• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Quad KT88 driver

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Previously known as kingden
Joined 2008
What is the best driver stage for a quad of KT88 tubes? I have been doing research and have come up with some ideas but largely drew a blank. Some of you probably have more experience in this matter. First off I have seen a circuit published in the 50's that uses a standard MKIII driver board with the quad.

[URL="www.the-planet.org/dynaco/Misc/Transformers.pdf"]www.the-planet.org/dynaco/Misc/Transformers.pdfwww.the-planet.org/dynaco/Misc/Transformers.pdf
[/URL]

The original Dynaco driver board does not look ideal but the Poseidon upgrade board looks more promising, as it sounds pretty good in my MKIII. My only question is it can it drive the capacitance and 50k ohms without incident?
 
hey-Hey!!!,
The general topology of the dynaco circuit is fine...but the implementation is not good. Consider a better pentode; 6EJ7 perhaps, followed by a 6AH4 for the split-load phase splitter. Run the pentode with 15-20k plate load and the 6AH4 with 10k loads. Set the pentode plate voltage at 20-22% of B+ available to the split-load.
cheers,
Douglas
 
Previously known as kingden
Joined 2008
Here is an amplifier I found on this site made by gingertube.

http://evatco.com.au/project_files/kt88.htm

http://evatco.com.au/project_files/dud120.jpg

http://evatco.com.au/project_files/dud120-2.jpg

Here are some changes I will make.

- Add a beefier B+ power supply by replacing the resistor with a regulator circuit.

- Replace the 6SN7s with E88CC's in the cathode follower. Is it better to use then the 6SN7s there?

- What about the Zobel network? This was designed to dampen an erratic ultrasonic ringing in the Hammond transformer (see The Set To Work description). Is it really important to correct that sort of anomaly? I am not equipped with a frequency sweep generator, so I how do I detect that erratic response if it is necessary?
 
Previously known as kingden
Joined 2008
Revision 2. Balanced

I recently completed my KT88 PPP amp. I am planning a revision 2. The new plan is to:

- Make the amp naturally more stable.

- Simplify the overall design.

- Introduced a balanced topology (with XLR inputs) to lower the noise floor.

What do you think of my overall new design? Anything you recommend I do different?

Spice projects a distortion figure of about 0.75% THD. This sound accurate?

Also, a 1:1 transformer on the input I heard helps reject noise better. What PC mount unit do you recommend?
 

Attachments

  • Output.png
    Output.png
    24.2 KB · Views: 272
  • Driver.png
    Driver.png
    17.6 KB · Views: 271
Maybe I'm just misunderstanding your question, but your circuit is already balanced. Hence, I don't get your line item of making it balanced.

You could just cap couple the two inputs to the (+) and (-) of the XLR connector and be in business (I think).

For input transformers, I like the Jensen Transformers. Specifically the JT-11P-1HPC would do the job nicely.

A differential or balanced input will not lower the noise floor of the amp, though. It will help in reducing the effects of EMI (electromagnetic interference) provided that the induced EMI is common mode on the input.

~Tom
 
What do you think of my overall balanced design? Anything you recommend I do different?

Spice projects a distortion figure of about 0.75% THD. This sound accurate?

Well, I did also rebuild my old Cary SLM-100 (auto bias to variable bias) who is similar to your project.

I planed for the second phase to go for XLR input while my Wadia 27 has XLR outputs.

I found some schematics on the internet who just use a differential input circuit, IMHO other words the simplest to have both symmetric or asymmetric and a phase inversion outputs. Then would use the driver (as in your current schematic) for the 2 * 2 KT88 or use a the phase shifter with symmetric connections. The question rises about the NFB ... what I would like to skip or connect them to phase shifter.

just my 2 cents to my planned project, what is in some regard very similar.
 
The accuracy of the spice models will determine how accurate your THD figure is. I use the MATLAB script that Norman Koren has on his website. It seems to produce models that are accurate for DC and AC operation below clipping - provided I take the time and care to read the numbers off the plate curves accurately. The THD behavior at clipping doesn't seem to be accurately modeled. So my spice sims tend to give slightly optimistic figures for max power out and such.

Your schematic looks like a brute-force approach. It looks like two half-circuits rather than a differential circuit now that I look closer. I would suggest using a differential pair (long-tailed pair) for the input stage. This would give you a differential input, common-mode rejection, and such. Of course an input transformer would do the same. The CMRR of the Jensen JT11 is hard to beat.

~Tom
 
Previously known as kingden
Joined 2008
I switched the driver to a differential stage. I am trying to apply the negative feedback. Is it possible to somehow inject feedback along with the audio input source into the same grid?

With an op-amp you can, but it is ineffective trying to set up the tube differential stage like an op-amp using global feedback.
 
I switched the driver to a differential stage. I am trying to apply the negative feedback. Is it possible to somehow inject feedback along with the audio input source into the same grid?

Sure. Same way as you would with an op-amp.

With an op-amp you can, but it is ineffective trying to set up the tube differential stage like an op-amp using global feedback.

What do you mean by "ineffective"? Converting the SRPP input stage you have to a diff pair would reduce the tube count by two tubes. Though, I wonder if you could build a combination. Sorta like a diff pair with active load. Way overkill for the application, probably, but it could be fun to try.

For an example of a differential amp with global negative feedback, have a look at Morgan Jones' "Crystal Palace" amp. It's featured in his book, "Valve Amplifiers" 3rd ed. I'm sure it'll be in the 4th edition as well, when that comes out that is.

~Tom
 
Previously known as kingden
Joined 2008
This is what I mean. I set it up the same way as you would a differential amplifier with an op-amp. With this circuit the FB is coming off of the 8 ohm transformer winding.

The gain is not reduced no matter what resistor combination I use. That is what I mean by ineffective.
 

Attachments

  • Untitled.png
    Untitled.png
    15.9 KB · Views: 120
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.