F5 power amplifier

[jacco vermeulen]

My bad, i meant just the PCS part.

I do recall noticing something of a price rate alteration => EU Mini Group Buy FQA19N20C / FQA12P20

[Magura]

There is nothing odd to the IRFP240 from IR.

The problem I guess Oscars is referring to, is the issue of the P-channel mosfets form IR (as in for instance IRFP9240).


So Oscars, you should be fine.


Magura

[osscar]

Yes, then I realized that it is almost the same and also found some info regarding these FETs in this topic. Thank you for answer!

[AndrewT]

The irf240 and the irf9240 are not well matched.
I think it hardly matters who makes to non-complementary pairs.

A better match is the irf240 and irf9140 for complements.

The 9140 is a 100V device and this limits the supply rails to <+-50Vdc. No problem for a ClassA amplifier.

[EUVL]

> Patrick built it, and it worked, but he didn't seem too impressed with it.
> Which suggests to me that something is not quite right.

I consider this remark misinterpretation and perhaps somewhat misleading.

Firstly the fact that I am currently not using the balanced F5 in my reference system, and that Peter likes his SE version, is not sufficient to draw any conclusion on the relative performance between the two.

I have only built the balanced version, with 2 different types of MOSFETs (Toshiba and Fairchild), and I like the Toshiba more. But it is personal taste. I have NOT built the SE version, so I cannot make any comparison for you.

In general, I only build balanced amplifiers, as they allow cancellation of even harmonics. Some like second harmonics which "sweeten" the sound. I don't. Again personal taste.

I do not recall saying the F5 is a not-good-sounding amp. There are many reasons why I am not using it in my reference system, the most important of which is lack of gain, which I need for my particular setup. On top of that, I have a very high bandwidth system (my power amp is 350kHz bandwidth, 120W Class A), so the additional bandwidth does not bring me any advantage. In someone else's system, it might be.

The fact that I choose another amplifier for my daily use does not make the balanced F5 good or bad. Just as one cannot compare a F5 directly to a F4 or an AXJ. They are all audiophile amplifiers. Which one you might like is, again, your personal choice.

I have optimised the balanced F5 circuit based on the devices I used, and as much as I can tell, the values are closed to optimum. That does not mean that there are no rooms for improvement. But it is unfair to imply there is anything fundamentally wrong with it, just as someone else tried to suggest with my SMD-B1.

In the end, you build the design which appeals to you. Guaranteed results can be had if you should build 100&#37; as Nelson says. I like to experiment and every now and then design my own. And publishing my circuit is no claim to any improvement over the original design. It is just my own implementation that I wish to share. Hopefully someone would also experiment and improve it further, for the benefit of all.


Happy experimenting,
Patrick

[jleaman]

I wish the aleph mini's were balanced, but then again, id have to go buy a few new goodies to make it work, I guess having balanced is nice, but to make it all work you have to have all the proper equipment.

Ie a balanced pre-amp either balanced out but se in or balanced in and out then you would need a source with balanced in ie cd player or some kinda converter.

I went se because i didn't have to mess around with that stuff, however if i had money and time i would try it with balanced, maybe next year ill build a f5 and try balanced, witch would mean i would have to build all new crosoovers etc etc, see it all adds up.

In personal learning through out life, i weed out certain things, i build amplifiers and stuff for me, not any one else, i learn alot and try alot of things also.

Every one is different every one has their own opinion too, mines beter than yours because ...... no no mines beter because i bought all gold pieces, oh yeah mines better because i spent 30k on cables, nooo mines better etc etc, i now weed out that stuff.

Build it enjoy it tinker with it learn from it

[EUVL]

My system is balanced from DAC (chip) to power amp.



Patrick

[jleaman]

Originally Posted by EUVL

My system is balanced from DAC (chip) to power amp.



Patrick

What about pre-amp ?

[EUVL]

Balanced, all JFET, no global feedback, 2MHz bandwidth, battery powered, drives 200 ohm load with ease, .....


Patrick

[EUVL]

One additional remark on the difference in the balanced F5 implementations :

> But I didn't implement the output protection circuits (Q4, Q5 in the original schematic) and used identical values for the source resistors (R1,2 =10Ohm, R11,12 = 0.235Ohm) and the feedback resistors (R5,6,7,8 = 100Ohm). The output devices are 2SK1530/2SJ201 from Toshiba.

In my published circuit, which is identical to what I built, I did use different resistor values for the top (2SK170 / 2SJ201) and the bottom (2SJ74 / 2SK1530) halves. This is because while the MOSFETs are perfect complementary devices, the JFETs are not. Using different resistor values would allow gain balancing between the two halves, such as they contribute equally to driving the load, and also allow better harmonic cancellation.

In a balanced system, this cancellation is less important as in SE, as the cancellation will take place again between the left and the right halves, assuming you have (near) perfectly matched devices.

And the reason I run mine at +/-16V, 4A is simply that I have low impedance speakers. As is, you get maximum output power at about 6 ohm. (I always leave 4V Vds as minimum, by design)


Patrick