Cleaning a power amp

[Phil Allison]

Hi,

ever since someone had the brain wave of putting "computer" fans inside power amps - service techs have had to deal with dust and fluff clogged heat sinks and filthy PCBs.

When such amps are installed in a music venue for years, cigarette smoke, artificial fog oil and moisture add to the mix making the result truly appalling.

In most cases thorough vacuuming and cleaning inside and outside surfaces with a damp cloth is sufficient. I know of techs using compressed air, but that seems terribly messy to me.

However, some example are so badly fouled the only way I have found to clean them properly is to submerge the PCB and heatsink assemblies in hot water and detergent and scrub them just like dinner plates and saucepans. In order to thoroughly dry the PCBs, I have used an oven set to low temp, hot air from a room heater or gun and simply the sun on a warm day.

This is obviously a very time consuming process requiring a lot of disassembly and reassembly, cleaning the fans themselves is one of the most tedious parts.

Recently I had to clean a couple of Crown XS700s that were completely fouled with black fluff that proved to be electrically conductive !!

Turns out they had been in the same rack with another amplifier that went up in smoke, filling them with carbonised fibreglass soot. This took more than the usual amount of time and care since a 2kW SMPS is built on the same PCB.

Anyone know an easier way or have useful comments?



..... Phil

 

[Phil Allison]

Rheilly Phoull wrote:

I suppose it depends on what sort of materials are on the pcb's etc but

I would be trying some suitable solvent like acetone or white spirit.

More costly but probably less time involved. It would need some capital

expenditure for a cleaning bay too.

** Acetone and white spirit attack many plastics - plus the fumes are harmful.

PCBs that have solder flux all over them need to be washed in solvent ( ie denatured alcohol) after being washed in hot water .

FYI: I have often imagined a bath full of Freon agitated by ultrasonics as the perfect answer.



..... Phil

 

[Phil Allison]

David Platt wrote:

However, some example are so badly fouled the only way I have found to clean them properly is to submerge the PCB and heatsink assemblies in

hot water and detergent and scrub them just like dinner plates and saucepans. In order to thoroughly dry the PCBs, I have used an oven set to

low temp, hot air from a room heater or gun and simply the sun on a warm day.



This is obviously a very time consuming process requiring a lot of disassembly and reassembly, cleaning the fans themselves is one of the most
tedious parts.


Recently I had to clean a couple of Crown XS700s that were completely fouled with black fluff that proved to be electrically conductive !!



Turns out they had been in the same rack with another amplifier that went up in smoke, filling them with carbonised fibreglass soot. This took

more than the usual amount of time and care since a 2kW SMPS is built on the same PCB.

Anyone know an easier way or have useful comments?



Bob Pease used to swear by the practice of putting PC boards into his

dishwasher, with "a full load of Calgonite" detergent (it's not full

of chlorine-based disinfectants), running them through the normal wash

and rinse cycle, and then drying. He swore that this was capable of

getting PC-board leakage down to levels below those that an expensive

"professional" solvent-based PC-board cleaning would deliver.



A manual rinse with deionized water before drying probably wouldn't be

a bad idea, if your local water is at all "hard".



I've heard of people using this approach on complete pieces of

equipment (after doing enough of a partial strip-down to remove

anything made of paper or cardboard) with some success. Depending on

how vigorously your dishwasher "sprays", it may be as effective at

removing build-up crud as a manual scrub would be.

** That is an interesting idea.

Likely Bill Pease used PCBs that were not smothered in fluff and disgusting contaminants like I mentioned - let alone conductive carbon soot.

Also, many amplifier PCBs have speaker and mains AC relays that are not water sealed plus water traps like DIL ICs in sockets and small plug and socket connectors.

The Crown amps were a major pain as any soot impregnated fluff left behind after washing and drying was likely to cause mayhem.




..... Phil

 

[Rheilly Phoull]

On 2/10/2014 8:34 AM, Phil Allison wrote:

Hi,

ever since someone had the brain wave of putting "computer" fans inside power amps - service techs have had to deal with dust and fluff clogged heat sinks and filthy PCBs.

When such amps are installed in a music venue for years, cigarette smoke, artificial fog oil and moisture add to the mix making the result truly appalling.

In most cases thorough vacuuming and cleaning inside and outside surfaces with a damp cloth is sufficient. I know of techs using compressed air, but that seems terribly messy to me.

