Jcm800 Downscale

[truckfighter]

Hello soundguruman,

thank you for your repley.

The correct way to determine if the amp is working right...
5 millivolts sine wave RMS 1 KHz at the input is supposed to drive the output stage clean and all the way to (full output) clipping.

Pleas let me verify if i get you right.

1 KHz, 5mV RMS presented at the input of the preamp will give me a clean, undistorted signal to a fully clipped signal depending on the setting of the master volum or the preamp volume? I'm guessing Master Volume.

Also, my main concern isn't the poweramp at this moment, it is only the preamp at this moment. My understanding is, that a guitar amp already distorts the signal in the preamp, i guess this distortion would be called gain and is controlled by the preamp volume. The distortion caused in the poweramplifier is controlled by the master volume.
So for now, I need to research what happens to the signal in the preampsection, with different gain setting (Gain knob turned to 1,2, ... ,10).

best regard,
florian

[J M Fahey]

Hi Florian/truckfighter
1) yes, I understand the rules were imposed on you.
My comment was that *an Engineer* , at this stage of his career, *should* be trusted to work under "real World" conditions, but , well, you must abide by their rules, of course.
Besides, I bet your suggesting a tube project must have been quite a shock for them.
I'm sure 90% of projects will be Arduino or PIC controlled gadgets or stuff like that anyway.
2) As was suggested above, the basic idea won't be to build a "regular" JCM800 preamp (you are not allowed to) but a functionally equivalent one.
That's a more reasonable approach.
Basically you should emulate it stage by stage.
Example : build the first stage, using 40 volts.
What you aim at is to have it clip with the same signal that would clip a regular 250V powered 12AX7 stage, and provide a similar waveform.
Then the next stage, and so on, including equalization, tone controls, etc. , until you reach the same waveform at the Master volume wiper.
That, feeding a mixed feedback chipamp will provide sound quite close to a real JCM800 head *not* used at maximum power (so as not to include power amp clipping), not a bad end result at all.
You'll have to find a way to introduce the effect of the presence pot though, as it is on the power amp itself and an important part of the sound.
Good luck.

[printer2]

OK, maybe a little change from the classic Marshall circuit, how about a pentode for the preamp stages rather than a triode? I just stuck a 6AU6 in a breadboard and tweaked around a bit. Used 40V supply, stuck some pots in place of the cathode, plate, and screen resistors. I have not listened to the result yet, just was looking for maximum output and symmetrical clipping at that voltage. Was kind of surprised at the result, Resistors, cathode - 2.5k with 2uF bypass, plate - 100k, screen - 355k with 0.1uF. Voltages were cathode - 0.7V, plate - 20V, screen - 12.7V. With 0.6V p-p input I could get 25V p-p out at clipping.

This gives a gain of about 40, used a 500k volume pot and with a 1M maybe a little more. The first stage gain of the Plexi is suppose to be in the range of 60, given the difference in headroom with a plexi at full voltage and the 6AU6 at 40V the gains don't look too out of line. Think the above is mostly correct, kind of late right now, can't wait to check it out tomorrow. Might even add a few more stages.

[soundguruman]

OK, maybe a little change from the classic Marshall circuit, how about a pentode for the preamp stages rather than a triode? I just stuck a 6AU6 in a breadboard and tweaked around a bit. Used 40V supply, stuck some pots in place of the cathode, plate, and screen resistors. I have not listened to the result yet, just was looking for maximum output and symmetrical clipping at that voltage. Was kind of surprised at the result, Resistors, cathode - 2.5k with 2uF bypass, plate - 100k, screen - 355k with 0.1uF. Voltages were cathode - 0.7V, plate - 20V, screen - 12.7V. With 0.6V p-p input I could get 25V p-p out at clipping.

This gives a gain of about 40, used a 500k volume pot and with a 1M maybe a little more. The first stage gain of the Plexi is suppose to be in the range of 60, given the difference in headroom with a plexi at full voltage and the 6AU6 at 40V the gains don't look too out of line. Think the above is mostly correct, kind of late right now, can't wait to check it out tomorrow. Might even add a few more stages.

6J7 rules.

[J M Fahey]

OK, maybe a little change from the classic Marshall circuit, how about a pentode for the preamp stages rather than a triode? I just stuck a 6AU6 in a breadboard and tweaked around a bit. Used 40V supply, stuck some pots in place of the cathode, plate, and screen resistors. I have not listened to the result yet, just was looking for maximum output and symmetrical clipping at that voltage. Was kind of surprised at the result, Resistors, cathode - 2.5k with 2uF bypass, plate - 100k, screen - 355k with 0.1uF. Voltages were cathode - 0.7V, plate - 20V, screen - 12.7V. With 0.6V p-p input I could get 25V p-p out at clipping.

This gives a gain of about 40, used a 500k volume pot and with a 1M maybe a little more. The first stage gain of the Plexi is suppose to be in the range of 60, given the difference in headroom with a plexi at full voltage and the 6AU6 at 40V the gains don't look too out of line. Think the above is mostly correct, kind of late right now, can't wait to check it out tomorrow. Might even add a few more stages.

*Very* interesting results.
Please keep us up to date.
That screen might be the ticket for reasonable gain at low voltages.

[printer2]

*Very* interesting results.
Please keep us up to date.
That screen might be the ticket for reasonable gain at low voltages.

