Clone Sorta Fender Pro Reverb problems?

[SoapboxSam1]

Hi all, I have built an amplifier using the Fender Pro Reverb AA165 schematic. It has only one channel with reverb and no vibrato. The preamp channel is coupled to the reverb circuit using the standard .02 capacitor. With this design, the volume was not near what I thought it should have been with P-P 6L6GCs as finals. So I took it a little further and also coupled the preamp channel into the PI circuit via a .047 cap and a 220k resistor connected at the reverb mixer 220k resistor like the original Normal channel. In essence, I’ve created a channel two with reverb with the addition of a channel one coupling capacitor also. I reversed the output leads of the reverb transformer so the signals at the two 220k resistors right before the .001 PI coupling capacitor would be in phase. The maximum volume level before clipping came up but still does not seem to be as it should. I am including measurements that I have made with and without a 1K sinewave at 200mv. applied, in hopes that someone can give me some advice as to what I am doing either right or wrong.

INFO:
Power Xfmr – P-TF22798
Output Xfmr – P-TF 22848 – 4 ohms
Choke – 22699

Oscilloscope probe X1

No Sinewave input:
Power input – (119.5 Vac)
PT xfmr output – (350 Vac)
Standby on B+ – (+490 Vdc)
Standby off B+ - (+435 Vdc)

2- 6L6GC Outputs
Output xfmr CT – (+435 Vdc)
Output tube Plates – (+433 Vdc)
Output tube Screens – (+434 Vdc)
Output tube Cathodes I – (48.4 ma)
Output tube Grids – (-40 Vdc)

Phase Inverter 12AT7
PI-2 – (+63 Vdc)
PI-3 – (+98 Vdc)
PI-7 – (+65.2 Vdc)
PI-1 – (+226 Vdc)
PI-6 – (+226.7 Vdc)
V at junction of 82k/100k – (+423 Vdc)

Signal Applied to preamp input: 200 mv.
PI-2 – (+61.6 Vdc)
PI-2 – (6 Vpk-pk)
PI-7 – (+63.8Vdc)
PI-7 – (5 Vpk-pk)
PI-1 – (+215 Vdc)
PI-1 – (45 Vpk-pk)
PI-6 – (+215 Vdc)
PI-6 – (45 Vpk-pk)

1-6L6G –5 – (-40.3 Vdc)
1-6L6G –5 – (45 Vpk-pk)
1-6L6G –3 – (+410 Vdc)
1-6L6G –3 – ( HUGE Vpk-pk)
2-6L6G –5 – (-39.8 Vdc)
2-6L6G –5 – (45 Vpk-pk)
2-6L6G –3 – (+407 Vdc)
2-6L6G –3 – ( HUGE Vpk-pk)

With only a 1X probe I could not measure the Vpk-pk on pin 3 of the output tubes but it was way over 45 Vpk-pk. I think I might be able to use a 10meg resistor in series with the probe and measure it as a 10X. I have a 1/10/20/100X probe on the way.

OUTPUT SIGNAL at Speaker Jack: ( 45 Vpk-pk)

I was using a 8 ohm passive load on the 4 ohm output so I would expect the output to be lower. So, is 45 Vpk-pk correct with this load? I have two output jacks wired in parallel so I can install two 8 ohm speakers. I went back and installed another 8 ohm passive load which gave me 4 ohms. Here are my calculations as to what the Power Output should be with both 8 ohm and 4 ohm:

Vrms = .3535 x Vpk-pk
Vrms = .3535 x 45
Vrms = 15.90

8 ohm load calculation:
P = Vrms2/R
P = 15.90 x 15.90 = 252.81/ 8
P = 31.60 W

I was thinking that I could get at least 40Ws out of this amp but the calculations don't give me that amount.

Outputs biased at 70%
Plate voltage 433 Vdc
Tubes 30W

.70 X 30 = 21
21 / 433000 = 48.4 ma

4 ohm load calculation:
P = Vrms2/R
P = 15.90 x 15.90 = 252.81/ 4
P = 62.20 W

The 4 ohm calculation just don’t seem right to me and I think I’m missing something in my way of thinking especially since the tubes should not be able to produce more than the 30W each since they are biased at 70%. When I went from 8 ohms to 4 ohms, the output sinewave did not increase so I used the same Vpk-pk to recalculate the Power Output with a 4 ohm load. Help… straighten me out!!

[Bruce / Mission Amps]

The power tube's idle current DOES NOT determine the finite output power, ...as viewed by your scope across a fixed non reactive load.
Get that one out of your mind for a while.
What frequency are you using? I never use the 1KHz standards of radio talk... I use 83Hz and 440Hz... that's about it unless checking for some crazy hi-freq oscillation issues at high volume or high level inputs.
4 ohms 15.9vac. AND 8 ohms .... it can not be the same thing with 8 ohms... something wrong with your test. It should have went down a bit at 4 ohms.

With it maxed out, are you looking at a clean sine wave across the dummy load, or?
200mv is pretty hot for a test like this and you probably have preamp distortion if you bring the volume control up too high.
Roll that input signal back or turn the signal generator output off, turn the volume pot up all the way (again, with zero mvac of signal injection), then with the probes across your dummy load, monitor your scope view, start increasing the signal injection amplitude.
When you just start to see a very tiny amount of flat topping on the sine wave, or the sine wave starts skewing, you have probably hit or exceeded the max clean output.
Also, most of those real Fender 6L6 amps, using the smaller, channel mount output tranny, can only make about 36-38 watts, "dead clean" at 83Hz.
With distortion products added in (great guitar tone) they will easily do well over 40watts @83Hz.

[SoapboxSam1]

Thanks for the reply. I hope you are still monitoring this post as I haven't been able to get back on the computer for a little while.

Please see my schematic at:

SortaCloneFenderProReverb.jpg picture by Soapboxsam - Photobucket

Please note the changes to the schematic are the removal of the .047/220K around V3A which was causing "out of phase" conflicts when mixing the signal from the preamp and the output of V3A. I have also removed the .1 cap between V3A pin 6 and the 220k resistor feeding the PI.

The power tube's idle current DOES NOT determine the finite output power, ...as viewed by your scope across a fixed non reactive load.
Get that one out of your mind for a while.

Got it!

What frequency are you using? I never use the 1KHz standards of radio talk... I use 83Hz and 440Hz... that's about it unless checking for some crazy hi-freq oscillation issues at high volume or high level inputs.

I am now using 440Hz/100mv

4 ohms 15.9vac. AND 8 ohms .... it can not be the same thing with 8 ohms... something wrong with your test. It should have went down a bit at 4 ohms.

You are correct as there is 5vpp less signal when switching from 8 to 4 ohms. The output transformer is a PTF22848 with 4 ohms output.

With it maxed out, are you looking at a clean sine wave across the dummy load, or?

With 100mv input, the output signal at max volume is very clipped with cross-over distortion.

I removed the output tubes and scoped the signal at Pin 5 of both tube sockets. V5-5 was 60Vpp and V6-5 was 62Vpp with both waveforms being symmetric. Replacing the tubes and scoping pin 5 of both tubes shows clipping on both pins with the waveform being pointed on the negative and flat topping on the positive. At this volume level, I scoped the output on pin 3 of both tubes. It was a larger signal but with a 1X probe I couldn't accurately measure the Vpp but could tell is was following the control gird input waveform. I suspect this is distortion could be caused by lag of PTF-22798 PT and saturation of the PTF-22848 OT being fed back to the PI via the NFB. What do you think?

200mv is pretty hot for a test like this and you probably have preamp distortion if you bring the volume control up too high.

Am now at 100mv input signal. You got my curousity up so I did some testing. I set the input volume to max, scoped pin 6 of V1B and started raising the generator level up until I saw a slight change in the waveform. I then connected the scope to the input jack and read 1.2Vpp. That's a lot of input signal before any distortion of the preamp. Heck, that's line level signal.

Roll that input signal back or turn the signal generator output off, turn the volume pot up all the way (again, with zero mvac of signal injection), then with the probes across your dummy load, monitor your scope view, start increasing the signal injection amplitude.
When you just start to see a very tiny amount of flat topping on the sine wave, or the sine wave starts skewing, you have probably hit or exceeded the max clean output.

I did this and found that the input signal was only 20Mvpp. That's not much signal input. Can you tell me why you wanted me to do this test?

Also, most of those real Fender 6L6 amps, using the smaller, channel mount output tranny, can only make about 36-38 watts, "dead clean" at 83Hz.

As it is right now, my calculations are:

Output Vpp measured just before clipping of output signal.

Output = 35Vpp with 4 ohm load

Vrms: .3535 X 35 = 12.019
Power: 12.019 X 12.019 = 144.456
144.456 / 4 = 36W.

Any increase in the volume control causes the output signal to go into clipping with no increase in the Vpp up to maximum volume control level (10).

With distortion products added in (great guitar tone) they will easily do well over 40watts @83Hz.

I am very interested in your opinion.

Thanks, Sam