Motor capacitor lifetime 1

[Hewlett]

Our room air handling units have a 3 speed fan (controlled by a 3 relay output honeywell controller) with a 2uF capacitor.
About a year after installation, we started to suspect "something wrong", and found that most capacitors (measured with a regular multimeter) was around and below 1uF. We found some spare capacitors and started comparison of speed with different (new and old) capacitors. We found that speed was significantly lower (20-40%) with the old capacitors.
Vendor was contacted, and duly sent a team and replaced all capacitors....with some that should be "much better" - these should stay "on top" at least 3 years(!!) Vendor considered capacitors almost a consumable, that they ideally should be replaced every couple of years, but most customers didn't care if they lost a little efficiency and "run them to the fan stopped".

I'm aware many capacitors have a "limited" lifetime, but was not aware that capacitance would drop as quickly as this, and can not find any info indicating they would/should have such a short lifetime in vendors manuals.

For an individual unit I wouldn't care much, but we have more than 1000 units, so we probably need to incorporate "something" to our maintenance schedules....

What is your experience with lifetime on motor capacitors like this?
What is your practice on replacing running capacitors?
By spot check measurement? By time? When unit stops?

 

[laundry]

Hello Hewlett
Capacitors normaly last many years without any faults or deterioration in uf.The main cause is over heating, this causes the cap to dry out and loose its value or even short.leakage is another fault caused by the above,Also transients do cause cap damage. Because manufacturers tend to use cap rated voltages to near to supply voltage instead of slightly over rating the cap.But its all down to £.

Barry.

 

[ScottyUK]

A 2uF cap is unlikely to be electrolytic, so drying out should not be a problem. For preference the cap should be a self-healing type; metallised polypropylene film (MKP) would be a good choice.

One possible way to lose capacitance is if the mains has a lot of transients which are causing the cap to self-heal. Self-healing causes localised burn up of the metallisation without causing total failure. If a lot of self-healing is occurring then this would explain gradual capacitance loss. As Barry says, cost is a factor: marginally rated components or poorly manufactured components are likely causes. Choose a higher voltage rated type and buy it from a reputable manufacturer. I'm a big fan of Evox-Rifa and ICAR capacitors because they are reliable and have good service life in demanding applications. They cost a little more, but the benefits are worth it.



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

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

 

[Skogsgurra]

Agree. Have Rifa capacitors in similar applications that have been running since 1981. Never had any problem. I guess that the vendor has opted for a very cheap solution that will cost you lots of unnecessary maintenance. Your best solution is to have them all replaced with better ones and then forget about maintenance cost.

Gunnar Englund

http://www.gke.org

 

[electricpete]

This type application is not something I work with much.

Out of curiosity, how are capacitors used in speed control?

It is a type D motor or type B motor?
3-phase or 1-phase?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Hewlett]

Thanks! I was not aware of this "self-healing" effect in this cap type, and have looked up some articles on that.
The original capacitors was: Arcotronics 1.27.6CCA MKP 2µF +/- 5, 420V 10.000 h/Class B, 470V 3.000 H/Class C. I'll try to get some transient measurements done.

Pete: It's a single phase motor with a permanently connected capacitor. I don't have the motor or documentation where I am at the moment, but I think this is a "Permanent Split Capacitor" motor (Diagram:

http://www.electrical-contractor.net/ubb/Forum15/HTML/000084.html

) There's a specsheet for controller here:

http://europe.hbc.honeywell.com/pro...well.com/products/pdf/a74-2959-3ge51r0699.pdf

 

[Skogsgurra]

Pete,

These motors are mostly used in ventilator applications in the 1/4 to 3 or 4 HP. They have a high rotor resistance and are stator voltage controlled (slip increases when voltage goes down). It is normally done with triacs, but it seems that this application uses relays to switch voltage (somehow). A thought: the switching is a transient per se. You may need to have a "switch tolerant" (German: "Schaltfest") capacitor.

Gunnar Englund

http://www.gke.org

 

[Hewlett]

I have not had transient check done yet, but noted on datasheet for capacitor there's nted a maximum rate of change f 15V/uS. Supply is 230V and capacitors are rated >400, which do not look like "underrating" to me(?)
Is it the "fast change" that create excessive self healing?

 

[Hewlett]

Sorry about the double pasting of link above, and typo's...time to go home...
With "fast change" I mean the rate of rise that eventual transients cause.

 

[electricpete]

Thx Gunnar.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[ccjersey]

PSC blower motors usually have main windings with different taps for the different speeds. I always thought there were different numbers of poles, but maybe not since under lower than rated load, the motor doesn't slow down when switched to "low". I know voltage input does not vary. Must be magnetic flux that varies allowing more or less slip depending on which winding tap is used.

Jim

 

[Hewlett]

Transient measurements done, and as predicted by skogskurra the switching creates significant transients often >600V seldom >700V. Based on these figures, does excessive self-healing because of transients sound like "THE explanation"?
I think we're on a good track to get maker realize that a redesign/ redimentioning is needed.
Could someone enlighten me on the different "operating classes" commonly marked on capacitors:....."420V 10.000 h/Class B", "470V 3.000 H/Class C".....what does it tell?
Does it straightforward indicate average lifetime with the specified voltage across...?

BTW, In my quest for answers, I came over a site with a lot of well explained power quality related issues that at least got my attention for some hours....duly bookmarked and hereby shared:

http://www.marcspages.co.uk/pq/index.htm