That is what I thought. Normally the 6AU6 is a very gainy tube, at low voltages not so much. I thought I would be using lower resistor values on the plate and cathode but rather than plugging in resistors I decided on the pot routine and had no clue on the values until I had it set up. I then turned off the power and measured them. Mind you there could be some variation from tube to tube and it may not sound like much but heck it was worth a look. I have way too much on my plate right now to spend time on it now, just had a hour to kill and did not want to start on any other project. I would like to do a whole amp at that voltage, I have a 1.1k/8ohm SE OT that would be interesting with a 6CW5 running at 4 ohms. Or maybe a SS amp instead with a tube preamp.

[J M Fahey]

Fine.
With a 40V limit I would use your tube idea for preamps, but an SS power amp.
To each its own.

[printer2]

Oh my!!




And I really mean it. I tried the 12AX7 and 12AU7 up to 60V in the starved plate thing, sounded horrible. I used the same values for both stages and a 5ook pot between them. Ran into a portable SS amp (20W car amp for power, runs on a 12V gel cell), 8" PA speaker, I built 15 years ago. I wanted to redo the preamp as it sound great clean but not great distorted. Clean the 6AU6's sounded fine, as you turned it up it got thicker and richer. I was really disappointing when the pot was at the end of its travel. Needs a little shaping on the low end for high gain but otherwise from clean to full output the SS amp never sounded so good.

Took it down to 20V and then 12V, less gain and output but still could get some hair out of it, I did adjust the screen though. Up to 60V (limit on my PS) a bit more gain and more output, screen where it was at 40V. Just using a Tele, not a loud guitar, the input stage will stay clean and the business end is the next stage. All in all a worthwhile couple of hours. Sure beats using a triode at low voltages.

[J M Fahey]

Thanks.
For a change listen to a single 12AX7 , driving a 20W SS amplifier, the whole enchilada fed from a 15V 2A wall wart.

[Chuck H]

Definitely has some "it" factor. The difference between SS and tube in the later stages is easy to hear though. It's a sort of blatty thing that I think is from the really flat clip of ss devices. But it's certainly a good useable tone. A tube pre and a SS PA is the obvious way to go for this project.

[Austin]

Why not try two 6080 tubes in pp for the power section? They can put out 250 ma at 40 volts with both halves paralleled. That would be ten watts per tube so a push pull output stage made with two of these could get pretty loud..
6as7 is almost the exact same tube.

http://www.mif.pg.gda.pl/homepages/f...27/6/6AS7G.pdf
http://www.mif.pg.gda.pl/homepages/f...049/6/6080.pdf

[J M Fahey]

Just checked 6080 datasheets.
They can *dissipate* 10W at 40 volts ...... but that's not the point.
Supposing we start with +40V idle on plates and biased to 0mA, just to extract maximum power possible, no matter what, plate voltage can swing down to 10V, at which point they pass 30 mA.
Two triodes in parallel would provide meager 60mA.
This would be the peak current capability, which coupled with a peak voltage swing of 30V adds up to a 1.8W peak power.
RMS would be half that, a meager 0.9W.
This is considering regulated 40V, zero ohms DC resistance in transformer windings, zero magnetic loss, pure class B; so in real world we would barely achieve half that, if any.
Now, using 2 octal triodes plus associated supply, including filaments), designing and building an OT, etc, for a meager 400mW of bland triode power, well, I dunno, don't dare to qualify it.
But yes, it probably *can* be done.

I can also ride on horseback from Buenos Aires to New York.
Which, by the way, *has* been done.
21500 Kilometers in 3 1/2 Years.(April 24 1925 to September 20 1928)


An example of things which are done just because they *can* be made.

[printer2]

Yeah but who is saying this needs to be practical, it is an engineering exercise after all. You can look at it two ways. It could be thought of a small bedroom amp, even half a watt can be loud with the right speaker, maybe not clean though. Or you can look at it as a beefed up pedal that can drive a speaker in a pinch. And at the end of it all, who knows, the supply voltage might get accidentally bumped up with a doubler or even a higher voltage transformer. Wonder what the OP has in mind?

[Austin]

Just checked 6080 datasheets.
They can *dissipate* 10W at 40 volts ...... but that's not the point.
Supposing we start with +40V idle on plates and biased to 0mA, just to extract maximum power possible, no matter what, plate voltage can swing down to 10V, at which point they pass 30 mA.
Two triodes in parallel would provide meager 60mA.
This would be the peak current capability, which coupled with a peak voltage swing of 30V adds up to a 1.8W peak power.
RMS would be half that, a meager 0.9W.
This is considering regulated 40V, zero ohms DC resistance in transformer windings, zero magnetic loss, pure class B; so in real world we would barely achieve half that, if any.
Now, using 2 octal triodes plus associated supply, including filaments), designing and building an OT, etc, for a meager 400mW of bland triode power, well, I dunno, don't dare to qualify it.
But yes, it probably *can* be done.

I can also ride on horseback from Buenos Aires to New York.
Which, by the way, *has* been done.
21500 Kilometers in 3 1/2 Years.(April 24 1925 to September 20 1928)
[ATTACH=CONFIG]19977[/ATTACH]
[ATTACH=CONFIG]19976[/ATTACH]
An example of things which are done just because they *can* be made.

Two of these tubes is really four triodes though because each tube has two sections, so it would be parallel push pull then.. If one tube can dissapate 10 watts how many watts could you get in push pull ab1? What about doing it the opposite of how you guys suggested with a tube based driver driving a solid state power section, but rather a solid state driver driving the push pull 6080 tubes in class ab2 for even more power? Maybe not so boring then..

[Chuck H]

So four triodes instead of just two?!? That changes everything. So if Juan is even fairly accurate in his considerations that would raise the meager 1/2 watt output to a whopping 1 watt!!! Better get some ear plugs.