dslreports logo
site
 
    All Forums Hot Topics Gallery
spc

spacer




how-to block ads


Search Topic:
uniqs
2331
share rss forum feed

TheMG
Premium
join:2007-09-04
Canada
kudos:3
Reviews:
·NorthWest Tel

1 edit

Difluoroethane gas duster - oops, I fried a UPS!

Does anyone know if difluoroethane (the stuff in cans of air duster) in liquid state is electrically conductive? I couldn't find a definitive answer from Googling.

I kind of screwed up. I was blasting the dust out of a UPS and forgot to disconnect the batteries. The can was almost full, so moving the can around too quickly would cause liquid to come out. Well, as I was blasting the dust out near the chopper MOSFETs, a little bit of liquid came out of the can which resulted in fireworks and 4 completely blown open MOSFETs.


lugnut

@look.ca

Re: Difluoroethane gas duster - conductive?

I think you may have answered your own question there, though if I were to speculate it may be that the propellant is not conductive, but rather the refrigerant properties of the escaping gas caused condensation of airborne water vapor which in turn shorted out your board.

Regardless, it serves as a good reminder to everyone to always power down the equipment before attempting any type of service procedure.


shortckt
Watchen Das Blinken Lights
Premium
join:2000-12-05
Tenant Hell
reply to TheMG
Have you read the MSDS? If i'm reading it correctly, it doesn't mention conductive but it is considered a flammable substance.

I have to agree with lugnut there, the fast temp drop caused condensation which provided an arc path, then the flammability of the duster gas expanded on the problem and gave you the fireworks.

»www.sisweb.com/referenc/msds/dustoff.pdf

»en.wikipedia.org/wiki/1,1-Difluoroethane

TheMG
Premium
join:2007-09-04
Canada
kudos:3
Reviews:
·NorthWest Tel

1 edit
reply to lugnut
said by lugnut :

Regardless, it serves as a good reminder to everyone to always power down the equipment before attempting any type of service procedure.

Yeah, normally I'm careful about removing all sources of power, but for some reason this time I forgot.

Hopefully it's just the MOSFETs that fried. My preliminary assumption is that the conductive path turned "ON" the gates of the MOSFETs, and since the primary winding of the transformer is a DC short, well, you know the rest.

When I get some time I'll poke around it and figure out the extent of the damage. If it's just the MOSFETs I'll order some new ones and attempt a repair.

said by shortckt:

I have to agree with lugnut there, the fast temp drop caused condensation which provided an arc path, then the flammability of the duster gas expanded on the problem and gave you the fireworks.

Sounds like a plausible explanation.

With regards to the flammability, there were no flames as such, only arcing from the MOSFETs themselves which were completely blown open. The batteries in this UPS are 4x 12V 7Ah in series, so 48V.


lugnut

@look.ca
reply to shortckt
said by shortckt:

Have you read the MSDS? If i'm reading it correctly, it doesn't mention conductive but it is considered a flammable substance.

»www.sisweb.com/referenc/msds/dustoff.pdf

Wow! That data sheet reads like something you'd use to fire up a BBQ, not an ostensibly safe household cleaning product.

You'd probably have been better off blowing your dust with a can of lighter butane. At least you'd know what to expect.

TheMG
Premium
join:2007-09-04
Canada
kudos:3
Reviews:
·NorthWest Tel
reply to TheMG
Well I poked around the board with the trusty old Huntron. Turns out all twelve IRF2807 in the battery converter circuit are shorted (not just the three that physically blew). Not really a surprise, I kind of expected that.

The good news is that all transistors on the gate drive board test fine, so there is definitely hope.

Time to place an order for twelve IRF2807. Hopefully that does it.


shortckt
Watchen Das Blinken Lights
Premium
join:2000-12-05
Tenant Hell
reply to lugnut
said by lugnut :

Wow! That data sheet reads like something you'd use to fire up a BBQ, not an ostensibly safe household cleaning product....

Yes, and I was surprised the first time I saw the MSDS on this "canned air" that I find used in every office I've ever been in!! I avoid the stuff, and insist on having a small air compressor in any IT shop I've worked in. Cheaper in the long run too. My neighbor who works in the HVAC service business carries a CO2 tank with regulator and coiled air hose in his van, but that 65 lb. welding cylinder can be cumbersome to haul around.

TheMG
Premium
join:2007-09-04
Canada
kudos:3
Reviews:
·NorthWest Tel
reply to TheMG

Re: Difluoroethane gas duster - oops, I fried a UPS!

Click for full size
Click for full size
Well I figured this thread would be better with pictures. So here they are.

Board looks worse than it is. It actually cleaned up nicely and there is no trace or pad damage.

R12 R17 R19 are open and D343 is shorted. This is probably what saved the gate drive board from damage (all semiconductor devices on gate drive board test ok).

