How to determine heatsink/bias - diyAudio
Go Back   Home > Forums > Amplifiers > Pass Labs

Pass Labs This forum is dedicated to Pass Labs discussion.

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
Reply
 
Thread Tools Search this Thread
Old 2nd January 2003, 03:26 PM   #1
Raj1 is offline Raj1  United Kingdom
diyAudio Member
 
Join Date: Jan 2003
Location: UK
Default How to determine heatsink/bias

Hi,

Someone please in the simplest of ways (for dummies), please explain the following:

I have a 0.4oC/W heatsink on a mono amp, the mosfets are dissipating 35watts as heat at a bias of 250ma. At this level I can just about keep my hand on the heatsink (must be around 45-50oC)

If I go for a 0.2oc heatsink, any rough ideas of how high I'll be able to set the bias?

Thanks
Raj
  Reply With Quote
Old 2nd January 2003, 03:52 PM   #2
RKC is offline RKC
diyAudio Member
 
Join Date: Dec 2002
Location: Dallas, TX
Watts dissipated, 35 X .4 = 14 deg C temp rise over ambient.
35 x .2 = 7 deg C temp rise over ambient.

I would say you have plenty of room to up the bias with your current heatsinks. You should be able to double the bias current and still be respectable on the temps.
  Reply With Quote
Old 2nd January 2003, 04:00 PM   #3
Thomas is offline Thomas  Denmark
diyAudio Member
 
Thomas's Avatar
 
Join Date: May 2001
Location: Right here
At 35 Watts a 0.4 deg/W heatsink would represent a temperature rise of 14 deg C wich are totaly acceptable. At an ambient temperature of 30 deg C your heatsinks would be 44 deg C wich also are accepable.

You must have very sensitive hands.

Doubble the heatsink capacity + double the bias = same temperature.

edit: RKC wrote the same at the same time as I did.
  Reply With Quote
Old 2nd January 2003, 04:17 PM   #4
diyAudio Member
 
Join Date: Aug 2002
Location: Toronto, Canada
Hi Raj,

You should also be mindful of the dissipation per output
device. Check this figure against the datasheet for
the device in question. A cool heatsink doesn't necessarily
mean the same for the output devices.

Cheers,
Dennis
  Reply With Quote
Old 2nd January 2003, 04:59 PM   #5
Raj1 is offline Raj1  United Kingdom
diyAudio Member
 
Join Date: Jan 2003
Location: UK
Default dissipation per mosfet

I think I've mis understood the manual for may amp.

There are two mosfets per (mono) pcb dissipating 35 watts each, so does this mean that the true rate of dissipation is 70watts per pcb?

I was lead to believe that the increase in quiescent current does not always mean that the watts dissipated will follow in a linear fashion, please clarify.

If I use my current heatsink 0.4oC/w accross the front of the mosfet and the 0.2oC/W at the back how more diissipation ability will I get, or is it better to just use the old hand trick on the heatsinks?

In this type of amp design I suppose I'd need really massive heatsinks to really bump the current up even to 1A.
I did however notice an immediate difference when I increased it from 210-250ma, this is why I am keen to get bigger heatsinks and push the current up some more.

Thanks to all for your help

Raj
  Reply With Quote
Old 2nd January 2003, 05:53 PM   #6
diyAudio Member
 
Join Date: Aug 2002
Location: Toronto, Canada
Default Re: dissipation per mosfet

Quote:
Originally posted by Raj1

....

In this type of amp design I suppose I'd need really massive heatsinks to really bump the current up even to 1A.
I did however notice an immediate difference when I increased it from 210-250ma, this is why I am keen to get bigger heatsinks and push the current up some more.

Thanks to all for your help

Raj
Hi Raj,

Beyond additional heatsink, I think you may need to scale
up the output stage by adding parallel output devices
if you want to substantially increase the bias current.
At 35W dissipation, the mosfet chip temperature likely
is fairly high. Just how high can it go and remain acceptable
is device-dependent.

