Classic running warm, now dead

[offgridQLD]

Dgd,
        that aluminum plate most likely wicks a lot of heat out of the case. Mind you if you don't have access to the rear of your classic you might not be able to feel the heat. My units I can put my hand around the back of them.  While the front feels warm the top left rear is where it's hot. The left side - top is reasonably warm to. Any where that inductor can radiate heat from  the location where its imbedded in the case.

As for the liquid cooling. I think it would stop the thing totally cooking but how I was testing it the thermal contact wasn't good enough to wick a moderate heat away from the case. Milling a flat section along with a O ring slot so that liquid could actually make contact with the case would be best. The other issue is it's another dynamic device that could fail and then you have no additional cooling if your relying on it.

It boiled down to the effort involved in getting it to perhaps do the job. It was less effort to just rewire the array to 2s and have it all work nice and cool  on its own (well as cool as a classic typically is = internal fans come on occasionally)

I guess in my view anyhow the classic will function at the high voc and low bat voltage (gap) with big arrays but is it happy doing so?  A bit like your car might have a 2 ton towing capacity (max) but it's much more comfortable with 1 ton if doing it regularly.

Then again perhaps I am to pedantic about temperatures and what I call hot isn't anything to worry about all all and as little to no effect on the components or service life.

Kurt

[wrt]

Hi guys,
I am back on solar power again, albeit in a limited fashion. The distributor looked after me after my retailer stepped up on my behalf. I now have a nice new one on the wall.

To Bob,

On the subject of the cause of the problem, I am interested in the temp limits of the classic, what temp does it shut down at? What temp would you normally see on the FETs on a35c day, or even on the odd 40c day (we get up to 7 days a year at this temp)

I guess what I am asking is, under normal circumstances should the classic have no problems under these conditions with panels putting out up to 120v. Perhaps non of this temp business makes a difference and I was just unlucky to get a crook FET that was waiting for the right combination of high temp and volts to fail.

Obviously I don't want to have to have my system reconfigured if it will no difference to the classics lifespan, but will if needed.

[Westbranch]

WRT, please add a system description to your Signature line in "Profile/Forum Profile / Signature" ..  after 61 posts all I can remember is a hooped  LiFePo4 Battery 

BW Have been delving into what Winston/Balquon have to say and separating out the chaff..  Mostly good as far as I can see, just need the cost to drop a bit.

[offgridQLD]

BW Have been delving into what Winston/Balquon have to say and separating out the chaff..  Mostly good as far as I can see, just need the cost to drop a bit.

I think you will find that the cost of the chinese lifepo4 batteries wont really come down. There is bugger all profit in them as it is. As it stands they are already less expensive than quality lead acid batteries with lots of benefits. Considering you cant really include the cost of cell management systems as thats a once off purchase.

Kurt

[wrt]

WRT, please add a system description to your Signature line in "Profile/Forum Profile / Signature" ..  after 61 posts all I can remember is a hooped  LiFePo4 Battery 

I have inserted the information you requested, though a quick read through my first three posts of the thread will soon jog the memory 

[dgd]

I guess what I am asking is, under normal circumstances should the classic have no problems under these conditions with panels putting out up to 120v. Perhaps non of this temp business makes a difference and I was just unlucky to get a crook FET that was waiting for the right combination of high temp and volts to fail.

Yes, under normal circumstances the Classic should have no problems.
I have just had two weeks of hot weather, 30 degrees C days and my Classic 150 with 90 to 110v PV input and output to a 24v battery bank.  FET temps  has been around 53, from 47 to 55
I would agree that you were probably just very unlucky to get a dud FET.
I'm sure that if this was a more common occurence then we would have heard more about it on this forum.
Also this recent trend here,  where getting input voltage closer to battery voltage = longer Classic life, is IMHO just nonsense and unproven.
In normal use input voltages up to the Classic's rated voltage should not be a problem.

dgd

[offgridQLD]

Also this recent trend here,  where getting input voltage closer to battery voltage = longer Classic life, is IMHO just nonsense and unproven.
In normal use input voltages up to the Classic's rated voltage should not be a problem.

Well the trend that most would agree with is cool power electronics = longer life.  And the for me anyhow considering I have two identical classic 150 charge controllers with the same size array on each but the only difference is one array is 3s 90 - 120v the other is 2s 70 - 80v. One runs a little hot and fans come on and off regularly + the case is soaked in heat from the inductor cooking away even under light loads . The other runs just warm with minimal fan time.

So the only question is how much less service life will I get out of the hotter classic .....time will tell ...actually it won't as I am switching it back to a lower voltage.

There is a big difference between 30c and 40C day .

Though I agree that a total failure due to running the classic at its upper voltage limits in hot weather must be rare. Though I still think it could reduce its life.

Kurt

[dgd]

Kurt,

I don't disagree with anything you say but I think normal circumstances was the issue.
I do not doubt that the extreme ambient temps you get and hot Classics are a concern AND that the lowering of input voltage WILL work for you.
However, I see your situation as exceptional and probably outside what would be normal for the Classic.  I think it was Vic that said keeping the Classic in an air-conditioned room would resolve the situation as lowering ambient temp was the best way to go.
That was why I keenly followed your Classic cooling experiments 

However, I would not see it as normal solution for an overheating Clssic to reduce input voltage. Lower the ambient or get a cooling system onto the Classic

dgd

[offgridQLD]

Yep I can agree with that. 

Perhaps I should put my view  another way as I didn't want to go around trying to scare people away from running high V input (it can be a handy option sometimes) just worth considering that if you're pushing the limits and or in a demanding environment its going to work against you when it comes to heat. Particularly if your controller is going to spend a lot of time idling along with a big array at its disposal that won't have its voltage pulled down much.

I mentioned my concerns to a electronic engineer friend of mine and he mentioned "I suppose if they use a passive freewheel diode, its losses would dominate, and they would increase with increasing array voltage. But I imagine that super-efficient MPPTs like the AERL models, would use another MOSFET to freewheel the inductor, and their losses might not be any different at high or low array voltages.

all went way over my head but he seemed to be confident in his idea.

Kurt

[boB]

The Classic will start current limiting if the FET temperature rises too fast in a couple minutes.

It will also limit current when the FET temperature rises to the high eighties degrees C.

The 250 starts current limiting at a wee bit lower temperature.  More like the lower eighties.
That might just be the 250KS though.

boB

[offgridQLD]

Thanks Bob,  handy to know.

Going by that it sounds like we are perhaps being a bit to pedantic about how hot our controllers run.

Kurt

[wrt]

Thank you indeed boB,

Given that Midnite chooses not to start limiting till the 80's I take this to mean that as long as my temp stays somewhere in the 50's it should be well within it's design parameters and would not have been a contributing factor in it's failure

Hopefully you do get a look at my unit. As you could imagine I am keen to find out what went wrong and do what I can to prevent it from happening again.

To dgd,
Giving it a little thought, I might lake a leaf from your book and move the unit a half inch out from the power board. I can do this with half inch thick ally mounting washers. This will allow the rear of the unit to convect heat adding a fourth surface.

Too simple not to do and has to achieve at least something 

[boB]

Thank you indeed boB,

Given that Midnite chooses not to start limiting till the 80's I take this to mean that as long as my temp stays somewhere in the 50's it should be well within it's design parameters and would not have been a contributing factor in it's failure

Hopefully you do get a look at my unit. As you could imagine I am keen to find out what went wrong and do what I can to prevent it from happening again.

To dgd,
Giving it a little thought, I might lake a leaf from your book and move the unit a half inch out from the power board. I can do this with half inch thick ally mounting washers. This will allow the rear of the unit to convect heat adding a fourth surface.

Too simple not to do and has to achieve at least something 

Yes, 50s and 60s in temperature is just nothing

I can take a look at your unit when it arrives.  Let me know if it is already here.

boB

[dgd]

My Classic 150 has been running with FET temp of 49 and pcb temp of 55 or thereabouts for the last few weeks.
ON Saturday I removed the top cover and secured it with one top bolt off to one side while re-routing some cabling.  I noticed after an hour or so the FET temp was down to 46 and the pcb to 42 even though it was a normal summers day with ambient 30.
So after finding some longer bolts and making four 1cm thick plastic standoffs that the bolts go through, I replaced the top cover.
So yesterday and today the temps are FET 45, PCB 40, ambient 29.
I can feel the warmish airflow out the gap at the top of the Classic.
Interesting experiment but will go back to normal setup soon as dont want insects or dust inside.
Might be useful if there was an alternate top cover available with easier airflow over the PCB.
Time now for cool beer at 3 degrees C 

dgd

[offgridQLD]

Interesting experiment.

I think where the thermocouple for the PCB  is placed might play a big roll in the temps shown. Often I have  BCB temps that are higher than my fet temp Say 45c FET 50c PCB. The issue with that I guess is that the fans are triggered by fet temps. So no matter how high the pcb temp are  the fans wont come on unit the fet temp gets to 48c or 58c for turbo fan.

When I see PCB temps higher than the fet temps is when the  inductor is wicking heat into the case under little load and high pv volts in . So the fetts aren't really doing much but the case is getting hot on the other side .

It's not like the classic is bug ,dust,insect proof in stock form. As the front cover has lots of open slots in it . Though I'm not sure how having a spacer stand off and a gap between the cover could affect the designed fan air flow path. That said when the fans are off ...(under 48c fet temp) with a space between  the cover it would most likely help get trapped hot air out.

All interesting.