What is the measured voltage swing peak to peak on steady tone 1KHz needed to bring a push-pull 6550/KT88 pair of tubes in UL connection with A431 output transformer?
Power supply B+ is 500Vdc
Each tube bias at 50mA
Question arised from measurement of a mono bloc Dynaco MK 3 and pushing only 40 Volts peak to peak at onset of visible clipping, on 8 ohms tap with a resistor.
Should go near 60 V p-p ??
Driver stage is a Poseidon with new tubes 12AX7, 12BH7
Outputs clips before the driver stage shows clipping measured at the input 6550 KT88 grid.
Thanks.
Power supply B+ is 500Vdc
Each tube bias at 50mA
Question arised from measurement of a mono bloc Dynaco MK 3 and pushing only 40 Volts peak to peak at onset of visible clipping, on 8 ohms tap with a resistor.
Should go near 60 V p-p ??
Driver stage is a Poseidon with new tubes 12AX7, 12BH7
Outputs clips before the driver stage shows clipping measured at the input 6550 KT88 grid.
Thanks.
40V P-P into the 8 ohm tap with an 8 ohm resistor is 14.1V RMS, which is 25W into 8 ohms.
Without an FFT or posting a photo of where you are deeming the amp to be clipping, it could be tough to say if there's anything wrong with the amp.
Yes, you should see 60V P-P, can you put enough signal into the amp to get there?
Without an FFT or posting a photo of where you are deeming the amp to be clipping, it could be tough to say if there's anything wrong with the amp.
Yes, you should see 60V P-P, can you put enough signal into the amp to get there?
Dynaco MKIII standard bias is (1.56/11.2)/2 or 70mA. You are short on bias hence less power at clipping.
KT88 Va=Vg2=500V and Ik=50mA means Vg1=-65V (can be found in the triode mode graphic)
To get at max Ia current the g1 has to raise to 0V, so input Vt=65V or 130Vtt.
Mona
To get at max Ia current the g1 has to raise to 0V, so input Vt=65V or 130Vtt.
Mona
Dynaco MK 3 output power 6550 or KT120
Ok some more information, either 6550 tubes or KT120 tubes, stetting the bias to lower operational class from AB to B, did not changed anything, i was thinking of getting max power just for testing purpose, but this did not proved....
So a pair of 6550 measured with scope to check onset of clipping points.
Input tube is a 12AX7 capacitor couple to the splitter.
My schimtt phase splitter 12BH7A have both phases equals and out of phase of course, I get 70 vp.p. at 400Hz, per leg.
This same drives gives 410vp.p. at the primary on plates to end up with
40v p.p. on the 8 ohms tap loaded with a dummy load of 8 ohms resistor.
Both 70v p.p. and 410v p.p. tracks exactly at clipping points on scope.
Freq. response is very good in the range of 30Hz to 40KHz plus.
Square wave response does have typical slight overshoot on 1KHz, 10KHz.
Same results with KT120 output tubes.
I will push bias to the 70mA, I was at 40mA and 50 mA for both tubes, did not changed.
My B+ is 500Vdc , with 5AR4 rectifier, Pi filter CLC.
Ok some more information, either 6550 tubes or KT120 tubes, stetting the bias to lower operational class from AB to B, did not changed anything, i was thinking of getting max power just for testing purpose, but this did not proved....
So a pair of 6550 measured with scope to check onset of clipping points.
Input tube is a 12AX7 capacitor couple to the splitter.
My schimtt phase splitter 12BH7A have both phases equals and out of phase of course, I get 70 vp.p. at 400Hz, per leg.
This same drives gives 410vp.p. at the primary on plates to end up with
40v p.p. on the 8 ohms tap loaded with a dummy load of 8 ohms resistor.
Both 70v p.p. and 410v p.p. tracks exactly at clipping points on scope.
Freq. response is very good in the range of 30Hz to 40KHz plus.
Square wave response does have typical slight overshoot on 1KHz, 10KHz.
Same results with KT120 output tubes.
I will push bias to the 70mA, I was at 40mA and 50 mA for both tubes, did not changed.
My B+ is 500Vdc , with 5AR4 rectifier, Pi filter CLC.
All measures with a TEK 2245 scope are from ground of course.
I suspect that my driver stage does not push enough voltage?
That's why I asked.
Ultimate test, use original driver PCboard with the 7199 tube.
I did a Poseidon version driver board with 12AX7 and 12BH7A point to point wiring without the LM134 current source IC.
I bought this pair in fair condition knowing all caps and other parts needed to be replaced.
Even damm original Dynaco power transformers had to be replaced(short primary) with new Hammond's 378X , with 237VA ...
I suspect that my driver stage does not push enough voltage?
That's why I asked.
Ultimate test, use original driver PCboard with the 7199 tube.
I did a Poseidon version driver board with 12AX7 and 12BH7A point to point wiring without the LM134 current source IC.
