Cooling an electromagnet with Deionized Water

[ChEMatt]

I have received an electromagnet that must be cooled with deionized water; the water passes directly through the square conductor wire so it is in direct contact with the copper throughout the coil. The spec given by the vendor for resistivity is 5 Mohm*cm.

The vendor selling me the deionized water system is telling me that the copper coil will be corroded by the DI water. The electromagnet folks are telling me I'll be "just fine". I have no experience with either.

I hope this is the right forum for this question. In you guys' experience, what kind of water is appropriate for cooling a copper coil electromagnet? How do I prevent corrosion but maintain resistivity?

Onwards,

Matt

 

[MikeHalloran]

I agree with the DI water guy.

Has the magnet vendor been in business long?

What are their warranty terms?



Mike Halloran
Pembroke Pines, FL, USA

 

[ChEMatt]

The magnet vendor has been in business for 40 years. They recommend using an oxygen scavenger and an algecide. The water guy says that the two chemicals will contaminate the DI water and I will "no longer have DI water".

The boss of the DI water guy said that the copper will passivate quickly. I could probably get a copper analysis of the water if needed to see what's happening.

Will the chemicals reduce the resistivity of the water?

Onwards,

Matt

 

[MikeHalloran]

Ah. You'll have a soup that comprises in part DI water, but it won't be DI anymore. Which is probably good. The reason DI is so corrosive is that it's not buffered, so tiny amounts of contaminants can drive the pH radically, even locally.

But if you are shooting for soup of a particular composition, then it's good to start with DI water, just for assurance that the water didn't add anything you didn't want.

So make the soup exactly as the magnet vendor says, and don't change it until they tell you to.





Mike Halloran
Pembroke Pines, FL, USA

 

[ChEMatt]

Part of the problem I'm having is that the DI water goes non-resistive as soon as it hits the tank in the chiller (started a separate thread on that one: thread164-282129 if you're interested) so I have this spec of 5 Mohm-cm that I can't meet to begin with.

If I could actually get a receiver full of resistive water to begin with, I would feel much better about adding the chemicals if the water then goes non-resistive.

Onwards,

Matt

 

[MikeHalloran]

You may have to pickle that tank to get it clean.

Does the magnet vendor ask that you mix the oxygen scavenger and algaecide with DI water (and then not worry about the mix), or does he actually require that the mixture be resistive?



Mike Halloran
Pembroke Pines, FL, USA

 

[itsmoked]

Part of the problem is that there are difference levels of "deionized". Jugs of DI you buy at the store and lab grade. The DI vendors often have to provide for commercial processes that demand seriously DI'd water that ends up being quite corrosive. They spend a lot of time fighting that specific battle.

On the other hand the copper guys are thinking, "just not any old tap water".

I'd find the lowest grade DI you can and add the required ingredients and proceed.


You would want to get that chiller tank cleaned well and then possibly pickled.

Depending on the cost to load up the system, you might just load it up and run it for a few days. Drain it and reload it for the long run. That will give anything that's going to be going into solution time to do it. Then the second time around there's little or none left. Sort of pickle-in-service.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[sreid]

You chiller manufacturer will have recommendations and he won't like DI Water. I use Hexid from ATC;

http://www.app-therm.com/product_hexid.asp

 

[sreid]

Also take alook at TECA Thermopile Chillers [solid state].

http://www.teca.us/

 

[Warpspeed]

The problem is not so much the water itself, but the electrical potentials between the ends of your electromagnet and various other parts of the (wetted) cooling circuit.

You end up with two electodes at high differential voltage, and the hopefully non conductive coolant, doing any ion migration on you. You definitely need a PH buffer !

It can be done, and is being done all the time in all sorts of very high power density water cooled electrical equipment.
But it gives me the horrors just thinking about it.

The last high powered RF induction heater I built used automatic transmission fluid as the coolant. No corrosion, no electrical leakage. The specific heat, and thermal conductivity of oil are lower, maybe 60%, but increased volume flow can more than make up for that.