YVAA high IGBT temperatures

[Chillmac]

Hi folks,

Just been looking at a YVAA chiller on a site and the IGBT temperatures are running high.
It's not causing a problem just yet, but its heading that direction.

When the chiller is running system # 1 IGBT temp could be about 80+ DegC and System # 2 would typically be at about 60 DegC.
Its strange that both systems are different when they are on the same loop.
Internal ambient temp is 30 to 35 DegC.

Glycol has been changed and there is no air in the system. Both cooling coils are very clean.
Pump is running fine so no problems there. You can actually feel the high temp differential entering and leaving the IGBT's on the pipe fitting on the back of the panel.

Only thing I can think of would the heat compound on the plates be causing an issue?
Any come across this problem before?

It seams to be getting steadily worse over time, and he heading toward the trip point.

Any help greatly appreciated.

[Nuclrchiller]

"Glycol has been changed and there is no air in the system"

Are you absolutely certain of that? We needlessly replaced a pump on a YS before we realized air was the problem. We were certain it was the pump.

"... high temperature differential entering and leaving the IGBT's ..."

Low coolant flow will result in a high differential. Higher than normal. Numbers here (actual temperature readings of coolant in and coolant out) would be helpful.

[Chev5372]

If a newer chiller , 3-4 years old, the cooling loop is now separate coils hanging on the side of condenser coil "V" and fans are driven off VFDs . So if subcooling is at set point fans will be running slower and glycol temps. could higher then what was seen with straight driven fans.
Get some temps. in and out. The copper block that the IGBTs are sitting on is a single pass of coolant and the pipe connections are only 3/8 inch. I would drain out coolant and try back flushing the plate use water not a lot of pressure top of plate is held on with a lot of small screw and a thin o-ring. Try to catch the out flow in a clean bucket to see if something was in there.


Sent from my iPhone using Tapatalk

[turbomaster]

80*C is a little high, curious what was outdoor ambient and machine load when you saw 80*C? Interesting, cabinet temps were ok. I had a pair of YVAA's that the VSD cooling loop became contaminated by some substance that caused complete fouling. We suspect someone added something other than JCI/York coolant or contaminated coolant. Wound up cleaning loops and changing out the coils. I wouldnt suspect the heat compound unless this is a new machine and the issue has been a problem from the start or the plates have been pulled for some reason.
I have seen where the coolant has been changed and needed to be topped off after a few days.

[waltervan]

Have IGBT been removed? If so was fresh transfer paste used
I recently worked on a YCIV with the same issue turns out the another company remover IGBT for a visual inspection not sure why
after i removed old paste and used new no more temp issue

[Chev5372]

Wow that's a lot of work just to look at the IGBTs. Really wonder about people sometimes.


Sent from my iPhone using Tapatalk

[Chillmac]

Thank you for all the feedback. Best to take it one step at a time..

Nuclrchiller - Definitely no air in the glycol system. When the machine in running, glycol level is about 6 - 8 inches below the filler cap. The pump was run as soon as coolant reached the top of filler tank to circulate it throughout the system.
I will try and get you actual number today and post later this evening.

Chev - Correct, the chiller is about 3 years old and does indeed have the separate coils hanging on the V and fans are all variable speed.
I'll get actual numbers and suggest back flushing the system. Might tell us if there is a blockage in it.

Turbomaster - Over here yesterday was about 10*C so pretty cold. Which should even help lower the coolant temp albeit if the fans are running slower due to subcooling at setpoint. What would you use to flush out the system if fouling is suspected?

Waltervan - No IGBT's never removed. This issue has been getting steadily worse over the past 6 - 12 months I guess. So its telling me that its either fouling or a blockage in the system somewhere.

I will try and get actual temperatures and load conditions today and report back.

Thank you all. Have a good one.

[Chillmac]

OK guys,
Quick update.
Apparently the wrong coolant has been used in this machine. The result is that the coolant is mixing into a sludge type liquid in the micro channel coolant coils on the side of the chiller.
This is then causing flow issues and heat transfer problem as you might imaging through the tubes in the coils.

Looks like it need to be drained out and coils taken off and flushed clean.

I'll keep you posted and thanks again for your help.

[Chev5372]

What did they mix to make it foul up so bad? Would be a help for others to know, keeps people from droping in to site and only reading part of the story. Might save someone else a headache.

[Chillmac]

Trying to find out some more history from the site guys. As soon as I do I will let you know.

[Chev5372]

Thanks


Sent from my iPhone using Tapatalk

[Chasespeed]

FYI, there are several types of antifreeze that are not compatible. This is something run into with cars and trucks, more than chillers I suppose. But, in case I'm referencing, green is not compatible, with blue, red, or orange. It turns to sludge.

Not suggesting that is the case here, just a thought: Maybe someone topped it off with some anti freeze from a parts store?

Otherwise, sub'd..... I'm curious as well.

Chase


Sent from my iPad using Tapatalk

[Chev5372]

The lengths people will go to not to buy , yes we charge a lot for the glycol, the right stuff has now cost way more due to repairs.


Sent from my iPhone using Tapatalk

[Chillmac]

Sorry for the late update guys.

The plot as thickened, excuse the pun.
Apparently someone use car coolant in the system. This looks like it has reacted to the aluminium in the coolant coil and has blocked them solid. And I mean solid. Not sure if whatever they used was the cause of the problem, but whatever it was caused it to block the horizontal tubed in the coil.

So turbomaster called it right.
Reckon the coils will need to be changed out, but at least we know what the issue is.
Many thanks guys for your input.

[KnewYork]

People don't want to pay the high JCI prices for their fluids and parts so they "improvise". Usually without any investigation of the compatibility of the material being introduced with the materials in the machine. Often with disastrous results as the owner of this chiller as now found out.

[Chev5372]

Hopefully it did not plug the passages in the plate under IGBTs , may want to keep an eye on those come summer time.


Sent from my iPhone using Tapatalk

[Chillmac]

Agreed KnewYork and like you say the expenses is now increasing due to this.
You are right Chev, it is possible that the passages in the plates could be blocked, but I would think that because the bars in the coil are so small that, that would be the first place to block up. We will know once we free them or change them and run the systems again.
If the temps come back to normal then we are good, if not we will need to take the plate out and inspect.

[Shakeel Khan]

Glycol Heat exchanger need to check should be choked i was faced same problem