I bought this pair in fair condition knowing all caps and other parts needed to be replaced.
Even damm original Dynaco power transformers had to be replaced(short primary) with new Hammond's 378X , with 237VA ...
70mA bias on tubes did not increased power.
Still think that i do not have enough gain in the 12AX7 ???
Still think that i do not have enough gain in the 12AX7 ???
More bias keeps the amp longer in classA and lowers the drive voltage (less -Vg1).
Increase power is done with higher supply voltage with appropriate Raa of the OPT.
Or driving the tubes into classB2.But then you have to get rid of the grid coupling capacitors as they get charged by the grid current.
Mona
A431 opt dynaco mk3
Found on the net:
Transformer specs are: 4300 ohms CT primary with screen taps, 4,8 and 16 ohm secondary. 20 Hz to 20 KHz response at 60W, within 1 db, 30 Hz to 15 KHz at 120 watts. Max DC per side (suggested) 120 mA.
Just trying to find if my design is able to push the 60 watts??
60v p.p at 8 ohms tap.
Poseidon design is not new, it is a current upgrade for a mk 3 which I did use their schematics and made a point to point input-driver circuit.
Otherwise do I have to doubt on OPT??
Found on the net:
Transformer specs are: 4300 ohms CT primary with screen taps, 4,8 and 16 ohm secondary. 20 Hz to 20 KHz response at 60W, within 1 db, 30 Hz to 15 KHz at 120 watts. Max DC per side (suggested) 120 mA.
Just trying to find if my design is able to push the 60 watts??
60v p.p at 8 ohms tap.
Poseidon design is not new, it is a current upgrade for a mk 3 which I did use their schematics and made a point to point input-driver circuit.
Otherwise do I have to doubt on OPT??
Why would you doubt the OPT?
Can you post a schematic of your design? I thought this was a Dynaco.
Can you post a schematic of your design? I thought this was a Dynaco.
Hi, yes it is Dynaco mk3, you may see my schematics on previous entry number 6.
Doubting OPT because , original power transformers were shot.
New Hammond PT , 378X in now.
Really do have A-431 installed, checked from size with new dynakit drop in replacements.
So, how much voltage swing is needed to push this amp to 60watts out?
Or 60V p.p. in 8 ohms??
Doubting OPT because , original power transformers were shot.
New Hammond PT , 378X in now.
Really do have A-431 installed, checked from size with new dynakit drop in replacements.
So, how much voltage swing is needed to push this amp to 60watts out?
Or 60V p.p. in 8 ohms??
Last edited:
If you put 1V of signal into the amp with an 8 ohm resistor at the output, how much AC voltage comes out of the amp?
60W out of 8Ω is about 22Veff sine or 31Vp.
To get that out of the transformer above, you need a supply of 450V with tubes in pentode.For UL 500V and as triode even more.
How much input voltage depends on the bias condition.
Mona
Looking at Dynaco MkIII , the amp uses UL, and it's output power 60w is documented,
this when plate and g2 voltages are 475Volt. Input sensitivity 1.6V for 60w out.
Output transformer used is A431
Thus 60w is possible with the authors hardware.
this when plate and g2 voltages are 475Volt. Input sensitivity 1.6V for 60w out.
Output transformer used is A431
Thus 60w is possible with the authors hardware.
220k as grid resistors are too large.
In addition, the grid resistors are usually connected not to ground but to
the negative bias source. your schematics ( dynacomk3-b.pdf) has
tvo sets of resistors loading down the phase inverter. Remove teh surplus one
and keep 100k , one side to grid and the other to negative bias.
Note that to follow spec this resistor should not be any larger then 51k
if 6550 is used, here is one point where dynaco actually cheated !!
Gain
To audiowize, yes good idea to measure gain.
I used 1 volt peak to peak at 400Hz with a scope.
Amp is warm with +500V
70mA per output tubes 6550
1volt peak to peak result :
Output loaded with 8 ohms load gives 15 v p.p.
12AX7 plate gives 7 v p.p.
12BH7A leg1 and 2 gives 25V p.p. each.
Plates of 6550's are at 200V p.p. each.
Measured THD is 0,1% with 1 v p.p.
To audiowize, yes good idea to measure gain.
I used 1 volt peak to peak at 400Hz with a scope.
Amp is warm with +500V
70mA per output tubes 6550
1volt peak to peak result :
Output loaded with 8 ohms load gives 15 v p.p.
12AX7 plate gives 7 v p.p.
12BH7A leg1 and 2 gives 25V p.p. each.
Plates of 6550's are at 200V p.p. each.
Measured THD is 0,1% with 1 v p.p.
To Ketje, yes i will mod this at 100Kohms.
Will test and report
Thanks to all helping me and submitting ideas, much appreciated.
Will test and report
Thanks to all helping me and submitting ideas, much appreciated.