Wattage Plot and questions

[casssax]

I picked up an Oscilloscope and have been playing around with it and the Stewart SL-21 amp I've been working on.


I was a little surprised that it's only outputting about 9W RMS.
(I measured the output wattage by sending a 1kHz sine wave measuring the voltage on the dummy load at the point the wave form started to distort)

I admit I don't really know what factors contribute to output wattage in an amp.
Maybe someone with more experience could tell me if based on the schematic above 9W is what they would expect.
The amp has two EL84's and originally I thought it would be similar to a Marshall 18W amp.


Below I plotted the wattage at different volume levels.
I fed a 1v P-P signal to the input and turned up the volume until I got 1V on the output and then recorded the voltage on the dummy load.
Then I kept turning up the volume until I had 2V on output and continued until the output voltage stopped increasing.
Then I plotted the observed voltage and the wattage. I did this twice, once with the tone control turned all the way down (max bass, seen as ObsVltg1/Watts1) and then with it turned al the way up (max treble, seen as ObsVltg2/Watts2).


The picture below is of the wave at full volume. The input is still 1V @ 1V/div and the output is at 5V/div.
The waveform starts to distort with the volume turned up 1/2 way.
I've never looked at the output of an amp like this before so I don't know if that is what you would expect to see or not.


I'm going to get some more pics going through the gain stages to see what is happening and when.
I started doing this before but didn't record what I saw. One thing I did notice was that the output of the phase inverter seemed to show a significant difference in the signal going to the two output tubes. Should I expect the signal going to the output tubes to be about the same or is this normal for a cathodyne phase inverter circuit?

Thanks.

[HotBluePlates]

Tremolo off, insert "whatever" input signal until you get 10v peak (or just a hair less) across the 1MΩ resistor from an EL84 grid to the footswitch. This is an upper limit of driving signal to the output tubes for your wattage numbers to be valid.

If you only jack up the input signal and measure speaker volts, you will keep getting more & more wattage until you top out around 2x the clean output power of the amp. At that point, it's so distorted you're throwing square-waves at the speaker, whose RMS power is 2x the RMS power of a sine wave with the same peak voltage.

[jjasilli]

Another thought.  It's good to know where the distorted waveform arises, power amp or preamp.  I.e, bypass the preamp; input clean signal directly into the PI until the scope shows a distorted waveform into a dummy load with proper ohms rating for this amp. Record how many signal volts the PI can take until the power amp has distorted output.  Also if you connect a 4 ohm OT tap to an 8 ohm load, you'll read half power.

Then measure how many clean volts the preamp can putout to the PI.  If the preamp is shy volts it will effectively reduce the apparent clean output of the amp.

[casssax]

I should have mentioned before that the highest voltage I saw before the wave started to deform was 8.92V.

I tried HotBluePlates suggestion and ended up seeing 8.8V.

So that would be about 9.7Watts

I'm going to try jjasilli's suggestion and see where the phase inverter starts to distort next.

[PRR]

> The waveform starts to distort with the volume turned up 1/2 way.

1V is very strong for a guitar signal. Half-up sure should be near/past output overload.

Yes, the phase splitter must have the same level both sides (below gross distortion). Unbalance there sure could be why output falls short. One '84 isn't pulling its share.

The "clean" output may be only 14W. The "18W" number is for enough distortion to easily see on the 'scope.

The waveform you show is into GROSS distortion, is typical, and is not of much interest. The asymmetry could be a sick driver, but a cathodyne in gross distortion will throw some awful funny waves.

[casssax]

I checked the two preamp stages. The first stage I get no distortion even at full volume (I'm using an input signal about 0.2V @ 400Hz)

The second stage starts to distort and turn a bit square after about 3/4 volume.
I noticed that the second stage I'm getting a sort of step pyramid shape no matter what the input volume.  There's not much going on between the first and second stage. Could a bad coupling cap be causing this?

Below the top wave is from the first stage and the bottom wave from the second.


On the phase inverter I'm seeing close to the same levels on pin6 and pin8 but pin6 has a strange dip in the bottom of the wave form and it is flattening out at the top of the wave. (this is with the volume about half way up, before the second preamp stage starts to distort.) I tried a different tube and got the same results.

Top wave form is from pin6 of the PI and the bottom is from pin8 with the volume set so the second stage is not starting to distort.

[HotBluePlates]

Tremolo off, insert "whatever" input signal until you get 10v peak (or just a hair less) across the 1MΩ resistor from an EL84 grid to the footswitch. ...

I should have mentioned before that the highest voltage I saw before the wave started to deform was 8.92V.  I tried HotBluePlates suggestion and ended up seeing 8.8V. So that would be about 9.7Watts ...

Assuming your numbers are volts-RMS at a speaker output, using an 8Ω load, then yes that all sounds correct.

I was a little surprised that it's only outputting about 9W RMS. ... I admit I don't really know what factors contribute to output wattage in an amp.
Maybe someone with more experience could tell me if based on the schematic above 9W is what they would expect.
The amp has two EL84's and originally I thought it would be similar to a Marshall 18W amp. ...

Normally, the clean output power of any amplifier is limited by the output stage. That is, they're designed so that the output tubes distort first. Otherwise, you're paying for power capacity (bigger tubes) but not getting your money's worth. Like running a KT88 to get 2w of output power.

Later guitar amps, often with master-volume circuits, are the exception to the above. Even so, they usually can make the output tubes distort before any other stage if the master volume is turned all the way up.

We don't know anything about your output transformer primary, though you could measure it to determine its impedance. That's how you'd calculate the power output possible from this amp. But PRR is probably correct that one (maybe both) of the EL84's aren't pulling their weight, assuming Stewart designed the amp to make reasonably-full use of the EL84's.