Motor run capacitor mystery

[Peabody]

My central air system stopped cooling last night, and this morning I picked up a replacement dual run capacitor for the compressor, installed it, and now everything works fine.  I didn't have a meter that measures capacitance, so my decision to replace the run capacitor was the same as looking for lost keys under the street light - it was the only thing I could do myself that might fix the problem.

But later today I found a "serial capacitance meter" sketch for Arduino, which actually works pretty well, and proceeded to test the old capacitor just to confirm it was bad.  But it tests fine.  The cap is nominally 35/5ufd, and tests 34/5ufd with the Arduino meter.

Since none of the original connections to the cap were loose, I'm wondering what's going on.  Of course the Arduino "meter" tests at 5VDC, whereas the cap will experience 240VAC in the compressor.  Is it possible for the cap to test ok at 5VDC, but still fail to work in service?  I guess the definitive answer would come from putting the old one back in, but I don't know if I want to know that badly.

Anyway, could someone with experience in this area explain what's going on here?  As should be apparent by now, I'm not exactly an HVAC kinda guy.  Should I keep the old cap as a spare, or toss it?  By the way, the old cap is only two years old.  It's an AmRad brand, made in the USA.  Short of putting it back in, is there a way to test it at higher voltage, at least at 120VAC?

[floobydust]

Is is oil-filled?

[Psi]

Maybe the capacity is ok but the ESR is bad and that is causing extra resistance at AC and limiting current.

That said, whenever I've had motor caps fail it's always been from low uF due to too many self-healing cycles blowing holes in the plates. Actually, now that I think about it that's not true. I once had a cap that had a loose internal connection inside the cap casing. The bolt terminal would only intermittently connect to the cap inside.
It was actually a standard looking smaller TH cap inside a bigger plastic casing with terminals on it. (Dodgy china motor cap)

[Xena E]

Not knowing your system I will offer as a possibility that there could have been an transitory stop start that overloaded the motor, and cycling the power would have been all that was needed for a reset, (particularly if your system has a thermal starter)
 

My advice is to just toss the old one in the can. I consider them service items as they will eventually fail anyhow.
X

[Benta]

Are you sure it's a run/run capacitor?
Those values indicate more a start/run cap, in which case the centrifugal switch could play a role. Perhaps your working on the motor got it working again?

[Peabody]

I think it is oil-filled.  It says "1P - non-flammable - 37ml".  It's one of these with a CPT terminal:

https://amradmanufacturing.com/products/motor-run-capacitors/

Well, I think the answer is to throw it away.  It's now a used capacitor that has been running for two years, and I don't think it would make sense to put it back into a unit after this has happened.  If the cap is good, it's still a mystery what might have happened, but if the new one runs ok this summer, then my conclusion is there's something wrong with the old one even though it tests good.

[Peabody]

Are you sure it's a run/run capacitor?
Those values indicate more a start/run cap, in which case the centrifugal switch could play a role. Perhaps your working on the motor got it working again?

I'm pretty sure it's run/run.  The second cap (5ufd) is for the fan motor, which ran fine even when the compressor wasn't starting up.

[SeanB]

Might measure 34uF, but likely that is not going to actually be usable with 10A of current flow, as likely there is a lot of the foil area burned away from self healing, and thus, if unwound, the films will appear to be mostly small areas of aluminium film, and largeareas with missing or transparent film from self healing.  They are not great these days, because the old ones were made conservatively, with thick films of foil as electrode and dielectric, but the modern ones use thinner films, and run them at a much higher stress, so they do fail with time, applied transients of starting, and running hot. I still have a few dating from 1977, near the end of the use of PCB oil, and they are still within spec, just that they are four times the volume of a modern one, and you need 2 in an AC outdoor unit with one for the compressor, and one for the fan. Modern ones will withstand 500V applied when new, after 2 years you apply 500V and can hear the capacitor doing self healing from the degraded dielectric inside them. Environmentally friendly though, the oil filled ones use coconut oil, and the dry ones use epoxy.

[antenna]

You can put the old cap back in and test it in circuit.  You will need a clamp on current meter though.  Take the current through the cap x 2652 / the voltage across the cap and that will equal the uF of the capacitor.  If what you measure/calculate is not within the tolerance printed on the can, replace it.

Of course, that 2652 is based on 60Hz, if your line frequency is 50Hz, then use 3183.  This is just ohms law where the ohms is replaced with the formula for capacitive reactance.  You don't need a capacitance meter to measure the capacitor.

[floobydust]

Oil-filled caps have a safety pressure switch inside. It might be defective, poor connection, disconnect under load etc.
CPT (compressor protect) terminal I'm not sure if that can go sour either, or if it's also pressure-switch based, or connected to anything in your A/C.

