I have two of these inverters, with only one installed. Its been running close on 3 years without a glitch. Got connected to that one 9x 250W panels, 20x120Ah SLA configured for 48V.

Installed a further 10x320W panels and also the other inverter. After connecting the solar panels and batteries, the inverter is giving error code 08. According to the manual it is " DC component in the output current exceeds the upper threshold"

I have no load connected to the inverter.

Any ideas on what the problem could be or who I can contact? My inverters are Growatt brand.

So was the other inverter sitting idle for 3 years, ir did you just purchase the second one recently?

It sounds like the former. Perhaps some of your electrolytic capacitors need reforming. The proper way to do this is to slowly let the volts on them ramp up. But this is impractical. Maybe just let the inverter run with no load, disconnect the PV input, and let it run for a few days. It might just come good with time.

Yes, it's been sitting idle in the box.

"Perhaps some of your electrolytic capacitors need reforming" --> Never heard of this before. can you please explain a bit more?

Do you mean to disconnect the PV, Battery and only leave it on mains?

 

Ivor

I'm a little hazy on the details, but electrolytic capacitors rely on a film (of oxide, I think) which can fade after something like 12-24 months. Ordinary use where the capacitor sees at least about 50% of rated voltage will keep the capacitor "formed". When it's not formed, it is very leaky, with most of that current going to the forming process. After a day or two, the leakage current should decrease to a very low value. But while the leakage is high, it won't have normal electrical properties, so it's not safe to use it "in action", e.g. inverting kilowatts of DC to AC. But if it can be powered up quietly, and if in that quiet state it gets the voltage it needs for reforming, then that should be safe.

I suggested removing the PV input so that no PV charging will happen, though it seems much more likely that the inverter itself is what has to be kept quiet. Perhaps remove the AC input as well, in case the inverter is used for AC charging (like in an Axpert). Hopefully just applying battery power will do the trick.

It's actually a neat feature of Voltronic Power inverters (Infinis and Axperts) that they have a comprehensive fault checking system. Though I suspect that all power converter manufacturers find quickly that such a system is well worth the effort, as so much can go wrong, and the results if not detected are often a loud bang and smoke.

[ Edit: "after a day" -> "after a day or two". ]

Edited March 18, 20196 yr by Coulomb

Thanks for the more detailed explanation.

I will leave the inverter on with only battery power supply. Will report back in a day or two.

Just a quick report back:

With just the batteries connected, the unit shuts down after few seconds.

So I powered up the unit with mains, then it remains on, still giving the error 08.

I will leave it likes this to see if it clears the fault. 

 

So quite a few days later and the 08 error is still present.

I did notice that I can in fact turn the unit "on" meaning it will switch the relay to connect the load to mains, sometimes it stays on like this for long time, other times it disconnects after a while.

I connected my PC with the solarpower software to both my units and noticed that the working unit has the regultion set to VDE4105 and the not working one is set to some other VDE???. I tried to change it but I get the famous factory password request.

What should the inverter be set to and could this be the cause of the problem?

 

Anybody know where and what is measured to get this error? Which output current? 

"DC component in the output current exceeds the upper threshold"

 

26 minutes ago, Mikkie said:

Anybody know where and what is measured to get this error? Which output current? 

"DC component in the output current exceeds the upper threshold"

In the Axperts (whose firmware I've read far more extensively), they literally average the current samples. Usually, they'll average the absolute value of the current samples, because otherwise you just get zero. Or ideally you should; you get the DC component of the AC current. This is designed to be zero; for various practical reasons it will never actually be exactly zero. But it should be a small value, and if it's not, then something has gone badly wrong, and some of the loads may be affected (e.g. those that contain step down or step up transformers, such as non-inverter style microwave ovens).

So that's what this error is about. The inverter switches the IGBTs in such a way that a symmetrical sine wave should result (spending almost exactly 50% of its time positive, and very nearly the same 50% negative. When the result is DC biased more than the upper threshold towards positive or towards negative, something has gone badly wrong, and the hardware needs repairing.