Dennis
  Reply With Quote
Old 2nd January 2003, 06:03 PM   #7
Raj1 is offline Raj1  United Kingdom
diyAudio Member
 
Join Date: Jan 2003
Location: UK
Default MOSFET TEMP

Hi Dennis,

In general at what sort of mosfet temp is 'safe', if these devices are currently dissipating 35watts each at 250ma, would it be fair to assume that 70 watts would be dissipated at 500ma? Is this still safe for most mosfets as long as the heatsinks and dissipation are adequate? The ones in use on my amp are double die types

Exicon ECW 20P20 and ECW 20N20

Thanks
Raj
  Reply With Quote
Old 4th January 2003, 06:07 AM   #8
diyAudio Member
 
Join Date: Nov 2002
Location: TX USA
Reducing the heatsink thermal resistance from 0.4C/W to 0.2C/W is the right direction to go, it will not reduce the junction temperature enough to stay within the the device maximum operating temperature of 175C and it will significantly exceed the derated power and quite likely fail.

Lets assume the 35 watt dissipation on a 0.4C/W heatsink(HSA), a maximum ambient temp of 50C (122F), an insulator(CHS) with a 1.0C/W and the IRFP044 device with junction to case(JC) of 0.83C/W.

Ambient 50C
HSA 0.4X35W=14C
CHS 1.0X35W=35C (typical Silpad)
JC 0.83X35W=29C
----------- -------------------
Max Junction Temp=128C

Its well below the maximum operating junction temp of 175C and also the derated maximum power at 128C is 56 watts. 180W-((128C-25C)X1.2W/C)=56 This is a safe and reliable design.

Ok lets assume the heatsink is improved and drops to 0.2C/W and the bias is increased to dissipate 70 watts.

Ambient 50C
HSA 0.2X70W=14C
CHS 1.0X70W=70C (typical Silpad)
JC 0.83X70W=58C
----------- -------------------
Max Junction Temp=192C

This clearly exceeds the maximum operating junction temperature and the derated maximum power. Your gonna let the smoke out!

Notice that the HSA temperature rise remains at 14C but both the CHS and JC temps double.

You could use a different device that has a lower junction to case resistance and a good thin Mica or Kapton insulator may be as low as 0.5C/W but it's still to hot to be reliable. Even directly mounting the IRFP044 to the heatsink at 0.24C/W you still exceed the derated power and then you will need to electically insulate the whole heatsink. It's best to add more devices and keep the per device dissipation within a safe range.

It's not just the heatsink temp that counts, you must figure in the thermal resistance of the insulator, junction to case, maximum ambient temp and the derated power of the device.

Bruce
  Reply With Quote
Old 4th January 2003, 07:19 AM   #9
Raj1 is offline Raj1  United Kingdom
diyAudio Member
 
Join Date: Jan 2003
Location: UK
Default further calculations

Hi Bruce,

I' have now decided to use a fan cooling method provided by Hugh Dean, which should provide heatsinks resistence of around 0.13 or lower. Also I generally use aluminium oxide washers, which are very thin, probably lower than 0.5 oC. But Hugh dean has suggested that if I use the method of fan cooling as described by him I could insulate the heatsinks from the chassis of the amp, so I will probably not need insulating washers at all. The actual device resistence is 0.5oC lower than other mosfet's (kindly posted details by Dennis Hui, Thanks mate!). One other MAJOR problem is that my amp is actually a class A/B design which has further heat problems. However I have a thermal cut-off so if I set this at a safe margin, WELL below the critical range I shouldn't have a problem of overdriving the mosfets, I'll place the thermal cut off device right next to the mosfet's to aid the speed of cut off.

Thanks
Raj
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
How to determine if class G/H? tktran303 Solid State 21 4th September 2008 11:48 AM
Can somebody please check my bias/heatsink calculations? spence Solid State 10 31st March 2005 08:29 PM
How to determine Q from a graph? JoeBob Parts 4 13th November 2004 04:12 AM
How to determine optimal bias class A current? Sunsun22 Chip Amps 2 30th August 2004 09:49 AM
How do I determine what this Transformer is? wotan2525 Solid State 5 21st March 2003 05:34 PM


New To Site? Need Help?

All times are GMT. The time now is 06:34 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2