question

muelli avatar image
muelli asked

Phoenix 24/1200 8 years old, ALL caps failed

Hello forum,

I am posting my experience with a rarely used (maybe 48/96h in total) Phoenix 24/1200 inverter.

It was connected to my Pb wet battery for 8y and after using it a couple of days ago, I suddenly could not turn it on anymore.

I removed it from the installation site and could hear something rattling inside. Through the vents I could see something that looked like parts of a TO-220 casing. Since warranty was long gone, I decided to take a closer look and upon opening the case I was shocked to what I saw:

ALL caps failed: 1 blew up completely and partly removed itself from the PCB along with some tracks on the backside, the rest are heavily bloated.

Additionally 1 MosFET exploded, which was the part I saw. Hard to tell what failed first, but as all caps are bloated I would point towards cap 1st, then FET waved goodbye.

What-the-f is going on here? I have never seen such poor quality caps that bloat/fail/explode after such relatively short time.

Neither sales@victornenergy.com (which resulted in a reply from a Bob Hopman, who pointed me towards the contact form at https://professional.victronenergy.com/support/?is_next=www.victronenergy.de ) nor filling the contact form gave me some usable result.

After filling out the contact form I was contacted by the initial dealer that sold this unit, who could understandably not help me and pointed me to the forum.

I need contact to someone at Victron who is willing to find out what is going on here. Systemic error with bad caps? Fluke?


As I am using several Victron devices (MPPT controllers, chargers, inverters, ...) as a private person and as a professional, I blacklisted Victron as a supplier until this matter is resolved and cleared up, on why all caps fail after such "short" time.


Pictures of the failure can be provided upon request.


Thanks and regards

Phoenix Inverter
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

6 Answers
Alexandra avatar image
Alexandra answered ·

@muelli

Look up DC ripple. This will kill any cap no matter the quality. What now made you think they are poor quality?

And what is usually what ages caps early. (Besides heat.) Blown caps can short which in turn will kill a mosfet.

Also not in use doesn't stop calendar aging. (I know I have a box of high quality completely mint caps here- dead) So there is that. Or bad pre charging which is also what alot of installers mess up.

And these are the two most common reasons I have seen. Besides short circuits either internally because of critters, or on the output.

FYI you won't get anyone from Victron giving your system personal attention, end user support is from local dealers. Your support period has ended as well (you won't get support from any manufacturer either, so not sure what you are expecting).

You will need to look for a person who repairs these for a living if you are really on repair. But component aging will make it not worth your while.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

klim8skeptic avatar image
klim8skeptic answered ·

@muelli Caps tend to fizzle and pop when exposed to higher voltages than they are rated for, do check that.

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

kevgermany avatar image kevgermany ♦♦ commented ·
This seems likely, given that all went at once.
1 Like 1 ·
muelli avatar image
muelli answered ·

DC Ripple is not relevant in my setup, according to my oscilloscope.


And yes, since I design PCBs myself and sell products with my design for cars (which is is an extremely challenging environment, as for temperatures and voltages/specifications), I know how to tell when caps are poor quality. In 10+ years none of "my" caps failed in these harsh environments.


>>Caps tend to fizzle and pop when exposed to higher voltages than they are rated for, do check that.

The caps are 35V/1000µF, so no problem here in my setup.


I would be sad if Victron did not care about this problem.

We have 2 suitable sayings in Germany for situations like these:
1. "Da hab ich Lehrgeld bezahlt!" - I learned my lessons/learning with money!

2. "Wegen Reichtum geschlossen!" - Company does not care because already rich enough.

If they really don't care, then be it. I will phase out all Victron products one after another and never look back. *shrug*

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

klim8skeptic avatar image klim8skeptic ♦ commented ·
@muelli Swapping the input power lead polarity will also tend to turn caps inside out. And disintegrate semiconductors.
1 Like 1 ·
muelli avatar image
muelli answered ·

>>Swapping the input power lead polarity will also tend to turn caps inside out. And disintegrate semiconductors.


Please, guys, the next person will remind me, that hitting the inverter with a hammer is also baaaad m'kay! :)
The setup has not been changed during the years. No polarity fault, no hammer, no aliens, nothing as far as I know.

2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Guy Stewart (Victron Community Manager) avatar image
Guy Stewart (Victron Community Manager) answered ·

Hi @muelli,

I haven’t seen enough capacitor or mosfet issues reported on this site for me to flag a systemic issue.

Being outside the warranty period of 5 years it won’t be getting logged via that process, so this community site is really the next best place for such things to collect.

There is one other similar complaint from 2021, also with a similar rarely used profile as your own - https://community.victronenergy.com/questions/87308/slightly-bulging-capacitors-on-victron-energy-quat.html

Given the long amount of time it has taken, even if there was an issue with a capacity parts supplier then, it would have already been changed just through other factors.

If you’re worried about Victron equipment longevity there is an option to increase the warranty period to 10 years at the point of sale, that should reasonably cover it.

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

muelli avatar image muelli commented ·

I was not aware of the option to extend warranty to 10 years :(

However thanks for the reaction.

0 Likes 0 ·
nesswill avatar image
nesswill answered ·

I am sorry to hear of this failure but to help/ understand why this happened we need more information please.

What where the environment conditions of where these were being used IE; hot, cold, varying temperatures, humidity, vibration??

Was It continually on or powered on and off when needed, what was the load??

There are lots more I can ask but to keep things short and understandable I will hold off until you provide more information.

FYI This is something that was posted elsewhere but it might help you understand the failure.

Once caps have reached this catastrophic failure, often they are now "shorted" and YES, creating a short in a circuit where there previously was no current being conducted can absolutely (and very likely) cause many other components to fail.

We are all trying to help and understand so the root cause can be identified and then posted to hopefully help others.

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

muelli avatar image muelli commented ·

>>What where the environment conditions of where these were being used IE; hot, cold, varying temperatures, humidity, vibration??


Varying, mounted inside a unheated shed.

>>Was It continually on or powered on and off when needed, what was the load??


Constantly connected to power, but switched off and only turned on if needed. Load: Around 50W

>>Once caps have reached this catastrophic failure, often they are now "shorted" and YES, creating a short in a circuit where there previously was no current being conducted can absolutely (and very likely) cause many other components to fail.


I am well aware of this, as I am designing PCBs myself. However I have never seen such a catastrophic result of a cap failing, I removed the failed cap today and it burned a hole into the PCB, dia around 5mm, removing everything else in this area (tracks). I really think it is beyond repair in a reasonable time... :(

0 Likes 0 ·