"During research, AmRad engineers saw that hard-start kits have enough capacitance to start the compressor without the motor-run capacitor attached. Hard-start kits are connected directly to the very terminals the compressors are attached to. In the event of capacitor failure, the capacitor’s terminals act as a terminal block, and the hard-start kit is STILL directly connected to the compressor. The compressor will restart every time the thermostat requests it, but without the motor-run capacitor in the circuit. Compressors are NOT designed to run without the motor-run capacitor and will quickly overheat causing refrigerant pressures to increase significantly and will be quickly damaged when run in that manner. The CPT breaks this connection, separating the hard-start kit from the compressor and preventing the hard-start kit from damaging the compressor by acting as a kill switch to take the hard-start out of the system in the event of a failed capacitor." source

I don't like that you've found nothing wrong with the old cap that has a lot of made in America pride, and a 5-year warranty. Phone'em up, send it back.
Measure resistances as well to see if there is leakage current too.
Motor run/start cap manufacturing has been decimated by low quality chinese products.

[Peabody]

If I put it back into service, and it works, then I'm left without an explanation for the compressor failing to start originally, but working fine now.  And I still don't know whether it might fail again soon.  It was enough trouble doing the switch-out that I don't want to do it again unless I have to.

So I think I need to throw the old one away.  The new one has worked fine all day yesterday (high = 95F) and today, so I'm not going to disturb it.  The old one is two hot summers old, and *may* have something wrong with it.  I don't think it would make sense to put it back into service in any compressor just to save $25.  That said, it still bothers me that it tests good, but mainly because it might indicate something else is wrong with the compressor.  There's just no indication of that.

The old cap was put in as part of general HVAC service two years ago when I bought the condo.  But that was done by the sellers, so I didn't pay for it.  According to the invoice, they were charged $240 to replace the cap.  I don't know why they replaced it.  The service was done in March, so the compressor hadn't been run in months.  It may have just been revenue enhancement by the tech.  Or that replacement, followed by this one, may indicated the compressor is becoming flaky.

[Zero999]

If the capacitor is good then, perhaps the compressor just got too hot and it worked once it had cooled down?

Either way, I think you're right to discard it. If it was replaced two years ago, then perhaps it's just a component which aught to be replaced every couple of years, as part of the service schedule.

[Psi]

If I put it back into service, and it works, then I'm left without an explanation for the compressor failing to start

You're also forced to disconnect HAL9000

[Peabody]

You're also forced to disconnect HALL 9000

I wonder why none of the makers of smart speaker devices (Apple, Amazon, Google) never offered a voice that sounded like HAL.  That would be my preference.  And I would want it to call me "Dave".

[jh15]

Probably get revoked by the movie company. I liked the HAL "EYE" on the Tesla S display when security recording was in use. Then the movie company had it removed. Some game's icon is in its place now.
     Of course, modern thieves probably never would have watched the movie anyway, but It looked cool.

[johansen]

If I put it back into service, and it works, then I'm left without an explanation for the compressor failing to start originally, but working fine now.  And I still don't know whether it might fail again soon.  It was enough trouble doing the switch-out that I don't want to do it again unless I have to.

So I think I need to throw the old one away. 

Yeess.. yyyeeessss... Please destroy any hope of finding out what is wrong.

[Jill Kent]

Here are the possible explanations and suggestions for your situation:

Possible Reasons
1. **Differences in Load Conditions**:
   - **Operating Voltage**: In actual operation, the capacitor is subjected to 240VAC, while your Arduino measurement was done at 5VDC. This voltage difference can cause the capacitor to test fine at low voltage but degrade in performance at higher voltage.
   - **Load and Temperature**: In real use, the capacitor not only faces higher voltage but also endures greater current loads and higher operating temperatures. These factors can affect its performance.

2. **Transient Fault**:
   - The capacitor might exhibit faults over a short period while appearing fine during testing. Such transient faults could be due to issues in the internal structure (like dielectric material aging or poor internal connections).

3. **Quality of Replacement Capacitor**:
   - You mentioned that replacing the capacitor solved the problem, indicating that the new capacitor performs well. Despite the old capacitor testing fine, it might have minor issues affecting its performance in actual use.

Testing Suggestions
1. **High Voltage Testing**:
   - The best method is to test the capacitor's performance at its rated voltage. You can use high-voltage capacitor testing equipment to simulate working conditions, but such equipment is usually expensive and requires professional knowledge to operate.
   
2. **Leakage Current Test**:
   - You can use a leakage current tester to check the capacitor's leakage current under high voltage. If the leakage rate is high at high voltage, it indicates a problem with the capacitor.

3. **Temperature Testing**:
   - Test the capacitor's performance under high-temperature conditions to simulate its working environment in the compressor. If the capacitor's performance degrades at high temperatures, it might have an issue.

