Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Solid State Talk all about solid state amplification.

Amplifier with only 1 type of transistors
Amplifier with only 1 type of transistors
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 28th June 2020, 11:47 PM   #31
steveu is offline steveu  United States
diyAudio Member
 
Join Date: May 2005
Location: Colorado Springs
Default bootstrapped and current source

Here is the npn circuit with a driver bootstrap and with a driver current source. I played with the driver current mirror on one to illustrate bias tweaking options. A real circuit would use a pot instead of the small series resistor.

I also remove one of the RC gain boosts because it seems I am pushing the OLG too much.

Note the simulation is driven to clipping to illustrate the clipping levels, ie not for THD estimates.
Attached Images
File Type: png npn-mirror-pi.png (31.1 KB, 96 views)
File Type: png npn-mirror-ci.png (31.4 KB, 100 views)
Attached Files
File Type: asc npn-mirror-pi.asc (9.4 KB, 1 views)
File Type: asc npn-mirror-ci.asc (9.6 KB, 2 views)

Last edited by steveu; 28th June 2020 at 11:50 PM.
  Reply With Quote
Old 29th June 2020, 02:30 AM   #32
steveu is offline steveu  United States
diyAudio Member
 
Join Date: May 2005
Location: Colorado Springs
Default VAS clamp

Both of these need a diode clamp (D5) on the VAS to remove the recovery / rail sticking glitch.
Attached Images
File Type: png npn-mirror-pi.png (31.3 KB, 91 views)
File Type: png npn-mirror-ci.png (32.8 KB, 96 views)
  Reply With Quote
Old 30th June 2020, 04:48 AM   #33
stumpamp is offline stumpamp  Luxembourg
diyAudio Member
 
Join Date: Jun 2017
Steve: Yes that is now becoming pretty awesome.

If I simulate at 0.75Vin 20kHz I get some crossover disturbance. If increasing C8 to 100p it is better but some oscillations come up. Therefore I remove C8 and put a 47pF cap between the collectors of Q11 and Q14 (just like on an earlier version). The value of this cap influences crossover from positive to negative cycle, it makes it more smooth. Together with below mirror circuit change the problem is gone.

I modified:
- current source 2mA at the differential (identical performance)
- bootstrap of Q7, replace by current source (2mA) (identical performance and no electrolytic anymore)
- RC R18/C5 not needed (more harm than good I think)
- R27/28 can be removed, they just add an identical offset on both sides of the current mirror. Of course re-tuning is required. R24=R25 =1k is pretty good. R24/25 is slightly dependent on voltage supply but so slightly just leave it 1k. I see no need for a trimpot.
- Value of R17/23 is so important. 10 ohm is somehow ok, 20 Ohm is worse, 1 ohm is good, at 20 mA load.
- I replace the current mirror with the "four transistor Wilson mirror" and this is really performing well in this circuit. For once that this thing makes a sense let's use it.

I did not understand the purpose of the following components
- Removed D1
- Shorted D4
- Shorted R19
I will put them back if I understand why.

I cannot find much other to improve. As perfect as it can ever be. Please comment.
Attached Files
File Type: asc 0-1 play npn-mirror-ci.asc (9.4 KB, 7 views)
  Reply With Quote
Old 30th June 2020, 05:09 AM   #34
stumpamp is offline stumpamp  Luxembourg
diyAudio Member
 
Join Date: Jun 2017
I shot this too quick. We should look into control of idle bias current of the lower transistor.
  Reply With Quote
Old 30th June 2020, 06:55 AM   #35
edbarx is offline edbarx  Malta
diyAudio Member
 
Join Date: May 2018
The very fact valve based amplifiers are still used and acclaimed by people who are very keen about music, is an indicator, bijunction transistors will continue to be produced. The manufacturer may change but some business person may start producing them. This happened with valves, there is no reason why bijunction transistors have to be an exception.

Remember people, there is a market law which states, if there is demand, someone will try to meet that demand.
__________________
Any fool can make things bigger, more complex, and more violent. It takes a touch of a genius, and a lot of courage, to move in the opposite direction. [Albert Einstein]

Last edited by edbarx; 30th June 2020 at 07:00 AM.
  Reply With Quote
Old 30th June 2020, 11:15 AM   #36
stumpamp is offline stumpamp  Luxembourg
diyAudio Member
 
Join Date: Jun 2017
Quote:
Originally Posted by edbarx View Post
bijunction transistors will continue to be produced
Agree. For sure that is not the motivation for this npn-only design.

However until now there was NO push-pull amplifier without complementary transistors, that was not relying on coupling transformers or electrolytics.
  Reply With Quote
Old 1st July 2020, 12:19 AM   #37
steveu is offline steveu  United States
diyAudio Member
 
Join Date: May 2005
Location: Colorado Springs
R27/28 move the common mode range of the current mirror down to cover the negative rail. If you remove them the output will clip sooner on the negative side.

D1 clamps Q9 from saturating and causing rail sticking. If you remove R27/28 then you will never get that close to the negative rail anyway.

D4 prevents the the need for the VAS to drive R33 through forward biased CB of Q17. This can happen on negative swings.

R19 moves the pole from Q20 down so that the feedback between Q19/20 is guaranteed stable. An alternate is adding a ~10nF cap across Q20 CE.

The improved current mirror is probably a good idea but there is some danger that the voltage on the collector of Q25 can not go low enough to turn off Q13/Q9.
  Reply With Quote
Old 1st July 2020, 04:18 AM   #38
steveu is offline steveu  United States
diyAudio Member
 
Join Date: May 2005
Location: Colorado Springs
Default driving Q8 off

Ah , so there are two ways to address the high frequency cross-over problem.
XX Use diodes to drive Q8 off.
XY Use a separate driver for the current mirror.
Attached Images
File Type: png npn-mirror-ci-XX.png (35.5 KB, 39 views)
File Type: png npn-mirror-ci-XY.png (35.3 KB, 27 views)
Attached Files
File Type: asc 0-1 play npn-mirror-ci-XX.asc (10.8 KB, 1 views)
File Type: asc 0-1 play npn-mirror-ci-XY.asc (10.6 KB, 2 views)
  Reply With Quote
Old 2nd July 2020, 07:32 PM   #39
stumpamp is offline stumpamp  Luxembourg
diyAudio Member
 
Join Date: Jun 2017
Quote:
Originally Posted by steveu View Post
I'm thinking to call this/these amps my "sow's ear" or "silk purse" amp from the expression about making a silk purse out of a sow's ear.
May I suggest sow's ear. It is more audio.

I did remove C5/R17. For the rest my only suggestion would be a NTC driver to pull down the idle bias on heat. I think this will be a very good amp.
Attached Images
File Type: png npnonly_prot.PNG (136.8 KB, 18 views)
 

Reply


Amplifier with only 1 type of transistorsHide 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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Where to find out what type of transistors (fet, bjt) a particular chip uses? sardonx Chip Amps 10 21st July 2018 02:27 AM


New To Site? Need Help?

All times are GMT. The time now is 01:53 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki