HP UHU at 250V on a BIG GAMBIN 3C Mill

[jcbm2004]

Hello to all!

I would like to share my experience on installing an HP UHU controller on a big Gambin3C mill.

I got this machine electrically out of order some years ago and I aim to convert it to CNC.
Initially all the feed movements were motorised using a big 5 KW CC motor under 250V. The original electronics was based on thirystors for slaving the motor's speed using a thachometer attached to the motor.
This system was out of order and was unsuitable for CNC use.
So I've build an HP UHU drive kit(ordered to Paul, Great thanks Paul!), changing the originals IRFP 264 by IXFH 30N50, suitable up to 500V and changing the original 100K resistors on IR21844 to 150K to increase the deadtimes and minimize cross conduction hazards.
Today the drive is mounted on the machine and the first tests including machining were OK.
You'll see some DIY boards used to generate suitable signals to the UHU and to control most of the functionalities of the machine in a conventional use.
The next step is to keep the actual motor and control for Z axis and motorize with servos X and Y axis.
But before that I need to find a ballscrew for the X axis and some time of course.
English is not my native language so be indulgent please!
Here are some photos of the set up on the link below.

http://www.usinages.com/avances-auto...n3c-t9761.html

Regards.
Jean-Claude

[H.O]

Jean-Claude,
Very nice! It's always nice to see others using the HP-UHU. Please keep us posted on your progress with this!!

What do you have the current limit set to on the HP-UHU? Any signs of heating? Reason for the question is that as far as I can see the IXFH30N50 has 2.5 times higher RdsON than the IRFP264N so it should generate considerably more heat. Shouldn't be a problem if you supply adequate cooling though.

Don't worry about your english!
/Henrik.

[jcbm2004]

Hello Henrik.
Yes RDSON is higher.
In standard use I have around 10A of current consumption so I set for moment maximum current at 15A.
There is a heat dissipator with vertical fins in the back of the UHU, a little smaller than the board. And I needed to add an electric fan, you can see it temporary mounted on the photo( of course the fan is better installed today....) with this fan the heat dissipator remains around 40 to 50 degrees while machining with 28 outside temperature.
I'm planning to add a temperature indicator on the dissipator and a DC current indicator to monitor all that under various machining conditions.
Roughly the total heat power losses would be below 50W in the above conditions.
I have to add that this engine is a rotor disc type, there is a 5 mH inductance in serie with the engine( located in the top left in the electric panel, looking like a transformer), probably compensating a low inductance rotor. Trials to drive the engine without it were not satisfying and the signal on motor connections was very much distorted .
I will continue to post on CNCZONE about the progress of this CNC conversion.
But be patient as my workshop is 700Km away from my living place so I'm going there only for some holidays.
Have a nice day!

Jean-Claude

[tenmetalman]

jcbm2004,
very interesting,keep us posted. be careful of servo feedback. I've been trying a similar change and thought it was going to work. was testing using low voltage on the servo & when I turned the servo by hand & blew one of the MOSFETS. the noise was impressive ! one of the legs was blown totally out of the mosfet !
Paul

[jcbm2004]

Hello Paul.

As you can see it took long time since I bought the kit to the setting on the machine, thank you for your job!
I don't understand what happened to your MOS. Was it a 30N50 also?
What kind of feedback do you mean? Is it the voltage regenerative feedback due to induction?
Did your bridge suffered from motor oscillations?

For the tests, I first tried at 50V, of course the response was poor.
Then I tried directly at 250V and started to send command to the motor without clutching on table movements. As everything was OK, I continued the tests moving the tables, including the Z axis which is the heaviest (the machine weight is 4350 Kg ) and then proceeded to some machining tests for an hour.
I want to say that for the moment , the machine is not working as a CNC.
The step and dir signals are generated via an homemade control board using a 16F887 using speed up and speed down ramps.
I didn't check for moment about any tracking errors and the motor probably needs better tuning using the UHU parameters.
All that I know is that for the actual settings, everything is working for using the mill in a conventional way, which was the first objective.
More tests will be done next time I go there (perhaps september).
I will keep you informed.

Many thanks to all people who participate to Uhu HP project, especially Kreutz!

Regards.

Jean-claude

[tenmetalman]

Jean-claude,
Hey, anymore progress ? Keep in mind that I barely know enough to keep myself in trouble. My project languishes for Looong periods of time due to honey dues & paying jobs then I try to pick up the strings and go at it again. I had tried the 75NF20. just soldered a set in and applied 12vdc. the servo wanted to turn slowly in one direction (reversible by changing the servo leads) and when I stopped the shaft my hand and turned it backward maybe half a rev. it blow the MOSFET. I didn't put any meters, etc on the board. I did have the terminal program running and looked at the wave. ? ? ?
Paul

[jcbm2004]

Hello Paul.

No news on that project.
I'm living 700km away from my workshop...And I think that I won't be able to go there again before september or october.
But I'm looking for a buying some place close to my home.
I don't understand why you MOSFET has blown. Perhaps Kreutz could explain that?
Regards.
Jean-Claude

[tenmetalman]

Jean-Claude, How you doing ? I was wondering if you've had a chance to do anything further with your CNC project. I found my problem when I tried the different MOSFET, I had a ground that I was unaware of the was causing the MOSFET's to blow. several other things going on right now so my project is dead in the water for now.
Later
Paul

[jcbm2004]

Hello Paul.
Finally I will go back there in december for few days.
It will be too short for working a lot on the CNC conversion.
Now I've got two machinetools ballscrews to adapt on the mill for X and Y axis.
But the adaptation will be a quite long work as it needs to dismantle the tables and machine some parts.
I'll be happy if I can find time to do some machining work on the mill .
I'll try first to fine tune the servo and then to make some test with faster acceleration and deceleration ramps.
I have also to install DROS, and a digital spindle tachometer.
Probably at least one half of a day will be spent to move a Gambin 10N mill that is sold.
I'm looking forward to the day I'll find a place to move all that machines close to my home...
Despite this, I've installed an Emco compact 8 lathe in a very small room (4m2) of my flat in Toulon, it's batter than nothing.
I Hope you'll be able to come back to your projects very soon.
Best regards.
Jean-Claude

[tenmetalman]

Jean-Claude, were you ever able to make any further progress with your changes to the UHU.HIGHPOWER drive. H.O. has a new chip & module that increases the speed at which the drive can process & thereby control the rapid travis & feeds. it has firmware that allows some lead & following error between the machine & program (which is adjustable) which to me would facilitate some fast roughing at minimum. not a very good description on my part, but go look at his thread "Servo processor "upgrade module" - looking for volonteers."
Paul

[jcbm2004]

Hello Paul.
Finally after pushing up the current limit to 24A my FETS has exploded with big noise and lights....
So I've decided to switch to the YAPSC board from MaxMod as the code is open source.
I'm doing some adaptations to the power stage anyway to withstand 250V and 25A. I don't have enough time to move forward fast on that project.
Sorry for the late answer to your question.
regards.
Jean-Claude

[tenmetalman]

Hey, How are you doing with the new drives ?

Paul