Retro Amp 50W Single Supply

Just having had a close look at the circiutry in #1 again: Shouldn't the voltage divider provide a potential of about half the supply voltage for the first transistor's base?


Best regards!
As you should see in my post above, I replaced the 270 k pulll up resistor with something else to make the voltage at the emitter of the input transistor idle about 1/2 the rail voltage. This minimizes 2nd harmonic distortion on large signals if the operating point stays near there.
As I had a very high gain input transistor and needed compatibility with a high impedance PAS2 tube preamp, I also increased the pulldown resistor to achieve about 500k input impedance.
 
PCB for # 278 . Done quickly, please point errors or ask me to do something differently.
thimios, is this what you are looking for:)?

Hope this isnt too divergent from Mr. Mile's subject thread:eek:.

otherwise, request mods to delete the post.
regards
Prasi
 

Attachments

  • penfold amp.png
    penfold amp.png
    74.4 KB · Views: 807
Last edited:
IHMO The AL125 sounded much better. Maybe a member who has some spare time could do a board layout of the AL125 using MJL21194 flatpacks and pushing the rails to 100v which should give you app 100w into 8 and about 150 into 4 ohm. Th double bootstrap on the AL125 is also unique not something I have not seen a lot of. I did try bridging the modules with 2x AL125 with 2n3773 at 80v and doing away with the speaker bypass caps it worked perfectly for a while but one of the modules would start drifting so the midpoint voltage would have to be re adjusted.
 
IHMO The AL125 sounded much better. Maybe a member who has some spare time could do a board layout of the AL125 using MJL21194 flatpacks and pushing the rails to 100v which should give you app 100w into 8 and about 150 into 4 ohm. Th double bootstrap on the AL125 is also unique not something I have not seen a lot of. I did try bridging the modules with 2x AL125 with 2n3773 at 80v and doing away with the speaker bypass caps it worked perfectly for a while but one of the modules would start drifting so the midpoint voltage would have to be re adjusted.

Thank you Bruce for your inputs. Much appreciated.
 
PCB for # 278 . Done quickly, please point errors or ask me to do something differently.
thimios, is this what you are looking for:)?

Hope this isnt too divergent from Mr. Mile's subject thread:eek:.

otherwise, request mods to delete the post.
regards
Prasi

Can someone tell, if this amp will work with modern more easily available devices? Just an exercise towards curiosity!:D.

regards
Prasi
 

Attachments

  • penfold amp with 2sk1058_2sj162_not tested.png
    penfold amp with 2sk1058_2sj162_not tested.png
    60 KB · Views: 726
  • sch_trial_not tested.png
    sch_trial_not tested.png
    15.4 KB · Views: 735
Hi Prasi, like I said I did build the circuit and it did work although I did double up on the fets. I should imagine it would work just as well, if not better with the BUZ series fets or perhaps with a few modifications IRFP240/9240. Personally I did prefer the AL125 it seemed to have more bottom end. It would be great if someone could simulate it in LT spice ( something I am not too familiar with ). I have since found the original article for the AL125 module in the October 1980 copy of Pratical Electronics. As soon as I can scan and post it
 
"-" is the "power ground". So connect the "-" to the VAS ground and all will be clear fast. Add a 4700uF -or so- output cap and you're good to go. The schem posted here is imho a more "true" one - an opinion based solely on my eyeballing expertise.

Google finds this: Archives: Vintage Audio – The Autona 125 module – Online Parts International

And this: Autona AL125 - Analog Ian


So if you're really into vintage ... imho a good idea would be to cobble up a derated 80W or so version with 4x TIP3055 and a BD139-16 instead of the BC441.

IF you are adamant to follow the autona circuit ...