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.

transistor amplifier design finalization
transistor amplifier design finalization
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 11th June 2019, 11:11 PM   #1
Fossilshark is offline Fossilshark
diyAudio Member
 
Join Date: Feb 2019
Default transistor amplifier design finalization

Hi! So i've been hacking away at this circuit for almost a year and it is very close to completion. The first circuit attached (named nobias) was made up by steveu on another thread where he helped me tweak the previous version of this design. I noticed it wasn't stable at 4ohms and also didn't have the greatest voltage output. So I modified it (citing Bob Cordell's Designing Audio Power Amplifiers) and at 16v it gives me the (roughly) 8ohm 15watt or 4ohm 30watt output I am looking for. My questions are:

I would at least like to know if the two sets of transistors above and below the VBE multiplier have the correct resistor-ing? The values dont affect voltage output or frequency response much so i'm assuming its just getting the current as low as possible without affecting output?

How can I get more low end response out of the amplifier? a 60hz cutoff isn't terrible but I would really like to aim for 10 (although it sounds overkill)

Is there anything I should change to make this more stable?

And finally, can I measure total harmonic distortion with spice?

Edit: I don't plan on using 3904's/3906's, they are just for simulation purposes.
Attached Files
File Type: asc AmpTest2.asc (7.0 KB, 34 views)
File Type: asc nobiasdt1.asc (6.5 KB, 15 views)

Last edited by Fossilshark; 11th June 2019 at 11:15 PM.
  Reply With Quote
Old 11th June 2019, 11:29 PM   #2
Mark Tillotson is offline Mark Tillotson
diyAudio Member
 
Join Date: May 2018
Location: Cambridge UK
Q15 has its base and emitter shorted together - you need a current sense resistor between the rail and Q8e/Q15b so that Q15 can act to protect the VAS from over-current.

Q10 and Q13 are upside down (collector/emitters switched). Again a current sensing resistor is missing from the constant current sink.

D6???

C2 very low value, you want it to be reasonably linear down to 20Hz, perhaps 2200F better choice? Cure the low end cutoff.

C5 is tiny. 22F would provide better fast switching of drivers.

R10/R11 ???
  Reply With Quote
Old 12th June 2019, 12:35 AM   #3
brian92fs is offline brian92fs  United States
diyAudio Member
 
Join Date: Jan 2017
Location: Northern California
transistor amplifier design finalization
A couple other thoughts...

DC Offset is 174mV. The mismatch between R20 and R9 is driving this.

Doesn't look like the bias has been set.

Seems odd to use a active current source for the VAS and a simple resistive source for the LTP.

R2 looks high - try 1k or 2k. As Mark pointed out, the emitter of Q15 needs a resistor - try 2k. I believe you will want a degeneration resistor for the emitter of Q8 as well if you're going to use current limiter Q15. Maybe something around 47r or 68r

R8 & R13 seem high. Is this intentional?

The Miller Pole frequency seems high - looks to be close to 4 MHz. I believe most designers keep this around 500kHz.

You're going to the trouble of more complex circuit blocks with the VAS, VAS current limiter and VAS current source, but the input stage is fairly simple. No degeneration resistors, fairly low current, resistor load. Is this intentional?
  Reply With Quote
Old 12th June 2019, 02:01 AM   #4
brian92fs is offline brian92fs  United States
diyAudio Member
 
Join Date: Jan 2017
Location: Northern California
transistor amplifier design finalization
I'd also suggest reading Mooly's excellent thread on LT Spice.
  Reply With Quote
Old 12th June 2019, 02:25 AM   #5
Fossilshark is offline Fossilshark
diyAudio Member
 
Join Date: Feb 2019
So heres what I found:

R8 and R13 are a little high just to protect the output devices. Not needed but its there.

removing r10 and r11 make it stop working.

how do I set the bias?

making c2 larger only cuts off frequencies after the 5kish area. I still have the 60hz cutoff problem.

what is the miller pole frequency?

the rest of the advice applied nicely it seems and I even have a little more wattage output.

Do I now have Q8, Q15, Q1, Q13 and Q10 wired correctly now?

Yes the input stage is fairly simple now that you mention it, I assumed Q4 + Q3 was the current source? can this be improved?

