Marshall JMP 25w Specialist

[tubenit]

I stumbled across this schematic yesterday & found it really interesting.  Never heard of this amp before and I am not aware of any Marshall that had a concertina phase invertor particularly with EL34's.

In some ways it has a gain stage topology like a Princeton Reverb.  Sort of an interesting idea to have a PR with EL34's.

However, it has a unique tone stack idea.

And I was intrigued by the ECL86/6GW8  reverb design and the insertion points that they used.............. and that they used a single 12AX7 triode for the EL34 vibrato.

http://el34world.com/charts/Schematics/files/marshall/Marshall_Schematics.htm

I found a terrible quality YouTube of the amp .......... sounds like kind of an old school blues vibe to it  (start listening at about :30 into it) :
Dave's Marshall Reverb Tremolo 25W

With respect, Tubenit

[tubenit]

Does anyone know if a Fender reverb trannie and a typical Fender reverb tank would work with this build?

Tubenit

[sluckey]

That does look interesting and tempting. They got a full featured amp with just 5 tubes! Are you considering building this? I like the reverb circuit. Better check the price of that 6GW8! A 6BM8 would probably work just fine and it's less than half the cost. The trem looks like a Princeton circuit and has a strong sound in the demo video.

I noticed a couple errors on your schematic. The 470K PI grid resistor is just hanging. It should connect to the junction of the cathode resistors. The top output cap is connected to node C but should connect to the PI plate.

I think the Fender reverb tranny and tank would work well in this circuit.

[tubenit]

Corrected schematic in first post.

THANKS for the reply and your thoughts about it.  You're right the 6GW8 is rather pricey compared to the 6BM8 which is still manufactured.

With respect, Tubenit

[sluckey]

Here's a bit of info from drtube.com...

"This amp was only shortly in production from 1972 up to 1973. It uses an ECL86 tube as the reverb driver/recovery tube. It was fitted with a Celestion 16Ω G15C speaker."

And a pic...

[tubenit]

That is a nice looking amp!

I redrew the schematic and changed the reverb to 6BM8.   And I changed the vibrato to a 6G9 style using a mosfet cathode follower. Part of my reasoning for that was: 1) it used more typical values as opposed to the 1.8M resistor values & 2) we know this is a strong vibrato design that will work with 6L6's also.

Not sure about the 6BM8 pentode section with pin 7 components?  I am not sure the .1 to ground is needed??

IF this seems like a reasonable design, I'll try to draw up a layout to go with it.

With respect, Tubenit

[HotBluePlates]

And I was intrigued by the ECL86/6GW8  reverb design ...

Had you tried anything like that with the 6BM8?

The triode section of the 6BM8/ECL82 is a bit weaker than the 6GW8/ECL86 (mu of 70 instead of 100) and the power tube section is lower Gm (6.4mA/V @ 200v for 6BM8, 10mA/V @ 250v for 6GW8). Both should work the same way, but I wouldn't know without trying whether the 6GW8 is critical the way Marshall is driving it right from the 1st gain stage's Volume control (as in, with weak signal voltage extra reverb-driver grunt is needed).

Otherwise, the 6BM8 is cheaper and available with new production tubes.

However, it has a unique tone stack idea.

The tone control is an interesting variation of the Fender 5E3/5F2 tone control. Maybe it was set up that way to facilitate mixing the reverb back in with the dry signal.

EDIT: got detoured after starting this reply; others have said similar.

[tubenit]

Had you tried anything like that with the 6BM8?

I personally have not but have wanted to.  I did find 3 different reportedly "proven" 6BM8 reverb schematics that leave me to believe this
one I drew might work also?

Yeah, I noted that the tone stack is sort of a "quasi, kind of, sorta" 5E3. 

I'm thinking with a careful build & the right speaker this may be a very very nice amp tonally!  I sure like the idea and design behind it.
Might be cool with KT66 power tubes also?

With respect, Tubenit

[tubenit]

NOTE:  I added a layout with schematic of modified "Specialist"  (see above - Reply #5).

Saw this post describing this amp on another forum:

This amp was one of the very first Marshall with PCB, loaded with one Celestion G15.
 Last year I replaced the noisy volume pot on one of these amps. Amazing clean sound, deep and warm that works well when you play a Telecaster for instance. The reverb and trem do both work well, with a good useful speed and depth range on the trem.

