Hurco BMC20 Dynomotion Retrofit

[Need TECH Help!]

So DMM has sent for a pickup of cooked drives for repairing them.
And i talked with Arturo at CNC4PC he said that motor RPM will be 3000RPM with 240V single phase. And 1500RPM with 120V single phase. Rep at DMM said it would be limited to 1500 RPM at 240V single phase. It is looking like these drives only require 240Vac either 240 single phase or 240 3 phase and not both. But iam surmising at this point. I have asked just how they should be connected a few times with DMM rep and either dont get answered or get a unclear answer. I sent an email to CNC4PC for a connection example of single phase and 3 phase. Will post when i found out.

[mmurray70]

That makes sense about 1500 on 120v. I would think 240 single phase would by far be the most common setup for these drives so it would be disappointing to only have 1500 rpm with this setup.

Maybe you should just wire these 240 single phase for your machine. Should be safe enough with that. If you have enough current in your supply line i guess you could just use one pair of wires for all three drives and leave one wire not used. Or maybe you could supply each drive with one of the three phases to keep things balanced. Might be a mess to wire it that way though.

If you use all 3 phases it looks like 140vac max according to the fine print in the DMM manual. So i think you can rule out your 240 3 phase unless you want to see more smoke. Can you give us more details on power supply to the machine? You do have 240 three phase right now right? Does the machine have an adjustable transformer built in?

[Need TECH Help!]

That makes sense about 1500 on 120v. I would think 240 single phase would by far be the most common setup for these drives so it would be disappointing to only have 1500 rpm with this setup.

Maybe you should just wire these 240 single phase for your machine. Should be safe enough with that. If you have enough current in your supply line i guess you could just use one pair of wires for all three drives and leave one wire not used. Or maybe you could supply each drive with one of the three phases to keep things balanced. Might be a mess to wire it that way though.

If you use all 3 phases it looks like 140vac max according to the fine print in the DMM manual. So i think you can rule out your 240 3 phase unless you want to see more smoke. Can you give us more details on power supply to the machine? You do have 240 three phase right now right? Does the machine have an adjustable transformer built in?

I plan to connect only the 240 single phase legs, to all drives, done with smoke.

Machine has 240 single phase and the 3 phase 240 from 3 phase rotary convertor. So yes, 3 phase going to machine.Total of 3 legs to machine.
There is no adjustable transformer only a 240 to 230 line conditioner and a 240 to 120 isolation transformer. The 240 line conditioner is for logic circuit of drives and for 120 isolation transformer.

[mmurray70]

Sounds like a good plan. Figure out which wire is the manufactured leg and avoid using it, be sure to use the two hots from your single phase supply.

Seems like your drives fried from over voltage, but i wonder if the phase converter could have been a factor too? I heard once that there can be some kind of an issue running vfds and servo drives from phase converter without a transformer in between. I didnt pay much attention to the issue because both my machines have built in transformers. Maybe someone else might have some more info on it.

[Need TECH Help!]

Sounds like a good plan. Figure out which wire is the manufactured leg and avoid using it, be sure to use the two hots from your single phase supply.

Seems like your drives fried from over voltage, but i wonder if the phase converter could have been a factor too? I heard once that there can be some kind of an issue running vfds and servo drives from phase converter without a transformer in between. I didnt pay much attention to the issue because both my machines have built in transformers. Maybe someone else might have some more info on it.

Yup, I know which is which. There is some "home built" converters that actually rob the leg being generated to run 3 phase motor which makes about like 40volts short. The converter i have is built and rated for CNC vfd duty, this is a 20hp from American Rotary. Its same converter i was running machine with before retrofit. The TECO VFD i put in machine with old control had no issues.
We have 2 of these same converters at work running shop 1 is 20hp and 2nd is 30hp. Running new Milltronics machines and new Hurco mill around 15hp VFDs, 1.2 to 2KW AC Yaskawa servos each and have no problems. These converters actually hold a better voltage tolerance (around 5%) than grid 3 phase 10%.

[Need TECH Help!]

Ok. Got a new drive and connected to only R and S of Main Circuit. Both R and S are 120Vac single phase making 240VAC. And no SMOKE!! Done some successful jog tests of Y axis with DMM tuning software. So far iam estimating around 450IPM moves, am expecting more, but dont know enough about tunning these drives. Going to contact DMM with news and see if they have any configuring recommendations for drive and my machine setup.
I also contacted them about testing new drive with 240VAC single phase only and they brought up my 3phase converter again. And said "Normally, you can just connect any two legs of a three phase output go get single phase." What?? Can someone tell me what iam missing? I have never used Delta Wye only Delta from a 3phase converter. Is Delta Wye able to do this?

Anyhow will post as i get more results on testing all drives and servos.

[mmurray70]

Good to hear you got it working. Are you using step and direction setup or 0-10v from Kanalog?

