The resistors are with 5percent tolerance and were quite closely matched, Need i bother over the variation? Or is it fine to run as such? Also, I did notice the Grid input resistor is different (1k and 10k) in the Triode section., Where as in Gingertube version,both are 1k.. Why is this different? Is it okay to replace the 10k to 1k, ( I don't use Global Feedback)
It is ok indeed. If you don't use gnfb, you can also try to swap the two primaries of the output transformer:
the unbalance could also be a difference in Rdc of one of the two sides of the output transformer.
In any case, we are talking about 30 uA difference between the two sides.
You can balance it with a trimmer on the cathodes, or just forget it and enjoy your new amp.
the unbalance could also be a difference in Rdc of one of the two sides of the output transformer.
In any case, we are talking about 30 uA difference between the two sides.
You can balance it with a trimmer on the cathodes, or just forget it and enjoy your new amp.
Thankyou,Very helpful.. One more question...I do notice some faint noise( no hum) when I rotate the volume potentiometer ( 100k logarithmic) The noise is completely absent in fully open and fully closed position.. Midway ,I notice the noise..- This is noticeable only when tracks are not played and paused...
I tried Input to ground and the amplifier becomes completely silent.. Do i suspect potentiometer issue or ground loop? This is a new alpha potentiometer of reasonably Good quality..Is there Dc passing? shall I use a input capacitor?
I tried Input to ground and the amplifier becomes completely silent.. Do i suspect potentiometer issue or ground loop? This is a new alpha potentiometer of reasonably Good quality..Is there Dc passing? shall I use a input capacitor?
Maybe the wiring of the pot? The source goes to the fixed 100K track, so a constant source impedance is presented to the input, and the wiper goes to the first stage.
I built the same amp with some of the cheapest transformers from Ask Jan First, and it performs very well. I had 5 ECL86s and I had to find the best 4 and make the best pairs to keep the voltages in step. Since there is a CCS then I don't think a bit of imbalance matters. I built the next iteration from Marc (bandol83) which has a slightly different CCS and includes a balance trimmer.
Yves, the originator of the design, made these comments ...
An amp like no other...
I built the same amp with some of the cheapest transformers from Ask Jan First, and it performs very well. I had 5 ECL86s and I had to find the best 4 and make the best pairs to keep the voltages in step. Since there is a CCS then I don't think a bit of imbalance matters. I built the next iteration from Marc (bandol83) which has a slightly different CCS and includes a balance trimmer.
Yves, the originator of the design, made these comments ...
An amp like no other...
... if the potentiometer in the cathodes of the phase shifter allows effectively to balance the static anode voltages, this can only degrade its dynamic balance.
Anyway, the counter-reaction hides all that under the rug.
The value of the volume potentiometer is limited and causes (a little) Miller effect that the counter reaction cannot correct, I got trapped too.
With "modern" sources you can go down to 20K without problems.
On the other hand I applaud the current source with bipolar transistors.
Thankyou, This clears the issue very well...I will change the potentiometer to 10k or 20k and try...I should also mention that, By grounding the body of the pot to star Ground,the noise reduced.. However,That makes Star Ground to Chasis ground connection now to be in two places..( since i connect Star Ground to Chasis ground through anti parallel diode) But many sources say that this should be the only connection between star Ground and chasis ground.. Any inputs?
I built the Pcl86 Yvesm version, with Jfet CCS. The difference in plate voltage in the Triode input section is 30 volts....Now, Do i need to change to Ian's Two Transistor Ccs and include a 1k trimmer to achieve equal DC voltage on triode plates? Is a difference of 30 volt across both input triode normal?
I was thinking other options to try not to affect dynamic balance.
Without gnfb we could reference the second grid to a slightly negative (if that triode is 30V below) or positive (if that triode is 30V above) voltage.
This should affect the dynamic balance less than adding a resistor (so local current feedback) to the cathode of the triode.
Without gnfb we could reference the second grid to a slightly negative (if that triode is 30V below) or positive (if that triode is 30V above) voltage.
