Modified Dynaco Mkiii

[TriodeLuvr]

The LTP with a SS ccs absolutely forces AC balance in the LTP stage. The global feedback should work on the output stage issues.
John

Yes, NFB helps, but balance and distortion can usually be improved further with the balance control. This can be demonstrated by balancing an amp and then transposing the two output tubes. Only a very small difference in the tubes (less than I've been able to measure by any other means) can result in a 10 dB or greater increase in even-order energy. Whether this issue is audible in isolation, I don't know, but I believe the sum of seemingly minor design differences like this can make or break an amp.

Jack

 

[spantou01]

These amps belong to my father and about 4 years ago he gave them to me. He was tired of messing and constantly adjusting them.

Thanks for your post! I enjoyed looking at the ADI CCDA conversion. To me it looks like a careful design that could be an improvement on the original Dyna circuit. If it were mine I would not be too hasty to abandon it, so my vote is for replacing those caps and then evaluate the performance. The EL34s will work well with the Mk III output transformers, so if you have enough power for your speaker system there is no need for the larger output tubes.

I’m not sure what needs “constant adjusting”, other than keeping an eye on the power tune bias. Am I overlooking something?

 

[Pio1980]

I dunno if 6CA7s will handle the voltages here as well as the EL34, but I do prefer them where appropriate with the Dynaco UL opts.

 

[kward]

I like the idea generally of differential amplifiers all the way from the frontend stage clear through to the output. If well executed it should produce very good signal to noise ratio. The specs on the original Mk III amp I believe said 90 dB signal to noise, which is already pretty darn good so I'm not sure differential amplifiers in the frontend is buying you a whole lot there. That's the thing I worry about in these types of mods. The design is a good bit more complex than the stock design, but what are you getting that makes the amp better because of it?

I don't care for the fact that it's got two stages of R/C coupling. That's just one more R/C stage to get in the way when tuning the amp.

I would have preferred to keep the original output stage, but EL34/6CA7/KT77 I think bring some interesting tube rolling to the table.

The real deal I think though, is in how well it's tuned. If the amp is stable both in HF and LF with arbitrary speaker loads, it could be a great design. If the OP has the means, I'd definitely test/measure it first before making a decision on what to do. It could be that just a few minor tweaks are needed to turn it into a real performer. You can always take it back to the stock design later, so I'd vote for exploring it to determine if it stands up to the test of stability and performance.

It better meet or exceed all the original specs (except for output power because of the EL34 conversion) or it's a step backward in my mind.

 

[Pio1980]

Have both a rebuilt otimized OEM Dynakit 70 and VTA 70, both w/6CA7s, I do find the VTA subjectively more "transparent", particularly "tightened up" with increased FB over the issued factor as per @peterh, taking a good amp and making it better.

 

[Pio1980]

I dunno if 6CA7s will handle the voltages here as well as the EL34, but I do prefer them where appropriate with the Dynaco UL opts.

The 6CA7 is structurally a downscaled 6550, I Wonder why he spec'ed the EL34 instead of keeping the 6550s.

 

[Lavane]

The 6CA7 is structurally a downscaled 6550, I Wonder why he spec'ed the EL34 instead of keeping the 6550s.

The OP stated the availability of good KT-88's at the time back in the 70's.
But as kward posted, as it is now, there are more options for output tubes.
EL-34's, 6CA7's, KT-77's and maybe KT-66's. I have a question though, can't 6550's and KT-88's still be used? The output wouldn't be any higher, but I would think they would bias up ok. I ran KT-88's in my ST-70 for a short time. All it took was a bias adjustment.

 

[rickl]

I have a question though, can't 6550's and KT-88's still be used?

The schematic posted by the OP does show 6550s (and 6cf7 PI and 12ax7 driver). I'd get it working as is and then decide if you want any changes. My guess it will sound pretty nice. I know my mark iii's improved a lot when I went to Curcio 'upgrade' boards. I have 4 mono blocks so it was pretty easy to compare the pair of originals and a pair of Curcio amps.

 

[Lavane]

The schematic posted by the OP does show 6550s (and 6cf7 PI and 12ax7 driver). I'd get it working as is and then decide if you want any changes. My guess it will sound pretty nice. I know my mark iii's improved a lot when I went to Curcio 'upgrade' boards. I have 4 mono blocks so it was pretty easy to compare the pair of originals and a pair of Curcio amps.

Then I would expect KT-88's and 6550's can be used as well as EL-34 types then.

 

[kward]

If the amp was tuned for say EL34s, that means it probably is not optimally tuned for 6550's or KT88s. I'm talking about LF and HF tuning, not about biasing the output stage (although that's also a consideration). Anyway, that's assuming it was even a consideration with the new driver circuit. As I mentioned above that's what worries me about these types of mods--not sure the amp was optimally tuned for the changes that were made. It's kinda like running regular gas in your Lexus. It may work but not optimally..

 

[Lavane]

Thats what I found with my ST-70. KT-88's just didn't sound quite right.
The same tubes sound just fine in the mono blocks I have that were designed to use them though.