Discrete Opamp Open Design - Page 178 - diyAudio
Go Back   Home > Forums > Source & Line > Analog Line Level

Analog Line Level Preamplifiers , Passive Pre-amps, Crossovers, etc.

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 17th October 2012, 01:53 PM   #1771
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
I would think you'll have more luck drawing heat out of the PCB than out of the top of the part. the thermal impedance of the casing will be significant, wouldnt an appropriately designed PCB with thicker than normal copper be more effective? this way you get the heat from the bottom of the part as well as that from the legs, which will be more than through the case.

how about mounting it upsidedown with thermal vias and using little BGA heatsinks on 'top' this way you take advantage of heat rising as well
  Reply With Quote
Old 17th October 2012, 02:44 PM   #1772
bcarso is offline bcarso  United States
diyAudio Member
 
Join Date: May 2005
Location: Canoga Park, California
Quote:
Originally Posted by qusp View Post
I would think you'll have more luck drawing heat out of the PCB than out of the top of the part. the thermal impedance of the casing will be significant, wouldnt an appropriately designed PCB with thicker than normal copper be more effective? this way you get the heat from the bottom of the part as well as that from the legs, which will be more than through the case.

how about mounting it upsidedown with thermal vias and using little BGA heatsinks on 'top' this way you take advantage of heat rising as well
With dissipations kept fairly low by cascoding, it's not that important IMO to draw heat away. The more significant reason to do thermal things is to provide coupling between parts, if they are matched. And as mentioned, biasing them near the zero tempco point makes life easier, although those currents may be inconveniently high and require additional mods to the overall amp.

The chip in the BF862, like many single JFETs, sits on the gate lead, and NXP bases their dissipation rating temperature on the temp of that lead. Pulling heat away or coupling via the plastic of the package is going to be less effective than coupling via the gate leads. However, obviously you can't just short them to one another

Again, as Scott points out, you can simply not worry about this and use a matched pair dual with electrically isolated channels like the LSK389, which is available in SO-8, or for the NOS crowd, the 2SK389 (funny package).
  Reply With Quote
Old 17th October 2012, 03:50 PM   #1773
qusp is offline qusp  Australia
diyAudio Member
 
qusp's Avatar
 
Join Date: Oct 2009
Location: Brisbane, Australia
Thanks I had presumed that was important too, but figured its going to be more effective through the leads and PCB going by my experience of heatsinking PCB mount stuff that doesnt already have a powerpad underneath. so just have the thermal vias through to copper area on the other side (can just be copper wire for DIY PCB) and bond with a little piece of copper on the other side with something like arctic silver, or even sekisui thermal tape (non-conductive so the additional piece of copper could be grounded, provided the capacitance from that is OK)

Last edited by qusp; 17th October 2012 at 03:52 PM.
  Reply With Quote
Old 17th October 2012, 05:28 PM   #1774
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: cambridge ma
With default values I found the worst case drift AFTER trimming the offset back to zero was about 20uV/C for both pairs of FET's having a 10% mismatch of Idss (combined in the worst way).
__________________
Silence is so accurate.
  Reply With Quote
Old 17th October 2012, 06:04 PM   #1775
bcarso is offline bcarso  United States
diyAudio Member
 
Join Date: May 2005
Location: Canoga Park, California
Quote:
Originally Posted by scott wurcer View Post
With default values I found the worst case drift AFTER trimming the offset back to zero was about 20uV/C for both pairs of FET's having a 10% mismatch of Idss (combined in the worst way).
Yes, and we do need to realize that this is an amp for audio applications. If d.c. drift were a huge concern there are ways, with some very good monolithic solutions.
  Reply With Quote
Old 17th October 2012, 06:06 PM   #1776
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: cambridge ma
Found my power on latchup possibility. If the minus rail comes on first the input is tilted all the way to the negative side and the upper cascode collapses. The simple diode fixes this by powering the cascode if the input ever tilts all the way during turn on.
Attached Images
File Type: jpg latch.jpg (49.2 KB, 334 views)
__________________
Silence is so accurate.
  Reply With Quote
Old 17th October 2012, 06:08 PM   #1777
bcarso is offline bcarso  United States
diyAudio Member
 
Join Date: May 2005
Location: Canoga Park, California
Quote:
Originally Posted by scott wurcer View Post
Found my power on latchup possibility. If the minus rail comes on first the input is tilted all the way to the negative side and the upper cascode collapses. The simple diode fixes this by powering the cascode if the input ever tilts all the way during turn on.
Good catch, as they say
  Reply With Quote
Old 17th October 2012, 06:55 PM   #1778
diyAudio Member
 
Join Date: Nov 2011
Location: Cooktown, Oz
Default Unobtainium components

Quote:
Originally Posted by kgrlee View Post
You can still get c254x, Toshiba c3329 & Rohm d786 in qtys for small production but their pnp complements are truly Unobtainium.
While we're discussing Unobtainium components, I'll ask forum members for advice and move back on topic.

The simplified Wurcer 990 I post in #1692 has excellent performance. On fas42's more stringent 100x 20kHz 20Vpp 600R test in #1694, its actually better than Guru Wurcer's more complex supa dupa circuit.

2nd -101.7dB 3rd -102.8dB Simplified 990
2nd -89.9dB 3rd -89.3dB Wurcer SupaDupa

But this is without the essential Clamping Diode Q8 It appears the evil non-linear leakage/capacitance of this diode is mucking things up big time.

If we could find a nice diode or diode-connected-BJT with very small leakage/capacitance, we could have Wurcer performance with much less complexity.

Any suggestions? Especially for diodes/BJTs with good SPICE models?

The voltage current requirements are quite small.

ricardo 'wannabe a SPICE guru' lee
  Reply With Quote
Old 17th October 2012, 06:56 PM   #1779
Waly is offline Waly  United Kingdom
diyAudio Member
 
Waly's Avatar
 
Join Date: Jan 2011
Location: London
Quote:
Originally Posted by scott wurcer View Post
Found my power on latchup possibility. If the minus rail comes on first the input is tilted all the way to the negative side and the upper cascode collapses. The simple diode fixes this by powering the cascode if the input ever tilts all the way during turn on.
Distortions due to the nonlinear diode capacitance?
  Reply With Quote
Old 17th October 2012, 07:29 PM   #1780
diyAudio Member
 
scott wurcer's Avatar
 
Join Date: Jan 2004
Location: cambridge ma
Quote:
Originally Posted by Waly View Post
Distortions due to the nonlinear diode capacitance?
No signal voltage across this point. Richard is talking about the JFET JE990 which has clamps that do see the signal. In IC's we had to use PNP's that had no reverse EB breakdown.
__________________
Silence is so accurate.
  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
Discrete OPAMP audio-gd Vendor's Bazaar 27 20th September 2012 04:02 PM
discrete opamp help blackpowderaudio Parts 0 16th December 2009 03:46 PM
THAT transistor headphone amp (250ma discrete opamp) design sanity check. Russ White Headphone Systems 19 13th December 2007 12:52 PM


New To Site? Need Help?

All times are GMT. The time now is 11:14 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