48:42] <tjtr33> pcw_home, sounds like Forth is in the works for it then ( fast big number crunching for astronomers :)
[21:50:04] <Valen> we can do it tjtr33
[21:50:07] <pcw_home> Fourth should be a easy port (though only have I think 8K of code space)
[21:50:26] <Valen> sorry jef79m we can cut G10
[21:50:44] <jef79m> Valen: you're the guy who was recommended to me, but it seemd like you hadn't been in here for a while.
[21:50:54] <Valen> ?
[21:51:15] <tjtr33> Valen ? PRU? garolite?
[21:51:25] <tjtr33> must be 'can cut G10'
[21:51:28] <Valen> we cut PCB all the time
[21:51:53] <Valen> and assorted fiberglass products so anything in that line should be doable
[21:52:16] <Valen> diamond grinder is a good way to do it btw
[21:52:54] <tjtr33> pcw_home, but you have to talk like yoda "lawn mow, fence paint, post fix" <<< old forth joke
[21:53:09] <jef79m> awesome, do you have an email address i could contact you at?
[21:53:29] <Valen> pm sent
[21:57:55] <pcw_home> Actually I kind of like Forth. Years ago we used it for hardware testing
[21:58:49] <tjtr33> skunkworks, thanks for the links to the Accupin, didnt see it till i hit the archives today, had bad cnxn on the day you posted.
[22:00:04] <tjtr33> pcw_home, I got boxes of forth pcs and 'capes' . i used them for early ATC's on EDM's (pop over drop swap :)
[22:00:25] <tjtr33> pc/microcontroller
[22:01:11] <tjtr33> yeah the code read like weird poetry after you got the words down
[22:01:42] <pcw_home> Postscript is similar
[22:02:00] <L84Supper> pcw_home: did you post specs on your cubie shield?
[22:02:10] <pcw_home> Not yet
[22:02:49] <L84Supper> pcw_home: did you see the new Beaglebone?
[22:03:19] <L84Supper> http://beagleboard.org/Products/BeagleBone%20Black
[22:03:41] <L84Supper> $45
[22:04:11] <pcw_home> Yes pretty amazing (well as long as TI subsidizes it)
[22:04:18] <L84Supper> no SATA and only 512MB
[22:05:17] <pcw_home> Yes the Cubie (especially with the A20) is more practical as a completer linux system
[22:06:22] <tjtr33> i got a 32G 10x sd for the cubie, but will look at a SATA also ( both are useful tools in any case )
[22:07:11] <pcw_home> If you want to do local compiles SATA is probably best
[22:09:30] <L84Supper> pcw_home: which IO board has analog +-5V differential out to drive galvos? open loop
[22:10:40] <tjtr33> with no external components? ( no 0-10 converted to +/- 5 ?)
[22:10:47] <L84Supper> 7i33
[22:11:30] <pcw_home> Normally drives have differential in but the A-D is single ended
[22:13:02] <tjtr33> hmm i just revisted a site for Encore Gardens in Taiwan, big outdoor laser show, drawn on a water screen. :( abandoned now, after the 05 earthquake )
[22:13:07] <tjtr33> galvos there
[22:13:09] <tjtr33> http://wordsilived.blogspot.com/2011/08/encore-gardens-taiwan-abandoned-theme.html
[22:18:10] <L84Supper> http://www.brandnew-china.cn/product/818183525-218400239/Medical_and_Beauty_Galvanometer_Scanner.html
[22:18:25] <L84Supper> some of these have analog and digital inputs
[22:19:22] <L84Supper> XY2-100 protocol
[22:19:30] <tjtr33> medical & beauty? tatoos or freckle removers
[22:20:26] <L84Supper> yeah
[22:20:39] <L84Supper> http://www.newson.be/files/TD_XY2-100_R0703.pdf
[22:22:09] <tjtr33> seems like a fast data stream
[22:23:19] <tjtr33> 500nS per pkt
[22:24:19] <pcw_home> Looks easy enough
[22:24:26] <L84Supper> looks like they keep the actual data quiet
[22:24:58] <tjtr33> ^^^ i must have got the math wrong :|, they say 10uS
[22:26:04] <L84Supper> uness it's just 16 bit position info, 0 is one end and 111111111111111 is the other extreme
[22:27:01] <pcw_home> Yes just 16 bit offset binary (and the three C bits whatever they are)
[22:27:09] <L84Supper> http://forums.parallax.com/showthread.php/90522-minimum-cycletime-for-XY2-100-protocol
[22:28:30] <L84Supper> 2MHz clock, synk and data, each uses rs422 voltage levels. total 20 bits are transmitted in 10usec. 3 control signals, 16 data and 1 parity
[22:29:15] <pcw_home> yeah so pretty easy (even a PIC could bit bang that)
[22:30:44] <L84Supper> anything IO -> XY2-100 protocol
[22:31:40] <pcw_home> Not much of a change from the SPI interface
[23:19:48] <ssi> I'll have a BBB in a couple days
[23:22:32] <ssi> pcw_home: for my limits on the 7i77, should I common the limits to field power, and then each switch to an IO pin?
[23:23:08] <ssi> It is suggested that inputs like limit switches use normally closed switches with one switch leg connected to field power and the other to the 7I77 input pin, so the normal machine state (not at limits) is to have the inputs activated. This way, a open switch wire or wire shorted to ground will cause a detectable machine fault.
[23:23:13] <ssi> yea I guess so