Xantrex Battery Charger Overheating - Frigoboat Fridge

[twistedtree]

Sailormannn, I think your easiest next experiment is as suggested - a temporary direct wire from the fridge to the battery. The way folks have described the danfoss compressor, it will generate a pattern of high spikes of power draw even though it's a "DC" device. Each of these spikes will cause a momentary voltage drop. The inductance in the cabling, which can be ignored for truly DC loads, will not only increase the momentary voltage drop, but will also create a kind of kickback effect following the spike.

By creating a direct, short cable run between the battery and fridge you will minimize the voltage drops at the fridge. Using heavy gauge wire is critical to low resistance and low inductance. #6 cable would be good. It's an easy experiment and will answer a lot of questions, in particular whether there is any hope of the fridge and Xantrex ever coexisting.

Interestingly, a typical charger doesn't output true DC either, but rather a train of pulses very much like the load of the fridge. My personal theory, which is worth what you are paying for it, is that there is an alignment occurring between the output pulses of the charger and the load pulses of the fridge. When the charger pulses, instead of meeting the resistance of the battery which would normally limit current output, it instead encounters the spike load of the fridge and encounters something closer to a short circuit than a resistant battery. If the charger frequency is some multiple of the fridge frequency, then every N charger pulses are essentially short circuited, leading to an overheat.

[s/v Jedi]

Tell us more about your theory... how much high-power energy gets transferred between charger and compressor that way, and how can the tiny control module of the compressor absorb all that extra power from the big battery charger without burning up, while the big charger overheats?!


ciao!
Nick.

[senormechanico]

I agree with Twisted Tree who filled in what I've been saying all along.
The pulses generated are of very short duration, but rather high voltage.
Averaged over even a second of time, it's just the power level written on the plate of the fridge, i.e. 5 amps @ 12 volts or whatever.

Taken at the millisecond sampling level, it might look like a rather high voltage
(30VAC or higher).
The electronic module is busy turning 12 Volts DC into 3 phase AC, usually around 22VAC to several times higher depending on the size of the box the compressor needs to cool.

I'll bet you've heard people on HAM or SSB or even VHF radios who have a hellacious hum in the background whenever they talk.
That's their refrigerator running and modulating the battery voltage so much that it modulates the radio output.

[twistedtree]

Originally Posted by s/v Jedi

Tell us more about your theory... how much high-power energy gets transferred between charger and compressor that way, and how can the tiny control module of the compressor absorb all that extra power from the big battery charger without burning up, while the big charger overheats?!


ciao!
Nick.

The great thing about my theory is that not only is it free, it can't be proven or dis-proven by anyone but Xantrex, and even if they care or take the time to figure this out, I'm guessing they aren't ever going to tell us about it..

But just for giggles.... the power would be absorbed by the fridge just as though the charger weren't there, with most of it going to the compressor, and some small amount consumed by the controller switching circuits. Each time one of the compressor phases is switched on by the controller it draws a big slug of power from the battery. With each slug the battery voltage dips then recovers.

Now throw the charger into the mix. It's trying to maintain a battery charge voltage - say 14V. But during those voltage dips caused by the fridge the charger puts out short peaks of current attempting to maintain voltage. It's internal power dissipation will be a function of it's internal resistance and the square of the current. If it happens enough, it overheats.

Or, the charger is just a piece of junk and the software in it forgets to turn on the fan, or falsely detects and over temp condition, or a million other possibilities.

What still puzzles me about all this is the apparent interaction between the charger and fridge. I can see how running heavy gauge wire directly between the fridge and battery will make the fridge happier, but I don't see how it will make the charger happier. Regardless, it's an easy experiment and will rule out a few things.

[s/v Jedi]

Originally Posted by twistedtree

Now throw the charger into the mix. It's trying to maintain a battery charge voltage - say 14V. But during those voltage dips caused by the fridge the charger puts out short peaks of current attempting to maintain voltage. It's internal power dissipation will be a function of it's internal resistance and the square of the current. If it happens enough, it overheats.

Ah, I see what you mean. Fortunately, that is not how it works. For the charger to overheat, it's average power output must be higher than rated.

We're talking about a 40A charger, right? That would indicate that it can continuously supply a 40A current during charging but will probably be maxed out at that 40A at 13.8V. That gives us 40 x 13.8 = 550W power output which should just not overheat the unit. For the fridge compressor to overheat the charger, it must consume an average (not peak) of 550W. It would burn before doing that (a BD50 takes 7-10A max).

