Tweet
Hello guys, i recently acquired a Mesa 50/50 on the cheap and since I want to be able to use it with both American and British flavored preamps and I will never use a stereo setup, i thought it would be a great thing to mod one side to be able to accept EL34s. I have done all of the calculations but would really appreciate your input in case I am missing something. This is going to be a long post, so by all means no need for each person to answer everything - every contribution will be appreciated. Let's jump right into..
For the record, my amp was made in '89, so it is version A. In version C they introduced a Deep mod from the factory. I have attached the schematic which includes both the original version and version C.
Transformers and choke:
Starting with the basics, Mesa is a bit cryptic about the P.T. specs, so going from an equivalent replacement from TAD, I will assume that the specs are the same, so:
350-0-350 @ 400mA and 0-6.3 @ 5A.
Quick calculation for heater current:
- Stock with 4x6L6GC, 3x12AX7 ~= 4.6A
- With 2x6L6GC, 2xEL34, 3x12AX7 ~= 5.8A
So, the first conclusion is that the stock PT cannot support the extra current if 2 tubes are EL34. For this reason and taking into consideration the space limitations inside the chassis, I can get to fit a Hammond 266M6 ( 3.15-0-3.15 @ 3A ), which I plan to use to power the EL34 pair exclusively (so let's say sockets 3 and 4 - i will disconnect the stock heater rails and power those two sockets with Hammond)
Moving to the H.T. winding, according to my calculations it should be O.K. For 450V on the plates, per the schematic and calculating for 100% dissipation for both types (30W 6L6GV - 25W EL34) and then giving a 150% safety margin for operation, the current demands are actually lower in case there are 2x6L6GC, 2xEL34, 3x12AX7 instead of 4x6L6GC, 3x12AX7 in the amp. So, since the HT is enough as is, it should be OK with the EL34s also.
Regarding the choke, again Mesa is cryptic but as far as I could gather, they are using a choke similar to Fender 125C1A, so 4H @ 90mA, 105 Ohm DCR.
Taken from TDSL Duncanamps, the screen draw + preamp draw for each case is:
4x6L6GC, 3x12AX7 = 2*(5.6...22) + 6*1.2 = 18.4...51.2 mA (... denotes a range)
2x6L6GC, 2xEL34, 3x12AX7 = (5.6...22) + (8...50) + 6*1.2 = 20.8...79.2mA
So again, even in worst case scenario, the draw should not exceed 80mA, so again the choke should do fine. (There would be a problem with 4xEL34s according to my calculations but with only one EL34 pair it should be OK).
Regarding the Deep Mod (which in reality involves only adding the choke, converting the first preamp cathode from 1K/0.15uF to 2K7/15uF and adding a 0.003uF 'resonance' cap in the Feedback Loop and a resistor after the 1st stage coupling cap) and the general voicing of the amp, do you see any reason it wouldn't sound good also with EL34s in there? This is one area that I miss experience - assessing the voicing of the amp.
Regarding bias, it will be made adjustable, with one trimpot setting the "overall" bias (affecting both the first 6L6 pair and the second EL34 pair) and a second trimpot used for fine-tuning the bias of the second pair. That means that the second pair can be also adjusted to 6L6 specs, when the trimpot wiper is at the edge, thus practically connected to the 6L6 bias point (thus making possible to use either 6L6 or EL34 on the second channel, as desired). According to my quick calculations, the 250k trimpot in parallel shouldn't affect the overall resistance too much and the 50k trimpot with the 50k in series should give enough room for the 6L6 adjustment. Does that make sense to you?
Regarding the Resonance pot I intend to install, as stated version C already had added a 0.003uF cap in series with the presence pot, providing more resistance to the lower frequencies and thus lowering the feedback applied for those frequencies, making the amp sound bassier/fuller. I thought about making this adjustable by wiring a 1MA pot as variable resistance parallel to the 0.003uF. This would act as a resonance/depth control right?
Added bonus: at zero the cap is shorted, so in essence i will also have the stock version.
Regarding the screen grid resistors for the EL34, i think it would be wise to increase them to 1K from 470 Ohms, right?
As far as control grid resistors are concerned, initially i thought the 220k was a typo from Mesa, but after opening the amp, they are indeed 220k! I guess there is no reason to change them right? According to what I read from Aiken, such large values would be mostly chosen to avoid parasitic oscillations etc, so it doesn't seem to me I should lower them to 5k6 for example. Would you agree?
Lastly, in version C there is an added resistor after the 1st stage coupling cap (R228 - 220k). What is the effect of this resistor? To limit the signal a bit? Should i include it?
