DIY F2 clone

Member
Joined 2006
Paid Member
I am getting some turn on thump (well speakers pop) with F2J.
Is there any reason the C4-C6 charging cap setup would not work as well with the JFET mods?
Changing capacitance doesn't appear to help. I guess my next thought is to change out a resistor. Not sure I am looking at the right thing.
 
F2J Built

Just finished my F2J and am enjoying the sound of this amp. Never heard a SET before but now I think I have a better understanding when someone talks about a sweet sound, or sweet top end, or smooth mids and highs. I didn't think it would have been this detailed as well, in fact it seems to have a bit more intimacy than the F5. I was scared to build this amp because I thought my F5(w/toshiba) would be much more detailed and I wouldn't like it. I'm glad I did build the F2J. Thanks Nelson.

I'm in the process of creating a network to mate with my FE167E stock drivers. It seems that without any network the amp sounds very detailed with a tiny bit of noticeably sibilance on vocals. With 47Ω//47Ω across the amp terminals the sibilance is reduced a bit, but it's openness in sound and details seem to be the trade-off.

Do I need to tinker a bit more with the network or could the sibilance be contributed to some of the capacitors/parts I used?

Build Notes: (Stock F2 with the following mods)
- Z1, R2 removed
- R6 = 47 kΩ
- Q1 = SJEP120R100
- C5 = 100µF Electrolytic generic (what I have on hand)
- C1=C2= 1µF Metallized Polypropylene Film Capacitors Panasonic Type:ECWF(A)
- C6 = 33000µF Panasonic T-HA series
- C4 = Not used
- Peter Daniel boards
- Bias = 12.2V after readjust
- 10 hours burn in
 
Member
Joined 2006
Paid Member
I agree it tops my F5 with Toshiba's, which occasionally feels darker and dull by comparison.

I used a couple of different networks, all without the 47R's in the amp itself.
10R-10uf
10R
25R

Settled on the 25R. You do need something there, or it will sound a bit spazzed out.
 
Last edited:
I agree it tops my F5 with Toshiba's, which occasionally feels darker and dull by comparison.

I used a couple of different networks, all without the 47R's in the amp itself.
10R-10uf
10R
25R

Settled on the 25R. You do need something there, or it will sound a bit spazzed out.

Does the network loading effect the bias?

Would it be advisable to remove the three 47ohm resistors outputs and place a pot there with a series resistor at 10ohms or so?
 
Last edited:
Member
Joined 2006
Paid Member
So yes,
Mine does not have the 3x47R resistors across the output, as in the default schematic. I just load them across the speaker terminals, or amp terminals. (preferably the former).

It does not affect bias, but does affect some of the gain I beleive you will hear. It may take it down some. It will take the output impendence down to what you set it for.

Did you read the old article about current crossovers on pass diy?
It's good, but I would really get a bunch of 10 or 15w sandcast resistor values and some bipolar caps and play with it till you reach good balanced sound.
That's what I did. There might be some tradeoff, in them, but such is life.
 
Member
Joined 2006
Paid Member
Offset probelm

Hi all,

I've built a clone of the F2J on Peter Daniel's boards with the SJEP120R100 and the recommended circuit changes as per Mr. Pass. I fired it up was going to check the bias when I noted that one channel had about 20mV offest but the other side has 5-6 volts at the speaker terminals. Adjusting P1 on the bad channel will decrease the offset but only to about 4 volts by the time the pot runs out of adjustment. I'm not sure P1 should affect offset and it doesn't seem to on the good channel. Any ideas?

Many thanks
Rick