It's starting to look as though one or more of the IGBTs has developed a high leakage. As far as I know, that's an unusual fault.

Is there any chance that you can get an oscilloscope (with a suitable high voltage probe for safety, of course) on the output? It would be interesting to see what it looks like. DC coupled and properly zeroed, of course, since that's what this fault is all about.

Thanks for the additional clarification.

Just for the record, looking at the HY3k service manual, error 08 has a different meaning. It states that the problem is "DC current sensor fail".

I took the front cover off, the unit looks fantastic inside, clean, no leaking caps, no (obvious) blown components and exactly like the  Infini photos in the manual.

I have an oscilloscope, probes at about 300V max. But I do not have circuit diagrams or assembly overlays, so a bit in the dark when it comes to measuring locations.

The error appears before any load is connected. The system is simply turned on with the addition of power.

What do you think about that VDE setting?

11 hours ago, Mikkie said:

What do you think about that VDE setting?

I would not know, but my guess is that it's not related to a current sensor.

The DC current sensor might be a blue hall effect sensor, it might have a part number with "LEM" near the front. These have an op-amp incorporated into them. These could be replaced, although they are not cheap. There might be some parts nearby that could be tested and/or given a more thorough inspection.

It's also possible that the sensor has become magnetised. You may be able to demagnetise it with one of those demagnetising wands that used to be used for demagnetising magnetic heads in old tape and cassette players.

Not much to go on, I know, sorry.

Hi, you sure know these inverters. It indeed has an HXN-20P (manufacturer LEM) current sensor, reference designator HCT3. 

Some pictures attached.

From what I can gather, it seems to be connected to the PV input. If that is the case, it really should not be providing an output with no PV connected. Oh, I would offer some body parts for a schematic...

Will try and find a replacement. I see Denver Technical Products Ltd is the agents for LEM. 

 

 

Edited March 24, 20196 yr by Mikkie

3 hours ago, Mikkie said:

It indeed has an HXN-20P (manufacturer LEM) current sensor, reference designator HCT3.

I was thinking more the two oddly-shaped blue blobs on the left of the photo below (outlined in orange), as opposed to the rectangular one on the right.

This is from a possibly different model, from this service manual:

https://adnsolar.eu/download/file.php?id=30

Of course, the "chocks" are actually chokes or inductors. There are some scattered partial schematics in there too, but you possibly have those already in whatever manual you are using now.

Alas, I found no mention of current sensors (other than in error 08) with a quick search.

It seems those two blue CTs are replaced in mine with a single VAC CT. See 3rd image below. The second image is just to show the lay of the land.

There is only one other CT component and it is a daughter PCB mounted vertically as shown in 1st photo. I see both live and neutral wires going through that CT. I am thinking that this would probably measure the current difference between the two wires where in the perfect world, the neutral will be exactly opposite of the live wire.

5 hours ago, Mikkie said:

I see both live and neutral wires going through that CT. I am thinking that this would probably measure the current difference between the two wires where in the perfect world, the neutral will be exactly opposite of the live wire.

Yes. I assume that if that test failed (there was a current difference of more than a certain amount), then you'd get fault code 16 (error 16), which is "GFCI sensor fail", at least according to the IGrid SS 3kW Plus service manual (page 23). So that's likely not the source of error 08.

I'd be inclined to guess that it's the one in your third photo, despite the grey colour. Perhaps it's from a different component manufacturer, and may not be as reliable as the LEM originals.

Good day, 

I would like to know did you solve the error 08 code on the Inverter as i just installed a new Growatt inverter and do get the 08 error 

Same here..Running 2x Dyness 2.4 Lithiumion

On 2020/02/01 at 4:28 AM, Ronnie Rudolph said:

Good day, 

I would like to know did you solve the error 08 code on the Inverter as i just installed a new Growatt inverter and do get the 08 error 

 

Hello Mikkie,

I seem to have the same inverter as you, t would it be possible to give me the value of some resistor in other R72 R74 on the back of the charger as well as R105 AND 106,
thank you for your return if you have the time to disassemble the whole.

cordially.