Orangebelt design halfway done; dedicated to bigun, the blackbelt designer amongst us.
Zeners will be added later. I want to test it without them. I am expecting a kaboom. Pray for me brothers, as this is my very first symmetrical work.
I'll post amp behaviour soon, whether smooth or erratic. I'm not afraid, I know I have many saviors here.
Zeners will be added later. I want to test it without them. I am expecting a kaboom. Pray for me brothers, as this is my very first symmetrical work.
I'll post amp behaviour soon, whether smooth or erratic. I'm not afraid, I know I have many saviors here.
Attachments
By habit, and if the printed board is ok, CF is rock stable at all gains. The only issue reducing feedback can be a decrease of damping factor and slight augmentation of THD, OMHO.
You are right, input level has to be 1V for full power output.
Initial test done.
Settings:
Bias 100mA total. Offset +-5mV(varies all the time). 100ohm gate stopper at board and FET. No source resistors added. Supply +-20V, 15000uFx2 smoothing caps.
Observations:
Stable bias. No oscillation(well, I don't have a o-scope, but my am receiver is very efficient in this job LOL, plus no overheating anywhere, the input BJTs are warm, the VAS BJTs are little more warmer, thats all and thats what I expected.) Offset isn't very "stable" but is limited to +-5mV, also is easily to set by the 1K pots after setting close to + or - 1volts with the 20K pot. NO audible noise with inputs open.
Currently playing The Look Of Love - Saskia Bruin
Settings:
Bias 100mA total. Offset +-5mV(varies all the time). 100ohm gate stopper at board and FET. No source resistors added. Supply +-20V, 15000uFx2 smoothing caps.
Observations:
Stable bias. No oscillation(well, I don't have a o-scope, but my am receiver is very efficient in this job LOL, plus no overheating anywhere, the input BJTs are warm, the VAS BJTs are little more warmer, thats all and thats what I expected.) Offset isn't very "stable" but is limited to +-5mV, also is easily to set by the 1K pots after setting close to + or - 1volts with the 20K pot. NO audible noise with inputs open.
Currently playing The Look Of Love - Saskia Bruin
Attachments
Last edited:
Esperado you've been attracted to this thread for some reason but was really not aware why at begining.
Your sch is classical example of people ingenuity to make progress on certain ideas. The SSA principle is very flexible and I can see how to implement it in your case turning the amp to current feedback SSA type.
Just remove R15-R18, D8,D9, split fedback loop like in SSA design (two times two feedback resistors) connect it to the emmiters resistors of input differential pairs, also lower 150 ohm resistors to 10 ohm. Leave the current sources Q3/Q9 like they are and because of cca. 1,2V DC potential between +/- diff-pair emmiter resistors you have to filter AC out with a good 100uF elco as much as possible.
Now you have your amp being SSA pimped.
Alive. One word is enough.
Well, mid bass to treble is perfect. Very good tonal balance, independent of volume; this I liked the most. A very transperarent sounding machine. I have yet to test the bass response as currently I lack a passive Sub.
I'll update transient production quality later; I think this needs more time than usual.
Cheers.
Last edited:
It started ...
You happy - me happy. Congratulations.
P.S. Now you can start to think to go SSA way - emmiter bias arrangement - it will give you even more stable DC conditions and less parts count.
It started ...
You happy - me happy. Congratulations.
P.S. Now you can start to think to go SSA way - emmiter bias arrangement - it will give you even more stable DC conditions and less parts count.
I'm already on it; my next project.
Special thanks to you.
Hi Bigun, start a TGM5 thread. If your going to design a PCB for this Ill have a go at it. I hate doing PCBs, no patience or time for it and so I have them designed and completly built up including components by pro companies that charge for it but for a once off amp it isnt viable.
I might not entirely end up exactly as your design is but a couple of wires or components off or below a PCB doesnt bother me.
My old design is a class A heater, with 4 pairs good old 2sj50 2sk135 in a cfp arangement as advocated by JLH at the time. Ill have to take it apart to get the exact schematic again, Ive forgotten what I did. I had some input given by JLH himself via my varsity prof and the idea was a amp with very low high order harmonics. It must have sounded pretty good otherwise I would have taken it apart or thrown it away but its still in my garage under a tarpole with all my other succesfull and potencially good sounding amp endevours.
I know why ;-) I'm dedicated at CF from years now, using it exclusively as far as audio is in concern. And your design is f... brilliant. As brilliant as your kindness is touching.
I was away from electronic since 25years, and it is hard for me, now, i don't feel any more the things, obliged to make calculations for everything, age is a nightmare, too.
That was my purpose, before to build from scratch YOUR definitive prototype.
See attachment, that's what you means ? I just wonder why you want me to change the emitter values from 150 ohms to 10, it increase current and open loop gain, right ? it need 4watts for the 330 ohms then, and, if i'm correct, the gain will be 33 if 330ohms are set to 660 now ?.
Where did you see this 100µF, where i have put the 1000µ ?
I'm lurning LT spice, right now, and it don't want to pimp-itNow you have your amp being SSA pimped.
But, when it will go, i think we'll have all the inputs architectures, now ? This one is complementary from the Shaan one. (Congratulations, Shaan)
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.