We have a Commander type CD 150 3-phase variable frequency inverter that started smoking under load. We opened it up and found this burned component. 
Does anyone know what it is and if it could be replaced?
Edit: side view, I removed the red wire
\$\begingroup\$
\$\endgroup\$
4
-
\$\begingroup\$ From the picture it may be a relay. Is there any visible part number on the component, or any labelling on the PCB next to the component? As for can it be replaced, would need to try and determine what caused the component to burn, e.g. was there a short on the output? \$\endgroup\$– Chester GillonCommented Oct 5 at 13:17
-
\$\begingroup\$ That looks like a RIFA, a brand of capacitor famous mostly for how frequently they fail like this. Replace it with a new film capacitor of a more reliable brand. \$\endgroup\$– HearthCommented Oct 5 at 15:39
-
\$\begingroup\$ there is writing on the component \$\endgroup\$– jsotolaCommented Oct 5 at 17:20
-
\$\begingroup\$ Given that it's some kind of patch (film cap mounted on the side with what might be hot glue), I would question the quality of the product. They've found some quality problem, patched the boards and decided to sell them anyway. And it seems that they didn't fix the problem with this patch after all... \$\endgroup\$– LundinCommented Oct 7 at 13:36
Add a comment
|
5 Answers
\$\begingroup\$
\$\endgroup\$
2
The PCB component designation of 'C10' suggests that it is a capacitor.
The oversized silk-screen component location suggests that it is designed to accept a much larger capacitor - possibly for a different model in the series or to allow for alternate, less compact capacitor types.
-
\$\begingroup\$ Hmm, I thought
C10was the label for an adjacent component which looks like had been bypassed with a wire. \$\endgroup\$ Commented Oct 5 at 13:19 -
3\$\begingroup\$ @ChesterGillon, the guys in forensics say to look at the wire's shadow. It indicates a change in elevation with it getting higher on the left. It appears to go to the top of the capacitor. Could they have used an axial lead component that was too long for the pad spacing? \$\endgroup\$ Commented Oct 5 at 13:44
\$\begingroup\$
\$\endgroup\$
C10? A capacitor. Possibly X-type. Most likely part of an EMC filter.
We will need better images of where it sits in the circuits and remnants of marking on the component itself to tell for sure.
\$\begingroup\$
\$\endgroup\$
2
The marking "C10" on the board means it's a capacitor, and the reason it is placed vertically instead of horizontally as outlined on the board is because the resistor "R8" (can't see the number clearly) below it is a high power resistor which gets very hot, and it seems to have caused the capacitor to melt.
I may be wrong about it, but the capacitor seems to be burned on the side towards the resistor.
It is likely the same value capacitor as the one next to the black wire, with 0.1uF printed on top of it, so it is likely a 0.1uF capacitor, but can't tell you the voltage, the other one should have it somewhere on its side, or this one could have that marking still intact.
It can be replaced if you have the replacement part and the tools and skill to replace it.
However, I would recommend to take additional measures regarding the hot resistor (white block): you could add some sort of a raiser for it which would lift it up and expose it more to the air and keep it a bit cooler, you could improve the air flow across it and you could additionally move the replacement capacitor a little farther from it. Basically, any measures to reduce the temperature in that area, especially for that capacitor.
It is also possible that the capacitor simply blew from an overvoltage or its insulation layers have simply failed and it then melted and popped. I don't see the PCB being darkened due to high temperatures from the resistor, which means that my assessment was incorrect, or that the resistor wasn't normally getting hot during operation, or that the inverter wasn't used for a long time so that the PCB didn't have the time to darken.
-
\$\begingroup\$ Possibly C10 overstressed by a noisy environment and/or overvoltage spikes? Or C10 not the appropriate type? Agree that most probable is simply overheating. \$\endgroup\$– user317139Commented Oct 6 at 11:59
-
1\$\begingroup\$ @Burglups Those are possible causes, but in this case it is almost obvious that heat from the big block resistor has caused its failure. \$\endgroup\$ Commented Oct 10 at 20:01
\$\begingroup\$
\$\endgroup\$
In cases like this where the part burns, I typically try and de-solder it and get a better look on the bit of the outside that remains. To me it looks like someone already de-soldered it and placed it sideways on the board with an additional wire going back to the pcb to complete the circuit.
All the brown goo around the part and the "C10" on the silk screen suggests it is a capacitor to me. If you can identify where in the circuit this is (ie. what other components it connects to) we may be able to reverse engineer what kind of cap it is and what type could replace it. the down side to just doing a direct replacement is that a lot of the time a cap self destructs because something else is wrong. however there is a good chance its just premature age due to being so near that big resistor.
Too much heat will boil off the capacitors electrolyte and char what is left. I also noticed there may be even older electrolyte "dust" around the same area suggesting someone already made this fix and the first caps electrolyte spilled everywhere. This may be why they placed the cap on its side and did a wire mode to get it away from the ceramic resistor.
\$\begingroup\$
\$\endgroup\$
As @Hearth mentions in the comments - the component in question is manufactured by Rifa. Unfortunately these capacitors are guaranteed to fail in the manner described and should be replaced.
As others have also suggested, replace this capacitor and the others of the same brand on the board with a X2 rated ‘mainscap’. The other similar capacitor we can see is a 0.1uF or 100nF 250VAC(check this) X2 rated capacitor. These are generally cheap and readily available worldwide.
Just for some background - these Rifa capacitors were considered at the time to be a premium component and used by many high end manufacturers like HP, Tektronix and Apple in the 1980’s. I, myself have had one of these capacitors fail on my Tek 1403 where it created an unpleasant stench and brown ooze.
Search for ‘Rifa’ and you’ll plenty of horror stories. The general consensus is to replace these parts regardless of whether they have failed or not.