Yes you can think of a 3 phase 240 supply as three separate single phase supplys that are 120 degrees apart in phase. So you can use any 2 wires of a 3 phase supply and have the same thing as normal single phase power. I mentioned this in post 142. It normally would not matter which two wires you picked and it might even be a good idea to use all three combinations between drives to keep things balanced. But considering your on a phase converter, i would just use the two mains and avoid the manufactured leg. No need of loading it down if you dont have too.

[Need TECH Help!]

Iam currently only testing with DMM tuning software. But when iam ready, will be using Kanalog 10v.

But your voltage would be higher.

Will be using converter for spindle VFD only now. And keep rest of machine on 240V single phase. This will allow me to run different attachments like high speed spindle without the need of running converter. Better on electric bill to.

[Need TECH Help!]

Ok.Been testing replacement drives on 240 single phase with DMM software and performing good, but will need regenerative resistor for stopping fast

Now am testing with Kmotion. Servo drive is set as Command input Analog. And is connected to Kanalog JP11. Servo mode of drive can be set as Speed Servo or Torque Servo.
If set to Torque Servo and enable drive's main and logic circuit power, servo immediately runs away.
In Speed Servo mode I have to set drive Main Gain, Speed Gain, and Torque Filter Constant down to 1 or servo will run away. And can send DAC1=XXX commands but servo will only move in one direction. In this same mode if I set gains back to where they work when testing with DMM software, servo will vibrate hard , but will respond when sending a DAC1=100 command or DAC1=0 and after about 15 or 20 second drive will fault over current. The Constant Torque setting cause s the hard vibration,but tried seeing it to 1 and servo slowly runs away in a incremental/jerking motion. Picture is screen shot of DMM software for parameter setting of drive.
Is there something obvious ?
Thanks,
Troy

[m_c]

Unless you have the KFlop configured to control the servo loop, nearly all analogue servos will run away if you just power them up and enable them. In analogue mode, you are reliant on the controller to monitor the encoder, and adjust the control voltage to maintain position. This will be far more noticeable in torque mode with no load.

Speed mode is usually the most stable operating mode, whereas Torque mode is usually more unstable, however Torque mode bypasses any drive internal speed control/filtering which means the motor should respond quicker, and potentially result in a better tune.

Personally, I'd suggest starting with Speed mode and any default drive settings. Get the KFlop controlling the servo loop to maintain position, ensure limit switches are disabling the servos, and you have a functional physical E-stop button near. Then connect up the motors to the ballscrews, as they need to be loaded for tuning, and start tuning via the step response screen.

[Need TECH Help!]

No KFLOP does not currently have encoder input from axis. So i need to do this before testing? Obviously i have never setup a Analog system.

Limit switches are connected to Konnect right now. But my Estop switch Disable/Enables my contactor which connects/disconnects drive logic and main circuit power.

[Need TECH Help!]

Now that i think about it, how could there be any accurate control with an analog +/- 10v without position feedback to the controller. I was thinking it worked like Step/Direction. Looks like i got some more wiring to do.

[mmurray70]

Unless you have the KFlop configured to control the servo loop, nearly all analogue servos will run away if you just power them up and enable them. In analogue mode, you are reliant on the controller to monitor the encoder, and adjust the control voltage to maintain position. This will be far more noticeable in torque mode with no load.

Speed mode is usually the most stable operating mode, whereas Torque mode is usually more unstable, however Torque mode bypasses any drive internal speed control/filtering which means the motor should respond quicker, and potentially result in a better tune.

Personally, I'd suggest starting with Speed mode and any default drive settings. Get the KFlop controlling the servo loop to maintain position, ensure limit switches are disabling the servos, and you have a functional physical E-stop button near. Then connect up the motors to the ballscrews, as they need to be loaded for tuning, and start tuning via the step response screen.

m_c, Ive been meaning to ask this basic question for a while... How does torque mode vs speed mode setup differ in kmotion? All i see is DAC servo output. Nothing specific about torque mode or speed mode? How do you specify which mode you are using?

Or do you even have too specify? Is it just a matter of selecting proper mode on the drive and then tuning accordingly? All the PID tuning info on the dynomotion website makes sense to me for torque mode, but i would have thought speed mode would have needed something different.

I bought my machine partially retrofitted and tuning was done so i missed all the fun. Trying to learn a little in case i ever have to replace a drive or motor with something different.

[Need TECH Help!]

Now that i am wiring encoder output of Servo Drives to Kanalog iam going to wire the Z index also, to JP2. When connecting to JP2, can there be 2 axis on one channel like this...

JP2-CHA_DIFFPLUS_4 ( X axis Z+ index)
JP2-CHA_DIFFMINUS_4 ( X axis Z- index)

JP2-CHB_DIFFPLUS_4 ( Y axis Z+ index)
JP2-CHB_DIFFPLUS_4 ( Y axis Z- index)

Troy

[Need TECH Help!]

