Re: Updated output
fab.
Okay, use the resistor values that will work.
You are the one that would know best.
And you can run the Simulations, too see the results.
My suggestion was Without Buffer transistors: Q36 Q37.
These buffers will allow higher value resistors, this is only a good thing.
I think your latest schematic looks good.
I downloaded and saved it in my /diyaudio-persons/fab folder
Where I have ALL the versions you have posted of this amplifier.
Regards
lineup
fab said:
fab.
Okay, use the resistor values that will work.
You are the one that would know best.
And you can run the Simulations, too see the results.
My suggestion was Without Buffer transistors: Q36 Q37.
These buffers will allow higher value resistors, this is only a good thing.
I think your latest schematic looks good.
I downloaded and saved it in my /diyaudio-persons/fab folder
Where I have ALL the versions you have posted of this amplifier.
Regards
lineup
Re: version w/o extra buffer
I think you should keep Q36 Q37 buffers for Gates.
It is not good to let Output Stage do high % of the amp total voltage gain.
Because this gain has got not same QUALITY,
as for example INPUT Stage.
The trick is to make the INPUT stage do as much of the gain
as possible,
But not that much of it that input stage start making a BAD JOB.
Because input stage transistors, are The Faster ones ( fT 300MHz ).
They are often also working in a more linear region
of amplification. At those currents they operates.
lineup
fab said:
I think you should keep Q36 Q37 buffers for Gates.
It is not good to let Output Stage do high % of the amp total voltage gain.
Because this gain has got not same QUALITY,
as for example INPUT Stage.
The trick is to make the INPUT stage do as much of the gain
as possible,
But not that much of it that input stage start making a BAD JOB.
Because input stage transistors, are The Faster ones ( fT 300MHz ).
They are often also working in a more linear region
of amplification. At those currents they operates.
lineup
Re: Updated output
Sorry I had a divider (input filter) in the input section too high for the impedance of the this CFP stage. I knew that the simulation was wrong the the theory of the gain was right:
CFP gain = 1 + R110 / R80 ( approximation since the open loop gain is not that high...)
So I get back to my divider with a gain of about 2 since I do not need 60V output and I want to limit the risk of oscillation as much as possible...Anyway, the power supply will drop problably at about 56v with both channel driven and if we add the Vds drop with a high RDSon (lateral mosfet) at high output current then 50Vac or so signal amplitude is quite OK.
fab said:
Sorry I had a divider (input filter) in the input section too high for the impedance of the this CFP stage. I knew that the simulation was wrong the the theory of the gain was right:
CFP gain = 1 + R110 / R80 ( approximation since the open loop gain is not that high...)
So I get back to my divider with a gain of about 2 since I do not need 60V output and I want to limit the risk of oscillation as much as possible...Anyway, the power supply will drop problably at about 56v with both channel driven and if we add the Vds drop with a high RDSon (lateral mosfet) at high output current then 50Vac or so signal amplitude is quite OK.
Attachments
Hi Fab & Lineup,
Have a look at this circuit...
It features opamp frontend and CB stage coupled with inverse driver to drive the Comp.MOSfets in Common Source configuration...A local and a Global feedback loop for linearization....All BJTs are small signal 2N5551/5401 types or similar....output devices are vertical mosfets IRF640/9640 ones
I have used this in my low power amps upto 500W with no cross-conduction at 50KHZ at all, also its slew rate is 75V/uS with +-65V rails....
K a n w a r
Have a look at this circuit...
It features opamp frontend and CB stage coupled with inverse driver to drive the Comp.MOSfets in Common Source configuration...A local and a Global feedback loop for linearization....All BJTs are small signal 2N5551/5401 types or similar....output devices are vertical mosfets IRF640/9640 ones
I have used this in my low power amps upto 500W with no cross-conduction at 50KHZ at all, also its slew rate is 75V/uS with +-65V rails....
K a n w a r
Attachments
Workhorse said:
Thanks Kanwar!
Sure is on topic. Using MOSFETs with Output taken from Drains.
About IRF640 / IRF9640, TO-220 HEXFETS:
They are probably the best TO220 IRF devices we can find.
I am surprised they are not more often used ......
I saved your simplified circuit, in my folder
/My documents/diypersons/Workhorse/
Questions:
1. Is there any special op-amp you have used and
can recommend for this application?
2. How much current do you use for driving the MOSFET Gates?
Regards
lineup
will never in his life have the money to visit India............................................................................................
India one of the oldest cultures of mankind
India have so many people, human resources, they can outperform any western country
India together with China will be major regions for future world economy!
lineup said:
Thanks Kanwar!
Sure is on topic. Using MOSFETs with Output taken from Drains.
About IRF640 / IRF9640, TO-220 HEXFETS:
They are probably the best TO220 IRF devices we can find.
I am surprised they are not more often used ......
I saved your simplified circuit, in my folder
/My documents/diypersons/Workhorse/
Questions:
1. Is there any special op-amp you have used and
can recommend for this application?
2. How much current do you use for driving the MOSFET Gates?
Regards
lineup
Thanks Lineup,
This circuit works best with OPA2134, NE5532, LM837, MC33079...works worst with TL074....
The Class-A buffer EF stage idle current is 25mA, since the drivers are in inverse follower modes they dissipate very little 35mW max...All the dissipation is served by the resistor itself...
For much lower power versions IRF540/9540 could also be used...
regards,
K a n w a r
fab said:
The driver stage in your amp is CE which is prone to oscillations with HF , also Cross-Conduction is likely to happen above 25khz and large clip sticking during saturation because of charge storage effects....
If you look in my circuit it uses CB which is much faster because of absence of Miller effect and there is no cross-conduction at HF...
regards,
K a n w a r
Also, The Turn-off of mosfets in your circuit is passive and Turn-ON is active, which again contributes to cross-conduction at HF, while you can eventually see in my circuit turn-off is faster and active while turn-on is slower and passive with simple resistor...
Also it improves the capacitive load handling
Also it improves the capacitive load handling
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.