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.

Blameless style lateral MOSFET amp
Blameless style lateral MOSFET amp
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 June 2019, 10:34 PM   #11
voltwide is offline voltwide  Ireland
diyAudio Member
 
Join Date: Jan 2010
The correct plot of loop gain is here
Attached Images
File Type: png LatFETamp08_loopgain.png (112.6 KB, 423 views)
__________________
tuby or not tuby, that is here the question!
  Reply With Quote
Old 16th July 2019, 12:01 AM   #12
voltwide is offline voltwide  Ireland
diyAudio Member
 
Join Date: Jan 2010
After tinkering with LTSpice for several weeks I will update my work in progress. Nothing really new, but loads of finetuning circuit details.
Attached Images
File Type: png LatFETamp4a_THD1k.png (159.2 KB, 360 views)
File Type: png LatFETamp4a_THD20k.png (160.9 KB, 332 views)
File Type: png LatFETamp4a_loopgain.png (113.3 KB, 321 views)
Attached Files
File Type: asc LatFET4_THD1k.asc (12.1 KB, 12 views)
__________________
tuby or not tuby, that is here the question!
  Reply With Quote
Old 16th July 2019, 12:05 AM   #13
voltwide is offline voltwide  Ireland
diyAudio Member
 
Join Date: Jan 2010
Today I started a pcb layout using KiCAD
Attached Images
File Type: png 1.png (94.7 KB, 334 views)
File Type: png 2.png (85.4 KB, 93 views)
__________________
tuby or not tuby, that is here the question!
  Reply With Quote
Old 16th July 2019, 11:34 AM   #14
Kay Pirinha is offline Kay Pirinha  Germany
diyAudio Member
 
Kay Pirinha's Avatar
 
Join Date: Oct 2011
Location: Somewhere in Germany
I don't get how the LTP and VAS current sources work. In my opinion they may share a common voltage reference. but in your schematics I'm afraid they'll interact heavily.
Best regards!
__________________
"Bless you, Sister. May all your sons be bishops." (Brendan Behan)
  Reply With Quote
Old 16th July 2019, 12:01 PM   #15
voltwide is offline voltwide  Ireland
diyAudio Member
 
Join Date: Jan 2010
Quote:
Originally Posted by Kay Pirinha View Post
I don't get how the LTP and VAS current sources work. In my opinion they may share a common voltage reference. but in your schematics I'm afraid they'll interact heavily.
Best regards!

Hmm, I do not get what makes this part special...
Can you check with LTSpice? This is extremely helpful in analyzing circuitry.
__________________
tuby or not tuby, that is here the question!
  Reply With Quote
Old 16th July 2019, 01:37 PM   #16
Kay Pirinha is offline Kay Pirinha  Germany
diyAudio Member
 
Kay Pirinha's Avatar
 
Join Date: Oct 2011
Location: Somewhere in Germany
Ok, I think I've got it: The VAS current source is a ring of two. The voltage at Q2's collector/Q5's base is 2 x Vbe wrt the positive supply. This also acts as the reference voltage for the LTP CCS.
Best regards!
__________________
"Bless you, Sister. May all your sons be bishops." (Brendan Behan)
  Reply With Quote
Old 16th July 2019, 07:19 PM   #17
steveu is offline steveu  United States
diyAudio Member
 
Join Date: May 2005
Location: Colorado Springs
Quote:
Originally Posted by Kay Pirinha View Post
Ok, I think I've got it: The VAS current source is a ring of two. The voltage at Q2's collector/Q5's base is 2 x Vbe wrt the positive supply. This also acts as the reference voltage for the LTP CCS.
Best regards!
There is a potential problem at clipping where Q2 saturates and all the current to R14 has to come from the base and R6, which turns off the LEDs. There are too many missing models for me to run the simulation to be sure this actually happens. I normally put a resistor in the base of ~Q2 so that if there is no collector current, the base current doesn't completely steal the current from reference voltage, or use a separate reference voltage. The cap C6 helps for a few microseconds.