Lastly, I assumed R20 just set input impedance, although changing it does affect the amplifier weirdly. should it be the same value as R9?
Attached Files
File Type: asc AmpTest3.asc (7.4 KB, 12 views)
  Reply With Quote
Old 12th June 2019, 03:31 AM   #6
Fossilshark is offline Fossilshark
diyAudio Member
 
Join Date: Feb 2019
disregard what I said about more wattage output. The total voltage out undistorted with a 4 ohm load is only 10v and its hard to adjust r16 so the amp does not clip using standard values.
  Reply With Quote
Old 12th June 2019, 03:41 AM   #7
Mark Tillotson is offline Mark Tillotson
diyAudio Member
 
Join Date: May 2018
Location: Cambridge UK
Quote:
Originally Posted by Fossilshark View Post
So heres what I found:

R8 and R13 are a little high just to protect the output devices. Not needed but its there.

removing r10 and r11 make it stop working.
Not for me, but I fixed the broken parts of the circuit first.
Quote:


how do I set the bias?
Set input to zero, adjust resistor R5 down till the current in R8/R13 about 50mA
Quote:


making c2 larger only cuts off frequencies after the 5kish area. I still have the 60hz cutoff problem.
That doesn't make sense to me.
Quote:


what is the miller pole frequency?
Set by the miller capacitor - its complicated, read up on compensation schemes.

Quote:


the rest of the advice applied nicely it seems and I even have a little more wattage output.

Do I now have Q8, Q15, Q1, Q13 and Q10 wired correctly now?
no, neither has the resistor in the right place. Q8 needs resistor between its emiter and the rail. Same for Q10
Quote:


Yes the input stage is fairly simple now that you mention it, I assumed Q4 + Q3 was the current source? can this be improved?

Lastly, I assumed R20 just set input impedance, although changing it does affect the amplifier weirdly. should it be the same value as R9?
R12 and R20 form a potential divider. R20 should match R9 for balance in the input pair.

Last edited by Mark Tillotson; 12th June 2019 at 03:49 AM.
  Reply With Quote
Old 12th June 2019, 04:07 AM   #8
brian92fs is offline brian92fs  United States
diyAudio Member
 
Join Date: Jan 2017
Location: Northern California
transistor amplifier design finalization
The questions you're asking aren't easy to answer in space of a forum. People have literally written entire books on them. I suggest Douglas Self or Bob Cordell's books on amplifier design. They're pricey but worth it. Rod Elliot's website is also a tremendous (and free) resource.

You may be better off starting with a known and proven design and tweaking it to put your own signature on it. Douglas Self's "Blameless" designs are good if you favor low distortion transparent designs. Rod Elliot's P3A is good if you like some color to your sound. It's also a simple design with a proven pcb available.

Regarding your questions on setting bias and analyzing THD, I encourage you once again to read Mooly's thread. I covers all these questions and more.
  Reply With Quote
Old 13th June 2019, 08:39 PM   #9
Fossilshark is offline Fossilshark
diyAudio Member
 
Join Date: Feb 2019
So I ran this design with a 10us timestep and with a 4 ohm load I get roughly .04% THD with a 1v input signal. I think i'm happy with this design for now. 15watts at 8ohms and 30watts at 4ohm at 18v+/- just like I wanted. It is not perfect though.
Attached Files
File Type: asc AmpTestFinal.asc (7.0 KB, 20 views)
  Reply With Quote
Old 14th June 2019, 04:00 PM   #10
brian92fs is offline brian92fs  United States
diyAudio Member
 
Join Date: Jan 2017
Location: Northern California
transistor amplifier design finalization
You look to be running a 16ma VAS current. This seems very high for a low voltage single output pair amp. I would expect this to be something close to 5ma. If you stick with 16ma, you'll need to ensure the device you choose can handle the heat dissipation.

Your gain and phase margin look OK.

The 100k Input and NFB resistors are high. Most designers choose something around 20k or 30k. This higher value will make life easier for the source driving it, but at the expense of increased Johnson noise.

The 2200u for NFB cap C2 is unnecessarily large. With your 10k R16, you'd only need 10u to achieve a 1.6Hz pole. If you update your R9 / R16 to a more reasonable 20k / 2k then C2 can probably be 47u. As C2 gets larger, PCB layout can be an issue due to physical size.

On a similar note, input cap C7 is too large for your 100k R20. You could drop this to 1u and still have a 1.6Hz pole. If you update R20 to 20k, you can still drop C7 to 4.7u and achieve a 1.7Hz pole.

Your Cdom pole frequency has issues. With the 47p you used, the -3db point is 4.6 MHz with a large hump. I suspect this would lead to stability issue (though I'm far from an expert here). To tame this, you'd need something like 470p.
  Reply With Quote

Reply


transistor amplifier design finalizationHide 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
Transistor amplifier design Fossilshark Solid State 27 2nd March 2019 06:10 PM
looking for a compact ~1W 8ohm transistor amplifier design DarkDeepRed Solid State 17 14th March 2016 06:43 PM
Transistor Preamp Design for ADC ksubieta Analog Line Level 6 21st October 2012 01:26 AM
Old transistor amp design - opinions please! pastedavid Solid State 9 21st March 2012 09:49 PM
active crossover check before finalization mazurek Multi-Way 0 30th December 2005 11:04 PM


New To Site? Need Help?

All times are GMT. The time now is 03:19 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