However, some example are so badly fouled the only way I have found to clean them properly is to submerge the PCB and heatsink assemblies in hot water and detergent and scrub them just like dinner plates and saucepans. In order to thoroughly dry the PCBs, I have used an oven set to low temp, hot air from a room heater or gun and simply the sun on a warm day.

This is obviously a very time consuming process requiring a lot of disassembly and reassembly, cleaning the fans themselves is one of the most tedious parts.

Recently I had to clean a couple of Crown XS700s that were completely fouled with black fluff that proved to be electrically conductive !!

Turns out they had been in the same rack with another amplifier that went up in smoke, filling them with carbonised fibreglass soot. This took more than the usual amount of time and care since a 2kW SMPS is built on the same PCB.

Anyone know an easier way or have useful comments?



.... Phil

I suppose it depends on what sort of materials are on the pcb's etc but
I would be trying some suitable solvent like acetone or white spirit.
More costly but probably less time involved. It would need some capital
expenditure for a cleaning bay too.

 

[David Platt]

However, some example are so badly fouled the only way I have found to clean them properly is to submerge the PCB and heatsink assemblies in
hot water and detergent and scrub them just like dinner plates and saucepans. In order to thoroughly dry the PCBs, I have used an oven set to
low temp, hot air from a room heater or gun and simply the sun on a warm day.

This is obviously a very time consuming process requiring a lot of disassembly and reassembly, cleaning the fans themselves is one of the most
tedious parts.

Recently I had to clean a couple of Crown XS700s that were completely fouled with black fluff that proved to be electrically conductive !!

Turns out they had been in the same rack with another amplifier that went up in smoke, filling them with carbonised fibreglass soot. This took
more than the usual amount of time and care since a 2kW SMPS is built on the same PCB.

Anyone know an easier way or have useful comments?

I suppose it depends on what sort of materials are on the pcb's etc but
I would be trying some suitable solvent like acetone or white spirit.
More costly but probably less time involved. It would need some capital
expenditure for a cleaning bay too.

I wouldn't ever use acetone for that purpose... it's too hard on too
many plastics.

Bob Pease used to swear by the practice of putting PC boards into his
dishwasher, with "a full load of Calgonite" detergent (it's not full
of chlorine-based disinfectants), running them through the normal wash
and rinse cycle, and then drying. He swore that this was capable of
getting PC-board leakage down to levels below those that an expensive
"professional" solvent-based PC-board cleaning would deliver.

A manual rinse with deionized water before drying probably wouldn't be
a bad idea, if your local water is at all "hard".

I've heard of people using this approach on complete pieces of
equipment (after doing enough of a partial strip-down to remove
anything made of paper or cardboard) with some success. Depending on
how vigorously your dishwasher "sprays", it may be as effective at
removing build-up crud as a manual scrub would be.

 

[Trevor Wilson]

On 2/10/2014 10:34 AM, Phil Allison wrote:

Hi,

ever since someone had the brain wave of putting "computer" fans
inside power amps - service techs have had to deal with dust and
fluff clogged heat sinks and filthy PCBs.

When such amps are installed in a music venue for years, cigarette
smoke, artificial fog oil and moisture add to the mix making the
result truly appalling.

In most cases thorough vacuuming and cleaning inside and outside
surfaces with a damp cloth is sufficient. I know of techs using
compressed air, but that seems terribly messy to me.

However, some example are so badly fouled the only way I have found
to clean them properly is to submerge the PCB and heatsink assemblies
in hot water and detergent and scrub them just like dinner plates and
saucepans. In order to thoroughly dry the PCBs, I have used an oven
set to low temp, hot air from a room heater or gun and simply the sun
on a warm day.

This is obviously a very time consuming process requiring a lot of
disassembly and reassembly, cleaning the fans themselves is one of
the most tedious parts.

Recently I had to clean a couple of Crown XS700s that were completely
fouled with black fluff that proved to be electrically conductive !!

Turns out they had been in the same rack with another amplifier that
went up in smoke, filling them with carbonised fibreglass soot. This
took more than the usual amount of time and care since a 2kW SMPS is
built on the same PCB.

Anyone know an easier way or have useful comments?

[DISCLAIMER] None of what follows will be remotely helpful.

**Ah, that brings back memories. I used to keep a 5 Litre container of
Freon TF™, along with several spray cans (they even supplied a nifty
clip-on brush thingy). Brilliant stuff. No residue, did not hurt
plastics, non-flammamble, disolved the nastiest greases and oils, not
harmful to humans and possessed a low vapourisation temperature.