I notice that the ~4V LED reference, cascode VAS and drivers, and lateral FETs saz that you don't care about wasting voltage swing and power efficiency. I suppose its better than a class-A efficiency.
  Reply With Quote
Old 16th July 2019, 08:23 PM   #18
steveu is offline steveu  United States
diyAudio Member
 
Join Date: May 2005
Location: Colorado Springs
Quote:
Originally Posted by Mark Tillotson View Post
I took the basic Hitachi MOSFET circuit and added various improvements from the Doug Self "Blameless" approach. Output inclusive compensation was one notable idea I thought would help push up the loop bandwidth and make up for MOSFET output non-linearities.

The results seem encouraging:

The rails are generated from simulated filtered ac supply and have 1V injected signals to investigate PSRR.

The response curve is nice and flat in the audio band:
You probably realize this:
1. the "VAS protection" is unnecessary in this circuit because there is no output protection that could shift the load to the VAS. ( but see #5)
2. When the amp clips +V, Q5 base current will try to feed R3 without any collector current and that will interfere with the Q9 base voltage. I put a resistor in the base of Q5 to limit this effect to less than the total current from R2.
3. a small cap directly on the base of Q9 improves the collector impedance of Q9. I usually add a resistor in series with the collector of Q9 for the same reason.
4. A capacitor directly on the base of Q5 , OR, a resistor in the base of Q2 guarantees stability of the current source feedback.
5. The collector of Q8 should be fed by a resistor or divider to limit the Q8 current during -ve clipping and reduce the Q8 collector voltage to allow using faster, cheaper transistor at Q8. This also improves "rail sticking".
6. a small cap on the base of Q1 filters ultrasonic input and improves the feedback pole at Q1 for better phase margin.
7. Tweaking the value of R7 can improve THD. 1K is a bit low.
8. Q2 creates a negative tempco at R13. Is that deliberate?
9. A Zobel network should not be necessary on a follower output. Normally a Zobel network is feedback compensation for CFP outputs.
10. A shunt resistor on L1 keeps the self resonance and C13 resonance under control in all situations, ie no load connected.

I would run your simulation to improve my comments but that would require adding all the missing models so forgive me for being lazy.
Cheers,
Steve
  Reply With Quote
Old 16th July 2019, 08:23 PM   #19
voltwide is offline voltwide  Ireland
diyAudio Member
 
Join Date: Jan 2010
It is always a compromise - this amp is not intended for efficiency, but for "academic" performance. Maybe I add some bootstrapped driver supply to increase output level.

I merged the cordell models and other 3rd party models into my "standard.bjt". So if you replace your "standard.bjt" by mine, you should have all models required
__________________
tuby or not tuby, that is here the question!
  Reply With Quote
Old 16th July 2019, 08:31 PM   #20
voltwide is offline voltwide  Ireland
diyAudio Member
 
Join Date: Jan 2010
trying to upload again-
-extension bjt rejected by uploader
-so renamed to txt
-file rejected by uploader: 346.9kb too big, 195.3kb allowed
shall I pm you my lib?
__________________
tuby or not tuby, that is here the question!
  Reply With Quote

Reply


Blameless style lateral MOSFET ampHide 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
BC550-560 Symmetric with LATERAL Output - Hiraga style lineup Solid State 32 22nd June 2011 12:58 AM
How can one tell if it is a lateral MOSFET Arroyo Solid State 4 18th June 2011 11:09 AM
"Blameless" style ~100W amp with CFP output. mnturner Solid State 40 17th May 2011 01:14 AM
Lateral MOSFET torture... megajocke Solid State 11 11th November 2007 07:08 AM
Lateral Structure Mosfet UltimateX86 Solid State 9 16th July 2004 09:50 AM


New To Site? Need Help?

All times are GMT. The time now is 08:34 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki