UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > General Vintage Technology Discussions

Notices

General Vintage Technology Discussions For general discussions about vintage radio and other vintage electronics etc.

Closed Thread
 
Thread Tools
Old 23rd Oct 2020, 8:47 am   #1
stevehertz
Dekatron
 
stevehertz's Avatar
 
Join Date: Sep 2011
Location: Rugeley, Staffordshire, UK.
Posts: 8,809
Default Heatsink compound - refresh or not?

I restore a lot of vintage hi-fi receivers and it goes without saying they all have power transistors mounted on heatsinks and the use of heatsink compound to provide a good thermal connection between the two. I have never refreshed or replaced the compound unless I have replaced a transistor.

On a true knowledge basis, not the mere notion that it may be good practice, should heatsink compound be changed at some point in order to maintain its efficiency ? I'm not talking about changing it when, for whatever reason, an associated power transistor is removed and it's easy to do so. That quite possibly makes good sense if only for the reason that fresh compound will compress and mould easier to the transistor and heatsink, making a more intimate joint.

I ask for fact based responses because in vintage hi-fi circles a lot of people habitually 'write off' heatsink compound that has happily been in place for decades. Does heatsink compound lose its efficiency as it ages in an installation? On an undisturbed joint, is so called 'hard', aged heatsink compound poorer than when it's fresh?

As an aside, on package types that allow it, I whizz the flat side of new power transistors over fine abrasive paper on a flat surface to remove any potential hot spots and provide for better thermal conduction via the compound and heatsink.
__________________
A digital radio is the latest thing, but a vintage wireless is forever..
stevehertz is online now  
Old 23rd Oct 2020, 9:10 am   #2
Andrew2
Nonode
 
Andrew2's Avatar
 
Join Date: Mar 2010
Location: Dukinfield, Cheshire, UK.
Posts: 2,034
Default Re: Heatsink compound - refresh or not?

It's a good question and one that I can't answer with any objectivity, but if ever I'm replacing a duff output device I always remove the old compound and clean the area with switch cleaner and dry thoroughly before adding new stuff. It just seems the 'right thing' to me, as the old stuff is always looking dry and gives the impression that it may not be quite so efficient any more.
__________________
Andy G1HBE.
Andrew2 is online now  
Old 23rd Oct 2020, 9:23 am   #3
GMB
Dekatron
 
GMB's Avatar
 
Join Date: Aug 2003
Location: near Reading (and sometimes Torquay)
Posts: 3,086
Default Re: Heatsink compound - refresh or not?

Well how about finding out, if you do this often?

I suggest that you carefully check out all measurable parameters including temperature done by pyrothermometer of an old system, then upgrade half of the output pair and see if anything changes. I think measuring the off-load base-emitter voltages after it has been working hard and getting hot, if done with a digital meter, should give an indication of whether it was useful to change it.

My experience is that the perhaps hardened compound seen on the outside is not a good indication of what is underneath the actual components. My guess is that I would not expect the compound to "dry out" (if that is even possible) where it is not exposed to the air.

I would not gratuitiously do anyway if it involved desoldering a part as I feel that is more dangerous than leaving it alone.

In my case there is an extra parameter, maybe for you too? I use so little of this compound that I am still using a tube of it that I bought about 50 years ago. So the "new" compound may be older than the "old" stuff on the unit itself!
GMB is offline  
Old 23rd Oct 2020, 9:36 am   #4
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Heatsink compound - refresh or not?

I honestly would not think of flattening the backs of power transistors, TO3 or not, The main problem is lack of flatness of the heatsink itself.

The problem is that any asymmetric extrusion - such as a finned heatsink - is very far from a flat back against which to attach power transistors. This is as a result of stresses induced during extrusion. I typically measure 0.2mm banana shape over a large heatsink, with the edges high and the centre dished cylindrically.

The Aavid catalog make this very specific, and say that if you want a flat back the heatsink should be post-machined. Or they will do it if asked and levy a significant charge to do so.

