Could anybody give me a hand and take a look at this circuit and give a little creative critisism?? I would greatly apreciate it!
An externally hosted image should be here but it was not working when we last tested it.
corrections
The original image post should now reflect changes, sometimes I start plopping in components faster then my brain works
yes Q7 should be NPN, R15,R19 are 4.7 ohm current limiters, also I made a mistake on R5 it should also be 4.7 ohm it provides the ground path for VR1 to set the offset.
Q8, VR2, R16,R17, D1 and R22 make up the bias cicuit, Q6 provides thermal tracking.
The amp will be biased into class A for a roll off into AB at about 20watts. Thanks and keep the observations coming........
The original image post should now reflect changes, sometimes I start plopping in components faster then my brain works
yes Q7 should be NPN, R15,R19 are 4.7 ohm current limiters, also I made a mistake on R5 it should also be 4.7 ohm it provides the ground path for VR1 to set the offset.Q8, VR2, R16,R17, D1 and R22 make up the bias cicuit, Q6 provides thermal tracking.
The amp will be biased into class A for a roll off into AB at about 20watts. Thanks and keep the observations coming........
amp design
To tell you the truth I dont know where it originated from, it was in one of my old books from my school days about 15 years ago, I drew it, but dont remember why, but it looks pretty decent. So I thought I might throw it together just been working out semiconductor choices.
Yes there is a connection from R3 to the juntion of R12 and C4. looks to provide a error correction for DC offset.
To tell you the truth I dont know where it originated from, it was in one of my old books from my school days about 15 years ago, I drew it, but dont remember why, but it looks pretty decent. So I thought I might throw it together just been working out semiconductor choices.
Yes there is a connection from R3 to the juntion of R12 and C4. looks to provide a error correction for DC offset.
Hi,
if this is an intented connection, it makes R9 absolutely useless and will serve a current of about 450mA through the base of Q7 and R14 to the negative rail.
This would in addition clamp the collector of Q7 near ground and the amp won´t work. The collector of Q7 must be able to swing close to the negative rail.
It has to be a misdrawing ( or the amp is designed for communication by smoke ).
Onra
if this is an intented connection, it makes R9 absolutely useless and will serve a current of about 450mA through the base of Q7 and R14 to the negative rail.
This would in addition clamp the collector of Q7 near ground and the amp won´t work. The collector of Q7 must be able to swing close to the negative rail.
It has to be a misdrawing ( or the amp is designed for communication by smoke ).
Onra
Hello! Let me ask, for C5 (I believe it is your dominant pole capacitor):
I have seen several fully complementary amplifiers use dominant pole capacitors around both transistors of the VAS stage (so in your case, it would be something like a small capacitor from collector to base of Q5 and of Q7). Have you considered this?
I'm just suggesting because I have never seen the dominant pole cap placed from output of VAS stage to feedback input. Are there advantages to this method?
I have seen several fully complementary amplifiers use dominant pole capacitors around both transistors of the VAS stage (so in your case, it would be something like a small capacitor from collector to base of Q5 and of Q7). Have you considered this?
I'm just suggesting because I have never seen the dominant pole cap placed from output of VAS stage to feedback input. Are there advantages to this method?
Miller feedback. Miller compensation is where the cap goes from collector to base of VAS transistors....I think
But yes it has to do with high freq. phase compensation stability and to inhibit occilation. I have found that using a regulated power supply, you may not even need it because it does affect high freq. responce. I guess PS noise can cause instability. (along with a lot of long other junk)
But yes it has to do with high freq. phase compensation stability and to inhibit occilation. I have found that using a regulated power supply, you may not even need it because it does affect high freq. responce. I guess PS noise can cause instability. (along with a lot of long other junk) a 5pair output stage with Re=0r1 will require an optimum ClassAB bias of ~20mVre.
This will result in 1A of output bias current and give a maximum ClassA output current of ~2A. Equivalent to 16W of ClassA power into 8r0 and transitions optimally to ClassAB for higher power. Technically this is a ClassAB amplifier when used with higher voltage supply rails. At lower levels it stays in ClassA for almost all music.
If you want 20W of ClassA bias then add a 6th output pair. Vre=18.7mV gives 20W into 8r0 before transitioning to ClassAB. If the supply rails are reduced to +-20Vdc to +-21Vdc while the bias is set to 1.12A then it becomes a full ClassA amplifier.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.