I'm thinking this could be a really great amp.   with respect, Tubenit

[HotBluePlates]

Might be cool with KT66 power tubes also?

If it were my build, I'd use EL34's first, then try plugging in the KT66 or 6L6 afterwards to see how it performs. Since all of the above are in the same power output class, B+ and OT primary impedance should not be an issue for tube swaps.

What might be more important is the other key consideration that's important in output tube swaps (and which we often ignore here): transconductance. At 250v plate and screen, the EL34 has a Gm of 11mA/V, the KT66 comes in at 7mA/V, and the 6L6 has a Gm of 6mA/V. What these numbers mean is that the KT66 (assuming the tube used emulates actual, original KT66's) requires slightly less grid bias and drive voltage for the same performance compared to the 6L6. But the EL34 has nearly double the Gm of the other two, so bias and drive voltage is much lower.

For a simple circuit, the EL34 may give more satisfying distortion performance. And given the trouble were familiar with regarding difficulty of getting sufficient bias-vary trem with 6L6's, the trem likely works much better with EL34's. That is, given their high Gm, a smaller trem signal will likely vary their bias sufficiently to get deep trem while the same trem signal on 6L6's might seem to lack depth.

[mresistor]

ECL86 aren't too bad used on fleabay.

[tubenit]

For a simple circuit, the EL34 may give more staisfying distortion performance. And given the trouble were familiar with regarding difficulty of getting sufficient bias-vary trem with 6L6's, the trem likely works much better with EL34's. That is, given their high Gm, a smaller trem signal will likely vary their bias sufficiently to get deep trem while the same trem signal on 6L6's might seem to lack depth.

HBP, 

Thank you, my friend!  I appreciate the useful information. That explains why the EL34 "tremolo" sounded so reasonably rich with one 12AX7 triode on the You Tube.

With respect, Tubenit

[TIMBO]

Hey Tubenit, I don't think you can go past the 6BM8 as a reverb driver. I am in the throws of getting the old Fi Sonic reverb sorted.
I think this is another example of a bit of bad engineering, but when working, an excellent reverb.
With the other fi sonics having the 500k pot position per schem gave plenty of reverb but only with the pot at about "3" on the dial anything above that was terrible.
With this build "3" is great but anything above "4-5" I got aching across the socket pin6&7. I don't know if that was the same as the other fi sonics. (This will be a project in the future to rectify this problem)
This issue is being look at by one of the forum members over here, as it is way over my head.

I am trying a mod of removing the 500k pot and installing a voltage divider , these values are measured off the pot as a best setting for the reverb.
It was suggested that a pot on the input of the recovery as per the Goldentone schem, this mod is great with a good range of reverb from a touch of FATTNESS to a full DEEP RICH TONE that is not too SURFY and TWANGY.
Cause I am a NUFF when it comes to the "MATH" behind this stuff, I can only hope I'm doing the right thing.
Overall, I think my voltages are bit high, so I'm in the process of tweaking that.

[Ritchie200]

From the History of Marshall book
"This amp was intended for the club or jazz guitarist, hence the 15" speaker.....it was the first Marshall to use a PCB,  but overheating problems finished it in 1973 after about a yeat and a half of production."

Jim

[Ritchie200]

Speaking of interesting designs, the Marshall 4001 Studio 15 was pretty unique  in its use of 6V6's, xlr out, and the ability to plug the speaker into the headphone jack for an attenuated volume on 10 output tube distortion.   But I dont want to get into the attenuation discussion again.....

Jim

[DummyLoad]

use an EL84 and add a 12AX7 - use the second 1/2 of the 12AX7 as a switchable gain stage.     


--pete

[jazbo8]

Does anyone know the specs for the reverb transformer and tank used in the Marshall (not from the re-drawn schematic)?

[tubenit]

I think the Fender reverb tranny and tank would work well in this circuit.

Sluckey's response to my inquiry regarding the original reverb design.

With respect, Tubenit

[jazbo8]

I think the Fender reverb tranny and tank would work well in this circuit.

Sluckey's response to my inquiry regarding the original reverb design.

With respect, Tubenit

Thanks, but I'm not sure if he was referring to the original Marshall design or your re-drawn version... in any case, is there a place where we can look up the specs of the transformers that were used by Marshall?