The perfect cleaner. Bummer about the ozone layer. The stuff was banned
decades ago.



--
Trevor Wilson www.rageaudio.com.au

 

[Phil Allison]

Nutcase Kook :

Only military grade ICs are specced to hermetic sealed packages.

** Horse manure.


For all

else, water is likely to migrate to the die with some probability

greater than zero, only needs going along one pin of one IC.

** Shame how nearly all PCBs are washed ( ie defluxed) in hot water these days.

Wot a bullshitting fool you are.


..... Phil

 

[N_Cook]

On 02/10/2014 03:24, David Platt wrote:

However, some example are so badly fouled the only way I have found to clean them properly is to submerge the PCB and heatsink assemblies in
hot water and detergent and scrub them just like dinner plates and saucepans. In order to thoroughly dry the PCBs, I have used an oven set to
low temp, hot air from a room heater or gun and simply the sun on a warm day.

This is obviously a very time consuming process requiring a lot of disassembly and reassembly, cleaning the fans themselves is one of the most
tedious parts.

Recently I had to clean a couple of Crown XS700s that were completely fouled with black fluff that proved to be electrically conductive !!

Turns out they had been in the same rack with another amplifier that went up in smoke, filling them with carbonised fibreglass soot. This took
more than the usual amount of time and care since a 2kW SMPS is built on the same PCB.

Anyone know an easier way or have useful comments?

I suppose it depends on what sort of materials are on the pcb's etc but
I would be trying some suitable solvent like acetone or white spirit.
More costly but probably less time involved. It would need some capital
expenditure for a cleaning bay too.

I wouldn't ever use acetone for that purpose... it's too hard on too
many plastics.

Bob Pease used to swear by the practice of putting PC boards into his
dishwasher, with "a full load of Calgonite" detergent (it's not full
of chlorine-based disinfectants), running them through the normal wash
and rinse cycle, and then drying. He swore that this was capable of
getting PC-board leakage down to levels below those that an expensive
"professional" solvent-based PC-board cleaning would deliver.

A manual rinse with deionized water before drying probably wouldn't be
a bad idea, if your local water is at all "hard".

I've heard of people using this approach on complete pieces of
equipment (after doing enough of a partial strip-down to remove
anything made of paper or cardboard) with some success. Depending on
how vigorously your dishwasher "sprays", it may be as effective at
removing build-up crud as a manual scrub would be.

Only military grade ICs are specced to hermetic sealed packages. For all
else, water is likely to migrate to the die with some probability
greater than zero, only needs going along one pin of one IC. Once water
is in there , how to expel, without leaving any trace? Is dry heating
all to <>100 deg C, going to do that?

 

[John-Del]

On Wednesday, October 1, 2014 8:34:03 PM UTC-4, Phil Allison wrote:

> However, some example are so badly fouled the only way I have found to clean them properly is to submerge the PCB and heatsink assemblies in hot water and detergent and scrub them just like dinner plates and saucepans. In order to thoroughly dry the PCBs, I have used an oven set to low temp, hot air from a room heater or gun and simply the sun on a warm day.

Back in the late 70s when TVs had large,hot running modules and it seemed like everyone smoked, I used to wash these boards before rebuilding them.

What I would do is remove any component that could trap water, such as flybacks, smps xfrs, large inductors, tuning coils, tuner, etc. then wash the board in the sink with hot water and a cleaning solution my brother (research chemist) used to make for me by the gallon. It was also great stuff to clean white walled tires with (some kind of detergent and stoddard solvent mixture).

I would let them drip dry, then put them in a large cardboard box with a hole in one end the diameter of a hair dryer, and let the board dry for an hour or so. I'd clean by hand the parts I had previously removed and reinstall them in the board then repair or rebuild the board as needed. The modules were squeaky clean and shiny. There was NEVER an issue with submerging the boards in water, and I did not rinse them with anything more than tap water.

 

I've had worse. In the RPTV days, some of them would get coolant leaks. The coolant is a mixture of 70 % ethylene glocol and 30 % glycerine. It starts out as nonconductive but when there is a voltage appled it become conductive and corrosive. It also soaks into the board itself.

I used to temperature cycle them a few times to "squeeze" the crap out of the board. Don'tcha think it might be a bad idea to make them our of porous material ? Oh yeah, sure is, wrecks the thing in some cases. In fact if it leaked on deflection or HV it could cause a fire, prompting them to put in a gutter.

