amp not outputting full wattage

[mozz]

I have seen 12at7 help the output tubes put out more wattage more than once. How is your screen voltage compared to the plate voltage?

[Jazz P Bass]

You set each tube to 45ma's quiescent current?

That is way too high for one tube.
How about shooting for 25 or so.

I have seen these amps do 80 watts pure clean.

[Chuck H]

You set each tube to 45ma's quiescent current?

That is way too high for one tube.
How about shooting for 25 or so.

I have seen these amps do 80 watts pure clean.

Great to have an actual number I can only assume they don't make a true hundred watts because most "100 watt" amps don't.

[nevetslab]

What frequency are you testing at? OH, you mentioned testing at 400Hz early on. The lower you go below, say, 50Hz, the power output starts rolling off some....in part from the O/T beginning to saturate, in part from now being below the charging frequency of the power supply. But still, 80W out into a known load is typical What IS your load resistance? OH, you did state that earlier, having your load in 1 ohm/100W steps up to 8 ohms. 4 ohms?

I agree with mozz with regards to the driver tube. The 12AT7 should be installed in place of the 12AX7. More for the plate current ability than the gain factor.

[J M Fahey]

Please retest with higher and lower load impedance.
If 4 ohm nominal try 2 and 8 ohm and post results.

My point being I want to know what´s limiting your power: lack of adequate voltage or adequate current.

Another separate experiment: remeasure power into the nominal load but *without* NFB, I want to make the Power amp "easier to drive".

All tests still with 400 Hz signal or thereabouts and just surpassing clipping on both peaks, usually one clips before the other.

Allow for a little clearly flat area on top and bottom peaks, as Enzo says , this is a Guitar amp, not an Audiophile one and even less a NASA thingie.

[Dave H]

Have you scoped the grid one's of the 6L6's to check that the clipping is not happening prior to the final stage?

I'd check this. At the onset of clipping at the speaker output the grid voltage of the 6L6s should be swinging between 0V on positive peaks and twice the bias voltage on negative peaks.

Bias was -43V in this example.

[Jazz P Bass]

Really nice scope shot.

It says it all.

[johnhoef]

Hi guys, and thanks so much for your continued help with this amp of mine! So, I made some progress today. I received the new 12AT7 in the mail today and was excited to install and see what happens. First I left the 12AX7 in the phase inverter and checked the sine wave input to the 6L6 control grids. I noticed it clipped at the same time as the output waveform. (using 1khz now). This is indeed a problem. Installed the 12AT7, and excitedly, checked the 6L6 grids. This time the sine wave stayed reasonably pure at the grids well past the output clipping. So, we are now getting sufficient drive on the 6L6 control grids. Mozz, you're right, the 12AT7 is absolutely necessary to provide sufficient drive to the 6L6s. I was now expecting higher wattage output, so went to measure the AC voltage at the amp output. With slight clipping at the output, I am getting around 20VAC at 8 ohms, 14VAC at 4 ohm, and 10VAC at 2 ohms. All three measurement translate to....you guessed it, 50 watts!! So, still some issue here. This amp just does not want to go past 50 watts!
Jazz P bass, thanks for your suggestion in reducing the bias to 25ma. I will do that once I have this amp working properly again.
I decided to take a look at the power supply, and see we have three diodes in series for each side. I replaced them with single modern diodes in hopes the power supply voltage will bump up some...no! So restored the original configuration. Wondering why they do this series thing? I also checked the two filter capacitors - they are OK.
JM Fahey, I would like to do this test for you, but what does "without NFB" mean?
Again, thank you all for your help!!
John

[Chuck H]

There has been some talk about the screen grids, but no reports on investigation.?. Already mentioned, if a screen grid resistor is open the amp will work, but that tube will be removed from service. Do ALL the screen grids have HV on them of somewhere around 400V???

If the screen grids check out it may be a good at this point to test the OT ratio, just to be sure.

I didn't go back through the thread but it's probable Juan mentioned the NFB loop because it will introduce a clipped waveform on the phase inverter simultaneously as the power tube plates begin to clip. Since you are looking for clipping at the power tube grids and plates for a comparison it might be a good idea to disconnect the NFB loop until the problem is identified.

[mozz]

As was said earlier, make sure all 4 tubes are working. Measure the actual resistance of your dummy load. One more thing, 20v output to 8 ohms, what is the highest distorted voltage you can measure with all vol and tone controls maxed?

[Jazz P Bass]

I would suggest, in the case of these two particular amps where the heaters are in parallel, that you test each tube separately into a dummy load.

That way you will know what each tube is capable of.

[Chuck H]

I would suggest, in the case of these two particular amps where the heaters are in parallel, that you test each tube separately into a dummy load.

That way you will know what each tube is capable of.

I might add, in that tubes socket respectively. Then if you do find a problem you can swap the tube/s to a different socket to see of the problem is following the tube or the socket.

[johnhoef]

OK, disconnected the negative feedback resistor. (I assume that is what you guys mean by "NFB"). No change in output....still 20vRMS. Reconnected that resistor, and did the individual tube and socket test. First, I tried each of the four tubes in the same single socket, with the other three sockets vacated. I got the same 10vRMS on the output for all four tubes as well as the same waveform and p-p voltage. Then used the same tube in all four sockets, one by one. Again, same result - 10vRMS on the output, with, of course, the positive side of the waveform on two of the sockets, and the negative side of the waveform on the other two sockets. So, this verifies the sockets as well as the tubes are OK. Further, all four tubes work well in other amps, so these four tubes are known good. Further, measured voltages on the tubes with no signal input. Plate Pin 3 = 391v, Screen grid pin 4 = 391v, control grid pin 5 = -48v
So, still no progress!! Further help is sure appreciated.
John

[alexradium]

OK, disconnected the negative feedback resistor. (I assume that is what you guys mean by "NFB"). No change in output....still 20vRMS. Reconnected that resistor, and did the individual tube and socket test. First, I tried each of the four tubes in the same single socket, with the other three sockets vacated. I got the same 10vRMS on the output for all four tubes as well as the same waveform and p-p voltage. Then used the same tube in all four sockets, one by one. Again, same result - 10vRMS on the output, with, of course, the positive side of the waveform on two of the sockets, and the negative side of the waveform on the other two sockets. So, this verifies the sockets as well as the tubes are OK. Further, all four tubes work well in other amps, so these four tubes are known good. Further, measured voltages on the tubes with no signal input. Plate Pin 3 = 391v, Screen grid pin 4 = 391v, control grid pin 5 = -48v
So, still no progress!! Further help is sure appreciated.
John

testing a push pull amp with only one side of the OT working is not good,you should use both sides balanced in current.
if you disconnect NFB and nothing changes i guess something is connected wrong or missing or burnt.
Is the output transformer 100% working?

[Reader]

- - - measured voltages on the tubes with no signal input. Plate Pin 3 = 391v, Screen grid pin 4 = 391v, control grid pin 5 = -48v
So, still no progress!! Further help is sure appreciated.
John

391V seems low. @Enzo alluded to this in his first post in this thread. I think it would be useful to know what the plate voltage droops to when you drive the amp to full power.