I have been having issues with my SE84DIY ever since I got it last October.  The problem is that it distorts when the output goes over about 1/2 watt (8 ohm speaker).  With a sine wave input, this translates into anything over 2 v rms (5.7 v ptp) measured across the speaker is distorted.

The problem is occurring on both channels.

All tubes have been replaced, and both sets of tubes behave the same way.

When the distortion is first heard (with a sine wave input) the input voltage at the grid of the input tube is 1.5v ptp (0.53 v rms) or about 4 times lower than the 2v input required for full output (as stated in the assembly instructions).

The distortion is first seen as the clipping of the top of the sine wave at the plate of the input tube.  It is then propagated through the output tube, transformer, and into the speaker.

A secondary effect (that happens later as the volume control is increased further) that may shed light onto the problem is that as the input level increases (up to 2v) the power supply increases.  As much as 50 vdc higher at a 2v input than with zero input.  The waveform at the input grid of the output tube is shifted way negative...normally its a nice sine wave centered about zero volts, but with high input level its a clipped sine wave going only 4v positive and 40 to 50 v negative.  This grid voltage is seen on both channels.  I believe the large negative voltage on the grids of the output tubes is turning the output tubes off to a point where they aren't conducting and this reduction in current is why the power supply voltage is higher.  I have no clue as to why such a large negative voltage is at the grid of the output tubes.

I have a sine wave generator, lots of volt meters, and an osscilloscope, so I should be able to setup, test, and report on anything further one might need to determine what the problem is.

Thanks is advance for the help

Dan

Steve (and all)

All testpoints are within the specs as stated in the assembly instructions.  I have not measured them to see how they change as the load changes from an 8 ohm to a 4 ohm speaker, but I can do that later if its important.

I suspect the spec you are thinking about is the 10.25v on the cathode of the output tube...Yes that's there, and any signal level up to 10.25 v peak (7.25 v rms) on the grid of the output tube produces clear, undistorted, output.  Trouble is that 7.25v rms on the grid of the output tube only produces 2 v rms across an 8 ohm speaker which is only 1/2 watt.  Both channels are doing this.

I've measured the following voltage gains:
input tube:  18
output tube: 20
transformer:  70:1

Could I have the wrong output transformers?

Dan

With an 8 ohm load and the input turned all the way down I measure the following:
TPA : 348 vdc
TPB : 312 vdc
TPC : 10.0 vdc
TPD : 2.6 vdc

When I change to 4 ohm load I get identical readings.

Also, with 8 ohm load, 1khz input turned up until clipping is just starting across the load, I measure the following:
V across load : 2.3 vrms (meter), 6.5 v ptp (scope)
V at output tube : 140 vrms (meter)
V at input to output tube: 7 vrms (meter), 20 vptp (scope).

This 2.3 vrms across a 8 ohm load is only 0.66 watts.  I would expect 3.8 vrms to get 1.8 w across a 8 ohm load.

The reason I suspect the transformers is that I don't think you can get any larger signal at the input of the output tube as you will exceed the voltage at TPC and that will turn the tube off.  The only thing left is the transformer or too low a gain in the output stage.

Dan