Magnetic Chuck Control Circuit Wanted

I was playing last night with mine and managed to fry my rectifier bridge when switching it off. Not a major problem as I have replacements on hand, but think I need to do some more research on inductive loads and improve the snubber circuit to protect the electronics.
 
Pete what were the specs on the rectifier you fried?
MS
 
markba633csi said:
Rick just look for an isolation transformer at about 300 watts and a bridge rectifier rated at about 400 volts at 10 amps. A couple small high voltage capacitors to protect the bridge from spikes. I can post a diagram if you need one.
Shouldn't cost more than a couple hundred, maybe less. The chuck should have a good solid ground.
If you need to adjust the magnetic force you would need to add a variac to the circuit (more cost).
Mark S.
ps a GFI would be a good idea too
Click to expand...
That would be GREAT
 
Ok give me a couple hours, I'll open my gimp program and whip up something.
MS
 
ps: I deleted my schematic, we have been discussing other ideas and I'm going to do another version.
MS 4/10/17
 
Last edited:
Mark,
I think you would need some sort of snubber circuit downstream of the switch or you will be getting arcing and voltage spikes when switching. An RC circuit or a Voltage Dependant Resister to provide an alternative path for the back EMF driven current on switching.
I would also add a large smoothing capacitor on the D.C. side - any AC ripple into the chuck is going to cause some magnetic hysteresis which will lead to additional heating. Not a consern for a short demag cycle, but may impact tolerances if using it on a surface grinder and the part is there for a few minutes. One of my goals is for the D.C. Output to be as smooth as possible for this reason.

My rectifier bridge was a KBPC5010. Good for 1000v and 50A so I thought it was overkilll and I have about 20 of them in a draw so they were on hand. And they were only a dollar each from memory. May of just been a bad one, or may of got over 1000v switching spike.

I was thinking of actually measuring the inductance of my chuck as in reality I'm not sure how big an inductor it actually is - I'm just assuming very big.
 
Pete: I agree about the need for a snubber. When you mentioned hysteresis, are you concerned about the part heating up or the chuck (or both)? Hadn't considered that.
The large cap would have to be in the circuit for DC mode only of course. It may have an effect on the waveform in AC mode though- here's where we start getting into second and third-order effects. Good old low efficiency linear. It would be good to know how big a kick the chuck makes when you cut the juice. I'm thinking a voltage clamping device would be better than an RC network.
Rick: If you need help picking out a variac, I saw several on Ebay last night for around 30-40$ incl. shipping. I can send you some links while Pete and me fine-tune the circuit. You would be building this thing in a box of some kind so you would want a "bare" variac not one that's already in an enclosure- saves money and allows you to see that the windings are ok. Variacs often get abused/overloaded/shorted and develop a burned spot on the wire-
I also saw a few iso transformers in the same price range.
Mark S.
 
Mark, wouldn't those caps in your drawing introduce ripple into the DC?
 
Mark, my added complication is im trying to use an scr on the ac side for the voltage control, and I'm chopping 230v ac to start with. To get down to an average 110Vdc D.C. I'm left with a very peaky waveform. Then when I include a capacitator to smooth it out it starts impacting the firing circuit.
Think I may need to step down with a transformer to closer to the voltage I need so I can tweak it with the scr and get a larger firing time - or change to a variac.

One of the disadvantages I read about electromag chucks was the heating and that causing expansion and difficulties in holding tolerances - have no idea how much of a problem it is in the real world, but thought any significant ac ripple on the dc may agrivate the problem.
 
You must log in or register to reply here.
Back
Top