Anyway, it doesn't sound like your boards would be impregnated with anything, at least it is all on the surface. I never submerged boards, just doesn't seem like a good idea. But I did take them to the sink and use hot water, and acetone, and alcohol. Of course I had the board clean out of the unit so any components that could be damaged I could avoid easily.

For scrubbing I used a small paint brush with the bristles cut short for stiffness. I started with acetone, then rinsed that off with hot water. Then doused it with 91 % isopropyl alcohol and kinda shook it down. Then a hair dryer. I found I had to repeat the process a few times because of the stuff in the pores but you will probably not have have particular problem.

You can get away with acetone if you don't drown the boards with it.

 

[William Sommerwerck]

There are foaming cleaners with citrus-based solvents (such as Curtisol --
which appears to be discontinued) that were used by antique radio collectors.
After sealing up components that could be damaged by water, the circuitry was
doused with the cleaner and allowed to sit for a while. It was then hosed off
and allowed to dry in a warm place.

 

[RobertMacy]

On Wed, 01 Oct 2014 18:52:06 -0700, Phil Allison <pallison49@gmail.com>
wrote:

...snip....
** Acetone and white spirit attack many plastics - plus the fumes are
harmful.
PCBs that have solder flux all over them need to be washed in solvent (
ie denatured alcohol) after being washed in hot water .

FYI: I have often imagined a bath full of Freon agitated by ultrasonics
as the perfect answer.



.... Phil

Years ago, we washed loaded PCB's in de-ionized water, then to accelerate
the drying process dissolved a bit of that water off the board with
alcohol. Later we found that inundateing the electrolytics with alcohol
did something to them that sounded a 'deathknell' for them over time. In
otherwords, the boards died a premature death, albeit a clean death.

 

I wonder if it would pay to use a conformal coating if you happen to know the customer is going to put the unit into the same environment again.

 

[Phil Allison]

jurb...@gmail.com wrote:

I wonder if it would pay to use a conformal coating if you happen to know the customer is going to put the unit into the same environment again.

** One maker of Mosfet power amplifier( Perreaux of NZ) had a major issue with PCB dust/fluff contamination - it made the amps crackle loudly or even blow up spectacularly.

Two mistakes;

1. There were signal tracks that ran parallel along the main PCB, spaced by about 0.4mm that differed by 240 volts DC.

2. The PCB was bare - ie not coated in the usual green resist.

After some years of service, "tracking" began across that 0.4mm gap - after cleaning, one could see the area was darkened in many places.

The usual fix was to thoroughly clean the PCB with alcohol, scratch away any darkened material and then coat the PCB with a clear lacquer sold for the purpose.

With really bad examples, I peeled off the offending tracks and fitted plastic coated wires in place.



..... Phil

 

[Jon Elson]

Phil Allison wrote:

Nutcase Kook :


Only military grade ICs are specced to hermetic sealed packages.


** Horse manure.


For all

else, water is likely to migrate to the die with some probability

greater than zero, only needs going along one pin of one IC.



** Shame how nearly all PCBs are washed ( ie defluxed) in hot water these
days.

Wot a bullshitting fool you are.

For new components, I agree with you. For boards that have run hot
for a long time, many power/thermal cycles, exposure to possibly
hideously contaminated air, I am not sure I agree. I think that
under such conditions, the sealing of the package definitely may
degrade. The chips themselves are passivated way better than in
the early days, too, and that helps.

On the other hand, getting things cleaned up well, and then make
sure to properly dry them before powering on, it should be OK.
Once the chips get warmed back up, the water will be driven out.
Gear that runs stone cold should definitely be dried carefully
(ie. some heat applied).

Jon

 

>"1. There were signal tracks that ran parallel along the main PCB, spaced by >about 0.4mm that differed by 240 volts DC. "

Thats fucking crazy. Really not, but it is pushing it, I cannot beleive they HAD to do that. They did it to induce hopefully unrepairable failures. Assholes, but oI do not say stupid or they fuxcked up, they are just assholes ot the world.


>"2. The PCB was bare - ie not coated in the usual green resist"

No solder mask ? (that is what we call that shit here)

With the cost of solder, most people figured out it was a good idea to use a solder mask, and this was true long time ago. Were they high or stupid ? Or can you discern ANY logical reason they did this ? Of course they could hve been incompetent, most of the people I have ever met were.

