I have a split unit air conditioner/heat pump

While i was away it stopper working, parents described the problem as "not working". I came in and tried to switch it on and flip between modes. Compressor is buzzing quietly but i can tell that it is not running. When i switch between modes it pulls a surge of high current (lights dim), and after ~20 sec of playing with it, tripped the breaker (C curve 16A)

Compressor is sealed unit pumping R-22. Compressor motor is single phase 240V 50Hz, 2 coils. voltage is applied directly across the main coil, and in series with 45uF 450V capacitor to the other coil. There is the relay that switches the phase to the entire setup (compressor + cap), but no "start relay" or such - unles they are inside the compressor

First i suspected the cap may be bad - cause bad cap indeed results in compressor buzzing but not starting. But i measured it and it measures as 42uF, appears to be ok....

Second i tested the coil resistance. Main is 1 ohm, other is 2.7 ohm. I tested the resistance on a working unit (a friend happens to have a similar model), the values are identical




Can a polyester film capacitor somehow go bad without changing its capacitance ?

If the compressor is stuck (locked rotor) - Why can it happen and how can it be released ?

I have a 40uf 420V/450V metal can (but polyester, not oil type) capacitor, which is rated as motor run (Epcos model B32332-I80). but it appears to be 1/2 the volume of the unit's original 45uF cap. I'd like to give it a try, however, the difference in volume made me thinking before i hook it up.... why is it such big difference in volume ?

The fact it tripped the breaker tells me the windings are roasted. When the rotor seizes in an induction motor it pulls way more current than normal and all that power is dissipated as heat (telltale sign it's cooking is the current draw gradually dropping over the course of a minute or so or so, alas associated with the humming noise changing slightly). Am I right by assuming the compressor was powered on before you got to it? Even if the windings have not shorted out, they may have shorted to earth, tripping the breaker. Can you measure the resistance between the windings and the case? A compressor is incredibly hard to repair, if not impossible! For a start, the only way of getting in is by grinding the weld off near the top. Also doing this will de-gas the whole system and spill oil everywhere.

I agree with Jonathan. It does sound like the rotors might be locked or coils shorted.
I had an old LG heat pump and this failed in the way you describe with the compressor humming. It didn't actually trip the breaker but the compressor just got incredibly hot.

If its an old R22 system then it could be a good time to replace it with a much more efficient R410a inverter system. They use much less energy and don't need a high starting current. The outdoor unit is so much quieter too!

Andy: I don't think that R410a is more efficient than R22: I've seen that R410A air conditioners here in Israel, takes more time to begin heating than the lastest R22 ones. When my Electra upper split A/C in my hostel, still had its original controller, it took a lot of time to heat up until the fan of the indoor unit, kicked in, and then the air wasn't so hot, and took a while to heat up (Sometimes the fan stopped once again and than kicked in back). This in contrast to the R22 upper split air conditioner, which take much shorter time to heat up.

The cause for breaker tripping is not the windings burning up, but that a 16A breaker trips in about 10..20 sec on few tens A, which are the current the compressor draws if it is not starting. It will trip if the compressor is not starting due to a bad capacitor, then work just fine when you replace the capacitor

I compared the coil resistance with an identical working unit, the resistance is identical : 1.0 ohm main, 2.7 ohm other, multimeter accuracy +/- 0.1 ohm. Problem does not seem to be electrical



The unit is top class for its time, and not THAT bad in regards to efficiancy or noise even for today. Replacing it solely for efficiency would be like replacing a T12 fluorescent fitting due to a bad ballast (when you can just replace the ballast). And if you suggested inverter, then putting up electronic ballasted one at that.....

So not an option as long as the issue can be resolved in a simpler way



Dor, would you say that HID is inefficient based on its light output at switch on, and cause it takes time to warm up ?

swap the cap first.if it went high esr it could still read ok.
otherwise its shot.a few hammer blows while it is on might get it going for a while.
you could double up the cap too.thats all a hard start is..

What you describe looks really a mechanical problem. It could be an excessive friction (missing oil,... - that could be due to not good pipe shaping, so the oil does not return into the compressor or just degraded oil) or some internal component get loose and have blocked the moving parts - it could be something broken, but as well as some loose part from lousy tube installation workmanship (traveling through the pipes and eventually reaching the compressor).
here the only thing you may try is really the hammer and don't worry that much: Either it helps, or the compressor would have to be replaced anyway...

The kW range motors really do trip breakers after few seconds when stalled (they do not rip the short circuit trigger, but the thermal overload trigger in the breaker), with all the winding in good shape.
As the capacitor is permanently connected, I would expect it to be a film type, not an electrolytic (the electrolytic are smaller and cheaper, but do not withstand long term loading, so could be operated just for starts, so need a disconnect relay; and yes, I really mean electrolytic - there are bipolar types on the market designed specially for motor starting). The ESR increase exhibit just the electrolytic ones, the film types exhibit either capacitance reduction or sometimes a short circuit. So from that I would expect the cap is OK.