Mag drive pump behaving inconsistently

[obidi]

Mag drive pump tested to 104 Volts (minimum. spec) fails in the field after a brief use. No visible component damage. When the 1400 Gauss magnet is replaced with another 1400 Gauss (or higher)magnet it consistently works. The replaced magnet consistently has a reduced Gauss reading following failure even though the pump passed rigorous tests prior to field placement.

I would appreciate any insight into this failure which we are finding randomly with the pumps.

Thanks.

tyO

 

[TenPenny]

What temperature are you running at?

Any chance of dry running? What are you pumping?

 

[Artisi]

Sounds like excessive heat being generated in the magnet.
Consider TenPenny's comments.

 

[obidi]

Thanks TenPenny, Artisi:

It fails at 70 deg. F, not just at elevated temp. And, in some cases, it fails as soon as it is installed, after successful pre-field test.

 

[Artisi]

I am not that experienced with mag drive units, but I know running dry can result in very high temp being generated in the mag-drive section.

 

[obidi]

I'm pumping aqueous solution with chemistry --nothing heavy. And, the pump is always immersed. I don't think it can run dry under such condition. Can it?

 

[MikeHalloran]

IMMERSED? Don't mag drives normally run in air? I mean the driving part?





Mike Halloran
Pembroke Pines, FL, USA

 

[TenPenny]

Every mag drive failure that I have seen falls into two cateogories:
-Dry running which overheats magnets or melts pump
-Product crystallization on shut down (ie, caustic soda)which destroys bearings

Your case seems to involve the magnets becoming demagnetized, which I would think could be caused by overheating, or some odd properties of the fluid, or an environmental/electrical issue.

Off the top of my head, at the bottom of this bottle of wine, I can't think of any obvious causes. Can you provide more details?

 

[ccfowler]

In your situation, I would want to stop and reconsider everything that I had been assuming or presuming to be "impossible." I would want to check everything, and not just the pump. What about start-up or transient operating conditions? What about piping loadings imposed on the pump? What about motor torque vs. pump inertia and torque requirements?

Could there be some conditions when there could be slippage? Perhaps a much stronger magnetic coupling is actually required for your application.

 

[EdStainless]

In most cases a small amount of the pumped fluid is circulated to keep the magnets cooled. There are usually both temp and flow switches in this loop since it is critical.
I have seen these reach some high temps very quickly.

The failure modes are either high temp or slippage.

How are you loading the units in the pre-test? Using a pump or a dyno? How are you measuring the magnet strength?

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[rmw]

Once you put the pump on line (after testing) is it possible that the process is dead heading the pump? That would cause the fluid in the magnet can to overheat and flash which would cause pump failure.

rmw

 

[jet1749]

Seen this kind of thing before on magdrives, several answers listed already. My experience included reduced magnetism due to the use of stainless steel containment cans. Low flow conditions leading to high temperature at the inner magnetic rotor (below a certain temperature the magnetic strength can recover, otherwise the magnets are toast) Also worth noting that the kw rating of the coupling should comfortably exceed both the hydraulic kw of the pump and the motor kw, otherwise decoupling of the magnetic drive will damage the magnets. May be worthwhile checking the coupling transmission torque capability (breakout torque)before installing in the field

 

[Artisi]

Good point from jet1749 - re decoupling of the magnets, a point you should check carefully and could well be the problem.

 

[obidi]

Thanks for your responses:
jet1749 brought up some points that we're yet to look into.

Thanks, all.

 

[pumpking]

What type of pump and who's manufacture ?

Most common things to effect magnetism of magnets is excessive temperature - we use Somarium Cobalt magnets on metallic pumps which handle tmperatures safely up to 320'C - they can however have a temporary loss in magnetism at temperatures exceeding 260'C and so we have to compensate by adding more magnets. This returns however when returning to lower operating temperatures.

If the pump was to be 'flashing' due to dead heading, you would see additional evidence of wear and damage to bearing material, and possibly face fracture of the ceramic or Silicon faces due to thermal shock etc. The pumped product not only has to lubricate the bearings, but also take away heat generated by eddy current effect (an energy generated by passing magnetic currents through stationery metals). Previous writer was correct, this is worse in Stainless Steel, but you will find most people use Hastelloy for the can as this gives the best efficiency against eddy current loss v cost ratio.

A cheap magnetic drive pump usually consists of low quality magnet materials, which have limits of temperature, limits of torque they can transmit and limits to product density. We often compete against 'cheap pumps' but again, win in the long term, as lifecycle cost is poor.

See if you can find out the material of the magnets from the manufacturer, also see if there is any bearing damage, and also is there is any discolouration of the isolation shell/can which can give many other clues as to what is happening

there is more information available on

http://www.cdrpumps.co.uk

if you need further assistance.