On Wednesday 03 February 2016 00:36:29 Jon Elson wrote: > On 02/02/2016 10:03 PM, Gene Heskett wrote: > > I believe they are. Would you like to see the .hal file? > > Hmm, not sure I could tell much from it. Is this using a > Pico Systems PWM controller, Yes > or a Mesa board? Yes, a 5i25. > (I've lost > track.) Following info assumes my PWM controller. When you > hit F2 to go from "machine on" to "machine off", then the > PWM outputs are disabled, and so the PWM output goes to a > low (zero Volts) state. That should turn off the transistors > of the PWM servo amp, assuming the PWM output goes to the > servo amp's PWM+ input, and the PWM- input is grounded.
The pwm signal is already off, long enough that if I want to start the spindle, I have to trigger the boot module to restart it. > When you hit F1 to go to the E-STOP state, then the PWM is > forced to zero, all SSR drive signals are forced to the Hi-Z > state. The SSRs have the + terminal connected to +3.3 V, > and the - terminal is grounded to turn the SSR on. > The SSR + is fed by a 12 volt supply, and the charge pump detector sinks the - terminal. The 12 volt supply is fed by its own transformer powered straight from the main switch on a power strip. The 12 volt supply is a small buck switcher I got from MPJones for about 6 bucks each, good for better than an amp. They looked to be handy as you could squirrel on away anyplace you needed a set voltage, even has a small display to show either the input or the output voltage. > >> But, maybe there is a momentary bobble on the commands to > >> the SSRs when you click the button. > >> I'm guessing that means the E-stop button here. > > > > Thats the second button in axis? Just for grins I also tried the > > first button, identical results, > > > > The charge pump itself is on full time, at 500 Hz. I am gating it > > with and gates driven by timers. That 500Hz is detected by charge > > pump detectors that have about 200 milliseconds of storage before > > they turn the SSR's off. So a wibble in the pump signal would have > > to be pretty drastic to effect them I think. Those I can look at > > with halscope. > > Charge pump? You are sending this out on one of the digital > outputs to an SSR? Yes, Peter insists that the disabled state of all his cards sits at a logic one when off. So without the charge pump, firing up LCNC means you have got to have curcuitry that considers a logic one as the off state, so all hell breaks loose for a while as LCNC starts if the psu starts first. Same with bringing up the PSU's, so without the charge pump detection, everything gets a hiccup if the control box is powered up after LCNC starts. And it does not do any good to program the 5i25's gpio output so off is logic low, it goes high when the 5i25 is disabled regardless. The WDT still disables all outputs to a logic one. Peter may know how to get around that, but I failed. So the only way to do this relatively non-timing critical psu switching is to use the carge pump, and a charge pump detector to do that control, it can absorb the noises without triggering the outside world for 1/4 second while its getting its shit in one sock. A crapload of such problems could be avoided if the 5i25 failed to a logic zero, same as powered off, when its disabled. I love the card otherwise. If BoB's inverted the signals, that would transfer to everything being off/low, solving many of these problems. But I haven't found a BoB that inverts. Working around that with charge pump detection seems to handle that PITA well, at least on paper. Too bad it doesn't do it in the real world. I think thats the first thing I check tomorrow, is the actual signal going up the wire to the charge pump detectors in the box containing that small psu, the regulator card, and two of those pump detectors to drive the SSR's. If that stops at the same time I click the off button, because the 5i25 has set all outputs high, then nothing I do with the off timings I set in hal means diddly-squat. Both will lose drive pretty close to simultaneously, and so fast I probably couldn't hear the <100 miliseconds it takes them to disable the SSR's, fairly close to simultaneously. Those two detectors have common SS si diodes in the signal detector, so they don't drive the hexfets to the least resistance state. No biggie as it only takes 4mils to enable the SSR's. The replacements I built, but haven't installed in that box yet, have schottky diodes so they develop another 3/4 volt to drive the hexfet with and have a slower off switch, I put one in the vacuum cleaner's control path, and its a good 3/4 second turning the vacuum off when I uncheck the flood button. Perhaps 1/2 second to start it. If that is the case, and my carefully timed signals aren't getting thru the 5i25 because its disabled already, then I am with Jackie Gleason, what a revolting development this is! The scope will tell that tale tomorrow, actually later today. I'm just up to recycle some contaminated water and will be back in the sack shortly. Thanks Jon. > > ---------------------------------------------------------------------- >-------- Site24x7 APM Insight: Get Deep Visibility into Application > Performance APM + Mobile APM + RUM: Monitor 3 App instances at just > $35/Month Monitor end-to-end web transactions and take corrective > actions now Troubleshoot faster and improve end-user experience. > Signup Now! > http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140 > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> ------------------------------------------------------------------------------ Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140 _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
