Bias and amp output

[JoshBernstein]

Hello again
Im posting about my newest build. Again. Im sorry.
Theres a description of the amp here: https://tubeamparchive.com/viewtopic.php?t=29373

Im starting to wonder about the wattage of this amp. Using a 1k hz sin wave fron my phone, im getting around 12 volts across a 16 ohm dummy load resistor which equates to about 9 watts with the preamp and power amp cranked. What the hell? I have 380 volts on the plates of the kt66's and im using a 6.6k trainwreck ot. I was expecting 15 to 20 watts out of this. Are my calculations wrong or is this actually only putting out 9 watts at max volume? Is there anything i can do? Ive attached a screenshot of my weber bias calculator results for reference. Thanks!

(I had measured my voltages on a dying batter before when i calculated 15 watt output in the previous thread, so these are more accurate.)

[martin manning]

Hello again
Im posting about my newest build. Again. Im sorry.
Theres a description of the amp here: https://tubeamparchive.com/viewtopic.php?t=29373

Im starting to wonder about the wattage of this amp. Using a 1k hz sin wave fron my phone, im getting around 12 volts across a 16 ohm dummy load resistor which equates to about 9 watts with the preamp and power amp cranked. What the hell? I have 380 volts on the plates of the kt66's and im using a 6.6k trainwreck ot. I was expecting 15 to 20 watts out of this. Are my calculations wrong or is this actually only putting out 9 watts at max volume? Is there anything i can do? Ive attached a screenshot of my weber bias calculator results for reference. Thanks!

(I had measured my voltages on a dying batter before when i calculated 15 watt output in the previous thread, so these are more accurate.)

Power into the load is V^2/R (where V is RMS or AC voltage) so yes, you have 144/16=9W.
For 380 - 25 = 355 Va-k, a 100% bias point would be 25/355 * 1.05 = 75 mA at the cathode, but the bias won't affect the output power very much. The theoretical (load-line) power for your configuration is about 30W, so you might expect about 25W in practice. If you reduce the load to 4k, the load line power would increase to about 45W (assuming screen voltage is close to anode voltage).

I suspect that you aren't driving the KT66 grids enough. Measure the AC voltage at the KT66 grids to see how much drive you have there.

[Stevem]

You have 380 volts on the plates at idle , or running with output stage at full clip?
In general if your dropping more than 45 volts off the plates at full clip in a AB output stage then you will never see the output wattage you think you should because your PT does not have the current output it should .

A amp that uses SS diodes and drops off more than 40 volts is not going to have much punch to it when it's fully revved up!

On the other hand if your running a class A output stage your plate voltage should not dip by even 8 volts!

[JoshBernstein]

I have the 380 on the plates at idle. Have not measured at full volume.
I will measure the the voltage at the grids of the kt66s when i get off work. What kind of voltage should i be looking for?

I fixed the issue i was was having from my previous thread where i thought i couldn't get any preamp distortion, but really i had just mixed up which knob was the master and which was for the preamp. Genious right?

Im getting a fair amount of preamp drive, so i dont really want to add another stage. If my voltage on the output tube grids is low, could i just increase the value of the master volume pot? Its a dual gang ppimv so its acting as the grid leaks.

[Littlewyan]

We really need a schematic so we can see how the amp is set up. You should definitely get more than 9W Clean power from 2 KT66s with 380V on the plates. Could be your transformer just can't supply adaquate current, your screen voltage is too low or it could just simply be you're not driving the KT66s hard enough.

[martin manning]

I will measure the the voltage at the grids of the kt66s when i get off work. What kind of voltage should i be looking for?

With signal applied, measure DC voltage on the cathode, and AC voltage at the control grid. The AC should be about 70% of the cathode voltage. That means the peak voltage at the grid is equal to the DC bias on the cathode and you are driving Vg1 to 0.

[brewdude]

How big is the 1kHz signal peak to peak?
Luke it be that you are not feeding a large enough signal from your phone to realize full power measured across the dummy load?

[dorrisant]

Martin, have you ever considered writing a book? I would be first in line to get it.
Where can I get the information you read to get you to where you are? I just seem to love your explanations and would like to see where you are getting it from, if that is possible.

Josh, I applaud you for your efforts and the questions you ask. Please continue both. Sorry to derail... I'm very interested in seeing a compilation of power tube input voltages and how the different tubes react as a reference. I don't know if such a thing/article exists, but it should.

[xtian]

With signal applied, measure DC voltage on the cathode, and AC voltage at the control grid. The AC should be about 70% of the cathode voltage. That means the peak voltage at the grid is equal to the DC bias on the cathode and you are driving Vg1 to 0.

Echoing Tony's applause. That's a new one for me, too!

Martin, is that AC RMS or p-p?

[martin manning]

The AC voltage at the grid is the RMS value of the signal, which is 0.707 times the peak. Accordingly if the RMS grid signal can reach 70% of the cathode voltage, then Vg-k is ~0 at the peaks.

dorrisant, Thanks for the compliment. I read the same stuff as everyone else, I guess (Blencowe, Kuehnel, etc.), plus watch You Tube videos (lots out there, some very good), and experiment to see how things work in practice.

[Stevem]

Any normally set up amp with two stages of gain before the PI ( and with good preamp tubes ) should with only 180 mv of input signal , drive its output stage to full clipping and peak wattage !

If you have a scope you can prove out when the PI output starts to clip and then when the output tubes start to clip.
If the output tubes do not clip then you have found your wattage issue.

If you do not have a O scope you need to get one if you want to keep doing amp things and still have any Hair left on the top of your head, or not be set off and committed , lol!

[JoshBernstein]

Well i have an update.
Cathode voltage at full tilt is 25.5 volts. Voltage on the control grid of one power tube is 33 volts, and 36 on the grid of the other. At idle the plate voltage on the kt66's is 380, and drops to about 377 with pre and power amp volume maxed. All of this with a 1k hz sin wave being played at .62 volts at the input of the amp.

And i really do need a scope. im not sure why i haven't bought one yet. But it's definitely on the shopping list.

[dorrisant]

And i really do need a scope. im not sure why i haven't bought one yet. But it's definitely on the shopping list.

Look here:

http://diyguitaramps.prophpbb.com/topic52.html

[martin manning]

Josh I agree you need to get a look at the waveform. If it is not sinusoidal, your meter is either assuming it is and giving you a false reading, or if it is a true RMS type, it is giving you the correct value, but then you can't assume that the peak is 1.414 times RMS. It would appear that you have enough drive voltage and power supply current though.

[Stevem]

.620 volts is HOT !and you just might be clipping the first gain stage with that , but for sure dependant on if and where you may have a master volume located your for sure clipping the second gain stage!!

Hook your meter set for ac volts to then output of your most used guitar and jam a hard hit first position A cord and note that voltage seen and the use that signal level yo drive the amp with.