The M2 Output Stage in Class A/B, and maybe a Power WHAMMY?

Next steps in next day's
Happy weekend 😁
IMG_20241116_131203.jpg
 
  • Like
Reactions: chermann and Dan K
Bootsstrap with opa604
some limitations:
1) my installed caps can only handle 25V , i tested it shortly up to +- 27V rails
2) if i change this cap limit my workbench power supply can only go up to +-32V

so have a look what we got now

+-24Vrails-Vin 300mVrms (pix DS204)--oscillations

RigolDS204.png


+-24Vrails-Vin 200mVrms (pix DS205)--oscillations

RigolDS205.png


+-25Vrails-Vin 200mVrms (pix DS206)--NO-oscillations
RigolDS206.png



+-25Vrails-Vin 400mVrms (pix DS207)--Zoom oscillations--look like 1.5MHz
RigolDS207.png


+-27Vrails-Vin 400mVrms --NO-oscillations

next steps--insert 22pF parallel to Rfb2 or change OPamp
 
  • Like
Reactions: Dan K
....really?....

this is developed yet!
more rail on the OPS-Output stage is useless because your opamp (OPA828) can´t provide the maximum swing to get near to the maximum output level.
remember the OPS has no voltage gain!

kr
chris
Perhaps I have the wrong end of the stick, perhaps not. What then is the use of the Bootstrap circuit, if not to keep the voltage across the v± of a lesser OpAmp within its limits whilst boosting the output. Who knows..
 
So...as a re read my post - it is maybe some line for clarify:


setup 1
We have a transformer with 12VAC and 200-400VA then you get after rectifier and cap filter about 17V voltage on both rails.
you can use your OPA828 without any restriction, because the safe operation area of the opamp is 18v on each rails.
the bias current should be set by your size of the heat sink of your amp.
the power of this amp is limited by the maximum rail voltage and the losses written in the datasheet of the OPA828. with 17V rail and a big heat sink you can set your bias about 1,7A without problems. 28,9 Watt per MOSFET´s.
chapter 6.5

1732371904697.png


so you loose about an extra volt (1,2V) on each rail on the output of the opamp. +17-1,2= 15,8V,
at 8R load ....about 11,17Vrms = 15,6WATT
at 4R load ...about 11,17Vrms = 31, 2 WATT (1,4A bias)

setup 2
We have a transformer with 14VAC or 15VAC and 400VA then you get after rectifier and cap filter about 19,7V - 21,21V voltage on both rails.
the M2 OPS is a set with an opamp and an OPS both are using the same rail. That means you do not care about OPS because, lets say, the MOSFET Fairchild or this famous Toshiba's can handle a lot of voltage and you have just to set your current according to your size of heat sink.
the tricky point is the opamp and its maximum operation / recommended operating voltage, e.g. OPA828 max +/- 18V .
--> so for the opamp, input stage and VAS- we have to "waste" 2-4Volts to have a safe operation of the OPA828.
a)

to "waste" this 2- 4Volt you have to look at the datasheet for maximum current of your opamp.
chapter 6.5
1732372535094.png


so we know that 7,1mA is the maximum current of the opamp on each rail- so to be on the safe side we want to "waste" 4volts at 7,1mA - so we need a limiter= resistor of 563,38 ohm ---> E96 1% resistor with 562ohm 1/4Watt.

b)
you can use a bootstrapping setup like the Lt spice file of Patrick in post 362 with a 17 (BZX84B17L) zener diode configuration instead of his example of 15V zener BCX84C15L

important is that we are not talking about the bootstrapping of Patrick in post 358 -this can really bootstrapping!..not tested!

patrick´s example in post 362 is a first step to use OPA828. to use it safe but without the advantage of more power.



setup 3
We have a transformer with 18VAC and 400VA then you get after rectifier and cap filter about 24-25V voltage on both rails.
So for the opa828 too much voltage. so we need to waste about 7V each rail.
if you want to use e.g. OPA551 - this opamp can handle +/- 30V so no problem.. no limiter resistor or bootstrapping (post 362)

power with 24Volt rail and a OPA551 is:
check the datasheet for voltage losses
1732373726985.png

so you loose about 2-3V , so rail 24V - 3V losses is 21V peak
at 8R load ....about 14,84Vrms = 27,55WATT
at 4R load ...about 14,84Vrms = 55 WATT..if you transformer and your bias (1,9 -2A) current is hugh and your heat sink is big!

for opa828 and other with less recommended operational rail voltage-->

a
you can use in the rails of the opamp with the reference part LM329 to "waste" about 7V- done by Vunce and DIYHarry..others...
I am not able to talk about noise in comparison of resistor like setup2_a or the Lm329 setup3_a..which is "better"

b
the same calculation like in setup2_a

so we know that 7,1mA is the maximum current of the opamp on each rail- so to be on the safe side we want to "waste" 7volts at 7,1mA - so we need a limiter= resistor of 985,9 ohm ---> E96 1% resistor with 1kohm 1/4Watt.

so you have with OPA828 not more power! nearly the same as setup 1


kr chris
 
Last edited:
Good morning

i used the design document in post 358 and play around with LT spice (attached). I use 100v rail voltage and the LT1793 (+/- 15V max) . R1 and R2 is set that the maximum swing at the +V and V- is not bigger than 30V (+15V to -15V) at the power pins of the LT1793.

the gain is about 22,3dB. it is just to play around it is not the M2OPS circuit.
so with 1V sine i got the maximum output of that opamp...about 13V swing .

It is not clear for me why the resistor Rfb must be bigger than 100k or Rfb1 not smaller then 12k otherwise i got no signal out....

kr
chris

1732429013162.png

It
 

Attachments

Last edited:
  • Like
Reactions: stv