Line Notching

Status
Not open for further replies.
M

MarkiusSchu

Member
Location
Chicago, Illinois, United States
Hello all,

I was recently asked to look at the power line feeding a 480VAC 1ø magnetic chuck, which at times refuses to work.
On attaching an oscilloscope to the line, I saw notching occurring, and eventually traced it back to a couple of induction heaters.
It was suggested to me to add a line reactor to clean up the signal feeding the chuck controller.

As a preliminary check, I hooked up a motor drive (sans motor) and took readings on the line with the line reactor, with an EMI filter, and with without either of these two things.
I have attached a picture of each.

Now, it appears to me that the notching amplitude has actually increased with both the reactor and the filter.
If you would, please guide me in the right direction.

Thanks in advance,
Mark
 

Attachments

  • EMI Filter.jpg
    EMI Filter.jpg
    144.7 KB · Views: 1
  • Line Reactor.jpg
    Line Reactor.jpg
    146.7 KB · Views: 1
  • Power Line.jpg
    Power Line.jpg
    146.5 KB · Views: 1
180410-0931 EDT

MarkiusSchu:

I would describe your notching as very short duration transients.

The transients you see are probably spaced 60 degrees apart. Thus, likely result from a 3 phase load that has switching elements, like diodes, SCRs, Triacs. ignitrons, etc., that switch on or near the peak voltage of their respective phase.

Note that the first and second spikes are negative, and the third is positive relative to your base sine wave.

I assume you measured voltage. Might be useful to see the full cycle.

Your magnetic chuck is really a large inductor. If its controller is simply a a set of relay contacts in series with the chuck, then I doubt you will see much of a blip in the chuck current from the voltage transients. The voltage transients are very short relative to the sine wave. Current thru the chuck is what does the chuck's work.

You are going to need to provide more information. The time constant of an inductive chuck is fairly long compared to 1/2 cycle, 8.3 mS. Is the chuck DC excited? Most likely. How is that DC produced?

Look at the AC current to the chuck. If a DC chuck, then look at its DC current as well. A DC chuck will have some type of rectifier between its AC input and the chuck.

.
 
Like gar said, need to see currents.

Have seen similar on diesel-gen supplied large SCR controlled battery charger, due to SCR commutation (e.g long reverse recovery of the SCR acting like a short). Indicates that the source is a high impedance? Is the 480 derived from a small transformer that hs too high of leakage inductance? Your added line reactor adds impedance which leads to the larger voltage drop in the notch.

Only thing I can think of that would cause the chuck to not work is noise causing the chuck control circuit to not work. Chuck model and type? e.g. details like gar asked for.
 
Gar/Junkhound:

Thank you both for the quick replies.
Yes, the induction furnaces (one 500kW and one 600kW) both have SCR banks, and I'm sure this is what is causing those transients.

The controller is an Earth-Chain MagVISE EEPM-C4.
I'm currently attempting to contact the chuck/chuck controller manufacturer to find out more information.
I am also waiting for the opportunity to monitor the chuck currents & voltage as they are being activated/deactivated (They are running a job in the machining center right now).
I will post more info as I get it.

Mark
 
MarkiusSchu said:
Hello all,

I was recently asked to look at the power line feeding a 480VAC 1ø magnetic chuck, which at times refuses to work.
On attaching an oscilloscope to the line, I saw notching occurring, and eventually traced it back to a couple of induction heaters.
It was suggested to me to add a line reactor to clean up the signal feeding the chuck controller.

As a preliminary check, I hooked up a motor drive (sans motor) and took readings on the line with the line reactor, with an EMI filter, and with without either of these two things.
I have attached a picture of each.

Now, it appears to me that the notching amplitude has actually increased with both the reactor and the filter.
If you would, please guide me in the right direction.

Thanks in advance,
Mark
Click to expand...
Are the induction heaters phase controlled with SCRs?
That's typically what you would see if they are.
 
Besoeker said:
Are the induction heaters phase controlled with SCRs?
That's typically what you would see if they are.
Click to expand...
I'd say from the scope traces that's most likely the case.

MarkiusSchu,
I agree with junkhound, the likely explanation as to the chuck having issues is that this line notching is causing problems in your control circuitry for it. That chuck appears to be a bunch of small inductive coils that likely use phase controlled SCRs themselves, so the line notching caused by the other large SCR controllers is interfering with the gating circuit for the SCRs in the chuck controller.

The problem is, putting a line reactor ahead of the CHUCK is not the solution, because the chuck is not the SOURCE of the problem. You will need to put 3% reactors ahead of the HEATING controllers. Adding the reactors will dampen the notching. And don't think that if 3% is good, 5% is better, you can create a new problem by having too much impedance.
 
MarkiusSchu said:
Jraef:

Yes, my original thought was to put the line reactor in front of each induction furnace.
However, I wasn't sure if the attenuation of even a 3% impedance reactor might cause problems/significant changes with our heating process.
Click to expand...

I doubt it will hurt it any but it won't solve your problem.
 
Status
Not open for further replies.
Top