Heatsink isolation with two chips

pmbrunelle · 11th February 2011, 01:39 AM

This is for a stereo amplifier with dual LM1875s.

Both chips are supplied from the same power supply, same rails. They're on the same board.

Is it a bad idea to bolt both chips to the same heatsink, without isolation? I understand that the cooling tab is electrically connected to the negative power supply rail.

I could put mica spacers, but that would compromise heat transfer.

Of course, cutting the heatsink in two would also work, but it's easier to make the amplifier more rugged when it's all bolted up together.

stratus46 · 11th February 2011, 07:14 AM

Use the mica. It's been used successfully for over 50 years.

G²

AndrewT · 11th February 2011, 10:22 AM

You can bolt the chipamps direct to the heatsink. It then becomes "live". You must ensure it can never be accidentally shorted to any other component.
You may have to add a decoupling capacitor from heatsink to -ve supply and/or to Power Ground.

pinkmouse · 11th February 2011, 10:45 AM

Frankly, I hate live heatsinks. It's far too easy to short something out whilst testing on the bench.

pacificblue · 11th February 2011, 04:36 PM

It is also inconvenient to touch them.

pmbrunelle · 11th February 2011, 10:19 PM

Both chips each have their own dedicated power supply decoupling caps located nearby.

I can relate with the "accidents" relating to a live heatsink. I blew a $220 MP3 by touching its headphone output to a live heatsink...

In this case, I'm not concerned about accidental shorting, since the heatsink will be inside an enclosure.

I was just more wondering about the effects of having power supply current being conducted via the cooling tabs and heatsink rather than the PCB copper tracks.

stratus46 · 12th February 2011, 07:40 PM

'Live' heatsinks also add possible 'ground loops'. Yes, I know it's the negative rail but the principle is the same. In general it's poor practice unless you have
a really good reason and it's very clear the heatsink is 'live'.

G²

Rainwulf · 13th February 2011, 12:23 AM

Quote:

Originally Posted by pmbrunelle

Both chips each have their own dedicated power supply decoupling caps located nearby.

I can relate with the "accidents" relating to a live heatsink. I blew a $220 MP3 by touching its headphone output to a live heatsink...

In this case, I'm not concerned about accidental shorting, since the heatsink will be inside an enclosure.

I was just more wondering about the effects of having power supply current being conducted via the cooling tabs and heatsink rather than the PCB copper tracks.

The pcb tracks are copper and will carry most of the current anyway, as aluminium has a higher resistance.

In my 6 channel amp, i just used seperate heatsinks, because the lm3875 has the same negative power on its tab.
I prefer much better heat transfer over the live heatsink, as all my heatsinks are inside a box.

pacificblue · 13th February 2011, 06:12 AM

You mean a massive block of aluminium with several square centimetres of conducting area has less resistance than a 35 µm thick copper track that is a few millimetres wide, if at all?

Rainwulf · 13th February 2011, 10:55 AM

I would imagine that any amp builder would build up the current carrying tracks on the board to hold those currents.

Though your point is very valid.
A large chunk of al would conduct pretty damn well heh.

11th February 2011, 01:39 AM	#1
pmbrunelle diyAudio Member Join Date: Sep 2005	Heatsink isolation with two chips This is for a stereo amplifier with dual LM1875s. Both chips are supplied from the same power supply, same rails. They're on the same board. Is it a bad idea to bolt both chips to the same heatsink, without isolation? I understand that the cooling tab is electrically connected to the negative power supply rail. I could put mica spacers, but that would compromise heat transfer. Of course, cutting the heatsink in two would also work, but it's easier to make the amplifier more rugged when it's all bolted up together.

11th February 2011, 10:22 AM	#3
AndrewT diyAudio Member Join Date: Jul 2004 Location: Scottish Borders	You can bolt the chipamps direct to the heatsink. It then becomes "live". You must ensure it can never be accidentally shorted to any other component. You may have to add a decoupling capacitor from heatsink to -ve supply and/or to Power Ground. __________________ regards Andrew T.

11th February 2011, 10:45 AM	#4
pinkmouse diyAudio Moderator Join Date: Apr 2002 Location: Chatham, England	Frankly, I hate live heatsinks. It's far too easy to short something out whilst testing on the bench. __________________ Al I conceive of nothing, in religion, science or philosophy, that is more than the proper thing to wear, for a while. Charles Fort

11th February 2011, 04:36 PM	#5
pacificblue diyAudio Member Join Date: May 2008	It is also inconvenient to touch them. __________________ If you've always done it like that, then it's probably wrong. (Henry Ford)

13th February 2011, 06:12 AM	#9
pacificblue diyAudio Member Join Date: May 2008	You mean a massive block of aluminium with several square centimetres of conducting area has less resistance than a 35 µm thick copper track that is a few millimetres wide, if at all? __________________ If you've always done it like that, then it's probably wrong. (Henry Ford)


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

11th February 2011, 07:14 AM	#2
stratus46 diyAudio Member Join Date: Nov 2009 Location: Los Angeles	Use the mica. It's been used successfully for over 50 years. G²

11th February 2011, 10:19 PM	#6
pmbrunelle diyAudio Member Join Date: Sep 2005	Both chips each have their own dedicated power supply decoupling caps located nearby. I can relate with the "accidents" relating to a live heatsink. I blew a $220 MP3 by touching its headphone output to a live heatsink... In this case, I'm not concerned about accidental shorting, since the heatsink will be inside an enclosure. I was just more wondering about the effects of having power supply current being conducted via the cooling tabs and heatsink rather than the PCB copper tracks.

12th February 2011, 07:40 PM	#7
stratus46 diyAudio Member Join Date: Nov 2009 Location: Los Angeles	'Live' heatsinks also add possible 'ground loops'. Yes, I know it's the negative rail but the principle is the same. In general it's poor practice unless you have a really good reason and it's very clear the heatsink is 'live'. G²

13th February 2011, 10:55 AM	#10
Rainwulf diyAudio Member Join Date: Feb 2009	I would imagine that any amp builder would build up the current carrying tracks on the board to hold those currents. Though your point is very valid. A large chunk of al would conduct pretty damn well heh.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Source for heatsink on TDA1543 chips?	drjlo	Digital Source	2	11th June 2008 10:18 PM
What isolation between chips & heatsink are people using?	taj	Chip Amps	25	15th August 2007 06:22 PM
Chips for Chumps aka "Intelligent Chips"	jackinnj	Solid State	0	15th April 2005 12:37 AM
LM4780 - Heatsink and isolation	Firebug24k	Chip Amps	15	30th March 2005 09:57 PM
Isolation	eeka chu	Tubes / Valves	4	6th December 2004 06:53 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read