Looking much more like a single pole response now. Good stuff. Try right-clicking on the V(b)/V(a) expression at the top of the graph, and putting a minus sign in front of it in the expression editor dialog. Then click OK. Left-click on the -V(b)/V(a) expression. That will activate the measurement cursor. Move it until you find the frequency at which the magnitude is as close to 0 dB as possible. What is this frequency? And what is the phase shift at that frequency?
Line up, there are 50 volts across R3, even when I reduce it to
8.2 k I get 4+ma in each differential , so removing it would
fry my LTP.!!!
I see your point, as you would have to modify the CCS to
take the place of R3...
8.2 k I get 4+ma in each differential , so removing it would
fry my LTP.!!!
I see your point, as you would have to modify the CCS to
take the place of R3...
ostripper said:
Looks like the amp is a bit overcompensated.
Try finding the frequency at which the phase shift of the loop gain is -100 deg. This will be a frequency larger than 344 kHz. That's what you want to be your new unity loop gain frequency. To increase the unity loop gain frequency, you can decrease the input stage emitter degeneration. Try decreasing it in the same ratio that you're increasing the unity loop gain frequency. That is, if the new unity loop gain frequency is e.g. 500 kHz, the new input stage emitter degeneration resistor should be 344/500 times the old one. Replace 500 by the actual frequency in kHz at which the phase of the loop gain is -100 deg.
50 volts, so what
the ccs transistor (BC546) can take 60 volt,
so taking 50 volts from R3 and let BC546 have it
is no problem
And if you need higher, you can use 2N5551 (160 Volt C-E) or some other
for constant current source transistor
the ccs transistor (BC546) can take 60 volt,
so taking 50 volts from R3 and let BC546 have it
is no problem
And if you need higher, you can use 2N5551 (160 Volt C-E) or some other
for constant current source transistor
To change CCS current in input you use R8, 220 Ohm
Increase R8 will lower current in LTP
Decrease R8 will give higher current in LTP
get away with R3, as D.Self says: it was a mistake (bad move) 🙂
Increase R8 will lower current in LTP
Decrease R8 will give higher current in LTP
get away with R3, as D.Self says: it was a mistake (bad move) 🙂
I'm confused. At 536 kHz, the gain should be less than 0 dB right? Which means you'd like to decrease the emitter resistors in the input stage to get it back to 0 dB there. Or did you already tweak the input stage emitter degeneration?
Mr. self has said other things ,too but do we always listen to him?? he said a balanced VAS was too much trouble, but I'm
not listening.
One thing at a time. your idea is good but the load on my LPT
is in the uA's (EF'ed VAS's)balanced as well. With a different amp, like frugal 1, I will try this.
BTW , I figured out how to import .models and made my own
LTspice libraries (custom)..thanks
OS
Great! Looks like you've really made some great progress here. I'm signing off for tonight, but if you're around tomorrow, maybe you'd be interested in playing with transient response in SPICE. With that tool, you can look at square waves and such, and verify that you have no overshoot or other problems which sometimes only show up in transient.
Thank you andy, for patience. Are you a teacher.? You have a
very professional and concise way of getting your point
across.🙂
OS
very professional and concise way of getting your point
across.🙂
OS
I am semi-retired at the moment, but have thought of getting into teaching for some supplementary income. I did do some teaching when I was a grad student back around 1980 and found it very rewarding.
Thanks for your kind words. I do appreciate it.
Thanks for your kind words. I do appreciate it.
ostripper said:
I have to agree, the info generated in this thread is thoroughly interesting, especially Andy's input
Great thread 😉
Soon I will build this (next week). Gives a real sense of satisfaction to start at conception and go to listening
in just a week.
Stay tuned, because I will go all out to show how to make a
near toner tranfer quality board with good layout the "frugal way"
(no computer involved).
in just a week.
Stay tuned, because I will go all out to show how to make a
near toner tranfer quality board with good layout the "frugal way"
(no computer involved).
Your not the only one ,Hugh. This amp is tricky...
First I wanted to keep the LTP very balanced..got that..
R17/13 do that.
(blue and green lines)
1/1000th of a MA... but that threw the unity gain point UP(1mhz)
R17 also changes OLG. yikes!! 😱
but with andy's pointers, changing Cdom and LPT degen. Resistors brought it back in line.
All this tinkering really smoothed out the slope as well as
creating real balance on both the LTP and VAS.🙂
First I wanted to keep the LTP very balanced..got that..
R17/13 do that.
An externally hosted image should be here but it was not working when we last tested it.
(blue and green lines)
1/1000th of a MA... but that threw the unity gain point UP(1mhz)
R17 also changes OLG. yikes!! 😱
but with andy's pointers, changing Cdom and LPT degen. Resistors brought it back in line.
An externally hosted image should be here but it was not working when we last tested it.
All this tinkering really smoothed out the slope as well as
creating real balance on both the LTP and VAS.🙂
lineup said:
Lineup
In the SCULD2 schematic that you posted, you will see that >270mW would be dissipated in the BC556 without that resistor. The BC556 is rated for 500mW maximum dissipation.
I believe that this would detract from the CS stability,unless at the very least it was fitted with a heatsink clip.
SandyK
Sandy, I see what you mean.
But in my book, 270 mW is not too much for 500 mW rated TO-92
it is a high level, but in no way critical
50-60% of max is msot times alright.
I personaly can think of running at 350 mW .. but preferably not more.
Why I posted that schematic, is because Douglas Self refered to it,
as to show, that copycats of his schematics, whithout their own knowledge,
will copy even when the original designer made a mistake.
And in this case D. Self called his including of that resistor:
a mistake (bad move).
And I also never have used such resistors from the LTP CCS.
Instead, if we need, we will use one transistor,
that suits our amplifier requirements.
Regards
Lineup
But in my book, 270 mW is not too much for 500 mW rated TO-92
it is a high level, but in no way critical
50-60% of max is msot times alright.
I personaly can think of running at 350 mW .. but preferably not more.
Why I posted that schematic, is because Douglas Self refered to it,
as to show, that copycats of his schematics, whithout their own knowledge,
will copy even when the original designer made a mistake.
And in this case D. Self called his including of that resistor:
a mistake (bad move).
And I also never have used such resistors from the LTP CCS.
Instead, if we need, we will use one transistor,
that suits our amplifier requirements.
Regards
Lineup
- Status
- Not open for further replies.