Mondo solenoid - up in smoke?.

[itsmoked]

I have an electromagnet that is about 2 feet in diameter.
It has run for years without shutdown.

Recently it had water sprayed on it for several weeks due to a nozzle failure.

The machine was shut down and an overhaul done.

Three weeks later it's turned on and promptly lays waste to the SCR based power supply controlling the magnet.

I fix the controller and suspect the magnet but don't have immediate access to it.

I do the math and come up with an expected resistance for the magnet windings ~ about 2.7 ohms. I ask the user to measure the resistance of the coil. They state it's about 2.5 ohms. (It's fed thru slip rings.)

Being the shewed untrusting field EE that I am, I add a fuse block to the power supply's output circuit supplying the magnet.

The user installs the supply, turns it on, and the output fuse blows simultaneously.

Here's my thinking. Do you agree with me?

I suspect the moisture got to the windings and dropped the resistance.

Measurement with junk meters show that the lead-to-case resistance is about 1Mohm.

Do you think a bake-out would get this thing back on its feet?

Since it runs at about 130VDC is there even voltage room for a bake-out?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[bklauba]

itsmoked,

Its worth a try. Moisture is almost certainly the culprit. One thing to be careful of, esp. during these transition periods from one season to the next, is to place the baked and now cooling magnet in a dry area to reach room temp. I've seen stators check AOK when pretty fresh out of the oven, only to have the megger readings go south if cooled in a damp locale.

Boy, this thread is a almost boring, compared to some of those where you related events during your honeymoon. Maybe boring is good once in a while??

BK

 

[edison123]

Why sliprings ?

If it is a regulated supply, then you can start with a small voltage and measure the current, which would also slowly heat the coil. I suspect turn shorts, not ground fault.



Muthu

http://www.edison.co.in

 

[waross]

After the bake out, consider dipping and baking or vacuum impregnation. If the insulation is shot, drying will leave the winding susceptible to moisture damage in the future.
Do you have a crank type megger Keith? If so start cranking slowly and increase the speed slowly. If there is damage that doesn't show with a junk meter, you may feel the crank resistance and see the needle drop when the breakdown voltage is reached.
2.5 ohm vs. 2.7 ohm: cold vs. hot, no problem.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

9VDC from a cheap-o VOM is not a good predictor of insulation integrity, it will only see catastrophic failures. Hygroscopic intrusion into winding insulation needs a good megger, at least a 500V version for that.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[itsmoked]

I always think I've provided every pertinent detail and you guys prove me wrong...

As to bake-out I really mean subtle warming via the magnet's leads. Removing and reinstalling it takes more than 4 days(about 80 man hours). So no dipping Bill!

So my concern about the voltage, if the thing is wigging out at 110V could a variac set to 20V even work at warming it up or would you guess the 20V would short out too?

Would pointing a bunch of heat lamps at it do the job?

Slip rings: The magnet has to rotate continuously so it is fed thru carbon slip rings.

Bill I do not have a hand crank megger. Er maybe I do.. I have a friend who found one a couple of decades ago. Not sure if he still has it. I do have a wall powered giant beast of a meggar.

jraef; You suggest I pull out the Beast and give it a go?
Warning I've never used it before. How do you not injure insulation with a meggar? And what test voltage would I use for something that is rated at 150V?



Keith Cress
kcress -

http://www.flaminsystems.com

 

[ScottyUK]

Keith,

Assuming it's wound with regular magnet wire then 500V would be fine. No different to a motor winding.

If you have a big variac, or can get control of the input to the thyristor firing controller, then try winding the current up slowly and warm the winding up a bit. I doubt heat lamps will help much, but a big DC source might be better than an AC supply as you will not have the reactance to worry about. On the other hand make sure you don't upset Mr. Lenz by changing the current too quickly otherwise you might have quite a firework show!

I'm curious: what does anyone want with a 2' diameter solenoid?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[itsmoked]

what does anyone want with a 2' diameter solenoid?

A very large grinder chuck.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MikeHalloran]

I've used Loctite 290 to quiet noisy inductors without evacuation or disassembly. Just let it wick in and wait for it to cure, or kick the surface with a catalyst.

But those were _much_ smaller inductors.

I was thinking it would provide some dielectric strength where now there's just air. It would keep the wires from moving around. .. ever.







Mike Halloran
Pembroke Pines, FL, USA

 

[edison123]

Normally these electromagnets are completely sealed. If it is yes in your case, then how did the water get in ? If is is not sealed, then may be you can take out the coil and have a looksee ?

Muthu

http://www.edison.co.in

 

[rhatcher]

itsmoked,

Your original post seems to indicate that the machine ran for several weeks with water spray on it. Obviously there were no problems with the power supply or magnet at that time. If this is true, why would the magnet have failed since then?

Do you think it is possible that some wiring got crossed during the rebuild?

 

[itsmoked]

Hello rhatcher.

why would the magnet have failed since then?

Because that's what windings do!

They fail 'eventually' because of water absorption. It can take time.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[itsmoked]

My megger is dead.

A borrowed megger that is an ancient tube type but seems to setup and calibrate correctly and has a test 1Meg button. It goes to a 1Mohm reading when the button is pushed. I hooked a 3.9Mohm resistor across it and it read exactly 3.9Mohm. I hooked a 10Mohm resistor across it and it read 10Mohm exactly. If I hook it to the electromagnet it instantly pegs towards the low end.(one side to case one side to slip rings)

So I made a full bridge rectifier with a capacitor across the DC side to provide variable DC.

There is a voltmeter across the DC side of the bridge.

Then the bridge's plus side goes thru an ammeter to a slip ring, (hooked to the magnet coils).

The return back to the bridge goes only to the magnet's steel casing.


As I turn the Variac up and the voltage reaches 100.0V I measure a current of 250uA, that varies linearly with voltage.

What say you all?



Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Did I do the math right? I get 400,000 Ohms. Not great but I would not see that as an immediate cause of failure.
Has anyone checked the brush holders and the wiring to them?
Have you rechecked the winding resistance?
Try your setup across the winding and look for a discontinuity as you ramp up the voltage. I expect that you will not be able to get near full voltage/current with your test set, but the coil may breakdown at a voltage within the capability of your test rig.
Are you able to feed the magnet with AC in series with a load resistor (light bulb?) and measure the current and voltage accurately enough to get a bench-mark impedance?
As for drying, I would use heat lamps. I am afraid that enough current through that winding to dry it will mean that there may be enough voltage to do more damage before the winding is dry. Consider a safety high limit thermostat.
BTW there isn't a suppression diode in backwards somewhere? That would blow the fuse.

Yours

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[edison123]

Have you considered using 100 V AC lamp test to check for coil contiuity to ground ?

Muthu

http://www.edison.co.in

 

[itsmoked]

Thanks guys.

I keep thinking of using AC but expect so much capacitive coupling to the body as to be very confusing. Also the peaks would be 1.4x higher.

Bill; This magnet only runs on 120V.

Also the way these things are set up is a center and a ring on the bottom with two round brushes pushing up from underneath. Then it's all sitting in oil with some water contamination.

I'm trying to figure out why it's over-amping the power supply. I was told the brushes were messed up. One had a collapsed spring. Of course it's handed to Me with new brushes and springs and looks great.

(time passes)
OK Bill..

Controller is good for 350W @ 120VDC.

I measured it before and it was about 12 ohms.
I should've stopped there!

I just went out and measured the coil resistance again and now it reads 0.8ohms!!

120V / 0.8 = A Lot.

I'd say problem found. Thanks.

That's why I like to bounce this stuff off others. I got myopic on insulation failure to the case... It was really inter-winding insulation failure.


Keith Cress
kcress -

http://www.flaminsystems.com

 

[edison123]

Keith - My 3rd Nov 07:49 post

"I suspect turn shorts, not ground fault."

I toldja. :>)


Muthu

http://www.edison.co.in

 

[edison123]

Forgot this.



Muthu

http://www.edison.co.in

 

[waross]

Hi Keith;
Going from 12 Ohms to 0.8 Ohms?? Did you reverse the meter leads for a reality check? I always try to remember to do that. Those readings would be consistent with a hidden diode and reversed polarity after a rebuild.
But if you reverse the meter leads and still get 0.8 Ohms, get out the heat lamps.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

Hi Keith. Got time for pictures? If not, I understand, just RF this post.

Bill
--------------------
"Why not the best?"
Jimmy Carter