Trying to destory my SCC's: I can't do it!

[Thafella]

So I took every EPEver/Victron/PWM/Renogy solar charge controller, and tried to fry it by connecting it only to solar panels, and not to a battery... guess which one survived?

EVERY SINGLE ONE! I couldn't destroy any of them, and I left them connected to a high voltage array (80 volts) for 4+ hours each.

I connected each one to a lifepo4 bank after the test, and they functioned perfectly. Tested output with a watt meter before and after as well, no change.

I read in the manuals that you must NEVER connect an SCC to an array without connecting it to a battery first, because the potential that it can fry. I know that having improper gauge wires on an SCC can cause overheating around the terminals (dedicated voltage sensing leads would probably fix this problem, but many SCC use the charging lines as voltage sensing, which makes it wise to over gauge them most times. The voltage drop of a long run of hot wire can be horrible). But to actually connect an array to an SCC, and not to a battery, in my experiments, no issue.

I am thinking that the feedback signal in the converter circuit needs a reference voltage at the output to work. Then it can track the power point between the panels resistance and internal resistance of the battery, then push current to the max. If you only have the input powered up but no current is flowing, what would it stress? The panel would be in an open circuit voltage, and possibly use some power to charge up caps on that side of the circuit, but thats about it. Should be fine to leave connected.

Am I wrong? I know some MPPT's circuit designs are pretty complex, but they are still a simple converter circuit with inductor and capacitor and basic logic system to modulate it. I know some current limited converters can handle amp sources at input/output, and others can't. but because the solar panel is open circuit in this instance, and the output is also voltage sensing of the battery, I don't think any damage can occur.

What do you guys think? How can I destroy one?

Just dropping by to add my data point. I fried an eBay mppt on first hookup doing this once. Perhaps it be because it was new out of the box and any first connection to a battery lets it set up a reference voltage. (It was 24-30v panels but a multi v 12/24/48 mppt)
However because it was new I can’t confirm it wasn’t just already dead.

 

[Dzl]

For reference, what was the brand?

 

[Thafella]

Just dropping by to add my data point. I fried an eBay mppt on first hookup doing this once. Perhaps it be because it was new out of the box and any first connection to a battery lets it set up a reference voltage. (It was 24-30v panels but a multi v 12/24/48 mppt)
However because it was new I can’t confirm it wasn’t just already dead.

For reference, what was the brand?

I was afraid someone would ask that. I don’t have anything of it left with me and I think it was unbranded.

Hopefully you recognize it from the standard body size. From memory of my research at the time it was a comodity board that everyone rebranded. It looked a lot like (may as well have been) this one:

MPPT 40A Solar Charge Regulator 12V/24V/36V/48V PV 130V Battery Panel Controller | eBay

Find many great new & used options and get the best deals for MPPT 40A Solar Charge Regulator 12V/24V/36V/48V PV 130V Battery Panel Controller at the best online prices at eBay!

www.ebay.com.au

I hooked it up to what would have been 24v at 15A which shouldn’t have even exceeded its wattage limit in 12v mode

 

[NeeYee]

So I took every EPEver/Victron/PWM/Renogy solar charge controller, and tried to fry it by connecting it only to solar panels, and not to a battery... guess which one survived?

EVERY SINGLE ONE! I couldn't destroy any of them, and I left them connected to a high voltage array (80 volts) for 4+ hours each.

I connected each one to a lifepo4 bank after the test, and they functioned perfectly. Tested output with a watt meter before and after as well, no change.

I read in the manuals that you must NEVER connect an SCC to an array without connecting it to a battery first, because the potential that it can fry. I know that having improper gauge wires on an SCC can cause overheating around the terminals (dedicated voltage sensing leads would probably fix this problem, but many SCC use the charging lines as voltage sensing, which makes it wise to over gauge them most times. The voltage drop of a long run of hot wire can be horrible). But to actually connect an array to an SCC, and not to a battery, in my experiments, no issue.

I am thinking that the feedback signal in the converter circuit needs a reference voltage at the output to work. Then it can track the power point between the panels resistance and internal resistance of the battery, then push current to the max. If you only have the input powered up but no current is flowing, what would it stress? The panel would be in an open circuit voltage, and possibly use some power to charge up caps on that side of the circuit, but thats about it. Should be fine to leave connected.

Am I wrong? I know some MPPT's circuit designs are pretty complex, but they are still a simple converter circuit with inductor and capacitor and basic logic system to modulate it. I know some current limited converters can handle amp sources at input/output, and others can't. but because the solar panel is open circuit in this instance, and the output is also voltage sensing of the battery, I don't think any damage can occur.

What do you guys think? How can I destroy one?