The good thing is all the bad components are still identifiable, since schematics for these UPS are pretty much unobtainium.


lugnut

@look.ca
The way those transistors blew open another possible theory comes to mind. Judging by the heatsinking and the size of the transistors those puppies run with an internal temperature about 100C. It's possible that the refrigerant effect of the canned air simply caused thermal contraction and the stress blew apart the transistors.

Either way, sudden temperature shocks are definitely not conducive to the well being of your electronic components.

b10010011
Whats a Posting tag?

join:2004-09-07
Bellingham, WA
Reviews:
·Comcast Formerl..
reply to TheMG
I like the stripped screw head on the third MOSFET.

That's a sign of some real quality workmanship there!
--
Bellingham Scanner Kicks Ass! »bhamscanner.kicks-ass.org/



SparkChaser
Premium
join:2000-06-06
Downingtown, PA
kudos:4
Reviews:
·Verizon FiOS
reply to lugnut
said by lugnut :

The way those transistors blew open another possible theory comes to mind. Judging by the heatsinking and the size of the transistors those puppies run with an internal temperature about 100C. It's possible that the refrigerant effect of the canned air simply caused thermal contraction and the stress blew apart the transistors.

Either way, sudden temperature shocks are definitely not conducive to the well being of your electronic components.

I've used a fair amount of freeze spray over the years and never saw that happen, not saying it can't happen.

I've switched to compressed air and a filter for any electronic blowing.
--
--
--
"Facts do not cease to exist because they are ignored." - Aldous Huxley

TheMG
Premium
join:2007-09-04
Canada
kudos:3
Reviews:
·NorthWest Tel
reply to b10010011
said by b10010011:

I like the stripped screw head on the third MOSFET.

No kidding. I was kind of surprised to see that, this was probably not a cheap UPS back in its day (Powerware 9125 2000VA).


lugnut

@look.ca
reply to SparkChaser
said by SparkChaser:

said by lugnut :

The way those transistors blew open another possible theory comes to mind. Judging by the heatsinking and the size of the transistors those puppies run with an internal temperature about 100C. It's possible that the refrigerant effect of the canned air simply caused thermal contraction and the stress blew apart the transistors.

Either way, sudden temperature shocks are definitely not conducive to the well being of your electronic components.

I've used a fair amount of freeze spray over the years and never saw that happen, not saying it can't happen.

Drop an ice cube into a pot of boiling water and watch what happens

Anyway, either one is a possible theory, either condensation shorting out or thermal stress blowing up, but judging by the appearance of those MOSFETS I'd have to say they literally exploded on contact.

TheMG
Premium
join:2007-09-04
Canada
kudos:3
Reviews:
·NorthWest Tel
said by lugnut :

but judging by the appearance of those MOSFETS I'd have to say they literally exploded on contact.

The damage to the MOSFETs was due to arcing. The arcing continued until I pulled the battery connector.

They didn't go "pop". It started with a sizzle, a puff of smoke, followed by arcs and sparks.


lugnut

@look.ca
In the immortal words of Emily Litella, (AKA Gilda Radner); "Never mind..."

public

join:2002-01-19
Santa Clara, CA
reply to TheMG
said by TheMG:

The damage to the MOSFETs was due to arcing. The arcing continued until I pulled the battery connector.

They didn't go "pop". It started with a sizzle, a puff of smoke, followed by arcs and sparks.

It is classic sustained ovecurrent damage where the die melts and the source lead separates.
When the current is somewhat limited it can take several seconds, and often results in a fire.

Kearnstd
Space Elf
Premium
join:2002-01-22
Mullica Hill, NJ
kudos:1
reply to TheMG
I am going to go with the spray actually having formed ice.

I honestly have never thought a UPS could get that dusty inside. I know I have to clean my PC. I check the air intake filters every few months as well.
--
[65 Arcanist]Filan(High Elf) Zone: Broadband Reports

TheMG
Premium
join:2007-09-04
Canada
kudos:3
Reviews:
·NorthWest Tel
said by Kearnstd:

I honestly have never thought a UPS could get that dusty inside. I know I have to clean my PC. I check the air intake filters every few months as well.

It's a double-conversion UPS, which means it dissipates heat all the time even when not working off the batteries, therefore there are two fans that are always running.

Quite different from typical "line interactive" or "standby" UPS units, which only run on inverter when there is a power failure, and either have no fan at all, or a fan that only runs when on battery. Those don't really accumulate any dust.

This UPS was left on the floor and pretty much neglected for years. The amount of dust inside it was absolutely incredible, the heatsinks were almost completely clogged with dust.

Kearnstd
Space Elf
Premium
join:2002-01-22
Mullica Hill, NJ
kudos:1
Ahh okay. so its constantly conditioning the power through the inverters rather than your typical bestbuy UPS that switches over kind of like how a generator cuts in for a whole house.

On a side note in relation to stating it must have cost a good amount when new.. the current model the 9130 the 2000VA unit runs around 1400 USD, And that is without the expansion batteries, just the internal ones.
--
[65 Arcanist]Filan(High Elf) Zone: Broadband Reports