I do not know. Iknow some shit but Iam not really an engineer. If soeone can figure ut anything about why they did shit like they did except for to make planned obsolescenc and shit to sell more product, tell me.

But all I see frm any of threm is fuck service, it is a $ 5,000 toaster.

If we want to have a service indeustry at all in the next few yrrars, we really have to get people to stop buying this new shit and get into not getting rid of the old shit. there is no choice.

 

[whit3rd]

On Wednesday, October 1, 2014 5:34:03 PM UTC-7, Phil Allison wrote:

ever since someone had the brain wave of putting "computer" fans inside power amps - service techs have had to deal with dust and fluff clogged heat sinks and filthy PCBs.



When such amps are installed in a music venue for years, cigarette smoke, artificial fog oil and moisture add to the mix making the result truly appalling.



In most cases thorough vacuuming and cleaning inside and outside surfaces with a damp cloth is sufficient.

Recently I had to clean a couple of Crown XS700s that were completely fouled with black fluff that proved to be electrically conductive !!

I'd try Plasti-Tac and/or Handi-Tak. Hard to describe, these are clay-like sticky
products, you rub 'em like an eraser over the surface to be cleaned, then
knead them like bread dough and the crud mixes in and never is seen again.

For some scorched and corroded gizmos, I've also used glass-bead
blasting.

For anything large, though, the old (before water-clean flux) solution was
an alcohol/water azeotrope (mixture). Rinse, distill the dirty solvent, and use it
to rinse again. As many times as needed...

 

[N_Cook]

On 02/10/2014 22:40, Jon Elson wrote:

Phil Allison wrote:

Nutcase Kook :


Only military grade ICs are specced to hermetic sealed packages.


** Horse manure.


For all

else, water is likely to migrate to the die with some probability

greater than zero, only needs going along one pin of one IC.



** Shame how nearly all PCBs are washed ( ie defluxed) in hot water these
days.

Wot a bullshitting fool you are.
For new components, I agree with you. For boards that have run hot
for a long time, many power/thermal cycles, exposure to possibly
hideously contaminated air, I am not sure I agree. I think that
under such conditions, the sealing of the package definitely may
degrade. The chips themselves are passivated way better than in
the early days, too, and that helps.

On the other hand, getting things cleaned up well, and then make
sure to properly dry them before powering on, it should be OK.
Once the chips get warmed back up, the water will be driven out.
Gear that runs stone cold should definitely be dried carefully
(ie. some heat applied).

Jon

Once the thermal cycling has broken the seal somewhere, then the gap is
of ideal capillary-action dimension to suck up any water

 

[William Sommerwerck]

"Phil Allison" wrote in message
news:2d6deca9-62e9-4e20-8ab7-dbd1df1d7b72@googlegroups.com...

** One maker of Mosfet power amplifiers (Perreaux of NZ) had a
major issue with PCB dust/fluff contamination - it made the amps
crackle loudly or even blow up spectacularly.

There were signal tracks that ran parallel along the main PCB,
spaced by about 0.4mm, that differed by 240 volts DC.

I remember Perreaux. I thought, with a name like that, their products had to
be dogs. Looks like they were.

 

[chuck]

On Thu, 2 Oct 2014 17:52:25 -0700 (PDT), Phil Allison
<pallison49@gmail.com> wrote:

jurb...@gmail.com wrote:

I wonder if it would pay to use a conformal coating if you happen to know the customer is going to put the unit into the same environment again.


** One maker of Mosfet power amplifier( Perreaux of NZ) had a major issue with PCB dust/fluff contamination - it made the amps crackle loudly or even blow up spectacularly.

Two mistakes;

1. There were signal tracks that ran parallel along the main PCB, spaced by about 0.4mm that differed by 240 volts DC.

2. The PCB was bare - ie not coated in the usual green resist.

After some years of service, "tracking" began across that 0.4mm gap - after cleaning, one could see the area was darkened in many places.

The usual fix was to thoroughly clean the PCB with alcohol, scratch away any darkened material and then coat the PCB with a clear lacquer sold for the purpose.

With really bad examples, I peeled off the offending tracks and fitted plastic coated wires in place.



.... Phil

Another fun problem that they had with a mid 80s model amp was an
intermittent popping that might show up once a week or so. This was
caused by a small signal diode that when heated or cooled didn't
exhibit the symptom. I spent weeks on the first one before resolving
this issue.