F4 power amplifier

[generg]

hi all
if I run 1 pair of the fqa parts per side (like the m2) can I just remove the unused mosfets and their .47/100r resistors without any other circuit changes

what power into 8 ohm can I expect after doing this ? same as m2
I will be running it of the ba3 frontend

Sheafer

I hear with this combination of "one pair F4" and BA-3 Frontend at the moment and yes I think we should have the same Watt,
Of course F4 3x200mV over the 0R47s (do I remember right?) I now drive the one pair with around 1.2A! Means 600mV over the 0R47.

 

[slr 5000]

Hi Gerd
fired it up today and had some smoke ,not sure why yet but I have melted the tracks coming off d3/d4

Sheafer

 

[Itsmee]

I have melted the tracks coming off d3/d4

Sounds like you have installed the diodes the wrong way round.

 

[slr 5000]

they are the correct way around but I think they may not be the right ones
I stole them off some other boards I have and presumed they were right

guess not
the schematic calls for 1n418 fast switching ,I think I may have installed zeners
Cheers Sheafer

 

[slr 5000]

yep
just checked with a magnifying lens and I have installed 1n4739 to d3/d4 instead of 1n418

damned rookie mistake but easily fixed
I just hope the tracks were all that burnet

Sheafer

 

[slr 5000]

got her running
after a little adjustment i have 1.25v across the source resistors and offset is near zero both sides

sounds good so far
time will tell if she,s a keeper though

Sheafer

 

[generg]

So you drive them with 2.5A, one pair?

Is this not too much? Or do you have 1R source resistors?

 

[slr 5000]

hi Gerd
yes 1.25v per device across the .47 3w resistor
its getting hot though just passing the 3 second hand test on 300x150mm sinks with one pair

what should I be running ?

Sheafer

 

[Zen Mod]

no more than 35W per mosfet

 

[generg]

Around the half ...

0.6V or 0.7V over the 0R47, means around 1.4 or 1.5A bias.

I am afraid one pair must work too much with your values

 

[slr 5000]

I,m just hardcore what can I say

I have now adjusted the bias to .6v per OR47
she is already feeling cooler

Sheafer

 

[generg]

Sometimes hardcore is not bad.....

 

[slr 5000]

I thought she was running hot !!!

lucky I was not playing my doors album
come on baby light my fire

did you see the schematic Antoinel did for me ?
its the ba3 frontend with the amplimos 2sk82/2sj28 output stage
should be fun if I can get past my usual mistakes and get it to work

 

[valvesound]

no more than 35W per mosfet

20V supply rail X 1.5A idle current = 35W per MOSFET

@SLR 5000 "did you see the schematic Antoinel did for me ?
its the ba3 frontend with the amplimos 2sk82/2sj28 output stage
should be fun if I can get past my usual mistakes and get it to work"

No I didn't see...please show us the schematic?

 

[slr 5000]

hi Valvesound
check out my other thread in solidstate

Sheafer

 

[Triodethom]

hi Gerd
yes 1.25v per device across the .47 3w resistor
its getting hot though just passing the 3 second hand test on 300x150mm sinks with one pair

what should I be running ?

Sheafer

6l6 I think in his build (f4) thread test distortion and and found that 400ma was more than enough to reduce 3rd 4th 5th harmonic to the noise floor thus that is as much bias as I would run . about 1/3 of what you have now. Hope that help. also reducing the resistor by 1/2 is common (.22 or 2 .47 in paired) .

 

[6L6]

Thom -

Yes, that is indeed correct. More interestingly, although going as high as 400mV bias (measured across the source resistor) did decrease the distortion, the gains were very low for huge amounts more current.

My amp is currently set at 0.275V across the source resistors. That feels like the sweet spot to me.

 

[Triodethom]

Thom -

Yes, that is indeed correct. More interestingly, although going as high as 400mV bias (measured across the source resistor) did decrease the distortion, the gains were very low for huge amounts more current.

My amp is currently set at 0.275V across the source resistors. That feels like the sweet spot to me.

Thanks for the input, What the size of source resistor ? I had been reading the f5 blog and was thinking about the 5 ohm fix and increase of idss of .8 ma for the curve imbalance between the j74 ans 170. Also noticed the 47.5 vs 100 ohm on the gate using the 19n rather than the 240 . Any thought on those points ?

 

[buzzforb]

I have built 4. Used 47r for gate resistors on all builds. P3 will handle jfet difference. The topology and active devices define the sound.

 

[lhquam]

I just finished an F4 build and think that it will be my preferred, everyday amplifier. I am currently listening to it driven by a Buffalo-IIISE DAC modded to deliver about 7 volts peak single-ended, which is about all that the LM4562 will do. That peak output is a bit low so I will need to add a gain stage such as the BA-3 FE to the F4.

The F4 really sounds great.