On Thursday 30 November 2017 22:45:06 Jon Elson wrote: > On 11/30/2017 12:29 PM, Gene Heskett wrote: > > 2. and this is for Jon. I have a fixed gain scale setting > > that prevents the pwm from exceeding a 98% duty cycle. The > > pwmgen is running at 20 kilkohertz, and when the pid is > > outputting 46 something it is up against this limit, with > > a 1.06 u-sec recharge time. The data that you get from the > > Pico site says half a microsecond is needed to recharge > > the gate drivers. So this should be enough. The 12 volts > > is nominally 11.6. BUT, its sitting out there right now at > > that minimum of 1.06 microsecnds off time, with the > > spindle turning about .8 rps. It ought to be at 3 grand. > > Clicking fwd, no reaction, add a + button click and it > > hits that 3 grand in about 1/2 a second. 30 seconds later > > it starts to slow, and in another minute its down to about > > .8 rps and running steadily. Motor is seeing a noisy 2 > > volts, and it has almost zero torque. While this is > > "percolating", I'm going to reduce that scale, wideing the > > off time to see what effect it might have. > > Well, I have seen some VERY odd behavior if the gate driver > chips decide the bootstrap capacitors are below safe gate > voltage. This is for the brush motor servo amps? They > always had good strong gate drivers, but I did up the value > of the bootstrap and power decoupling caps to 1 uF on later > units. (These are, unfortunately, on the back of the board.) > If you can get the +12 V supply up to 12 or even a little > above, that may help. I forget the exact under voltage > lockout level, but I think it might be 10.6 V. So, with an > 11.6 V supply, minus .7V for the bootstrap diode, there > REALLY isn't much left before the UVLO triggers. > > Note that the optos in the servo amps alter the duty cycle > some, and if the LED series resistors are too high a value, > it makes it worse. But, that would NARROW the on-time, and > make the OFF-time wider, so that is not likely the problem. > > Jon
This almost acts as if the input is upside down. If I limit it to say 10% on time, it boots the thing up and runs the motor wide open, for perhaps 1 to 2 minutes, then ever so slowly the speed will oooze down until a couple minutes later its down to about 40 rpm, with virtually zero torque, and will do that for a couple hours. At that point I can take it back to a 98% duty cycle drive at 20 KHz, leaving 1.06 u-secs for cap recharging and it doesn't make a detectable change. I'm thinking its losing high side charging time because I've blown a chip. This is the second one I bought, about 3 years ago now. The first one, after I'd put a bigger toroid in the fwd leg, is still running TLM just fine. But I may have blown something in the inputs as I was trying to replace the bob, which had a long string of slow opto's with one that was all buffers with a 10 MHz bandwidth both ways, and had forgotten exactly how I had arrived at the 12 volt enable signal. I got that sorted eventually so I know I've got good signals to it. That 12 volts is from a wall wart, all it powers is this enable signal, and I measured it a couple times today while it was miss-behaving, and got around 12.9 volts both times, so that s/b ok. Regardless, can you repair it, and if so, the turn around time? I can paypal you the charges since its at least 99.9% my fault, and not exactly a new one now. 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> ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
