1 . Schema + "preferred BOM" .
2. Theory of operation ... with all the variables on some values
to fine tune for rail voltage or output currents. What should be
matched , all these little "things".
3. Pretty pictures (E -porno) , to give a visual "backup".
4. Objective actual operating readings with a little subjective input.
Bonsai does it best (what a marketer !) hifisonix.com: Audio Amplifier Design and Music Reviews
His "style" seems to be the way to go - excellent presentations !!
OS
2. Theory of operation ... with all the variables on some values
to fine tune for rail voltage or output currents. What should be
matched , all these little "things".
3. Pretty pictures (E -porno) , to give a visual "backup".
4. Objective actual operating readings with a little subjective input.
Bonsai does it best (what a marketer !) hifisonix.com: Audio Amplifier Design and Music Reviews
His "style" seems to be the way to go - excellent presentations !!
OS
Guys,
Could one of you test Kypton C and Slewmaster with +-30V PSU please?
I have a lab PSU with that voltage and with current limiter, because I will test with low PSU as possible before applying -+50V.
But I can only get a voltage measured directly on R22 + R27 at 0.17V, I have adjusted R18 +R19 to 1.7K. and R25 set to 25Kohm.
I take the feedback from emiters between Predrivers (Q105 and Q106)
Zeners are working at 12V and there are only 5mv DC offset measured from ground to input Q101 base or "temperary feedback point"
I measured close to zero voltage between PD and ND, why is that?
Anyone help?
Could one of you test Kypton C and Slewmaster with +-30V PSU please?
I have a lab PSU with that voltage and with current limiter, because I will test with low PSU as possible before applying -+50V.
But I can only get a voltage measured directly on R22 + R27 at 0.17V, I have adjusted R18 +R19 to 1.7K. and R25 set to 25Kohm.
I take the feedback from emiters between Predrivers (Q105 and Q106)
Zeners are working at 12V and there are only 5mv DC offset measured from ground to input Q101 base or "temperary feedback point"
I measured close to zero voltage between PD and ND, why is that?
Anyone help?
Last edited:
Kypton c are exept above built after schematic post 6323, measured 0.25V around all emitter resistors in CCS mirrors
Any inputs here?
Any inputs here?
Test the input separate from the output first. Put a 1k resistor from PD to the feedback connector and another 1k resistor from ND to the feedback connector. You should have around 11 volts between PD and ND and your DC offset measured at the feedback connector should be only a few mV. You can put a signal to input and read it at the feedback connector like this. Get your input board operating properly first before connecting to the output board. You should be getting 7.7mA across R22 if everything is working right too.
The output board can be tested by putting a 15k resistor from PD to V+ and another 15k resistor from ND to V-. Install 1 output pair and you should be able to set up your bias. I'm sure OS's feedback resistor scheme should work but I'm not familiar with it.
You should use a bulb limiter when you are testing. It's a lot easier to see a 50 watt bulb out of the corner of your eye than to watch an ammeter on a limited supply. As you turn up your bias you should be able to see the bulb starting to glow.
After both halves are operating properly on their own, put them together and start testing again.
Slewmaster board disconnected from Kypton C.
Measurents on Kypton C PCB:
I have mounted 1K from PD and ND back to feedback point between R15 and R16
Offset measured to -5V DC
Only 5mV between PD and ND
62mV DC around R23 and R26
-5V DC from Q1 emitter to ground.
-6V DC from Q2 emitter to ground.
0.19V around R22 + R27
2mV around R20 and R21
D7 + D8 lights with 2.68V around each.
What can I tjeck, please help?
I removed servo mounted in socket, measured offset 0.4V, Q2 emitter 0,27V, q1 emitter 1.1V, 0v Q1 and Q2 between collector to emitter.
Other measurements from last post still the same
Other measurements from last post still the same
Leave the servo out for now. I've found that R4 and R5 1 meg resistors need to be matched very closely for the servo to operate properly. That's likely what's causing the huge offset. 400mV is still kind of high but worry about that later.
You need to raise your VAS current. If you have 470R resistors in R20 and R21, try putting 680R in there and see what your voltage drop is across R22 and R27. What values do you have installed for R22 and R27?
You need to raise your VAS current. If you have 470R resistors in R20 and R21, try putting 680R in there and see what your voltage drop is across R22 and R27. What values do you have installed for R22 and R27?
R20 and 21 = 470R, R22 + R27 = 68R, r25 reduced to 20k. Measured about 50v around R25. I don't Think it do any change to raise r20+r21 as voltage are close to zero around r22 + R27 and base current IR23 and R26 are High as 2.6mA??
Thanks for your help🙂
Thanks for your help🙂
You need higher VAS current. If base current is already high on Q9 and Q10 try lowering R22 and R27 to 47R. The actual current on the base of Q9 and Q10 is set by the voltage drop of LEDs themself I think. Raising the value of R20 and R21 will increase emitter current on Q8 and Q11.
Last edited:
It did not make any change to set r22+r27 to 47R. But the current throught R25 equals current throught R22+R27, what does that mean?
Please help again anyone.
Are you sure you have the correct transistors in Q8 and Q11? It almost sounds like they aren't conducting.
That the cascode is not conducting and whatever current there is goes
to the leds.
R25 usually only conducts 1/5th the total R22 current.
Another scenario would be where the cascodes are not shunting through the
OPS Vbe,
OS
Q8 and 11 are corrects. I have the same error on both on my Kypton C boards, both using 2sc3503e together with KSA1381e in VAS stage so maybe this is the problem, I will try changing one vas transistor at time and then report back.
Try to change them to bd139/140, sometimes the faster KSA/KSC types can be hard to make stable.
Do what thimios/Jwilhelm did . Post pictures and voltage readings.
I have 2 working Kypton C's in front of me.
They have been rock solid with 38,65,73v rails. Even worked with
the reverse led and R6 screen errors.
It sounds like you are not biasing the VAS at all.
Nothing wrong with 3503/1381 E or D. I have one kypton with E/E , the
other with E/D. One just had 50mv more offset before I dropped the servo in. 😕
Since you reworked the PCB's your self , maybe an error. Photo of top/bottom
voltages .....
Edit - the 38V test required 2.2K (R18/19). This amp is so simple , it has to work -
some bad trace , or other physical error.
OS
Last edited:
With 3503/1381 , I even put 10pF comp. on mine for laughs - no oscillation.
This one might not even blow up with no comp. 😀
OS
If VBE on Q9 and Q10 look correct it's just a matter of raising R20/21 values and lowering R22/27 values. Voltage drop across Q8 and Q11 is less than the LEDs so the LEDs are controlling base current on on Q9. It's only operating at 30 volts so there is no current available there. Also Q9 is PNP, so raising the voltage at it's base will decrease base current so raise resistance at R25 back to it''s center and adjust your VAS current at the rails, not at your VAS driver gain.
Personally I would raise the test voltage if nothing is getting warm. You're going to need to alter the circuits to get it to operate at 30 volts, then go back the opposite way again to run at 50 volts. It's actually counterproductive.
Personally I would raise the test voltage if nothing is getting warm. You're going to need to alter the circuits to get it to operate at 30 volts, then go back the opposite way again to run at 50 volts. It's actually counterproductive.
Last edited: