Go Back   Home > Forums > >

Solid State Talk all about solid state amplification.

Towards a wideband non switching Auto Bias power amp
Towards a wideband non switching Auto Bias power 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
Old 17th September 2021, 11:42 AM   #81
padamiecki is offline padamiecki  Poland
diyAudio Member
 
Join Date: Mar 2004
Location: Krakow
Thank you very much Ian!
__________________
regards, Pawel
  Reply With Quote
Old 17th September 2021, 02:24 PM   #82
padamiecki is offline padamiecki  Poland
diyAudio Member
 
Join Date: Mar 2004
Location: Krakow
Default unfortunately symbol missing

prnscr attached
Attached Images
File Type: png Obraz1.png (10.7 KB, 135 views)
__________________
regards, Pawel
  Reply With Quote
Old 20th September 2021, 05:29 AM   #83
IanHegglun is offline IanHegglun  Australia
diyAudio Member
 
Join Date: Mar 2017
Default Simulating Thermal-Lag in the MOS-diodes

Hi Pawel,

Sorry, my last post subcircuit was forgotten in the zip. The missing ones are in the latest one below.

This one shows heating up from idle with 40W, then change at 10s to 1W and it cools. It is the version for 8 ohms and includes electrothermal models for the main MOSFETs and the MOS-diodes (4 MOSFETs). Sorry about the clutter from the thermal model bits. M4 is coupled to M3 via Ts3 label, likewise M6 is coupled to M5 via Ts5 label, and M3 and M5 are not thermally coupled (but the could be linked by changing the labels). The thermal resistance for cooling M3 and M4 is 50C/W, assumed to be that of two TO-220's coupled together.
Towards a wideband non switching Auto Bias power amp-diamond-splitter-ztc-spreader-mosdnc-elth-nsb-irfp240-elth-cct-png
Towards a wideband non switching Auto Bias power amp-diamond-splitter-ztc-spreader-mosdnc-elth-nsb-irfp240-elth-png
.
This show the thermal time constants in the main MOSFETs and the MOS-diodes. The idle current (t=0) is 295mA, t=1s is 318mA, t=10s is 347mA, t=11s is 325mA, t=20s is 301mA.

With 40W, once the heatsink reaches final temperature the crossover current is 50mA more than at idle, the HS rise is 17C from idle. About half the current change occurs in the first second.

The FFT was plotted for the first 5 cycles from t=0, 1s, 9.5s, 10s, 11s, 20s. There is a check box for the in the FFT dialog.

At t=0 40W the dominant harmonic is -70.9dB, then at t=9.5s it is 70.8dB so the increase in idle current is not causing more distortion.

At t=10s 1W the dominant harmonic is -98.5dB, then at t=20s it is 93.7dB so the increase in idle current from 40W reduces the distortion compared to after the heatsinks cool back to their normal idle current of 295mA. This is because an idle current of 295mA is "under-biased", deliberately to reduce the idle heat to a "reasonable" level.
BTW The "optimum" idle current for this output stage is relatively high because there are effectively no source resistors (the MOS diodes have only about 20m Ohms of resistance, much less than the usual 200-300m Ohms.)

In this simulation I have used guestimate values for the thermal properties of the MOS-diodes. They are not on a heatsink, just back-back with a thermal pad. Once bench measurements are made the thermal linkages can be updated for real life simulations. In the mean time we can see the general effects such as thermal lag in the MOS-diodes and the effect on distortion.

BTW D1 and D2 are added across the MOS-diodes so LTspice can find the correct operating point. They do not normally conduct in this circuit and are (probably) not needed in a real circuit.
__________________
IansHomePage; betterVDMOS; PAKproject
  Reply With Quote
Old 20th September 2021, 05:44 AM   #84
TNT is offline TNT  Sweden
diyAudio Member
 
Join Date: Apr 2003
Location: Sweden
Towards a wideband non switching Auto Bias power amp
An observation... whats is so bloody special about 20kHz? A lot of circuits you see start to drop of FR wise or has a rising presence of distorsion around 10k.. Is this just a coincident or is it dues to our 20-20k thinking that make it end up like this.... But its not all in our minds - there seem to be som sort of physical limit also that kicks in. Most class-d has a dist jump at 7-9 k. Regulators start to keep the grip (output impedance etc) at 10k.

Why isn't this figure 63k instead so it could be well outside our pass band?

//
__________________
More distortion to the people! Timing . . . . is paramount.
  Reply With Quote
Old 20th September 2021, 01:12 PM   #85
spladski is offline spladski  United Kingdom
diyAudio Member
 
Join Date: May 2012
It could be 63KHz. You would need less gain. Gain is a function of impedance. Higher gain, higher impedance and parasitic capacitance determines bandwidth. Operating at higher currents in signal stages is a way around this but leads to higher distortion.
That is why open loop gain drops with frequency, which reduces feedback and causes distortion rise.
  Reply With Quote
Old 20th September 2021, 08:48 PM   #86
TNT is offline TNT  Sweden
diyAudio Member
 
Join Date: Apr 2003
Location: Sweden
Towards a wideband non switching Auto Bias power amp
I think I do know this but I'm so psyched that the limits seem to coincide with the human hearing boarder - why isn't physics just 10x better and we would not have any problems... really

//
__________________
More distortion to the people! Timing . . . . is paramount.
  Reply With Quote
Old Yesterday, 06:58 AM   #87
IanHegglun is offline IanHegglun  Australia
diyAudio Member
 
Join Date: Mar 2017
Default Simulating Thermal-Lag in the MOS-diodes

Quote:
Originally Posted by IanHegglun View Post
Hi Pawel,
This one shows heating up from idle with 40W, then change at 10s to 1W and it cools.
Oops. There was a mistake in the signal source. The corrected file is attached for the stated power changes. My apology.

In addition I have added thermal compensation so the idle current remains more constant with the main MOSFET temperature changes, particularly from startup. R10 is now a 22k thermistor mounted on M1 case. It is optional.
__________________
IansHomePage; betterVDMOS; PAKproject
  Reply to this post

Reply


Towards a wideband non switching Auto Bias power ampHide this!Advertise here!
Thread Tools

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
Fixed bias vs. auto bias HFGuy Tubes / Valves 36 2nd March 2021 11:16 AM
Auto-Bias vs Manual bias vs EFB lugnut Tubes / Valves 25 23rd January 2020 07:45 PM
DC Automatic Bias Control - Auto fixed bias Ceglar Tubes / Valves 21 11th October 2014 09:21 AM
Class AB, Class AB Non-Switching, Class AB Soft Non-Switching steph_tsf Solid State 4 15th March 2010 03:39 PM


New To Site? Need Help?

All times are GMT. The time now is 08:28 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Copyright ©1999-2021 diyAudio
Wiki