What we see here is a bug in the Xantrex firmware that does not allow the fan to operate in float charge phase. Without that air-flow, the rating of the charger gets reduced to a fraction of the 550W (say 50W), which is not enough to feed the fridge compressor.

The proof for that is that the overheat/overload symptom immediately stops when the charger is kicked into another mode (equalize) which triggers the fan to run again.

ciao!
Nick.

[twistedtree]

I think it all depends on the how big the peak currents are, and what the duty cycle is. The charger has some internal resistance which causes it to dissipate power and heat up. To the extent that internal resistance is a constant (which is a simplification that isn't entirely true for a switching power supply), the power dissipated internally will vary as the square of the current. If the charger doesn't do a good job of current limiting, it doesn't take much of a duty cycle of spikes to generate the same amount of waste heat as 40A DC.

Anyway, it's just a theory. Might as well get them all out there.

I missed that the fan doesn't run in float mode, but does run in other modes. If that is consistent, then I agree with you, and that's a much simpler, and hence likely explanation.

But I also understand that other loads equivalent to the fridge DO NOT trigger the overload - it's just the "special" load that the fridge presents. So what is it about the fridge load that's different? And that brings us back to the pulsed load from the controller driving the AC compressor.

[SeaweedSam]

I'm with BTrayfors... DUMP THE XANTREX! New unit not working properly? Where is your manufacturer support? Why did you even have to Query the forum? BS!!! Several others may do the job (IOTA eg). I don't wait in long restuarant lines, don't ever give return business to places with bad customer service or low quality product, don't buy products solely on their "brand name" reputation . Too many good products and businesses out there want your trade!! Good Luck!

[motion30]

Originally Posted by SeaweedSam

I'm with BTrayfors... DUMP THE XANTREX! New unit not working properly? Where is your manufacturer support? Why did you even have to Query the forum? BS!!! Several others may do the job (IOTA eg). I don't wait in long restuarant lines, don't ever give return business to places with bad customer service or low quality product, don't buy products solely on their "brand name" reputation . Too many good products and businesses out there want your trade!! Good Luck!

It is general knowledge xantrex has very poor product

[s/v Jedi]

Originally Posted by twistedtree

I think it all depends on the how big the peak currents are, and what the duty cycle is.

Exactly. When the duty cycle (I was worried to use the term but you seem familiar with it :-) is high enough to overload the charger, it would burn up the controller (which is a HF 3-phase inverter) of the compressor because it's electronics have just a fraction of the power ratings of the charger components.

ciao!
Nick.

[senormechanico]

Imho, what everybody is ignoring here is whether or not the charger is ACTUALLY overheating, or if it just THINKS it is.
I think it's probably not overheating, just getting fooled into that mode by the hi freq. noise from the fridge's inverter and exacerbated by the (probably) long wiring to the (still questionable) battery.

[DotDun]

Originally Posted by senormechanico

Imho, what everybody is ignoring here is whether or not the charger is ACTUALLY overheating, or if it just THINKS it is.
I think it's probably not overheating, just getting fooled into that mode by the hi freq. noise from the fridge's inverter and exacerbated by the (probably) long wiring to the (still questionable) battery.

+1

I'm also wondering if there is an old coil type shunt in the path keeping the batteries from filtering the AC.

New wire runs would answer a lot.

[Captain Bill]

The way I see it we have 3 working theories here. The first is that feedback from the fridge controller is causing the overheating. The second, that there is a firmware glitch that stops the fan from running in float mode. The third is that the unit is actually drawing much more power for brief instances that it is actually indicating causing the overheating. S/V Jedi, I would agree with your assessment that the low power draw from the fridge could not be causing a 40a charger to overheat, that is if it was being properly cooled. If the fan is not coming on, then it might be possible for the components to overheat. I think your assessment of a firmware error is the likely culprit, though I would call it a design error. My guess is that Xantrex tried to save a few cents per unit by not putting an actual temperature sensor in the unit, but deciding by some firmware algorithm that the fan needs to come on. I would guess that the designers never thought about the pulsed draw requirements of the danfoss controller. Why would this be? If one looks at the instructions accompanying most refrigeration units it says to connect the unit on a separate circuit directly to the battery. This allows the battery to filter the pulsed requirements and reduces feedback to other electronics. So we know that this fridge is not on a separate circuit, so why is it feeding back into the battery charger. I am beginning to wonder how the battery charger is hooked up. All of my charging sources connect to their own set of cables, through my pathfinder and then to the batteries. The power to the DC buss comes off the batteries through a separate set of cables, so the battery sits between the chargers and any loads. Is this the case in this instance or does the charger connect to the DC buss along with all the loads so the charger is exposed to the noise and pulsed draw of the fridge without the battery being in between? I doubt that the noise/feedback from the fridge is actually causing the overheating but It would make some sense that the pulsed nature of the fridges draw and the pulse width modulation of the charger might be fooling the firmware on the charger into deciding that the load was not enough to require the fan to cool the components. I think the idea of a test running directly from is a good one. I would guess all of this is a result of Xantrex trying to cut out a few components and do everything in software.

