tristar mppt60

[Eccentric Anomaly]

As a fellow MPPT 60 owner (though currently it's in storage) can I ask what failed? I have to admit to being somewhat dubious about the things - the software seems to have a lot of odd quirks such as giving funny/impossible readings.

[Tinbum]

There are 3 bidirectional diodes (across the battery terminals, I think) and one of those had shorted. There was no external sign of any damage to it. The other 2 seemed ok but I'm replacing all 3 as they are so cheap.

One of the things with these controllers is to make sure the PV voltage never goes above 150v. I think they sometimes show faults for overvoltage shutdown even though the displays readings show they are less than 150v. Mine has been good so far. It is strange though that they shut down when after uploading new settings and don't reboot themselves. They also don't show any shutdown sequence in the manual. They say turn the battery on first then pv but nothing about turning off.   

[rogeriko]

By pulling the old one apart you have now voided the warranty. You will have to pay customs duty and 20% VAT on the new one when it arrives. You had better be sitting down for that one.

[Eccentric Anomaly]

Ta - fear it might be a useful hint to squirrel away.

Yes, it seems the failures tend to be on power up or reset. It could be related to the panels being at Voc rather than nearer Vmp but if that was the problem you'd expect failures when batteries filled up. My very hypothetical hunch would be that there can be a bit of a weird case in the start up where two FETs which shouldn't be turned on at the same time are, or something like that.

[Tinbum]

By pulling the old one apart you have now voided the warranty. You will have to pay customs duty and 20% VAT on the new one when it arrives. You had better be sitting down for that one.

Morningstar told me to do whatever I wanted with the old unit so no warranty to void. I've had things from the US in the past that were warranty replacements and not had to pay vat or duty.

[Tinbum]

Ta - fear it might be a useful hint to squirrel away.

Yes, it seems the failures tend to be on power up or reset. It could be related to the panels being at Voc rather than nearer Vmp but if that was the problem you'd expect failures when batteries filled up. My very hypothetical hunch would be that there can be a bit of a weird case in the start up where two FETs which shouldn't be turned on at the same time are, or something like that.

These diodes are only rated just above battery voltage so I suspect that somehow on start up some of the PV at about 110v (Voc) leaked through?

[rogeriko]

OK that's good about the warranty, normally an item has to be returned first with a special export paper and only then do they allow an equivalent import duty/vat free. Its a big hassle, lots of paperwork. I used to have this problem with broken Studer inverters. They had to go back to Switzerland (outside of the EU) to be repaired and then reimported to the EU. The paperwork was so complicated that I used to drive them there myself, have a skiing holiday, and then drive them back.

[Tinbum]

I hope I don't, but if I can repair the old one I won't be too bothered.

It would be interesting to see a circuit diagram of the controller. I found this as a general principle of how an MPPT controller works.
 
http://320volt.com/wp-content/uploads/2011/03/MPPT-circuit-solar-charger-circuit-gunes-panelli-sarj.jpg

The MPPT 60 has 3 induction coils potted in the metal case and it's interesting that there are 3 diodes across the battery and 3 across the PV inputs, so I suspect it has 3 MPPT circuits paralleled up.

[Eccentric Anomaly]

Guessing wildly - maybe the MTTP 45 has two of the same?

Do you really mean “across” the battery or in series with it?

[Tinbum]

Yes across the battery, they are designed to stop damage due to overvoltage and are also bidirectional.

New controller just arrived- sent Friday received Tuesday- wow. Customs documents shows replacement for faulty unit and also duty and taxes payable by exporter.

[Tinbum]

New parts soldered in and controller works fine.

[Eccentric Anomaly]

Excellent, well done. Good to know they're fixable, at least sometimes. Are they zeners for voltage protection or just ordinary diodes for reverse voltage protection?

Seems difficult to imagine how one on the battery side would get an over voltage, though. If it was just connected to the PV panels, without the battery, it might happen but you said your battery MCB blew. Hypothesis: disconnected from battery but connected to PV somehow the output voltage got too high blowing this protection diode and leaving the output capacitor charged to a much higher than battery voltage. Then connect the battery at which point the capacitor discharges into it tripping the MCB. Plausible?

Do you use the remote sense wires to allow the charge controller to see what the battery voltage actually is (without the voltage drop on the leads)? I have a Maplin bench power supply of a type which can be blown up by connecting the remote sense leads without connecting the output:

http://edavies.me.uk/2013/07/lifepo4-charge/

Could it be that the MPPT 60 suffers from the same problem?

[Tinbum]

This is the part,

http://uk.farnell.com/littelfuse/smcj78ca/diode-tvs-78v-1-5kw-bi-5-smc/dp/1886372

Controller was connected to the PV while the battery was disconnected so maybe it leaked through, shorted the diode out and then I turned the PV off. Then when I turned the battery back on the mcb blew straight away. The diode was totally shorted inside but showed no sign of any damage from the outside. perhaps if I hadn't had a mcb it may have blown up!!

Yes I do have the battery sense wires connected but they are very thin so wouldn't pass much current.

[Eccentric Anomaly]

Tinbum, yes, I think we agree on a likely sequence of events.

Yes I do have the battery sense wires connected but they are very thin so wouldn't pass much current.

The problem with that lab power supply is not current in the sense wires as such. Instead what seems to happen is that it sees voltage from the battery on the sense wire but below the set voltage (because the battery isn't fully charged). Therefore it turns the output voltage up a bit to try to correct the sensed voltage. If the output isn't connected then that, of course, doesn't work so it tries to turn the output voltage up a bit more. It keeps doing that until it overvolts and the capacitors (or something meaty) blow up. It's stupid not to have a voltage limit but it seems to be the case and, apparently, it's not the only such supply to suffer from this problem.

[Tinbum]

Arh, yes see what you mean- could be a possibility. Arh but just thought the PV seemed ok connected even with the shorted Diode, but then I didn't measure the voltage across the PV when connected like that.