Got some tests done in Kmotion. I am not getting much velocity or acceleration and when watching the DAC ouput it looks like it tops out at about .8 V. I tried entering a feed forward V of 1 or 2 and yee hah! It traveled in + and - direction REAL fast! Very thankful for estop switches. Anyone got any ideas? Here is some plots of Y axis.

[m_c]

m_c, Ive been meaning to ask this basic question for a while... How does torque mode vs speed mode setup differ in kmotion? All i see is DAC servo output. Nothing specific about torque mode or speed mode? How do you specify which mode you are using?

Or do you even have too specify? Is it just a matter of selecting proper mode on the drive and then tuning accordingly? All the PID tuning info on the dynomotion website makes sense to me for torque mode, but i would have thought speed mode would have needed something different.

I bought my machine partially retrofitted and tuning was done so i missed all the fun. Trying to learn a little in case i ever have to replace a drive or motor with something different.

In terms of difference in Kmotion/KFlop, ultimately in either mode the kflop looks at the motor position, and alters the output voltage to try and achieve the position the KFlop wants. The only difference is in tuning to allow for the different ways each mode responds.

I'm by no means any kind of expert in tuning, I've just learnt enough to understand the basic principles, and manage to get a reasonable tune on my machines.

[m_c]

Got some tests done in Kmotion. I am not getting much velocity or acceleration and when watching the DAC ouput it looks like it tops out at about .8 V. I tried entering a feed forward V of 1 or 2 and yee hah! It traveled in + and - direction REAL fast! Very thankful for estop switches. Anyone got any ideas? Here is some plots of Y axis.

The Output value is what limits the maximum DAC voltage. Maximum voltage limit is 1023, so 200 should limit voltage to about 2V. On your second screen grab, the left hand legend is for Position Error, and the right hand for Output. If you're ever unsure what each legend is for, uncheck the box at the top, the relevant scale will be hidden along with the plotted line.

Leave the Feed Forward values at zero. It should only ever be increased as a very last step, and even then only if all other tuning methods fail.

I start by setting the Output, Integrator, and Error to 1023.
Start by increasing P, until I get noticeable oscillation during/after a move (the move has to be big enough, so that the motor reaches full speed)
Then increase D to reduce the oscillation.
Repeat the above two steps until I can't anymore, then back P of until things become stable again.
Add a little bit I (0.05 is a good starting point - too much and things become very unstable very quickly!).
Keep increasing I until things go unstable, then back of P.

That should get you in the ball park for fine tuning things, at which point it's a case of doing lots of moves,and adjusting parameters to see how things respond, with the aim of reducing following error.
It's normally at this point if I can't seem to get a reasonable tune, I'll save the step/response data, and post it up with the settings in the hope Tom will give me some pointers.

[Need TECH Help!]

Ok. I have tried many different settings now and this is the best i have got so far. Axis retruns to 0 but there is about .010" overshoot that i cant stop. If i make 'I' gain 0 than there is no overshoot and plot looks smooth, but i can rotate servo with my hand and move axis about .003" before servo pushes back. With current setting this is as fast as axis will move without drive faulting with a Over Voltage, i need to get a regenerative resistor to stop this fault and be able to go faster.Here is plot with overshoot and a screen shot of DMM drive current config.

[mmurray70]

In terms of difference in Kmotion/KFlop, ultimately in either mode the kflop looks at the motor position, and alters the output voltage to try and achieve the position the KFlop wants. The only difference is in tuning to allow for the different ways each mode responds.

I'm by no means any kind of expert in tuning, I've just learnt enough to understand the basic principles, and manage to get a reasonable tune on my machines.

Ok so Kflop doesnt need to know which mode your using, you just tune accordingly for each mode. That makes sense.

So would speed mode be more likely to have larger position errors at higher speed? Im just thinking that it would need some error to drive output high just to cruise at a high speed without any load on it. Kinda sounds like a poor setup to me. Or maybe integral gain would correct for this in practice.

[mmurray70]

Ok. I have tried many different settings now and this is the best i have got so far. Axis retruns to 0 but there is about .010" overshoot that i cant stop. If i make 'I' gain 0 than there is no overshoot and plot looks smooth, but i can rotate servo with my hand and move axis about .003" before servo pushes back. With current setting this is as fast as axis will move without drive faulting with a Over Voltage, i need to get a regenerative resistor to stop this fault and be able to go faster.Here is plot with overshoot and a screen shot of DMM drive current config.

Your overshoot and lack of regenerative resistor could be related. If your running at close to max speed the drive cant really do much at all to slow motor with going over voltage. Wouldnt hurt to try current settings with the resistor added and see if it changes.