I'd only clean and use new heatsink compound if I was replacing power transistors because of failure. Even the loaded white compound is silicone grease based, and that does not evaporate. So no reason to replace it as a matter of course. Besides which, the purpose of the compound is just to fill in surface roughness (assuming the heatsink is flat!). As such only a very thin layer is needed. I typically apply it with an artist's spatula and apply a bare amount. You just need enough so that there is a spare squeeze out at the correct bolt torque for the package.

Craig
Craig Sawyers is offline  
Old 23rd Oct 2020, 9:36 am   #5
CambridgeWorks
Nonode
 
CambridgeWorks's Avatar
 
Join Date: Jun 2010
Location: Spalding, Peterborough, Cambridgeshire, UK.
Posts: 2,851
Default Re: Heatsink compound - refresh or not?

Can't answer the question, but as an aside, before I retired, I repaired vhf and uhf tx/rx. The biggest was 265W AM, VHF. Gemini mosfets in output stages.
The most common was the 50W PA, with just a pair of MRF174 O/P.
Replacements were always lapped using I think 400 grade wet and dry. I applied the compound as a slightly thicker smear from the tube than required and always used the shiny, wax like side of the backing paper as used on sticky labels to wipe it across. This then gave a nice even and light coating. Screws were then tightened to correct torque.

It will be interesting to see if there is any evidence for degradation given in any replies here.
Rob

In response to GMB , i have seen on a lot older, non work related gesr where upon the removal, a lot of a powdery substance is left, rather than a sticky grease.
Craig, in our manufacturing, QA went to great lengths checking surface flatness on every PA heatsink. These were allcustom made and machined.
PA Transistors I think were just batch sampled.
I recall once a MRF174 where the gold plated flange was visibly warped!!
Usually, lapping would soon show the previous high points that had been rubbed away.
__________________
Apprehension creeping like a tube train up your spine - Cymbaline. Film More soundtrack - Pink Floyd

Last edited by CambridgeWorks; 23rd Oct 2020 at 9:46 am. Reason: Added responses
CambridgeWorks is online now  
Old 23rd Oct 2020, 9:36 am   #6
barrymagrec
Octode
 
Join Date: Dec 2017
Location: Morden, Surrey, UK.
Posts: 1,552
Default Re: Heatsink compound - refresh or not?

I would obviously always use new heatsink compound if the device had been removed /replaced for any reason but I wouldn`t interfere with existing compound if there was no other reason to.

Obviously the compound hardens over time but this is mostly the excess which is doing nothing anyway.

On new builds I use the thin silicone pads - much less messy.
barrymagrec is offline  
Old 23rd Oct 2020, 9:39 am   #7
stevehertz
Dekatron
 
stevehertz's Avatar
 
Join Date: Sep 2011
Location: Rugeley, Staffordshire, UK.
Posts: 8,809
Default Re: Heatsink compound - refresh or not?

Just to re-iterate what I have already said, I do, and it's easy to replace heatsink compound when replacing a power transistor. My point/question is, does heatsink compound lose its efficiency over time, especially a few decades.
__________________
A digital radio is the latest thing, but a vintage wireless is forever..
stevehertz is online now  
Old 23rd Oct 2020, 9:43 am   #8
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Heatsink compound - refresh or not?

To answer that - I personally have no idea.

Craig
Craig Sawyers is offline  
Old 23rd Oct 2020, 9:43 am   #9
stevehertz
Dekatron
 
stevehertz's Avatar
 
Join Date: Sep 2011
Location: Rugeley, Staffordshire, UK.
Posts: 8,809
Default Re: Heatsink compound - refresh or not?

Quote:
Originally Posted by Craig Sawyers View Post
I honestly would not think of flattening the backs of power transistors, TO3 or not, The main problem is lack of flatness of the heatsink itself.

The problem is that any asymmetric extrusion - such as a finned heatsink - is very far from a flat back against which to attach power transistors. This is as a result of stresses induced during extrusion. I typically measure 0.2mm banana shape over a large heatsink, with the edges high and the centre dished cylindrically.

The Aavid catalog make this very specific, and say that if you want a flat back the heatsink should be post-machined. Or they will do it if asked and levy a significant charge to do so.

