APC Smart UPS SUA1500 does not detect AC / doesn't turn on

[CapLeaker]

I found an SOT23 diode marking A4 with a sideways "F" and it is a dual diode, common cathode. I actually wonder what that sideways printed letter business is... If that shitty lil diode makes the whole UPS not work... man...

[CapLeaker]

O.k. I think I am getting somewhere. I started to go the reverse way and concentrated on why I am not getting any power on the outlets when on AC. Sure and behold I came on RY3. That is the relay on the very top right on the picture. I tried to energize it with my bench PSU and got nothing, but yet it consumed all the power that I had preset (200ma). Sure enough I got a 75 Ohm short across the coil? Well no wonder that doesn't work! Relay out she goes and??? Relay works fine but I got still a 75 Ohm short across the pads where the relay coil is. Traced it to an SMD diode D13. Took that diode out and short disappeared. I guess I need a new diode for now.

Diode marking A4 with a sideways "c" probably something 75V to 100V dual diode in some config.

[CapLeaker]

Replaced C75 and C109 de to high ESR values (but showed capacitance all right). Hooked up the battery pack and plugged it in. Hear a couple of relay clicks, press the on button and the unit goes beep. nothing happens on the display. If I press the on button for 3 seconds, the unit will go on battery and supply power o.k. Basically same thing what it was before. No bueno...

[CapLeaker]

I was thinking the same thing about what to check I would not necessarily trust a ESR meter for a definitive answer to weather or not a small value UF capacitor has become I have ran into case where the ESR readings were not that far off of a new one but for some reason after recapping the device it started working again and I really do not have an answer for this

That depends on the ESR meter. Good ESR meters can set the frequency. Better ESR meters can set like 4 testing frequencies and you use them depending on cap size. Larger caps use low frequency like 100 hz. Small caps use 100khz. That's why some of them are so darn expensive. At the end you get what you pay for. I didn't run into such problems with my handheld Agilent LCR meter.
The biggest problem are the high voltage capacitors. They can show good ESR but yet leaking voltage. Look at my plasma cutter thread.

[redwire]

A new 22uF 50V is rated say ESR 1.0Ω but the old one measures 2.0Ω Does that pass your criteria? It's 2x rated but looks close sorta.
It depends on what the cap is doing - say a time delay or mains filtering is low stress compared to filtering power for an SMPS IC, it's a big difference in the stress on it and ESR needed for the circuit to work.

I find the small, low value electrolytics, under 47uF don't last, they seem to dry out fast or the seal is worse. In production, if the reflow/wave solder was a bit hot and cooked parts, the caps would last maybe 2-3 years only. It does affect the bung.
So some products you'll see the same brand of caps failing early.

[sam_sam_sam]

It looks good, worth checking some of the small caps at the 23 year mark.
Usually, two relays for voltage tap change, one for load disconnect, two for transfer. I don't know what the 20uF 150VAC cap is doing, but that relay must get beat up, contacts roast.

I have seen they do a self-check where on power up, the load is disconnected from mains, UPS runs for a second it looks for output voltage, then is either good to go (connects mains to load) or it sits there like a zombie if it saw ACV before running, which means the transfer relay is stuck and it would go boom if it ran.
Check C109 for the "transfer relay contact weld detect" with that opto IC20, and a few caps around there.

I was thinking the same thing about what to check I would not necessarily trust a ESR meter for a definitive answer to weather or not a small value UF capacitor has become I have ran into case where the ESR readings were not that far off of a new one but for some reason after recapping the device it started working again and I really do not have an answer for this

[redwire]

It looks good, worth checking some of the small caps at the 23 year mark.
Check Elektrotanya if there is a schematic for it. Your board marking is APC 640-7732D Rev. 5 but I did not see if with all the confusing bullshit names, model numbers then there are different power versions using same board, which products use it.

Usually, two relays for voltage tap change, one for load disconnect, two for transfer. I don't know what the 20uF 150VAC cap is doing, but that relay must get beat up, contacts roast.
I have seen they do a self-check where on power up, the load is disconnected from mains, UPS runs for a second it looks for output voltage, then is either good to go (connects mains to load) or it sits there like a zombie if it saw ACV before running, which means the transfer relay is stuck and it would go boom if it ran.
Check C109 for the "transfer relay contact weld detect" with that opto IC20, and a few caps around there.

[CapLeaker]

I had the same problem with my Powerchute UPS. It's like it refused to go online, acted dead yet worked off battery if I hit the power button and it was charging ok.
A while back I'd replaced just the larger electrolytic capacitors, and this time when I went in it was a small 22uF cap that went low value, it was across the relay coil or in the circuit for a relay.
Must be the transfer relay. I think the relay pulse went too short for it to pull in or something. So replacing all the small caps I did.
These are a nice UPS because of the metal enclosure.

This is a picture of the same version board I have (mine is at work). I went through the obvious yesterday and so far I came up with squat. The only thing I didn't do is was using powerchute and connect the UPS to my computer. I do realize this thing is 20 years old, but I do want to put it back in service.
The whole board got only like a dozen electrolytic capacitors. Would be easy enough to change them. I did press that chip with the firmware firm in its socket, no change.

[redwire]

I had the same problem with my Powerchute UPS. It's like it refused to go online, acted dead yet worked off battery if I hit the power button and it was charging ok.
A while back I'd replaced just the larger electrolytic capacitors, and this time when I went in it was a small 22uF cap that went low value, it was across the relay coil or in the circuit for a relay.
Must be the transfer relay. I think the relay pulse went too short for it to pull in or something. So replacing all the small caps I did.
These are a nice UPS because of the metal enclosure.

[CapLeaker]

reading through this might help
https://community.se.com/t5/APC-UPS-...er/td-p/301697

Thanks, but nothing there that I didn't try. He ended up buying another one from fleabay.

[petehall347]

reading through this might help
https://community.se.com/t5/APC-UPS-...er/td-p/301697

[CapLeaker]

Sooo... I have an interesting victim today. Our special on the menu is a APC Smart UPS SUA1500. Looks like it's got Alzheimer or isn't smart anymore (dead, dumb as f@ck), but at the same time it isn't.
Battery is about a year old and they are good at a resting voltage of 26.5VDC. If I press the on button for 3 seconds it does turn the inverter on and I get 120VAC no problem at the back and the battery shows full. So it isn't the battery. It also charges the battery no problem when plugged in add that as a twist!
However if I plug it in with the battery attached it won't do squat. I (short) press the power button, get a short beep and nothing on the LED's. I do hear a delayed click from a relay, but that is about it. When on AC, there is no way in any shape or form that I get more than 1.5VAC out of the AC sockets on the back, but at the same time it does charge the batteries.
The main filter cap tested good, I don't see anything burnt, or leaked, the over load fuse is o.k. (otherwise it wouldn't charge the battery), battery is o.k., battery connections are good, so WTF am I missing? It's like the UPS doesn't see any AC input power. I've repaired a stack of UPS's but this is something I haven't come across. Maybe a relay?
Anyone seen a strange behaviour on an SUA like that before?