On Wednesday, May 30, 2012 10:58:17 AM Alex Joni did opine: > http://en.wikipedia.org/wiki/RC_circuit > > from there: > http://en.wikipedia.org/wiki/File:RC_Series_Filter_(with_V%26I_Labels).s > vg > > and the response: > http://en.wikipedia.org/wiki/File:Series_RC_capacitor_voltage.svg > > Regards, > Alex > That sort (T=RC) of a solution puts a pretty decent initial acceleration into this as can be seen in the last graph, tapering off such that the accel load is very much reduced as the target speed is neared. If a digital version of the "miller integrator" could be made by adding a fixed amount to the D/A input value each servo cycle, this could be reshaped into a much straighter accel/decel. That part I believe can be cobbled up out of existing hal parts. It would need to be inserted into the signal chain before the PID module, and would need access to the S value requested so that it would know when to stop incrementing the desired speed being sent to the PID module. It may also be possible to use one of the PID inputs to slow this 'rate of change' which should, I'd think, be done at a slower rate than the PID control itself, so that the PID module is in charge of things at all times.
The miller runnup or integrator was an analog circuit that was first used to generate the time accurate horizontal sweep in the first generation of triggered oscilloscopes, and with a high enough gain tube in the socket, can develop a saw tooth waveform that is within .001% of a straight line. It was also used as part of the audio modulation in GE tv transmitters, and was able to modulate with less than 1% distortion in that application. > ----- Original Message ----- > From: "John Thornton" <[email protected]> > To: "Enhanced Machine Controller (EMC)" > <[email protected]> Sent: Wednesday, May 30, 2012 3:45 PM > Subject: Re: [Emc-users] Commutating Reactor > > > Hi Dennis, > > > > I have a 2.5" diameter facing mill that I use and I don't notice any > > difference when using that. My machine lights do not dim during > > acceleration which I assume is a good thing. I'm in the same canoe as > > you with 240v single phase my only option to work with. > > > > Before all the experiments I could get 0 to 2k with no problems and > > would structure my programs around that or use the G4 pause and ramp > > up to the rpm I wanted above 2k. > > > > I too modified my program using a pause between S values but only had > > to do that going up in rpm. From 6k to 0 I've never had a fault. > > > > The resistor/capacitor would be a low cost experiment. How do you > > calculate what the values are for the resistor and the cap? Did you > > put the resistor in series with one of the control lines? Does the > > cap connect to the upstream side or the downstream side of the > > resistor? Maybe a stick drawing would help me understand... > > > > John > > > > On 5/30/2012 7:05 AM, Viesturs Lؤپcis wrote: > >> I am helping Dennis out and reposting his message to mailing list: > >> ===================================================================== > >> = > >> > >> Hi All, > >> > >> I have been following this thread and want to put in my 2 cents. I > >> had the same issue as John with my Bridgeport Interact 412-V. It's > >> spindle is similar to John's: 7.5kW 60-6000RPM. My rotary phase > >> converter is also similar to John's: 15HP TEFC with start and run > >> caps; legs are balanced within a few volts phase-phase. > >> > >> When I set the spindle RPM to 6k and performed an M03, the mains > >> would dip in voltage and the halogen light powered by the machine > >> would get dimmer. The power meter for the spindle would peg at 180% > >> of load as it was starting. The same occurred when stopping the > >> spindle but the halogen machine light would get bright as the > >> spindle dumped the load back into the grid. In each case I would > >> get a voltage fault and for my machine this was a critical error. > >> Rebooting every time I started or stopped the spindle just would not > >> do. > >> > >> Also want to interject that a loaded spindle may be worse. John, can > >> you get to 2500RPM consistently? Try putting a heavy tool in the > >> spindle - like a 5-10 pound boring head or something. Will it trip > >> out below 3k RPM? Is it spindle load related? > >> > >> The thing I noticed was the dimming and brightening of the machine > >> light. This told me there was too much impedance between the VMC and > >> the AC mains, i.e. the rotary phase converter. If bolted directly to > >> 208V 3-phase this would not occur. Even though there is a 30A 3-pole > >> breaker in the panel (with 3-phase), you are connected to a > >> transformer that can supply 200A+ so the impedance is very low and > >> voltages tend not to bounce around much. But I have single phase > >> 240V in my garage so I had to make this work. > >> > >> I started experimenting and modifying my programs to work around the > >> problem: step from 0-2k RPM, wait a bit, step from 2k to 4k RPM, wait > >> a bit then up to 6k RPM. Same with stopping the spindle. My > >> machine's memory is only 3.5k so this was cutting into my program > >> storage and it was a pain. I dug through the machine schematics and > >> found the spindle control signal was an analog 0-10V line. If I > >> slowed the ramp up and ramp down command to the spindle controller it > >> would perform the same function as gradually changing the RPM in > >> code. > >> > >> I cut the 10V command line and inserted ~ 4.7k resistor in series. > >> > >> On the spindle controller side I put something like a 10uF capacitor > >> to ground. This creates a low pass filter so fast transients like a > >> 0-10V command to "get going" take a few seconds to actually make it > >> to 6k RPM. The same with changing the voltage from 10V to 0 - it > >> takes a few seconds for the command to reach it's final value. That > >> did it!! Now my halogen machine light stays at a constant brightness > >> through spindle acceleration and deceleration. > >> > >> John, I think I also read if jogging too fast with one axis your > >> machine would trip. Don't remember what axis this was but on my > >> machine the head is pretty heavy and the Z-axis servo should pull > >> more current than the others if it does. I never had a > >> jogging/voltage trip but you may want to look at rewiring the power > >> circuit for the servo amplifiers to use the lower impedance 240V > >> legs of your rotary phase converter. The servo amps may also be > >> single phase within your machine but still on the machine side of > >> the reactor and input transformer to make the machine work with > >> universal voltage inputs. Does not mean it cannot be rewired. > >> > >> My fix was a few cents worth of parts to modify the CNC. Took less > >> room than a 50kVA transformer and I still think the wild leg of the > >> 15HP rotary phase converter will still move around even though there > >> is a delta-wye transformer in between. That is a lot of money and > >> effort to be experimenting with. > >> > >> > >> Dennis > >> > >> --------------------------------------------------------------------- > >> --------- Live Security Virtual Conference > >> Exclusive live event will cover all the ways today's security and > >> threat landscape has changed and how IT managers can respond. > >> Discussions will include endpoint security, mobile security and the > >> latest in malware threats. > >> http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > >> _______________________________________________ > >> Emc-users mailing list > >> [email protected] > >> https://lists.sourceforge.net/lists/listinfo/emc-users > > > > ---------------------------------------------------------------------- > > -------- Live Security Virtual Conference > > Exclusive live event will cover all the ways today's security and > > threat landscape has changed and how IT managers can respond. > > Discussions will include endpoint security, mobile security and the > > latest in malware threats. > > http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > > _______________________________________________ > > Emc-users mailing list > > [email protected] > > https://lists.sourceforge.net/lists/listinfo/emc-users > > ------------------------------------------------------------------------ > ------ Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. > Discussions will include endpoint security, mobile security and the > latest in malware threats. > http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users Cheers, Gene -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) My web page: <http://coyoteden.dyndns-free.com:85/gene> Who loves me will also love my dog. -- John Donne ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
