juma said:
How stable is Iq with U4 not working (PNP with collector more positive than emitter)
Aha! You noticed...I made that mistake early on but left it in the schematic to see if anyone was actually paying attention.
Seriously.
It uses the 2SC3503 for Vbe and tracks fine. I'll post an up to date schematic just as soon as I get this computer straightened out (my screen capture utility - MS Paint is missing)
Wavebourn said:
The schematic in post #1 does not have the bell and whistles: bypass, decoupling or RC filters and neither does the prototype - just the basics. The final design will have it all (eventually...time, never enough).
kipman725 said:
Simulated THD is low and should give a good indication of the real thing (it has in the past with other amps).
Besides, add GFB and it becomes just another slightly different amp.
Member
Joined 2009
Paid Member
Am I understanding this amp ?
It is single ended at the front, but not at the output. It's Class A all the way through. With no global nfb the use of an LTP offers no benefit. The nice thing about dumping the LTP is that also gets away from dominant H3 that I see popping up in my sims.
But simple single ended front end usually aren't too linear. So you added a CCS to help out there. You added a common emitter amplifier to the input device to reduce the loading on it, thus providing a phase change and a simple chance to put some -ve feedback in place which does wonders for the linearity. The VAS is not bootstrapped, but linearized with a CCS too.
Why does Rick call it a differential transconductance - I can't see the "differential" part ?
It is single ended at the front, but not at the output. It's Class A all the way through. With no global nfb the use of an LTP offers no benefit. The nice thing about dumping the LTP is that also gets away from dominant H3 that I see popping up in my sims.
But simple single ended front end usually aren't too linear. So you added a CCS to help out there. You added a common emitter amplifier to the input device to reduce the loading on it, thus providing a phase change and a simple chance to put some -ve feedback in place which does wonders for the linearity. The VAS is not bootstrapped, but linearized with a CCS too.
Why does Rick call it a differential transconductance - I can't see the "differential" part ?
Member
Joined 2009
Paid Member
Bigun,
Here's the story on the "differential transconductance amplifier". I'll try to explain, but think about it before you say I'm crazy.
In the discussion that follows, the voltages and currents I'm talking about are AC values. Assume the base of Q1 is the + input and the emitter is the - input of the amp. If the base to emitter voltage of Q1 increases (slightly), then the collector current of Q1 increases. This causes the currents through Q2 to increase. In turn, the voltage across R13 increases as does the voltage across R21. The (AC) currents through R13 and R21 are approximately the same, and the voltage at the emitter of Q1 is nearly the same as the voltage at the base. Note that this assumes a very high gain transistor, but it is a reasonable approximation for discussion purposes. Therefore, (Vc2-Ve1)/R13=Ve1/R21. After a little algebra, we get Vc2/Ve1=(R13+R21)/R21.
Rick
Here's the story on the "differential transconductance amplifier". I'll try to explain, but think about it before you say I'm crazy.
In the discussion that follows, the voltages and currents I'm talking about are AC values. Assume the base of Q1 is the + input and the emitter is the - input of the amp. If the base to emitter voltage of Q1 increases (slightly), then the collector current of Q1 increases. This causes the currents through Q2 to increase. In turn, the voltage across R13 increases as does the voltage across R21. The (AC) currents through R13 and R21 are approximately the same, and the voltage at the emitter of Q1 is nearly the same as the voltage at the base. Note that this assumes a very high gain transistor, but it is a reasonable approximation for discussion purposes. Therefore, (Vc2-Ve1)/R13=Ve1/R21. After a little algebra, we get Vc2/Ve1=(R13+R21)/R21.
Rick
Member
Joined 2009
Paid Member
Rick,
Now I know what you mean. I had thought you meant something else, as you had said "In effect, Q1, Q2, Q3 and the associated components form a differential input transconductance amplifier" and I was off thinking that you thought Q3 was involved in a capacity other than a current source for Q1 and Q2.
I was thinking 'local negative feedback' and not 'differential' amplifier. I've been looking at LTPs for too long !!!
We're on the same page now.
Now I know what you mean. I had thought you meant something else, as you had said "In effect, Q1, Q2, Q3 and the associated components form a differential input transconductance amplifier" and I was off thinking that you thought Q3 was involved in a capacity other than a current source for Q1 and Q2.
I was thinking 'local negative feedback' and not 'differential' amplifier. I've been looking at LTPs for too long !!!
We're on the same page now.
Wavebourn said:
I don't have anything against (global) feedback. I just wanted to do this amp without it. Seems to work well.
Final schematic for now - could be how I leave it:
I spent a bit of time playing with different output stage configurations without appreciable improvement. I think simple maybe best.
Member
Joined 2009
Paid Member
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.