What's new
What's new

Balancing my RFC, I keep blowing a start capacitor. Could voltage be an issue?

setlab

Aluminum
Joined
Aug 19, 2007
Location
Florida
I've built a 30hp RFC using the Fitch Williams converter plans posted here and am running into an issue when balancing it. As I gradually add Cp capacitance and record the voltages based on the plans chart I'll reach my recommended stop point when Vbc is approximately 1.03Vab (for capacitor Cp). At that point during startup the first start capacitor in my start capacitor bundle blows its top off.

Reviewing my table of final voltage values I am getting: Vab = 243 Vac = 249 and Vbc = 229.6

My thinking is that my first start capacitor is blowing up because it is rated between 220v-250v and maybe for a moment there is a greater than 250v going into it on startup?

Another theory I have is during the table tabulation I started the RFC too many times and the capacitor simply was not allowed proper time to cool off? At the time the capacitor blew I was on start number 6 with maybe a minute between starts.


What do you guys think could be going on?

If it matters I'm using about a 2500uF start capacitor bundle, roughly 85uF per motor HP.

Plans I used: Fitch Williams Converter Design & Balancing Instructions
 
I believe it's usually recommended to use ~370VAC capacitors with a 240VAC supply voltage, as the combination of inductance and capacitance can cause each component to see more voltage than the original supply voltage due to resonance. Running components very near to their limits is not recommended.

I assume you know to use AC caps, not DC.
 
Yes, the higher rated caps are the better ones to use.

But if the same one is failing each time, you might look at connections. The start cap is only supposed to be "in-circuit" for a short time. If it stays in too long, it may indeed pop.

Be sure the whole lot is getting disconnected after the start.
 
The voltage rating isn't great, but I think the amount of starts per hour is the bigger issue. It's passing too much current for too long for it to handle. How much capacitance does a single capacitor have?
 
I believe it's usually recommended to use ~370VAC capacitors with a 240VAC supply voltage, as the combination of inductance and capacitance can cause each component to see more voltage than the original supply voltage due to resonance. Running components very near to their limits is not recommended.

I assume you know to use AC caps, not DC.

+1.

I use 4XX rated.

They cost a good deal more as a "percentage", but not MUCH more as to it still being rather "small money" ..... when bought from the major HVAC industry distributors, Southern USA, Gulf Coast (and Florida..) that move them by the carload every season to the commercial and residential Air-con and cold-storage reefer installation and maintenance trade.

May not be an RPC under every roof.. but fridges, freezers, supermart food cases, heat pumps, air-con, summer cottage to skyscraper?

"Ubiquitous" do yah?

The goods will be "fresher", and even American Made.

:D
 
As to the root cause for you detonation, probably "both." Excessive starts in a short time, and undersized (working voltage that is) component.
 
Be aware that the higher capacitance a single capacitor has the more heat it is dissipating. If you have high capacity, physically small capacitors they'll have a worse service life.
 
Higher voltage and lower duty cycle.

Electrolytic start caps are intended for short-time operation. They're made as cheaply as possible with a very low 'Q' factor. Cycling them repeatedly will cause them to overheat and vent.
 
Looks like the problem is solved now, looking threw some of my auction score boxes I found some higher voltage capacitors and so far my RFC has worked just as it should with them installed.
 
The voltage rating isn't great, but I think the amount of starts per hour is the bigger issue. ..........

That's a point, although the starts are presumably un-loaded, so not "hard starts", and have some time between to adjust capacitance, etc.

I've never had any issues that way, although I am sure it has reduced overall life expectancy of the part to do a lot of starts in a short time..

The OP is having apparently new parts last only a relatively few starts. And it may be the same position in the wiring that gets the failure every time.

At that point, one wants to know if the others have extra resistance in their wiring, or if maybe they are not even working.... they might have an open wire or mis-connection that does not even put them in the circuit.

If that first one can just barely start the thing, it may be an extended start that really takes it out of the one part. I can see that lasting only a short time, especially with repeated starts.
 
That's a point, although the starts are presumably un-loaded, so not "hard starts", and have some time between to adjust capacitance, etc.

I've never had any issues that way, although I am sure it has reduced overall life expectancy of the part to do a lot of starts in a short time..

The OP is having apparently new parts last only a relatively few starts. And it may be the same position in the wiring that gets the failure every time.

At that point, one wants to know if the others have extra resistance in their wiring, or if maybe they are not even working.... they might have an open wire or mis-connection that does not even put them in the circuit.

If that first one can just barely start the thing, it may be an extended start that really takes it out of the one part. I can see that lasting only a short time, especially with repeated starts.

Those high capacitance small size caps pass a lot more current and have higher resistance, so dissipate an exponential amount of power. (Assuming his capacitors are physically small).

Starting a 30hp RPC will draw some considerable power I think, even unloaded.
 
I'm kinda wondering why the RPC is being started many time in a short period? Is it an RPC or 3 phase motor being run on single phase and the Capacitors a static conversion?
 
If he is trying to balance the thing, he will be starting, measuring, then shutting down, changing the balance parts, then starting again etc. I'd think no more than a few times, but he may be either trying to get it perfect , or otherwise "chasing his tail" and be starting a lot more.

