"SiC Puppy SE Amp Using Semisouth 085 JFET - Page 5 - 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 19th July 2011, 01:22 AM   #41
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
Next thing to look out for - resistors with overly optimistic wattage ratings.

Everyone wants things to be smaller and cheaper. With resistors, this can be counterproductive. I have on hand 3W resistors that are not much bigger than a 1/2 W part, and really should be rated at 3/4-1W. The smaller resistors just get hotter for the same dissipation. This may not hurt the resistor much, but it can be pure hell for adjacent components and the PCB. What started this rant were "2W" resistors derated by 50% that were still cooking at 130-150C (a thermal imager shows up a lot of sins). To make a long story short, several resistors are now physically much larger, and spaced away from critical components (mostly capacitors). I may post some thermal camera images to illustrate my point. The problem children are R3,4,10 and 11. I'm waiting for Mouser to deliver some aluminum oxide TO-247 insulators, which will help the thermal interface between the heat sink and the Semisouth FET, as will as improving HV isolation.
  Reply With Quote
Old 19th July 2011, 01:50 AM   #42
flg is offline flg  United States
diyAudio Member
 
flg's Avatar
 
Join Date: Jul 2005
Location: North East
This is a common enlightenment I learned a long time ago 1W on a 2.5W resistor is not going to cut it! I now use a minimum rating of 5X and I'm ussually happier with 10X the quiescent power for a "hot" Resistor. I do use alot of 5W, 10W, 12W, 25W and 50W Rs. The thermal distortion or at least the Tc is much lower typically also.
__________________
"It was the perfect high end audio product: Exotic, inefficient, expensive, unavailable, and toxic." N.P.
  Reply With Quote
Old 19th July 2011, 02:12 AM   #43
flg is offline flg  United States
diyAudio Member
 
flg's Avatar
 
Join Date: Jul 2005
Location: North East
Hey, thanks for sharing a shematic!
__________________
"It was the perfect high end audio product: Exotic, inefficient, expensive, unavailable, and toxic." N.P.
  Reply With Quote
Old 19th July 2011, 02:48 AM   #44
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
Given the cheating in power rating that goes on these days, this would be one application where steel leads make sense, limiting the amount of heat conducted to the PCB and solder joint from the hot resistor. I've never understood why some inrush protector manufacturers use copper leads....Others will no doubt quibble on sonic grounds - everything involves compromise. Spacing the resistor off the PCB helps. I used ceramic beads made for that purpose at my last job - can't find a source for them now. Some resistor manufacturers squish the leads or put a jog in them to get the spacing.
  Reply With Quote
Old 19th July 2011, 08:53 PM   #45
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
Well, I changed the resistors, and they are not running any higher than 65C now, which is quite a change compared to the 100C+ temperatures on the first revision of this design. Bias current drifts up about 10ma (from 150 ma to 160 ma) as the amp warms up, probably due to change in leakage current. The gate is at ~1.5V with a 10k gate resistor, meaning there is ~150uA of gate leakage. I may incorporate a source follower in the next pass of this design to eat the leakage current. A really ammbitious follower is not needed - I may be able to get by with a pair of 35 cent jfets, at the cost of adding some low voltage rails.

The 085 die temperature is at ~90C after sitting a while. An aluminum oxide washer helps the thermal resistance of the Semisouth part - the heat sink actually starts to get warm now... At any rate, with 302V of drain voltage and 160ma bias current, the power dissipated in the 085 is 48.3W. The die temperature ris is ~65C (25C to 90C), so the total thermal resistance of the interface is 65/48.3 = 1.35 degrees C/Watt.

This is probably good enough for the time being to try driving the amp into a load of some sort - more later.
  Reply With Quote
Old 20th July 2011, 01:28 AM   #46
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
Some psychedelic thermal photographs - the first is the 085 temperature after running a while at 300V B+. The heat sink is starting to get toasty, as the aluminum oxide thermal pad is doing its job well.
Attached Images
File Type: jpg 085_temp.jpg (15.8 KB, 722 views)
  Reply With Quote
Old 20th July 2011, 01:29 AM   #47
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
The seconnd photo shows the temperature of R3, one of the load resistors for the DN2540 input fet.
Attached Images
File Type: jpg R3_temp.jpg (15.5 KB, 690 views)
  Reply With Quote
Old 20th July 2011, 01:31 AM   #48
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
This one shows the temperature for R10, one of the source resistors for the 085 SiC jfet.
Attached Images
File Type: jpg R10_temp.jpg (16.3 KB, 576 views)

Last edited by wrenchone; 20th July 2011 at 01:37 AM.
  Reply With Quote
Old 20th July 2011, 05:52 PM   #49
The one and only
 
Nelson Pass's Avatar
 
Join Date: Mar 2001
SiC is supposed to theoretically operate to 600 deg C., but it appears that
the ratings come from what the package can withstand. Perhaps the
die bonding fails above 175 deg.

  Reply With Quote
Old 20th July 2011, 06:30 PM   #50
diyAudio Member
 
wrenchone's Avatar
 
Join Date: Feb 2004
Location: Silicon Valley
I would worry about the solder bond to the package and the epoxy at temperatures much above 150C. A more important issue is the stress from thermal cycling, which can delaminate or crack the chip. The 085 has high thermal rresistance as it is.... I wonder how they do the bond-down anyway? I'll ask someone downstairs at work.

Another issue is that pesky gate leakage, which I believe is responsible for the bias current creeping up as the device heats up. I'll monitor the gate voltage next time I have the amp on the bench to see if this really is true.

I'm happy that I appear to be able to get away with blowing 45 watts in a TO-247 without melt-down. The aluminum oxide washer I'm using as an interface helps a lot. They're thick enough so that cracking doesn't appear to be a big problem. I need to now invest in some conical washers and a torque wrench, which will help keep the mounting screws tightened consistently. I'll probably use these in some of my other Class A projects.
  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
SemiSouth SiC JFET from $16/pc, shipping NOW! zhoufang Vendor's Bazaar 42 24th July 2012 06:11 AM
SemiSouth SiC Power JFETs RocketScientist Solid State 105 19th February 2012 08:02 PM
Semisouth SiC JFETs: SJEP120R100 or SJEP170R550 better for building amp? zhoufang Pass Labs 32 2nd September 2010 05:41 AM
new SemiSouth SJDP120R085 "normally on" JFET zhoufang Pass Labs 11 8th June 2010 04:41 AM


New To Site? Need Help?

All times are GMT. The time now is 11:40 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