That 35 watt rating was one channel driven, one channel idling. Running both channels, you'll get twice as much voltage sag over the single 5AR4 and choke than you would in a Mark IV.
Yes, but what if I make two sections of RC filters supplied from common transformer winding?
To be clear: I want to design my own PCBs and power supply for this amplifier and not use ready-made ones, so I can change something a little bit. I am even wondering whether to stabilize the voltage of the first and second stage, and the bias voltage.
To be clear: I want to design my own PCBs and power supply for this amplifier and not use ready-made ones, so I can change something a little bit. I am even wondering whether to stabilize the voltage of the first and second stage, and the bias voltage.
Last edited:
Ok, I thought you were comparing the stock amps. Anything you can do to reduce power supply sag will be an improvement. That could be a bigger power transformer, solid state rectification, lower resistance choke or resistor in the supply. Bigger power supply caps give you lower ripple and lower impedance in bass frequencies. Splitting the supply for two channels will also help because the current will be halved and less interaction between channels. As you approach theoretical perfection, all these things make a smaller difference. Duncan amps PSU designer is great for seeing this, after you get a supply drawn up, you can simulate a current jump to see how the B+ voltage reacts. I usually use something like a 50% increase in current over idle. With a good enough power supply, a stereo amp should perform similarly to two mono amps, but in the case of the original Stereo 70, the power supply was compromised for cost in comparison to their mono amps.
Ok, so let's have a look at this ST70 circuit: http://tubezone.net/pdf/diy-st70-manual.pdf
1st stage should be supplied with stable 300VDC, and the 2nd with 350VDC, right? I won't be stabilize B+ bacause of current.
And once again: can I remove DC balance? (R29/R30 and R27/R28) And connect EL34's cathodes together and to ground via 10R for bias current control?
1st stage should be supplied with stable 300VDC, and the 2nd with 350VDC, right? I won't be stabilize B+ bacause of current.
And once again: can I remove DC balance? (R29/R30 and R27/R28) And connect EL34's cathodes together and to ground via 10R for bias current control?
Those are unrelated questions to what I thought we were discussing, namely the power tubes' supply.
But If I read the manual correctly, those are the voltages for the 1st & 2nd stages yes. You could remove the DC balance, but personally I think it's a good feature, supposedly distortion in the output transformer is reduced by having the current match as close as possible.
If we are talking about Dynaco, there is no need for DC balance as Dynaco uses a concertina phase splitter
The above uses a LTP which is inherently AC unbalanded ( usually compensated with AC balande pot)
DC balance is most likely taken care of with the ccs in the cathode circuit.
I found something like this: https://www.audioamp.eu/_obchody/te...o-st-70-ccs-vcetne-modulu-autobias-49d594.pdf
I think I can do it in this way. I also noticed that there is B+ lower about 50V.
Also the description on its website mentions about some LED that shows when this autobias cannot set the bias current at required level, but I don't see any LED at the schematic :/
I think I can do it in this way. I also noticed that there is B+ lower about 50V.
Also the description on its website mentions about some LED that shows when this autobias cannot set the bias current at required level, but I don't see any LED at the schematic :/
Last edited:
So an LTP with a CCS is AC balanced as the delta current going up one leg is the negative of the other as no current goes flow though the grids. The DC balance may be off as the valves are not matched. The can be corrected with a trim pot in the cathodes. Degenerating the cathodes also reduces gains if you have too much.
You mentioned connecting the EL34 cathodes together. It better to have separate cathode resistors so you can measure bias current in each. It's also important they are not too watty. You want them to blow and go open if the valve develops a short rather than damaging the OPT. It also better to have separate bias pots as these valves are often not as well matched as you might hope. This can affect the 2'nd harmonic distorsion.
Well, I have not enought knowledge to build my own amp, that's why I'm looking for something like Dynaco. But I'm going to design my own PCBs so I can make some changes in the source project.
To be honest, the best for me would be if the amp sets bias and everything by itself, I'm lazy 😀
It is with a resistor tail, but with a CCS the delta current in one arm is exactly the opposite of the other as the total has to be equal to the CCS. No current goes to the grids so if the plate resistors are equal the delta voltages also balance.
That DC balance pot is for balancing the current in the output tubes, not the phase splitter.
I'm wondering how the automatic bias circuit should work? Maybe I could make some opamp measuring the bias current and sets the bias voltage related to measured value? Sounds simple, but this current will change with audio signal, will it?
What “automatic bias circuit” are you talking about? Something like this?
https://www.audioamp.eu/kategorie/modules-adaptive-bias-control/
https://www.audioamp.eu/kategorie/modules-adaptive-bias-control/
Yes, but I don't want to mount separate PCB with this circuit, but I'm afraid I won't find a schematic of this module.
BTW - I thought maybe doing some good transistor amplifier with built-in tube preamp will be better? It's of course not real tube amp, but what do you think about this idea?
BTW - I thought maybe doing some good transistor amplifier with built-in tube preamp will be better? It's of course not real tube amp, but what do you think about this idea?
Looks intresting, but can you explain the wiring for me? I mean, I see there are 4 channels, each for one tube. VBIAS is connected with grids (via resistors, of course)? So why is it connected to the voltage divider (R5/R6)?
J5 is connected to the catode? So what's J4? And J3?
J5 is connected to the catode? So what's J4? And J3?