• 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.

Vacuum Tube OTL power amp!!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I will try using a direct connection...and see what happens.

I think you suggested 10K..I have seen some other schematics with a direct "no resistance connection"..what value do you use in your amps?

Regards
M. Gregg
I suggested 10K or lower , I usualy 1k - 2,2 K . By the way the data of 6SN7 gives 2,2 - 2,6 pF input capacitance . Here is my post number 442 - Well , I talk about input capacitance of the 6SN7 and not about input impedance , in this circuit the 470KΩ resistor with the input capacitance of the 6SN7 Cgk wich is 3pF plus the capacitance of the tube socket plus the capacitance of the circuitry , makes an RC filter wich its cutoff frequency is from 42,5Khz to 56Khz ( Fc = 1/ 2 * pi* R * C ), remember that the output impedance of the previous stage will added to this resistance and then the cutoff frequency will be smaller , so there is no need to lower the frequency of the amp ( in open loop ) that much .With a more powerfull tube , I didn't mean output power tube , maby the use of low output power tube , and I say it only to reduce more the upper output ( anode ) impedance ( with smaller Ra and Rk ) of the phase spliter to reduce the Miller effect of the output tubes , and have a better frequency response , and there is no need to drive the output tubes in classAB2 . Thanks Dim.
 
Last edited:
Disabled Account
Joined 2010
Just for interest,

Looking at this there should be an improvement with the change in input resistor.. Page 7 & 8 of this link ..put togeather..

http://www.kennethkuhn.com/students/ee351/square_wave_testing.pdf

I have some components on order and will try it on the next strip..just for fun of course... :)

Run out of solder..LOL..solder sucker needs a new tip..

Regards
M. Gregg
 
Last edited:
Full 6c33c..

At 8/9/10/11/12V...I did manage 13V however the 1.6A T fuse blew..so no picks ...didn't want to try it again..drawing 1.7A
Again X10 probe so you have to multiply..


Regards
M. Gregg

Just want to check I'm reading this correctly. You are getting a maximum of about 12V peak-to-peak, into 7.9 ohms? That would be about 2.8W continuous power. That seems pretty low. Are you sure about the voltages? You mentioned blowing a 1.6A fuse at 13V peak-to-peak, but that would be only 4.6V rms, which doesn't seem like it should be enough to blow the fuse.

Could something be wrong with your voltages?

Chris
 
Here are a couple of traces for the Tim Mellow OTL, with a 1KHz sinewave input, driving an 8 ohm resistive load. Scope is set on 10V per division. The first trace is just before clipping, and the second is when clipping is just beginning. A DVM across the load confirms about 14V rms (i.e. about 40V peak-to-peak) just before clipping. This implies about 25W continuous power just before clipping.

Chris
 

Attachments

  • photo1.JPG
    photo1.JPG
    39.5 KB · Views: 405
  • photo2.JPG
    photo2.JPG
    41.2 KB · Views: 405
Something is wrong the ouput RMS voltage is 4.6V...That can't be correct...Don't know if I have a fault...

Alastair can you check your RMS voltage out?

Regards
M. Gregg

Well 4.6V rms is certainly consistent with your scope trace showing 13V peak to peak.

I believe autobias could be expected to give a lower output power than fixed bias, but I'd be surprised if it should be as low as you are finding.

Chris
 
Disabled Account
Joined 2010
Well 4.6V rms is certainly consistent with your scope trace showing 13V peak to peak.

I believe autobias could be expected to give a lower output power than fixed bias, but I'd be surprised if it should be as low as you are finding.

Chris

I agree,

I am surprised..I find it hard to believe its this low by comparison..
I can get 10W out of EL34..auto bias..
I am going to look closer at the driver stage... :)
Also the distortion on the scope trace (on one side) is mystery at the moment...

Regards
M. Gregg
 
Last edited:
Have to check in and enjoy the progress.

Think it´s time for Alistair to fire up the scope and check if he gets the same reading.

M Gregg can you cut the feedback and measure the different stages to see whats going on ?

Life´s slowly returning to normal so soon i have time to finish my amp and can participate in the measuring.

Looking forward to more readings of driver and PI.
 
OK Right--Here goes....

Yup--I Boobed, I'm an idiot, a total fool--Go on, you might as well all larf at me.....

Today I did some semi-proper testing using my Tektronix 454 Scope and a pukka sig-gen, and some much,-- now obvious-- Tuning Up!.
(This is why its taken me a while to do this, Setting all this gear up when I have the place to myself....)

I found that the PI cannot currently swing more than around 20V p-p before flat-topping/distortion, similar to what M-Gregg was having.

I have increased the values of the two PI resistors to 47K and now have 60V p-p before any mishaping/distortions, just a nice sine-wave.

PI +B Supply is 320V...

I've also dropped the Grid resistor to PI to 100K.

Running PL509, a single pair, the amp will drive 11V p-p of CLEAN, Perfect sine wave @1KHz with No sign of distortions or flat-topping, into 8 ohms.

Increasing the O/P to apprx 11.5V, we get the usual flat-topping (Symetrical) And Crossover Distortion, Caused by the Cathode-Caps charging and altering the bias of O/P valves.

This was done into an 8 Ohm non-inductive wire-wound resistor.

So, The Peak Watts value I currently have from a single channel with only a single pair of PL509 is 15.125W --Undistorted.

This equates to--If my maths is good-- to 10.6 Watts, RMS.

--Which means the O/P valves are supplying 1.375 Amps of Current.

--So, Although the O/P pair COULD make More Power if Fixed-Bias, the Self-Bias is effectively limiting the Current Passed to just a Little Under the Max Pulse value in the data-sheet (1.4A for the '509), which cant be bad.

I left it running just before the point of clip like this to see what happened to the O/P valves.

--The answer was, Nothing! They got fairly warm but no red-plates or glowing screens, just worked...

I did a Full Power Frequency test (sine-wave). I was unable to determine the LF rolloff as my sig-gen only goes to 3Hz, so its below that.....
The -3dB upper freq limit was 200KHz, Good enough even for the Batmen among you...

The Only slight issue I have which I have not yet addressed is a Slight Overshoot on Square-waves but happening at 30KHz.....

So--Go on, have a good chuckle, Yup--I boobed....
 
OK Right--Here goes....

[...]
Running PL509, a single pair, the amp will drive 11V p-p of CLEAN, Perfect sine wave @1KHz with No sign of distortions or flat-topping, into 8 ohms.

Increasing the O/P to apprx 11.5V, we get the usual flat-topping (Symetrical) And Crossover Distortion, Caused by the Cathode-Caps charging and altering the bias of O/P valves.

This was done into an 8 Ohm non-inductive wire-wound resistor.

So, The Peak Watts value I currently have from a single channel with only a single pair of PL509 is 15.125W --Undistorted.
[...]

This equates to--If my maths is good-- to 10.6 Watts, RMS.
....

Hang on, there's something wrong here. 11V peak-to-peak means 11/(2 sqrt(2)) V RMS, i.e. about 3.9V RMS. This means the continuous power (RMS power) is about (3.9)^2/8 which is about 1.9W.

Chris
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.