Sorry for the huge post, any input is really appreciated. I didn't include all my calculations but of course I can provide them if something is not clear.
Thanks for taking the time to read.
For the record, my amp was made in '89, so it is version A. In version C they introduced a Deep mod from the factory. I have attached the schematic which includes both the original version and version C.
Transformers and choke:
Starting with the basics, Mesa is a bit cryptic about the P.T. specs, so going from an equivalent replacement from TAD, I will assume that the specs are the same, so:
350-0-350 @ 400mA and 0-6.3 @ 5A.
Quick calculation for heater current:
- Stock with 4x6L6GC, 3x12AX7 ~= 4.6A
- With 2x6L6GC, 2xEL34, 3x12AX7 ~= 5.8A
So, the first conclusion is that the stock PT cannot support the extra current if 2 tubes are EL34. For this reason and taking into consideration the space limitations inside the chassis, I can get to fit a Hammond 266M6 ( 3.15-0-3.15 @ 3A ), which I plan to use to power the EL34 pair exclusively (so let's say sockets 3 and 4 - i will disconnect the stock heater rails and power those two sockets with Hammond)
Moving to the H.T. winding, according to my calculations it should be O.K. For 450V on the plates, per the schematic and calculating for 100% dissipation for both types (30W 6L6GV - 25W EL34) and then giving a 150% safety margin for operation, the current demands are actually lower in case there are 2x6L6GC, 2xEL34, 3x12AX7 instead of 4x6L6GC, 3x12AX7 in the amp. So, since the HT is enough as is, it should be OK with the EL34s also.
Regarding the choke, again Mesa is cryptic but as far as I could gather, they are using a choke similar to Fender 125C1A, so 4H @ 90mA, 105 Ohm DCR.
Taken from TDSL Duncanamps, the screen draw + preamp draw for each case is:
4x6L6GC, 3x12AX7 = 2*(5.6...22) + 6*1.2 = 18.4...51.2 mA (... denotes a range)
2x6L6GC, 2xEL34, 3x12AX7 = (5.6...22) + (8...50) + 6*1.2 = 20.8...79.2mA
So again, even in worst case scenario, the draw should not exceed 80mA, so again the choke should do fine. (There would be a problem with 4xEL34s according to my calculations but with only one EL34 pair it should be OK).
Regarding the Deep Mod (which in reality involves only adding the choke, converting the first preamp cathode from 1K/0.15uF to 2K7/15uF and adding a 0.003uF 'resonance' cap in the Feedback Loop and a resistor after the 1st stage coupling cap) and the general voicing of the amp, do you see any reason it wouldn't sound good also with EL34s in there? This is one area that I miss experience - assessing the voicing of the amp.
Regarding bias, it will be made adjustable, with one trimpot setting the "overall" bias (affecting both the first 6L6 pair and the second EL34 pair) and a second trimpot used for fine-tuning the bias of the second pair. That means that the second pair can be also adjusted to 6L6 specs, when the trimpot wiper is at the edge, thus practically connected to the 6L6 bias point (thus making possible to use either 6L6 or EL34 on the second channel, as desired). According to my quick calculations, the 250k trimpot in parallel shouldn't affect the overall resistance too much and the 50k trimpot with the 50k in series should give enough room for the 6L6 adjustment. Does that make sense to you?
Regarding the Resonance pot I intend to install, as stated version C already had added a 0.003uF cap in series with the presence pot, providing more resistance to the lower frequencies and thus lowering the feedback applied for those frequencies, making the amp sound bassier/fuller. I thought about making this adjustable by wiring a 1MA pot as variable resistance parallel to the 0.003uF. This would act as a resonance/depth control right?
Added bonus: at zero the cap is shorted, so in essence i will also have the stock version.
Regarding the screen grid resistors for the EL34, i think it would be wise to increase them to 1K from 470 Ohms, right?
As far as control grid resistors are concerned, initially i thought the 220k was a typo from Mesa, but after opening the amp, they are indeed 220k! I guess there is no reason to change them right? According to what I read from Aiken, such large values would be mostly chosen to avoid parasitic oscillations etc, so it doesn't seem to me I should lower them to 5k6 for example. Would you agree?
Lastly, in version C there is an added resistor after the 1st stage coupling cap (R228 - 220k). What is the effect of this resistor? To limit the signal a bit? Should i include it?
Sorry for the huge post, any input is really appreciated. I didn't include all my calculations but of course I can provide them if something is not clear.
Thanks for taking the time to read.
Comment