Hey All,
I'm planning to go back into my Mark VI's again. These were the first amps I ever worked on and I'm sure I didn't a very good job. I've learned a lot since then. I have three questions.
The first is the rectifier. The amp uses a full wave rectifier (see attached) the diodes are original and I would like to replace them. What type should I use?
The second question regards the G2 connection. I'm using KT88's and have read on another thread that a good mod is to put a Zener between the UL tap and pin 4. Is that a good mod? TDSL shows 553 volts on plate but 300 volts on G2. My amps run 545 B+ How much should I reduce G2 voltage?
And lastly, I'm using a 12AU7/12AU7 LTP front end. I want to stay with the LTP but is there a better tube I could be using? Possibly a pentode preamp tube? EF86? And a 6CG7 phase splitter?
Thanks, Kevin
I'm planning to go back into my Mark VI's again. These were the first amps I ever worked on and I'm sure I didn't a very good job. I've learned a lot since then. I have three questions.
The first is the rectifier. The amp uses a full wave rectifier (see attached) the diodes are original and I would like to replace them. What type should I use?
The second question regards the G2 connection. I'm using KT88's and have read on another thread that a good mod is to put a Zener between the UL tap and pin 4. Is that a good mod? TDSL shows 553 volts on plate but 300 volts on G2. My amps run 545 B+ How much should I reduce G2 voltage?
And lastly, I'm using a 12AU7/12AU7 LTP front end. I want to stay with the LTP but is there a better tube I could be using? Possibly a pentode preamp tube? EF86? And a 6CG7 phase splitter?
Thanks, Kevin
Attachments
So you changed the front end to a "long tailed pair", or is it still the stock concertina, which originally used the 7199?
jeff
I ordered MUR4110 diodes, 1000PRV 4 watt and 1n5948 Zeners 91 volt 3 watt to knock down G2 voltage. I'll see how those work.
I would replace the original silicon diodes with UF4007. The originals had two in series but the UF4007's are rated for 1000 volts so you do not need two in series.
The screens of a KT88 are rated for 600 volts max. I think they would be fine without zeners. My PPP amp just uses 100 ohm screen stopper resistors and 4.7k grid stoppers. Be sure those are there to stop RF anomalies.
My driver uses a 12AX7 and a 12BH7 phase splitter and I am getting 0.25% THD with a KT90 quad at 138 watts out.
I am not sure of the leakage inductance of the MK VI's transformer but keep an eye on stability when tweaking the driver.
The screens of a KT88 are rated for 600 volts max. I think they would be fine without zeners. My PPP amp just uses 100 ohm screen stopper resistors and 4.7k grid stoppers. Be sure those are there to stop RF anomalies.
My driver uses a 12AX7 and a 12BH7 phase splitter and I am getting 0.25% THD with a KT90 quad at 138 watts out.
I am not sure of the leakage inductance of the MK VI's transformer but keep an eye on stability when tweaking the driver.
No, its my own design. As far as I am concerned, A MK VI with proper mods can deliver similar numbers assuming the iron is good enough.
From what I can tell, Dynaco used 8417 tubes to meet a price point with their trademark pentode-triode driver. The cathodyne phase splitter produces limited headroom so they used 8417's which require less swing to drive them to full power.
If you ask me, the MK IV was done on the cheap. Looking at the original SPEC, the MK IV transformer was cheapish also. Only up to 15k at 120 watts, I'd be suspicious of the iron.
If you ask me, the MK IV was done on the cheap. Looking at the original SPEC, the MK IV transformer was cheapish also. Only up to 15k at 120 watts, I'd be suspicious of the iron.
12BH7 is one of my favorite tube types. The aesthetics is one reason.
I can have my 12BH7 be short and my 12BH7 tall. I love to lay out my chassis so everything is nice and symmetrical.
The pictures are of my "Dynaco on Steroids" amplifier. I just completed the second block and with my KT90 quad I am getting 0.16% THD at 140 watts at 1k. The other delivers 0.25% THD. Changing around the tubes statistically alters the numbers slightly as all tube characteristics vary.
I can have my 12BH7 be short and my 12BH7 tall. I love to lay out my chassis so everything is nice and symmetrical.
The pictures are of my "Dynaco on Steroids" amplifier. I just completed the second block and with my KT90 quad I am getting 0.16% THD at 140 watts at 1k. The other delivers 0.25% THD. Changing around the tubes statistically alters the numbers slightly as all tube characteristics vary.
Since the transformer is FWCT, the diode voltage rating needs to be at least 2*sqrt2 times the transformer secondary. 1000V rated parts may be marginal/undersized.
If you have a 370-0-370 transformer; 370 X 2.82=1043V. With a 400-0-400, 400 X 2.82=1128V.
I would consider 1200V 8A Fairchild Stealth diodes or something equivalent.
Mouser P/N 512-ISL9R8120P2
isl9r8120p2 Rectifiers | Mouser
There are also CREE schottky 1200V diodes
Digikey P/N C2D05120A
Hey Guys,
I ordered 8 of the 1200v 8a Stealth diodes from Mouser. I also ordered 8 51 volt 5 watt zener diodes from Mouser. That gives me options on g2 voltage. Can I get away with just two of these for each amp? On another post I found mention of a zener to knock down G2 voltage to minimize IM distortion. Since I am fighting IM distortion, (I think) is this a good idea? Can anyone tell me which of my choices, 51 volts or 91 volts might work better? or would the 91 volt just cost me more power?
Kevin
I ordered 8 of the 1200v 8a Stealth diodes from Mouser. I also ordered 8 51 volt 5 watt zener diodes from Mouser. That gives me options on g2 voltage. Can I get away with just two of these for each amp? On another post I found mention of a zener to knock down G2 voltage to minimize IM distortion. Since I am fighting IM distortion, (I think) is this a good idea? Can anyone tell me which of my choices, 51 volts or 91 volts might work better? or would the 91 volt just cost me more power?
Kevin
Hey Guys,
I've heard it said that the Mark VI output transformers are not the highest quality. But I wonder? The Audio Research D75 is a great tube amp and its frequency response is 20-15k with 1% distortion at 75 watts and it sounds great. The McIntosh MC275 is rated at 16-400k at rated output with .1% distortion and it sounds dry an lifeless.
Sy suggested an EF86 and 6CG7 combination Ala Mullard 5-20. But NOS EF86 tubes often suffer serious microphonics.
Isn't it possible to design the amp to use different dual triodes like 12AT7, 12BH7, 12AU7 or 6CG7? Then I could listen to what sounds best? Taking a page from Broskie? I guess the thing to do would be to design the front for each possible tube and see how really different they would be?
Kevin
I've heard it said that the Mark VI output transformers are not the highest quality. But I wonder? The Audio Research D75 is a great tube amp and its frequency response is 20-15k with 1% distortion at 75 watts and it sounds great. The McIntosh MC275 is rated at 16-400k at rated output with .1% distortion and it sounds dry an lifeless.
Sy suggested an EF86 and 6CG7 combination Ala Mullard 5-20. But NOS EF86 tubes often suffer serious microphonics.
Isn't it possible to design the amp to use different dual triodes like 12AT7, 12BH7, 12AU7 or 6CG7? Then I could listen to what sounds best? Taking a page from Broskie? I guess the thing to do would be to design the front for each possible tube and see how really different they would be?
Kevin
How are you getting those numbers with the MC275? I own an MC30 and it indeed sounds dry and lifeless. I totally know what you mean.
The ratings you posted for the MC275 are fallacious, unless you are referring to something else like IMD.
McIntosh MC275 Amplifier, Vacuum Tube Amplifiers, Tube Amplifiers Mcintosh MC275
Maybe your own tested with .1% THD with the right tube combination, etc.
Mr. Frank McIntosh's circuit was innovative but his intent was never really to deliver the "tube sound." He was for the time more "forward thinking" like fender was with their blue solid state guitar designs of the late 60s. Consequently, his designs sounded dry and transitory. Note this is just an argument and my take on hi-end audio history. Maybe some of you will see things through other lenses.
A think to try perhaps is to reduce the negative feedback the Macs use. They may open up more. In many designs this seems to happen.
Regarding your driver:
- I recommend a 12AX7 with a 12BH7 phase splitter.
- My universal driver topology allows for any tube. You just have to change the resistor combinations and driver B+ voltages for proper biasing, etc.
- Pentodes in driver circuits sound too dry and harsh for hi-fi, so I stay clear of them.
The ratings you posted for the MC275 are fallacious, unless you are referring to something else like IMD.
McIntosh MC275 Amplifier, Vacuum Tube Amplifiers, Tube Amplifiers Mcintosh MC275
Maybe your own tested with .1% THD with the right tube combination, etc.
Mr. Frank McIntosh's circuit was innovative but his intent was never really to deliver the "tube sound." He was for the time more "forward thinking" like fender was with their blue solid state guitar designs of the late 60s. Consequently, his designs sounded dry and transitory. Note this is just an argument and my take on hi-end audio history. Maybe some of you will see things through other lenses.
A think to try perhaps is to reduce the negative feedback the Macs use. They may open up more. In many designs this seems to happen.
Regarding your driver:
- I recommend a 12AX7 with a 12BH7 phase splitter.
- My universal driver topology allows for any tube. You just have to change the resistor combinations and driver B+ voltages for proper biasing, etc.
- Pentodes in driver circuits sound too dry and harsh for hi-fi, so I stay clear of them.
Last edited:
kingneb,
The figures for the Mcintosh are from the original manual that I found on Hi Fi engine. They seemed a bit outrageous to me but that's what they published. The best sounding amp I ever built was a Marantz 8b clone. It used a 6BH6 preamp tube and a 6CG7 phase splitter. The 6CG7 is a nine pin version of 6SN7. I'm new to designing but I want to see what I can come up with that will allow multiple triodes.
The figures for the Mcintosh are from the original manual that I found on Hi Fi engine. They seemed a bit outrageous to me but that's what they published. The best sounding amp I ever built was a Marantz 8b clone. It used a 6BH6 preamp tube and a 6CG7 phase splitter. The 6CG7 is a nine pin version of 6SN7. I'm new to designing but I want to see what I can come up with that will allow multiple triodes.
Last edited:
There's nothing magic about EF86. Any high gm sharp cutoff pentode will work in the first position.
it is not the tube mu alone that determines circuit voltage gain, choice of operating points and load resistors does...
- Status
- Not open for further replies.