In the March 2001 issue of Audio Xpress there was an article titled "The Cure for the Common Dynaco MKIII". Kara Chaffee described a modification for the Mark III which replaced the 6AN8 driver. It was supposed to be a relatively inexpensive mod which improved the sound.
Does anyone have any specific details about this article? I'm curious to know exactly what the mod involved.
Does anyone have any specific details about this article? I'm curious to know exactly what the mod involved.
The modification involves the front end of the amp and a slight mod to the feedback circuit. The input 6AN8 is replaced by one half of a 12AT7 driving a 6CG7/6FQ7 Schmitt phase splitter. The feedback resistor is changed to a pot to allow adjustment. ( I see this technique shown on a number of mod circuits to tube amps. I wonder how many folks realize that changing the "feedback" in this fashion, also varies the bias on the input tube.) Additionally the author adds another filter stage to the power supply. These are probably all worthwhile changes to improve the MkIII.
hey-Hey!!!,
The amp is a complete package. If you improve a piece of it, the shortfalls of the rest become more obvious. If you insist on using the Iron, put it on a chassis that has room to work in. Once that is done, there is far less limit to the sorts of mods available to you.
cheers,
Douglas
The amp is a complete package. If you improve a piece of it, the shortfalls of the rest become more obvious. If you insist on using the Iron, put it on a chassis that has room to work in. Once that is done, there is far less limit to the sorts of mods available to you.
cheers,
Douglas
SY said:
I have a pair MkIII, and need new power-trafos...for 230V.
Can you please suggest were I can find a suitable upgrade?
I will probably use a SS rectifier with relay/delayed HV.
and new 550V JJ caps...or a cap-upgrade board, and have already finished a pair "diytube.com" "Poseidon" driver boards.
I have been tempted to run the two 115V trafos in series...but afraid that one will "over-power" if the loads are different?
Arne K
You probably don't want to put the primaries of two different transformers in series.
Two major PS upgrades are desirable: first, larger reservoir capacitance. That will be the biggest bang-for-buck. Increasing the decoupling capacitance for the driver stage is also very worthwhile. Second, regulation of the B+ and the bias supply. Bias supply regulation is easy, needing nothing fancier than a Zener diode. B+ regulation is a bit more complex, but still not overwhelming- check out Joe Curcio's reg for the Stereo 70 published some years ago in Glass Audio.
If it were me, I'd use something a little simpler like a Maida reg or a simple series regulator with a one transistor error amp. At minimum, regulate the driver stage supplies, but best practice is to regulate the output stage supply as well.
One thing important to note- output and bias supplies should both be regulated or neither should be. Don't just do one.
Two major PS upgrades are desirable: first, larger reservoir capacitance. That will be the biggest bang-for-buck. Increasing the decoupling capacitance for the driver stage is also very worthwhile. Second, regulation of the B+ and the bias supply. Bias supply regulation is easy, needing nothing fancier than a Zener diode. B+ regulation is a bit more complex, but still not overwhelming- check out Joe Curcio's reg for the Stereo 70 published some years ago in Glass Audio.
If it were me, I'd use something a little simpler like a Maida reg or a simple series regulator with a one transistor error amp. At minimum, regulate the driver stage supplies, but best practice is to regulate the output stage supply as well.
One thing important to note- output and bias supplies should both be regulated or neither should be. Don't just do one.
You'll want to excise that 525V quad cap before replacing the 5AR4 with a bridge. The higher B+ also means KT-88s not 6550 for outputs. I was able to fit ten 47µF, 450V radial caps on 2.5" x 4" board that fits nicely under the PT of the Mark III to give me 50-25-25-25 µF at ~900WVDC. I left the liberated quad cap in place as a reminder of what a flawed component it is. With 122VAC coming out of the wall I see pin #8 of my 5AR4 at 525V all day long.
I just finished the cleanup / upgrade of my first pair of MkIII, and WOW !!! -this is a totally different "beast" than a ST-70 !
New power-trafos for 230V, "Poseidon" pcb,( diytube.com :: View Forum - poseidon ) and PSU-cap-boards.
If I had known the difference (from ST-70) was that big, I would have finished them long ago!
Arne K
New power-trafos for 230V, "Poseidon" pcb,( diytube.com :: View Forum - poseidon ) and PSU-cap-boards.
If I had known the difference (from ST-70) was that big, I would have finished them long ago!
Arne K
I know what you mean. I had never heard Mark IIIs before I finished reworking the set that I rescued from an outdoor shed a few weeks back. They sound nothing like any ST-70 I have heard
I'm helping someone who bought a good used pair as his first tube amps for a nice price with recently new boards and can caps which look like the stock circuit and are probably dynakit replacement boards.
The previous owner recommended he replace the output tubes as they were failing so he purchased reissue TS 6550's. One immediately red plated a bit and did the same when tried in different positions. He said the tube seller will take it back and a new one is on the way.
He put one of the best of the old KT-88's in it's place and the amp works biased at 1.40.
I told him I would have bought KT-88's as they can take a bit more.
I recommended that he bias it really low and turn it up till it sounds good. He found it's not bad at 1.36. Sounded best at the original factory setting he thought (1.56) Maybe as tubes break in they will get better sounding at a lower setting.
Will these tubes last at a lower setting? I read a comment a few years ago on this form from someone who owned a pair yrs. ago and thought they were hard on output tubes so got rid of them.
I realize it's a dissipation issue related to the high B+ (close to 500V) on these amps and lowering bias helps to a point.
Is he best to sell the 6550's for KT-88's or do as I thought and lower the B+ by using a inexpensive Antec toriod for about $40. They have more Amperage in that price range for better PS headroom too.
I thought he could keep stock PS trans for fil. heaters and Bias V making them run cool and run forever.
