|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
50.80.146.114
In Reply to: RE: Opinions please: need recommendations for a Triode driver for a 2a3 posted by dave slagle on January 17, 2017 at 04:59:57
Dave,
I forgot to mention the following in my implementation of triode mode with the 6C6 tubes. The output is low. I have to turn up the volume control all the way to get good output. That said, I am still using 2.6V on the bias for the 6C6. I believe it should have more, correct me if I am wrong.
Pat
Follow Ups:
OK..
My pre-amp has 27 db / 17.4V RMS output.
That opens up the window to look at tubes with lower gain.
My pre-amp has 27 db / 17.4V RMS output. The pre-amp also has an internal setting whereby I can reduce the output by 6db if necessary. The pre-amp's manual suggests running the output at the lower output level.
Hmmm... if you are only biased at 2.4V and turning it all the way up gives you closer to 15V something is not adding up. Lets assume you get a gain of 15 from your triode 6C6, that menas you need 3V in to get to the 45V bias of the 2A3 but since you are only biased at 2.4V this means your driver will actually hit A2 before it clips the output. This tells me you need more bias and a third battery will start you in the right direction. It doesn't explain the difference between the alleged preamp out and the amp input bias voltage unless your sources only put out a quarter of a volt or less.
I have been using a battery bias on my 6c6 tubes. If possible, I would like to continue to use the battery bias on the new driver tube.
Getting back to the assumption that you can drive this with a clean 10V source. This suggests you can look at the lower gain tubes and possibly even a DHT but as your mu goes down your bias will increase and at some point you *may* find that the stack of batteries becomes a sonic liability rather than an asset. (I found getting over 5V of bias with batteries or LED's to lose the magic and that a tradiitonal RC combo to sound better ymmv)
I would rewire the amp for a 6J5 and see where things end up and then you can explore getting some more bias. As a quickie with the 6C6 you could try lowering the plate resistor. The schematic says 75K and I would cut it to 30-40K so you can keep the current up as you increase bias.
dave
Dave, not sure how this affects our direction, but I have already lowered the plate resistor to a 27K. So, my current listening reflects the 27K resistor.
Pat
OK... how much current are you running?
going to a higher bias will reduce the current and at some point you will be running things too lean which means you need more voltage plate to cathode.
you could get a few more volts by replacing the 15K dropping resistors with a choke and of course a plate choke will put you in the other direction of having possibly having too much plate to cathode voltage.
dave
Dave, I was looking at the EML website. While pricey, the 20B seems to be a good option. The 20B has 20x amplification which is exactly what the 6C6 has in pentode mode. The recommended voltage is 380V, so not sure if I could run it with the existing power transformer which puts out only 325V. The load impedance is 12..20k ohms.
Here is the link to the specs. thoughts?
You have to realize that that 380V is plate to cathode and the 12K load is assumed to be a transformer (or plate choke). If it were resistively loaded at 12K and drawing 24ma of current the voltage drop across the resistor alone would be 288V meaning you would need a B+ in excess of 650V.
dave
Update as of 1/18/17:
I spent the day migrating the amp back to the 6C6 pentode driver. Listening carefully. Then migrating back to the 6C6 triode strapped. I can say with confidence that I prefer the 6C6 triode strapped. The noise floor is lower, which results in a much deeper soundstage. High and mid frequencies have greater texture and a smoother presentation. So, I want to continue on my quest to implement a triode driver for the 2a3 output tubes.
Thank you to those who have posted their input so far.
Here is where I stand and question that need to be answered:
1. The only issue with using the 6C6 triode strapped is that to get near full output, the preamp volume knob needs to be turned all the way up. With the preamp at full throttle, the output is about 90% of max output (just a guess, but close enough for the sake of this conversation). So, is there a way to get the 6C6 triode configuration to get more output from the power amp? If so, how?
2. The 20B driver tube has an amplification factor of 20. The 6C6 data sheet says that the 6C6 has an amplification factor of 20 in Triode mode, and does not list an amplification factor for pentode mode. So, is an amplification factor of 20 enough gain and therefore the 20B might be a suitable driver? Don't know, but your input would be appreciated. I have a near ready schematic of the 20B --> 2A3. Going with this solution requires more $s and modifications. The 20B --> 2A3 requires either a plate choke or a plate transformer that acts as an interstage transformer.
Ideally, we would find a solution to option (1) and if it works nicely, then I could move towards a more permanent solution (2) that leverages the choke and possible interstage transformer.
Any input would be appreciated.
Thanks,
Pat
1. The only issue with using the 6C6 triode strapped is that to get near full output, the preamp volume knob needs to be turned all the way up. With the preamp at full throttle, the output is about 90% of max output (just a guess, but close enough for the sake of this conversation). So, is there a way to get the 6C6 triode configuration to get more output from the power amp? If so, how?
If your preamp puts out 17V like you mentioned and your 6C6 is giving a gain of 15 (fair guess for a resistive loaded tube with a mu of 20) then something is seriously amiss if you can get anywhere near full volume on the linestage since 3V into the amp will run the 6C6 into A2 current and clip the output tube.
Your removal of the cathode bypass cap on the 2A3 may be what is "saving you" since the clipping behaviour looks a lot more polite without it in place. (even though the distortion is still upwards of 10%)
In any case, going from tube A with a µ of 20 to tube b with a similar µ isn't going to change your lack of gain.
dave
I took some measurements using a 1k sine wave input at 0.683VAC. FYI: instead of 100k ohms, there is a 220k ohm resistor at the RCA input (I have been playing with different input resistors, FWIW).
Here are the results:
Input signal: 0.683V
6C6
grid 145Vdc
plate 145Vdc (grid and plate are tied together for triode)
filament 4Vdc
2A3
grid 8.77Vac (using signal tone input)
plate 286Vdc
filament 2.6VAC
VDC P-K 237Vdc (target should be 250-265)
The 2A3 plate voltage is low at 286Vdc, typically I see it at 300Vdc.
The 6C6 grid voltage is 145Vdc, I typically see 175-180Vdc when in pentode mode.
I am unsure why the 2a3 plate voltage is so low and is the 6C6 grid 145V voltage OK since this is triode strapped?
Pat
see you upped the bias to 4V and the driver is giving you a gain of 12.8. This puts you at just over 50Vrms (70V peak) into the grid of a tube biased at 50V so everything looks to be in line and I don't think you will get much more out of this situation. (unbypassed 2A3 inot quad esl63's)
This doesn't explain why a preamp that puts out an alleged 17V needs to be turned up all the way since that should cause some pretty severe clipping to an input stage biased at 4V with a gain of 13
dave
9
Pin 4 and Pin 2 (plate) are tied together.
where are you measuring 4volts on the 6c6?
The filament supply is 4V. I am using 2 LEDs in series to produce the 4V.
The heater of the 6c6 should have 6.3 volts across it.
I assume the two LEDs are between the cathode and ground.
The 4 volts is your bias voltage (cathode) not you heater (filament) voltage.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: