Power Amplifier for 100kHz.

On 2023-12-19 22:21, JM wrote:
> On Tue, 19 Dec 2023 21:53:41 -0500, Phil Hobbs
> <pcdhSpamMeSenseless@electrooptical.net> wrote:
> 
>> On 2023-12-11 15:32, JM wrote:
>>> On Monday, December 11, 2023 at 6:49:30?PM UTC, Glen Walpert wrote:
>>>> On Mon, 11 Dec 2023 15:01:25 +0000, Clive Arthur wrote:
>>>>
>>>>> On 11/12/2023 14:05, legg wrote:
>>>>>> On Wed, 06 Dec 2023 23:22:26 GMT, Glen Walpert <nos...@null.void>
>>>>>> wrote:
>>>>>
>>>>> <snip>
>>>>>
>>>>>>> Some years ago Jim Thompson posted an audio amplifier design which
>>>>>>> used current mirrors to provide bias to the output transistors for the
>>>>>>> express purpose of keeping crossover distortion low over a large
>>>>>>> temperature range. He claimed it was the bees knees, but a quick
>>>>>>> search failed to turn it up. Perhaps someone else saved it or
>>>>>>> remembers the thread?
>>>>>>>
>>>>>>> Glen
>>>>>>
>>>>>>   From old LTSpice trash here;
>>>>>>
>>>>>> http://ve3ute.ca/query/Half_Bridge_for_77_280Z_thompson.pdf
>>>>>>
>>>>>> RL
>>>>>
>>>>> Thanks!
>>>>>
>>>>> Looks like the top output Darlington is AC coupled and when the
>>>>> comparator detects a quiescent current through the output resistors
>>>>> transitioning to less than some value, it pumps the upper Darlington
>>>>> base voltage up a bit, otherwise, the upper Darlington base voltage
>>>>> drifts down.
>>>>>
>>>>> Is that about right?
>>>> There was a lot of discussion of this circuit when it was posted, and Jim
>>>> posted some models and simulations possibly still available on Phil's
>>>> archive. I don't have time to actually think about it right now, but here
>>>> are some post snips with comments and model links, sorry about the length:
>>>>
>>>> ------------
>>>> Here's half of the full H-bridge amplifiers that I built for my 1977
>>>> 280Z... Image scanned in quarters and pieced together for easier
>>>> understanding...
>>>>
>>>> <http://www.analog-innovations.com/SED/Half_Bridge_for_77_280Z.pdf>
>>>> -- 
>>>>> If both output transistors are briefly off or very nearly off while the
>>>>> output is increasing through crossover (zero or near zero current through
>>>>> both emitter resistors), then the LM311 goes high due to the lag through
>>>>> the RC on the negative input, delivering additional current to the
>>>>> current mirror with illegible designations through D1, pulling current
>>>> >from the 20uF 10V capacitor, increasing it's voltage thus increasing the 
>>>>> bias offset provided by Q5 and Q6 until there is enough bias voltage
>>>>> difference to insure some small overlap in the on time of the output
>>>>> transistors. Q1 and Q2 appear to keep the bias voltages centered between
>>>>> the rails, and possibly Q8 pulls the negative input of the comparator
>>>>> down enough to prevent noise from turning it on with no input?. (Not at
>>>>> all sure about Q8, it might do more than that).
>>>>>
>>>>> Am I close? Hints on Q8?
>>>>>
>>>>> Regards,
>>>>> Glen
>>>>
>>>> You are virtually on the money!
>>>>
>>>> Q8 is just a current mirror operating on R13 to establish the bias current
>>>> at the zero crossing (your observation that corrections only occur while
>>>> passing thru the zero crossing are dead-on... except that the Q8 current
>>>> prevents both off).
>>>>
>>>> ...Jim Thompson
>>>>
>>>> The Q5/Q6 Darlington is simply to knock down the base current so that a
>>>> long R/C time constant dominates.
>>>>
>>>> ...Jim Thompson
>>>>
>>>>> What's the SPICE quiescent bias? Back of the envelope, I get 75 or 80mA.
>>>>>
>>>>> How do you pick R15/R16/C4? Looks like it's to bootstrap the bias above
>>>>> the 13.3V rail with a time constant longer than the roll-off of the
>>>>> amplifier.
>>>>>
>>>>>
>>>>> Best regards,
>>>>> Spehro Pefhany
>>>>
>>>> In a later life I might have used a diode. We improve our skill-set over
>>>> the years... at least some of us do... some just bloviate >:-}
>>>>
>>>> ...Jim Thompson
>>>>
>>>> See...
>>>>
>>>> As requested, entered into PSpice and simulated....
>>>>
>>>> <http://www.analog-innovations.com/SED/My_1977_Z_Amp.pdf>
>>>>
>>>> for the simulation (and a readable schematic).
>>>>
>>>> Betwixt the "honey-do", I ran intermod distortion, comparing class-B to my
>>>> class-A-B method, zip file now updated...
>>>>
>>>> <http://www.analog-innovations.com/SED/JimThompsons_A-B-
>>>> Bias_Amplifier.zip>
>>>>
>>>> ...Jim Thompson
>>>>
>>>>
>>>> To go along with that schematic, here is the subcircuit that should work
>>>> in all modern flavors of Spice...
>>>>
>>>> <http://www.analog-innovations.com/SED/My_1977_Z_Amp.zip>
>>>>
>>>> To simulate my circuit in LTspice, open a text editor and type the
>>>> following...
>>>>
>>>> * Jim Thompson's 1977 Z Amplifier *
>>>> ** Analysis setup **
>>>> .tran 0 10m 0 100n .OPTIONS ITL1=1500 .OPTIONS ITL2=2000 .OPTIONS
>>>> ITL4=1000 .OPTIONS STEPGMIN .OP X1 IN OUT VCC 0 My_1977_Z_Amp VCC VCC 0
>>>> 13.3V VIN IN 0 SIN 0 4 1K 0 0 0 .INC
>>>> "C:\InsertYourPathToCopyOf\My_1977_Z_Amp.sub"
>>>> *
>>>> .END
>>>>
>>>> Save as whatever name rings your chime, say...
>>>>
>>>> "JimThompson'sMarvelousAmplifier.cir" >:-}
>>>>
>>>> Then open LTspice. On the Tools/Control Panel/Save Defaults section check
>>>> both Save Subcircuits... check-boxes.
>>>>
>>>> Then Open "JimThompson'sMarvelousAmplifier.cir"
>>>>
>>>> Then Run
>>>>
>>>> View whatever node voltage or device current you like.
>>>>
>>>> Irrespective of Larkin's stone throwing, it doesn't fail for several
>>>> reasons... one specifically because it was 1977. Can anyone guess what
>>>> that was?
>>>>
>>>> Interestingly it takes LTspice _much_longer_ to run this circuit than it
>>>> does PSpice, particularly the bias point calculation is butt slow.
>>>>
>>>> Note that you _do_not_ need to draw a schematic in LTspice (or any other
>>>> Spice, for that matter) to simulate someone else's circuit. Many of my
>>>> clients only have LTspice, so I just pass them a PDF schematic and a
>>>> netlist, and they can verify my work just fine.
>>>>
>>>> ...Jim Thompson
>>>>
>>>>
>>>>> Latest version....
>>>>>
>>>>> <http://www.analog-innovations.com/SED/JimThompsons_A-B-
>>>> Bias_Amplifier.zip>
>>>>
>>>> Turns out that my A-B bias is STUNNINGLY better than the conventional
>>>> diode-biased class-B... almost 30dB better on intermod distortion!
>>>>
>>>> Intermod is what gives you those nasty atonal ear-piercing sounds when you
>>>> play a Mozart wood-wind ensemble with French horn accompaniment.
>>>>
>>>> After 36 years, revisiting my scheme, and fixing the bias droop, it's time
>>>> for me to go back and roll my own sound system from scratch... like I did
>>>> up until my late 30's... then I got "too busy" ;-)
>>>>
>>>> I'll toss the TL091 and put some discrete's in there... maybe even use my
>>>> TL431 diff-pair >:-}
>>>>
>>>> ...Jim Thompson
>>>
>>> Here is a link to Mr Thompson's file (JimThompsons_A-B- Bias_Amplifier.zip) should anyone want to simulate the circuit.
>>>
>>> https://1drv.ms/u/s!AkjNCaVyTfIag2w_3ym3b15HkMyP?e=PWxkbm
>>>
>>
>> I went do download this, but it wants a login.
>>
>> Cheers
>>
>> Phil Hobbs
> 
> I've emailed it to you.  Links posted here will only be active for a
> few days.  If anyone else needs it try :-
> 
> https://1drv.ms/u/s!AkjNCaVyTfIahAKhXsApoGWJlEtu?e=ey7AEC
> 
> I did actually build a variation of this when it was originally
> posted.
> 
Thanks!

Cheers

Phil Hobbs

-- 
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com