An available 350-0-350 250 mA or better high amp one 360-0-360 450 mA ( but out of stock for a while) would give a B+ of 430-450 V depending on line V too. Stock is 400-0-400 @ 200mA.
The big 450 mA 325-0-325 one would work with SS diodes in place of 5AR4.
The 6.8K R to the 6AN8A might have to be changed.
Toriods could be enclosed in boxes or round covers over metal plates and hidden behind amps.
Is the above a bad idea? It would allow him to roll many different tubes and even get longer life out of KT-88's.
After all HK CIt II had a lower B+ 0f about 460-470 I believe. Maybe a bit lower than this which one could end up with will sound good too?
Owner doesn't mind a bit less power if tubes last.
How dependent on B+ is sonics? I think from reading many new production tubes can't take what some of the NOS could take.
I thought of a big choke , but that wouldn't help much and would just raise the DCR of the PS. Maybe a better choke with lower DCR could be used too. Say 10-20 ohms, and it would add 5-7 volts too the B+ if needed.
Thanks very much in advance,
Randy
The previous owner recommended he replace the output tubes as they were failing so he purchased reissue TS 6550's. One immediately red plated a bit and did the same when tried in different positions. He said the tube seller will take it back and a new one is on the way.
He put one of the best of the old KT-88's in it's place and the amp works biased at 1.40.
I told him I would have bought KT-88's as they can take a bit more.
I recommended that he bias it really low and turn it up till it sounds good. He found it's not bad at 1.36. Sounded best at the original factory setting he thought (1.56) Maybe as tubes break in they will get better sounding at a lower setting.
Will these tubes last at a lower setting? I read a comment a few years ago on this form from someone who owned a pair yrs. ago and thought they were hard on output tubes so got rid of them.
I realize it's a dissipation issue related to the high B+ (close to 500V) on these amps and lowering bias helps to a point.
Is he best to sell the 6550's for KT-88's or do as I thought and lower the B+ by using a inexpensive Antec toriod for about $40. They have more Amperage in that price range for better PS headroom too.
I thought he could keep stock PS trans for fil. heaters and Bias V making them run cool and run forever.
An available 350-0-350 250 mA or better high amp one 360-0-360 450 mA ( but out of stock for a while) would give a B+ of 430-450 V depending on line V too. Stock is 400-0-400 @ 200mA.
The big 450 mA 325-0-325 one would work with SS diodes in place of 5AR4.
The 6.8K R to the 6AN8A might have to be changed.
Toriods could be enclosed in boxes or round covers over metal plates and hidden behind amps.
Is the above a bad idea? It would allow him to roll many different tubes and even get longer life out of KT-88's.
After all HK CIt II had a lower B+ 0f about 460-470 I believe. Maybe a bit lower than this which one could end up with will sound good too?
Owner doesn't mind a bit less power if tubes last.
How dependent on B+ is sonics? I think from reading many new production tubes can't take what some of the NOS could take.
I thought of a big choke , but that wouldn't help much and would just raise the DCR of the PS. Maybe a better choke with lower DCR could be used too. Say 10-20 ohms, and it would add 5-7 volts too the B+ if needed.
Thanks very much in advance,
Randy
hey-Hey!!!,
The 360-0-360 Antek would work but it has no 5V winding. I'd call Heyboer Trans and ask for a 360-0-360 HV with 6.3 and 5V with capacity based on how thick a lam stack you think can be squeezed in there. You can probably go a half inch thicker. I'd also ask for a 125V primary. capacity will probably come out on the high side of 250 mA...
cheers,
Douglas
The 360-0-360 Antek would work but it has no 5V winding. I'd call Heyboer Trans and ask for a 360-0-360 HV with 6.3 and 5V with capacity based on how thick a lam stack you think can be squeezed in there. You can probably go a half inch thicker. I'd also ask for a 125V primary. capacity will probably come out on the high side of 250 mA...
cheers,
Douglas
"I thought he could keep stock PS trans for fil. heaters and Bias V making them run cool and run forever."
I should have said rect. heater too. That should work leaving the stock PS trans using the Antec just for the B+. I would think? I think the antec's are a lot less $ than the new reproduction ones with all the heater & bias taps. The ST-70 clones from Triode Electronics & Dynakit At about $130-135 ea would be perfect with 360-0-360 @300mA and I don't think anyone else would be much less. The Mark III clones are closer to $100 and they would bolt in and if Heyboer could make them to spec maybe he would except that. The only thing as I mentioned is that the Antec's would have to sit in boxes etc. behind amps. This would save drilling new holes in amp base too.
He's just wondering what kind of life would he get out of KT-88's in the stock amp say he set bias @ 1.40 ?
The reissue Tungsol KT-120 was another thought for the stock amp! I bet circuit wouldn't be up to it and so maybe sonics would suffer. PS tans may not be up to them either!
Thanks very much Douglas!
Regards,
Randy
I should have said rect. heater too. That should work leaving the stock PS trans using the Antec just for the B+. I would think? I think the antec's are a lot less $ than the new reproduction ones with all the heater & bias taps. The ST-70 clones from Triode Electronics & Dynakit At about $130-135 ea would be perfect with 360-0-360 @300mA and I don't think anyone else would be much less. The Mark III clones are closer to $100 and they would bolt in and if Heyboer could make them to spec maybe he would except that. The only thing as I mentioned is that the Antec's would have to sit in boxes etc. behind amps. This would save drilling new holes in amp base too.
He's just wondering what kind of life would he get out of KT-88's in the stock amp say he set bias @ 1.40 ?
The reissue Tungsol KT-120 was another thought for the stock amp! I bet circuit wouldn't be up to it and so maybe sonics would suffer. PS tans may not be up to them either!
Thanks very much Douglas!
Regards,
Randy
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.