This should affect the dynamic balance less than adding a resistor (so local current feedback) to the cathode of the triode.
Thankyou. I will add the Trimmer and update. For clarity- 30 volt is voltage difference across anodes ( triode anode 1-102 volt triode anode 2- 72 volt.).
I like the way it sounds,but i feel there is more midrange shout and something is not right about the timbral quality ( previously used a SE amplifier with PL504 and i absolutely loved it). Given the rave reviews for the PCL 86 in BabyHuey topology,I belive Something is missing and want to make sure everything works right
General rule for RC coupling cap:
Before the tubes warm up, B+ rises to a much higher voltage than when the tubes are warmed up.
Whatever the B+ is Before the tubes warm up, then . . .
usually one end of the coupling cap will be connected to the driver plate load resistor (that resistors other end is connected to B+), and the other end of the cap is tied to:
Either the grid resistor to ground (0V),
Or the grid resistor is tied to the negative grid bias supply (EL84: -7V, -10V; 2A3: -45V, etc.).
Suppose when the amplifier is turned on, the 350V B+ rises to 400V, and the EL84 grid bias is at -10V.
Then, the coupling cap sees +400V, plate resistor, coupling cap, grid resistor, -10V bias.
That is probably a bad example, but could be true, if so, the coupling cap sees + 400V - (-10V) differential = 410V across the cap.
I use choke input B+, and solid state rectifiers, with operating B+ at about +300V; but Before the tubes warm up, they go to about 390V.
I use 600V rated coupling caps.
Check what your amplifier circuit does, especially at power-up, to see if your coupling caps might be stressed to overvoltage.
That goes for 45, 2A3, 300B, EL84, 6L6GC, KT66, etc. amplifiers.
check the coupling cap voltage ratings, and do not forget to check the B+ filter cap voltage ratings too.
Designer Beware (I do not know the Latin term for that one).
Before the tubes warm up, B+ rises to a much higher voltage than when the tubes are warmed up.
Whatever the B+ is Before the tubes warm up, then . . .
usually one end of the coupling cap will be connected to the driver plate load resistor (that resistors other end is connected to B+), and the other end of the cap is tied to:
Either the grid resistor to ground (0V),
Or the grid resistor is tied to the negative grid bias supply (EL84: -7V, -10V; 2A3: -45V, etc.).
Suppose when the amplifier is turned on, the 350V B+ rises to 400V, and the EL84 grid bias is at -10V.
Then, the coupling cap sees +400V, plate resistor, coupling cap, grid resistor, -10V bias.
That is probably a bad example, but could be true, if so, the coupling cap sees + 400V - (-10V) differential = 410V across the cap.
I use choke input B+, and solid state rectifiers, with operating B+ at about +300V; but Before the tubes warm up, they go to about 390V.
I use 600V rated coupling caps.
Check what your amplifier circuit does, especially at power-up, to see if your coupling caps might be stressed to overvoltage.
That goes for 45, 2A3, 300B, EL84, 6L6GC, KT66, etc. amplifiers.
check the coupling cap voltage ratings, and do not forget to check the B+ filter cap voltage ratings too.
Designer Beware (I do not know the Latin term for that one).
I did attach a trimer and tried to balance the anode voltages.. I did not notice any sonic differences and I went back to original schematic.. In a post, Yvesm has mentioned that Trying to equalize the voltages can do harm than good as he says that difference will induce only Even harmonic distortion.. However, I did manage to Tame the midrange glare, I just added the 2k2 grid stopper to the pentode G1. after the coupling cap. Ginger tube version has it, but Yvesm version does not have one... I think that Grid stopper is really required, for optimal sound.. I also use individual bias pots to set the Current through tubes identical.. I added 10ohm resistor in series to measure the current draw... surprisingly, Removing the 10ohm and connecting straight to ground, Kills the sound. So,the 10ohm is retained there... I always thought this resistor does not have anything to do with the sound.. May be there is a feedback thorough it?