Hello.
I have read about these successful variants with pcf in different amps and my q is if it is possible to implement pcf in my F4-amp and how that circuit look like ?
/ tord
I have read about these successful variants with pcf in different amps and my q is if it is possible to implement pcf in my F4-amp and how that circuit look like ?
/ tord
The F4 does not have feedback so I would think not. Not necessary. My very limited understanding of PCF is that it is used to dance with the negative feedback.
You could implement PCF in an F4, but it is not as simple as in other amps.
You need some kind of voltage-gain. You could use a step-up transformer between the output and input to increase the voltage sensed by r-sense.
I would consider it better to build another amp or add a voltage gain stage to the F4, but this will make it into a whole other amp.
Cheers,
Johannes
You need some kind of voltage-gain. You could use a step-up transformer between the output and input to increase the voltage sensed by r-sense.
I would consider it better to build another amp or add a voltage gain stage to the F4, but this will make it into a whole other amp.
Cheers,
Johannes
Well... I think it is not a good idea to try pcf in an F4 nonfeedback amp but I had to ask...
/ tord
/ tord
The problem would be that you have to throw away some of the voltage
gain, so that the follower becomes less than unity gain.
😎
It is amazing to me how you can confuse me with such a short sentence.
Wikipedia
Unity Gain: When a signal passes through a system or a device without changing level (amplify or attenuate) it is said to be at “Unity Gain”. Unity Gain is fundamental in setting up any audio chain or signal path.
Wikipedia
Unity Gain: When a signal passes through a system or a device without changing level (amplify or attenuate) it is said to be at “Unity Gain”. Unity Gain is fundamental in setting up any audio chain or signal path.
I was not clear?
You can set up a unity gain follower for less than unity gain using an input
voltage divider, and then bootstrap the bottom of that divider to the resistor
which is sensing output current to ground.
This will lower the output impedance of the follower.
😎
It's not necessarily a bad idea it's just not the best use of pcf since the circuit is already quite respectable and can easily be further improved simply by increasing the bias as high as your heatsinks will allow eg 2A instead of 1.3A.
Last edited:
Some might think that better damping factor without the need for big heatsinks and higher bias current IS a very good use of PCF.
Yes good point.
Some people like to warm their house in winter and make full use of the performance of their existing heatsinks. 🙂
You can also achieve the same sought of improvement at 1.3A by increasing the number of output pairs from 3 to 6 pairs.
Many ways to skin the cat.
Bear in mind though going from a damping factor of 40 to 100 is not as Audible as going from say 4 to 100. So it really depends on the performance of the circuit in stock configuration and your type of speakers whether it's worthwhile.
Some people like to warm their house in winter and make full use of the performance of their existing heatsinks. 🙂
You can also achieve the same sought of improvement at 1.3A by increasing the number of output pairs from 3 to 6 pairs.
Many ways to skin the cat.
Bear in mind though going from a damping factor of 40 to 100 is not as Audible as going from say 4 to 100. So it really depends on the performance of the circuit in stock configuration and your type of speakers whether it's worthwhile.
Last edited:
I just meant I am not knowledgeable enough to keep up. I am trying to learn. No doubt I have the best teacher there is.
🙂
This is not meant as a serious proposal, just some LT-spice play with PCF in a simplified F4 source follower amp.
With these resistor values the gain actually increases from 0,8 times the input voltage to 1,08 times the input voltage while the simulated distortion drops by almost an order of magnitude - and this for a higher power output.
Cheers,
Johannes
In simulation it exhibits a slightly negative output impedance as the output voltage increases from 4,328 volts over 8 ohms + 1 ohm R-sense, to 4,652 volts over 4 ohm load + 1 ohm R-sense.
I must have made something strange since I can not reproduce the large drop in distortion when comparing with and without PCF.
Please ignore the "order of magnitude - distortion" statement above.
Cheers,
Johannes
Please ignore the "order of magnitude - distortion" statement above.
Cheers,
Johannes
That's simulators for you....
Nevertheless, that was a useful and imaginative example.
😎
Well... this IS interesting !
Maybe I will try something like that.
/ t
Edit: Have just finished playing an old Yamaha B2 that was in need for some service... Hmm Nice old amp, I think I want to buy it.
Maybe I will try something like that.
/ t
Edit: Have just finished playing an old Yamaha B2 that was in need for some service... Hmm Nice old amp, I think I want to buy it.
Last edited:
Here is a much better version. Less unnecessary stuff leftover from fooling around with Nelson Pass suggestion of bootstrapping a voltage-divider.
Thanks Nelson Pass! I can't take any credit from this though. It was your idea. I just fooled around with some variations on how to bootstrap the voltage divider when it occurred to me that I could bootstrap the signal source with positive feedback from the R-sense (R8 in this example) instead.
I must thank you for this idea. I am currently playing with an IRLB3813 as a source follower with an autotransformer as source-load and output "transformer". I will test this idea with bootstrapping the PCF through the signal source tomorrow.
Cheers,
Johannes
- Status
- Not open for further replies.