I have a bit of hands on experience here from back in the day when I used Morningstar and EPever MPPT SCCs - the scenario that destroys the SCC is not so much applying PV without battery connected first, rather, a sure fire way to destroy a Morningstar SCC is to disconnect the SCC from battery when the PV array is producing power at or close to the 'maximum clamp current' of the SCC - the Morningstar and Epevers have a maximum short circuit current they can tolerate - disconnect the battery while the PV array is able to produce more than this short circuit current and you will release the magic smoke.

I have had this happen twice with Morningstar TS MPPT 60A SCCs and the one Epever 60a (Etracer BND series) - I can confirm that the Victron SCCs do not suffer this flaw.

The risk of SCC destruction is higher when the PV array is oversized relative to the SCC's max short circuit current rating - so in my case, I had a 4500w array connected to the Morningstar CC with rated 60A max current - accidentally flipped the battery disconnect breaker while performing inspection/maintenance on a bright sunny afternoon (CC was making about 58Amps at 59v) and the SCC blew up immediately with a loud bang and some fire.

I stopped using DC breakers as battery disconnect after that first incident.

I since switched to Victron as I needed a compatible SCC for my Pylontech US2000s - Morningstar offered a lot of cautionary statements around battery disconnects and did not offer native support for Pylontechs or other Lithium brands at that time.

To be clear, you need to test the SCC for this failure mode by disconnecting the SCC from battery while the SCC is in operation and producing power at the max rated battery current (current limiting the PV array) and connected to an oversized PV array e.g. one that can make 90A while the CC is rated for 60A and clamping the solar current to within the SCC tolerance.

 

[OffGridGuy]

NeeYee, thank you for updating this thread...

1. To return it as current discussion
2. Add some VERY useful information.

I've be using and installing Morningstar Controllers since the late 90's....
Over the years I've had EVERY one fail... They all sit on my 'Wall of Shame' .
The usual response from Morningstar Customer Service was 'damaged due to lightning'.....

Thru the years we've replaced all those controllers with Midnite 150's, or on the smaller systems Victron SSC's.
None have failed.....

I definitely agree with NeeYee... Maybe the next round of testing should include disconnecting the battery, while the SSC is actually charging.....

This will be the failure mode when the BMS disconnects the battery pack, after detecting an over voltage run-away cell.....

This is now a common problem case..... It would be interesting to see who passes and who fails......

Enjoy your testing !!!!

 

[upnorthandpersonal]

So, I use a MUST PC18-10015F 100A charge controller. Since I've been in the process of rewiring my system to it's permanent location (finally!) I did two things: first, a test exactly like you proposed, disconnect the battery while it's charging simulating a fault/BMS shutdown - no issues. Secondly (yes, in that order) I contacted MUST and explicitly asked them if this could cause any harm, or if it needs the battery connected before solar. The answer: battery can be disconnected, no damage will occur.

 

[rickst29]

I'm so glad that someone made a new post into this Thread. This is 100% relevant for my not-yet installed "Warm the Battery from the Charging Circuit' scheme, possibly allowing me to add the EpEver "Battery +" Connector into my battery charge input selector Relay. A temp controller switches any available 14.5V "battery charging Voltage" into a set of motorcycle seat heater pads, when the temp controller has determined the battery bay temperature to be "too low" for charging.

Since the EpEver can apparently survive "hot" reconnection of the genuine LFP batteries while PV Power is present, I can simply join the battery-charge wires from the SCC and 120-VAC Power Cnverter Converter together, before entering that "switcher" Relay. Simple. At the cost of possibly blowing up a relatively cheap EpEver and being forced to replace it with a Victron afterwards I'm going to try it.

 

[durval]

So I took every EPEver/Victron/PWM/Renogy solar charge controller, and tried to fry it by connecting it only to solar panels, and not to a battery... guess which one survived?

EVERY SINGLE ONE! I couldn't destroy any of them, and I left them connected to a high voltage array (80 volts) for 4+ hours each.

Wow, @Will Prowse! Amazing, thank you very much for testing this and then sharing the result with us!

Just one question: can you please tell what EPEver model(s) you tested? I would very much like to be sure my EPEver/EPSolar Tracer3210AN (or some other model in the same XX10AN series) was tested and survived, so I too can dispense with the complication/cost of adding a relay in the PV input of my SCC...

Thanks in advance,
-- Durval.

 

[Symbioquine]

I don't have much personal experience with replacing SMD chips, but the strategy EEVblog demonstrates (not solar-specific, but both using ChipQuik) in this video seems more compelling;

 

[Berseker]

What do you guys think? How can I destroy one?

I carelessly swapped the battery out and pv in cables to the charge controller.
This released the holy blue smoke + some gray ones as well?

Lessons learnt, dont be thinking about an upcoming expedition to everest, while working on sensitive stuff

 

[curiouscarbon]

I carelessly swapped the battery out and pv in cables to the charge controller.
This released the holy blue smoke + some gray ones as well?

Lessons learnt, dont be thinking about an upcoming expedition to everest, while working on sensitive stuff

thanks for sharing the holy smoke with us for learning! ?

note to self… use white tape and black pen to label cables..

(have been messing around with a small diy lifepo4 charger with xt30 cables and sometimes i plug the battery into the pv input,instant facepalm when that happened. luckily it’s 3.2 volts to the pv input so it didn’t break in this case but i still cringed upon realizing the error)

cheers!

 

[MurphyGuy]

My chicken coop has a 5 watt solar panel and a Morningstar SL10 charge controller.. Their purpose is to charge a 12 volt battery for the automatic door mechanism.

The cheap little battery went bad so I disconnected it but forgot to disconnect the solar panel.. We didn't raise chickens for three years and it just sat there.

No damage.

 

[Cal]

I've seen pwm controller output voltage rise to the panel open circuit voltage when the battery is disconnected. In a 12V system, that would be about 22V.

Another destructive source comes from the fact electrons want to keep flowing when the circuit is opened at the battery. The moment the circuit is opened, inductance in the circuit generates a voltage spike. The greater the current, or the inductance (length of cable), the greater the voltage spike. The spike can get large enough to take out other electronics. Often power relay contacts get pitted and fail. This is due to the generated voltage spike which induces a spark across the contacts as they open. The spark burns the contacts.

The voltage spike may not take out the SCC, but it could destroy sensitive electronics that are still connected to the buss.

 

[MikeBou]

My Victron Smart 100-20 got fried. Victrons mppt calculator showed it would be ok on 2x 325W (ReneSolar Poly) at 24v. but i left it connected to solar whilst doing maintenance. it stopped charging at 24v totally. was just-about usable on a 12v system, but more like a PWM, low power for 2x 120W panels (a BSC3048 PWM gives more, depending on load). Firstly the Victron 100-20 had stopped switching off the load; later the current measuring of the load stopped. It is going back for repair (hopefully under warranty - and that is why they are so expensive). i also used a 100-50 on 4x 325W 2S2P got 45.5A max, but burned the battery positive (Positive Earth SCCs have an advantage there); also used an Epever 6415AN on that, but it did get very warm. Now using 2x Epever 6415AN each with just 2x 325W (2S seems the same as 2P) they dont get hot, i like the wiring box. Victrons all seem to have too small terminals; which do not take square crimped ferrules for the specified max size wire, which in any case is smaller than i would like

 

[acdoctor]

I disconnected the battery at the SCC in a charging system and smoked everything connected to the system but the SCC was fine!
I turned the breakers off from left to right instead of right to left.

 

[circus]

My Victron Smart 100-20 got fried. Victrons mppt calculator showed it would be ok on 2x 325W (ReneSolar Poly) at 24v

Are you sure? I'm too cheap to be personally aquainted with Victron but 750w/24v=31a... not 20a. Do charge controllers limit output no matter how much go's in, or will it just burn out?

 

[MikeBou]

Are you sure? I'm too cheap to be personally aquainted with Victron but 750w/24v=31a... not 20a. Do charge controllers limit output no matter how much go's in, or will it just burn out?

The Victrons will limit charge current, to their rated output. it is 650W max from panels, say into 26v 310Ah Lifepo4; so about 25A max. i was surprised the Victron mppt calculator suggested the 100-20, which i was using on a smaller 360W system, but had a spare BSC3048 for that.

 

[sunshine]

The Victrons will limit charge current,

Does this particular model limit charge?

https://www.victronenergy.com/upload/documents/Datasheet-BlueSolar-and-SmartSolar-charge-controller-overview-EN-.pdf

One line "Over temperature protection and power derating when temperature is high." doesn't give me confidence that this particular model has the same current limiting capacity as more expensive, much heavier, Victrons have.

 

[BradCagle]

I have a blown EPEver 6415AN

Only in service a few months. The battery did get fully discharged several times during overcast days.
Now, one of these overcast events (that lasted several days) the SCC died.

Was it from the battery/BMS disconnecting at night, and the sun hitting it in the morning with battery missing? I dunno, but that's what I assumed.

I tore it apart (after getting no support reply from EPEver) checked all the mosfets, fuse, and diodes. All fine, no shorts.
I pretty much traced it back to the micro controller board being dead.

 

[zorrobyte]

What happens if my battery's BMS kicks in protection due to overvoltage, or any other reason? Is this the same as the SCC being disconnected from a battery?