I'd only clean and use new heatsink compound if I was replacing power transistors because of failure. Even the loaded white compound is silicone grease based, and that does not evaporate. So no reason to replace it as a matter of course.

Craig
I think you've over thought my flattening of the backs of power transistors Craig. I only use very, very fine abrasive paper and the part is only rubbed for a short time to remove any obvious high spots. There is no heavy 'flattening' going on.

Regarding the heatsinks themselves, there's not a lot you can do about surface anomalies unless you go to the extent of removing them and performing some kind of surface flattening. And why anyway if all work well? But I take the point that they're not always flat.
__________________
A digital radio is the latest thing, but a vintage wireless is forever..
stevehertz is online now  
Old 23rd Oct 2020, 9:50 am   #10
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: Heatsink compound - refresh or not?

Quote:
Originally Posted by stevehertz View Post
Just to re-iterate what I have already said, I do, and it's easy to replace heatsink compound when replacing a power transistor. My point/question is, does heatsink compound lose its efficiency over time, especially a few decades.
It depends on what type it is. There are oxide compounds and silicon compounds out there. The oxide compounds crack and flake out of the DUT is vibrated. The silicon compounds are actually glass-like and will drip out eventually. They both have a finite lifespan really.

I tend to replace if I’m in stuff. The newer oxide compounds are pretty good and have much better thermal conductivity too. I’ve also found the stuff isn’t necessarily well applied to start with!

Can recommend Arctic Silver MX4 which is about £5 for a reasonable syringe full on amazon. Designed for CPUs but works very very well compared to much more expensive stuff.

I’ve never sanded or polished a surface. The point of the thermal compound is so you don’t have to do that.
MrBungle is offline  
Old 23rd Oct 2020, 12:02 pm   #11
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Heatsink compound - refresh or not?

These are worth a read. They do not address the topic of longevity of interface greases, but they do include quite a bit on surface flatness, roughness, interface thermal resistance using different materials greased or not, and tightening torques for different packages.

I use a torque screwdriver (with a 1/4" square fitment for screwdriver and sockets). It is always surprising how little torque is needed; the tendency of using the standard engineering grunt is that semiconductors are over-tightened to the extent that the flange is distorted - particularly with TO220 and related family packages.

Craig
Attached Files
File Type: pdf AN1040-D semiconductor mounting considerations.PDF (262.5 KB, 36 views)
File Type: pdf Semiconductor mountin AND9859-D.PDF (692.3 KB, 38 views)
Craig Sawyers is offline  
Old 23rd Oct 2020, 12:15 pm   #12
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,923
Default Re: Heatsink compound - refresh or not?

I don't ever recall seeing compound dry out or run out.

But I suppose there have been many different products over the years, and when in service, some running at quite low temperatures and some really quite hot. With such diversity, some compounds may have survived very well and others quite poorly.

Perhaps one question to consider is how hard is the device being run, how close to it's limits? Does it really need the the paste to be at it's very best?

B
__________________
Saturn V had 6 million pounds of fuel. It would take thirty thousand strong men to lift it an inch.
Bazz4CQJ is offline  
Old 23rd Oct 2020, 1:14 pm   #13
broadgage
Nonode
 
Join Date: Jan 2010
Location: North Somerset, UK.
Posts: 2,129
Default Re: Heatsink compound - refresh or not?

Whilst I appreciate this is not answering the original question, I suggest great caution if disturbing old heat sink compound.
It may contain beryllium oxide, which is very toxic and is no longer in general use due to this toxicity.
Some old power semiconductors also contain beryllium oxide to assist in heat transfer from the silicon die to the outer case.

Disposable gloves should be the minimum precaution. If any of the material gets onto overalls or other clothing, these should be removed and thoroughly machine washed.
broadgage is offline  
Old 23rd Oct 2020, 2:58 pm   #14
Beobloke
Heptode
 
Join Date: Oct 2007
Location: Southampton, Hampshire, UK.
Posts: 816
Default Re: Heatsink compound - refresh or not?