In any case, that should not kill capacitors normally.

.............

Starting a 30hp RPC will draw some considerable power I think, even unloaded.

But unloaded they start faster, so the total energy dumped into the cap is smaller.

With just barely enough capacitance to get it rolling over, the start takes a lot longer, in any condition. That could easily explain why he is killing capacitors.
 
Well I got everything balanced as well as I can with my capacitor supply on hand. I'm surprised I got it to run on a 50a breaker. When I start it up all the shop lights dim for a breif second but when it's running with no load it's only drawing a little less than 8 amps if my cheap clamp on amp meter is to be trusted.
 
Well I got everything balanced as well as I can with my capacitor supply on hand. I'm surprised I got it to run on a 50a breaker. When I start it up all the shop lights dim for a breif second but when it's running with no load it's only drawing a little less than 8 amps if my cheap clamp on amp meter is to be trusted.

It actually is NOT to be trusted, as a matter of physics!

That, too, has been "done to death", and "Right here, on PM", w/r power-factor and true "POWER" measuring, not just Volt & (apparent) Amp measuring.

No fear.

It doesn't actually MATTER enough to actually matter.

:)

Now that you "got it"? JF "Run what you GOT!"
 
If you need any more caps I have some run capacitors I bought for a welder single phase conversion that I’m not planning to use now. 6- 35uf and 2- 60uf all 440v. Pm me if interested.
 
It actually is NOT to be trusted, as a matter of physics!

................

Nonsense...... if it measures current, that current is actually flowing.

If the meter is inaccurate, that is a different matter. But do not be fooled.... an accurate meter, whether average or true rms, is telling the truth about the current it is designed to see.
 
Strostkovy...... BS is to be squashed.... We do not DO BS here.

................

The physics haven't changed.

Correct...... Clamp on measures the field resulting from the current. If it does so accurately, then that current IS FLOWING.

You don't agree? Argue with the physicists, not with me. You and Rozen are off base.

The breaker is going to respond to the current, whether that current is in-phase, or out of phase with the voltage. It cannot help doing so... that is what it does.

Ok, back out of la-la-bullcrap land.

What you CAN do about excess current in the breaker, which you may already have done, is to install power factor correction capacitors (PFC capacitors).

If you did that, then the out-of phase current is nearly all gone, the "PFC" capacitors cancel it. That is a valuable technique for getting an RPC to operate on a limited breaker, so long as the breaker is not popping from the capacitor surge at turn-on. It takes the idler motor magnetizing current out of the picture as far as the breaker is concerned.

The "idle" current in a motor that is running with no load, is mostly "magnetizing current" that is out of phase with voltage. it can be 30 to 50 percent of full load amps. Power factor may be as low as 0.1, so it represents very little actual power.

A 30 HP motor running no load might draw as much as 25 to 30A, possibly more, depending. If your idler is drawing 8, you likely have the power factor corrected up considerably, per the Fitch Williams design. A W.A.G. would put the PF around 0.6 or 0.7 looking from the breaker. Not bad. You do NOT want to go over 0.9.

Incidentally, if you have a motor controller on that motor, the PFC capacitors have to go before it, because the overload is figured for the bare motor current, not the PFC-corrected current.

How about your capacitor blowing issue? Any movement there?

Another thought came to me, late, but relevant.

That voltage issue...... it may be a symptom, if true, of a problem.

The way the "balance" capacitors work is to cancel the inductance of the idler motor. But, the criterion for cancellation is that BOTH the inductance and the capacitance "disappear" when the cancellation is exact. That condition is also known as "resonance", in this case a series resonance.

You DO NOT WANT that, because it causes a near short across the line, with high line-to-line voltages on the capacitor, and heavy current flow. Just what you do not want. It can cause the exact problem you see.

Is that what you have (or had)? We don't know, but it sounds possible. Only you can know for sure.

If you have that condition, it means you have over-corrected, with too much capacitance, and have gotten too close to "resonance" in the unloaded condition. That can happen if trying to correct too far in the LOADED condition, with the load motor running.

The load motor "damps" the resonance, so you will NOT have any idea what the unloaded condition will be. It is easy to become surprised by the high voltages, and blow parts.

The usual system, and I think the Fitch Williams plan says this, is to correct UNLOADED, to no more than 10% over the incoming voltage, and accept whatever drop occurs under load. A better number is 8% over.

if you try to correct too perfectly in the loaded condition, you may have radically high voltages unloaded, and very heavy currents. Always do your correction UNloaded.

If you feel you MUST correct more, then really, you probably actually need a larger idler. But you CAN do one thing...... you can connect the added capacitors, the ones in addition to what does 10% high unloaded, AFTER the start contactor of the LOAD motor. Not after the overcurrent sensor, but after the start contactor.

In that way, you only have them connected when the load motor is on, and there is probably no issue with the no-load voltages of the idler, because when the load motor is disconnected, so is the added capacitance.

(the contactor does need to be able to handle the inrush to them)
 








 
Back
Top