Handling Suggestions
1. **Keep or Discard**:
   - **Keep**: If you don't have other spare capacitors, you can temporarily keep the old one as a backup, but be aware that it might be unreliable.
   - **Discard**: If you can easily obtain new capacitors, it's better to discard the old one to avoid similar problems in the future.

2. **Future Prevention**:
   - **Regular Inspection**: Regularly check the capacitors in your air conditioning system to ensure their performance under high voltage and high temperature.
   - **High-Quality Replacements**: Choose high-quality replacement capacitors to ensure their reliability in actual use.

In summary, it's possible for a capacitor to test fine at low voltage but fail to work at high voltage. You might consider performing more professional high-voltage testing or decide whether to keep the old capacitor as a backup based on your actual situation.
It is also important to choose the right capacitor supplier.

[Spam content removed]

[Peabody]

Thanks very much.  In the end I decided to dispose of the old capacitor.  I was concerned that someone might be tempted to use it as a replacement someday, and didn't want that to happen.  The supply house is half a mile from me, and the cost of the new capacitor was $24.  So there's just no reason to risk reinstalling the old one.  Meanwhile, over a month later now, the compressor continues to work fine with the new cap.  I'll never know exactly why the old cap tested good, but the possibilities you suggested no doubt contain the answer.  But I think it's a useful cautionary thought - a cap that tests good might not perform correctly in actual service.  So before replacing a compressor, it might be a good idea to try replacing a "good" capacitor, just in case.

[johansen]

Meanwhile, over a month later now, the compressor continues to work fine with the new cap. 

when the capacitors fail, the compressor won't start, it gets hot, trips the internal circuit breaker. resets, cycle repeats every 1 to 10 minutes.

was that happening?

if not the failure was elsewhere.

[floobydust]

Here are the possible explanations and suggestions for your situation:

Possible Reasons
1. **Differences in Load Conditions**:
   - **Operating Voltage**: In actual operation, the capacitor is subjected to 240VAC, while your Arduino measurement was done at 5VDC. This voltage difference can cause the capacitor to test fine at low voltage but degrade in performance at higher voltage.
   - **Load and Temperature**: In real use, the capacitor not only faces higher voltage but also endures greater current loads and higher operating temperatures. These factors can affect its performance.

2. **Transient Fault**:
   - The capacitor might exhibit faults over a short period while appearing fine during testing. Such transient faults could be due to issues in the internal structure (like dielectric material aging or poor internal connections).

3. **Quality of Replacement Capacitor**:
   - You mentioned that replacing the capacitor solved the problem, indicating that the new capacitor performs well. Despite the old capacitor testing fine, it might have minor issues affecting its performance in actual use.

Testing Suggestions
1. **High Voltage Testing**:
   - The best method is to test the capacitor's performance at its rated voltage. You can use high-voltage capacitor testing equipment to simulate working conditions, but such equipment is usually expensive and requires professional knowledge to operate.
   
2. **Leakage Current Test**:
   - You can use a leakage current tester to check the capacitor's leakage current under high voltage. If the leakage rate is high at high voltage, it indicates a problem with the capacitor.

3. **Temperature Testing**:
   - Test the capacitor's performance under high-temperature conditions to simulate its working environment in the compressor. If the capacitor's performance degrades at high temperatures, it might have an issue.

Handling Suggestions
1. **Keep or Discard**:
   - **Keep**: If you don't have other spare capacitors, you can temporarily keep the old one as a backup, but be aware that it might be unreliable.
   - **Discard**: If you can easily obtain new capacitors, it's better to discard the old one to avoid similar problems in the future.

2. **Future Prevention**:
   - **Regular Inspection**: Regularly check the capacitors in your air conditioning system to ensure their performance under high voltage and high temperature.
   - **High-Quality Replacements**: Choose high-quality replacement capacitors to ensure their reliability in actual use.

In summary, it's possible for a capacitor to test fine at low voltage but fail to work at high voltage. You might consider performing more professional high-voltage testing or decide whether to keep the old capacitor as a backup based on your actual situation.
It is also important to choose the right capacitor supplier.

[Spam content removed]

You are forgetting to mention oil-filled (motor run) capacitors have a safety pressure-disconnect switch inside, which can also malfunction. If the cap mysteriously goes open-circuit.

CapXon are very well known shit quality. Why is this? AiSHi is also terrible I like how they heat up from high ESR and then dry out. The reputation is bad.
Many chinese capacitor companies are scams, they go public on the HK stock exchange then pump and dump the stock price, all the while making garbage quality- which was the point all along. Then they get de-listed.

[johansen]

You are forgetting

that a person with 2 posts suddenly shows up to make an almost AI generated list of generic bullshit.

[floobydust]

I know, it's a spam standby/marketing account but figured to ask a few questions to flush that out.
I like how chinese cap manufacturing companies are registered in the Cayman Islands as well. Surely a bit more capacitance for free.