The last time I repaired a power amplifier a few years back, it required new output transistors as one channel's devices had blown. When I removed them, there was merely some powdery remains of what used to be heatsink compound left under them, so I lifted the other channel's devices whilst I was in there and replaced the compound under them, as it was just as bad.

As a result, whenever giving an amplifier a general service, I always have a poke round in the general direction of these devices to try and ascertain what sort of condition it is in. More often than not, it's in a bad way and benefits from renewing.

As an aside I've just had a brainwave -audiophile heatsink compound!
Beobloke is offline  
Old 23rd Oct 2020, 3:21 pm   #15
Peter F4VSA
Tetrode
 
Join Date: May 2019
Location: Limoges, France.
Posts: 60
Default Re: Heatsink compound - refresh or not?

I would very much hope that no heat sink compound would ever contain beryllium oxide, far too dangerous. I have never heard of that. When I was involved in the design and production of rf power transistors in the 70's we used loads of beryllium oxide but only ever within the package. I've made a lovely contact cooled valve linear with big blocks of the stuff, safe IF used properly and using standard heat sink compound on the heatsink

Peter
Peter F4VSA is offline  
Old 23rd Oct 2020, 3:34 pm   #16
MrBungle
Dekatron
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
Default Re: Heatsink compound - refresh or not?

Too late on the audiophile heatsink compound - someone thought of it already!

https://www.overclockers.co.uk/therm...th-02j-tg.html
MrBungle is offline  
Old 23rd Oct 2020, 3:39 pm   #17
stevehertz
Dekatron
 
stevehertz's Avatar
 
Join Date: Sep 2011
Location: Rugeley, Staffordshire, UK.
Posts: 8,809
Default Re: Heatsink compound - refresh or not?

OMG!
__________________
A digital radio is the latest thing, but a vintage wireless is forever..
stevehertz is online now  
Old 23rd Oct 2020, 3:42 pm   #18
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,923
Default Re: Heatsink compound - refresh or not?

Quote:
Originally Posted by Beobloke View Post
I always have a poke round in the general direction of these devices to try and ascertain what sort of condition it is in. More often than not, it's in a bad way and benefits from renewing.
I think this leads to the question about does it matter if the "medium" in the compound (i.e. the base fluid) is present or not? Isn't it the powdered oxides that actually do the work of providing thermal conductivity, and these are manufactured as a paste simply for easy of handling and enabling them to flow to form a perfect interface?

I think that I'd always tended to assume that most heatsink compounds were probably based on silicone fluids/grease, and the idea that silicone grease would evaporate completely from far underneath a bolted-down power transistor comes as a surprise.

But especially where it is easy to do and on transistors pushed hard, I guess that the idea of "upgrading" to a modern heatsink compound is tempting.

B
__________________
Saturn V had 6 million pounds of fuel. It would take thirty thousand strong men to lift it an inch.
Bazz4CQJ is offline  
Old 23rd Oct 2020, 3:47 pm   #19
Philips210
Nonode
 
Join Date: May 2007
Location: Redruth, Cornwall, UK.
Posts: 2,562
Default Re: Heatsink compound - refresh or not?

Quote:
Originally Posted by Peter F4VSA View Post
I would very much hope that no heat sink compound would ever contain beryllium oxide, far too dangerous. I have never heard of that.

Peter
Hi.

That was my understanding too, I always thought it was mostly zinc oxide.
BeO heat transfer pads do seem to have been used fairly widely though. They have the appearance of ceramic. I have a few of the white and also pink coloured pads salvaged from old equipment, safely stored in polythene bags.

Regards,
Symon
Philips210 is offline  
Old 23rd Oct 2020, 3:56 pm   #20
Philips210
Nonode
 
Join Date: May 2007
Location: Redruth, Cornwall, UK.
Posts: 2,562
Default Re: Heatsink compound - refresh or not?

Quote:
Originally Posted by MrBungle View Post
Too late on the audiophile heatsink compound - someone thought of it already!

https://www.overclockers.co.uk/therm...th-02j-tg.html
Blimey, and look at the cost of it!

Symon
Philips210 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 3:08 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.