[adrac]

If any one cares before I dump this whole unit, I tested it in several different modes.... Yes the charger gets quite warm putting out minimal aperage in float mode (20% or less of its output capacity) You can manually program the charger to put out less amperage but in float mode it doesnt matter what you set it at it still flashes its overheating and is warm to the touch... It seems the fan will kick on in either equilization mode, or by manually setting battery temp to cold on the panel (it puts out more voltage than when its on hot or warm settings), and it does not seem to overheat when you put the charger on 2 stage... meaning the batteries drop to below 12.5 volts before the charger kicks on... I have a straight run from the charger to the batteries on the positive side and on my boat the charger overheats with or without the refer on. Unfortunately I assumed they had better quality like when they were called Stat Power.. I also believe I read an article in Practical sailor a few years back praising Xantrex, but take it for what its worth cause I dont specifically remember...On the humorous side the certificate on the box claims HALT (highly accelerated life tested) tested under extreme thermal and mechanical conditions.... I guess sitting at a dock drawing three amps of DC power is extreme.... IMHO the charger does overheat and it is due to pos components and a glitch in the programming and once the fan does kick on the faults and warnings stop. But what do I know Im just the proud owner of this flashing LED light machine. Any of you guys have any experience with pro mariners Pro Nautic 1230P charger?
Thanks for reading

[Sailormannn]

Originally Posted by adrac

If any one cares before I dump this whole unit, I tested it in several different modes.... Yes the charger gets quite warm putting out minimal aperage in float mode (20% or less of its output capacity) You can manually program the charger to put out less amperage but in float mode it doesnt matter what you set it at it still flashes its overheating and is warm to the touch... It seems the fan will kick on in either equilization mode, or by manually setting battery temp to cold on the panel (it puts out more voltage than when its on hot or warm settings), and it does not seem to overheat when you put the charger on 2 stage... meaning the batteries drop to below 12.5 volts before the charger kicks on... I have a straight run from the charger to the batteries on the positive side and on my boat the charger overheats with or without the refer on. Unfortunately I assumed they had better quality like when they were called Stat Power.. I also believe I read an article in Practical sailor a few years back praising Xantrex, but take it for what its worth cause I dont specifically remember...On the humorous side the certificate on the box claims HALT (highly accelerated life tested) tested under extreme thermal and mechanical conditions.... I guess sitting at a dock drawing three amps of DC power is extreme.... IMHO the charger does overheat and it is due to pos components and a glitch in the programming and once the fan does kick on the faults and warnings stop. But what do I know Im just the proud owner of this flashing LED light machine. Any of you guys have any experience with pro mariners Pro Nautic 1230P charger?
Thanks for reading

Interesting. It sounds like you have already done the test with the straight run from the refrigerator to the batteries and it hasn't made a difference. That's not very encouraging. I wonder if there are any users out there with a Xantrex TrueCharge2 running a small Danfoss refrigerator on float where the fan does come on to eventually cool the unit successfully. None of the three Xantrex chargers I have installed have done this. Hard to believe this is the case with every unit as suggested; wouldn't almost without every one of these units being returned defective. The guy that installed mine did the exact same install in another boat and he says it's fine.

Anyone out there found a charger and remote panel that will mount where the Xantrx remote was.

[bob_77903]

I have a Xantrex TrueCharge2, and a AB cold machine with Danfoss compressor. I have had no problems with the charger at all. The cooling fan runs fine. Only time I see the red temp light flashing is if I try to charge in the cold battery temp setting on a very hot day. When its set in the warm or hot battery mode, it functions fine. I suppose the naysayers here would want me to throw the Xantrex overboard and get an Iota. Haven't had any problems with the Xantrex, if I ever do I'll consider that option.