Setting bias.
The amp uses fixed bias which may be adjusted at each output tube socket from under the chassis with great care using a mini slotted screwdriver. The pots are only supported by the circuitry itself. A DVM should be placed on millivolts DC range across each 10 ohm cathode resistor while setting the bias. I set for 35 mA quiescent (350 millivolts across the cathode R) when using Mullard 6CA7's. At that setting with these tubes the amp is capable of 28 watts RMS each channel and the power transformer does not overheat if the dress tube cover is left off. That cover causes a lot of radiant heat to impact the one side of the transformer enclosure from the output tubes. There is no ventillation to cool the transformers inside.
The amp uses fixed bias which may be adjusted at each output tube socket from under the chassis with great care using a mini slotted screwdriver. The pots are only supported by the circuitry itself. A DVM should be placed on millivolts DC range across each 10 ohm cathode resistor while setting the bias. I set for 35 mA quiescent (350 millivolts across the cathode R) when using Mullard 6CA7's. At that setting with these tubes the amp is capable of 28 watts RMS each channel and the power transformer does not overheat if the dress tube cover is left off. That cover causes a lot of radiant heat to impact the one side of the transformer enclosure from the output tubes. There is no ventillation to cool the transformers inside.
Re MC34B
Hi
Ive had this amp for just over a year now and find its sound quality very good considering the price.
Ive swapped out the 6n1p's for Svetlana winged C's and what an improvement! Now i want to swap the Sovtek 6L6GC's for something of much higher quality...what do you suggest / recommend?
Ive been thinking about Svet 6L6 winged C's or Phillips 5781a's.
Do you recommend tube dampers like the ones made by golden Dragon?
This is my first post on DIYaudio.com, Im a valve newbie and will need to find an Amp technician in my area to rebias...
is there a lot of info in the archives on this little amp and indeed on rebiasing in general?
Best Regards
Jack.
Hi
Ive had this amp for just over a year now and find its sound quality very good considering the price.
Ive swapped out the 6n1p's for Svetlana winged C's and what an improvement! Now i want to swap the Sovtek 6L6GC's for something of much higher quality...what do you suggest / recommend?
Ive been thinking about Svet 6L6 winged C's or Phillips 5781a's.
Do you recommend tube dampers like the ones made by golden Dragon?
This is my first post on DIYaudio.com, Im a valve newbie and will need to find an Amp technician in my area to rebias...
is there a lot of info in the archives on this little amp and indeed on rebiasing in general?
Best Regards
Jack.
You live in England. The famous Mullard valves came from there. Get some Mullard 6CA7/EL34's. Forget 6L6's. I used 6BK7B's to replace the long tail phase splitters (outermost small tubes on the chassis) and picked up a noticeable improvement in fine detail.
I tried 6DJ7's and 6922's. They sounded lifeless in either holes. I now have 6CG7's in the innermost small holes (input amp) and although I lost about 3 dB sensitivity which is no problem I have very good tone. There is probably a more magical tube that will go here but I haven't found it although I haven't tried.
I tried 6DJ7's and 6922's. They sounded lifeless in either holes. I now have 6CG7's in the innermost small holes (input amp) and although I lost about 3 dB sensitivity which is no problem I have very good tone. There is probably a more magical tube that will go here but I haven't found it although I haven't tried.
I don't recommend substituting 6CA7/EL34 for 6L6 without some other changes to the amplifier and some good scope work to make sure the amplifier is still stable. The 6CA7 has considerably higher transconductance than a 6L6, higher filament current, and requires a very different bias point for proper performance. 6L6 bias levels if fixed biased will put a 6CA7 very close to class b operation.
6CA7 are an option if the transformer filament winding has sufficient reserve to run them safely, but not without the precautions mentioned above.
I would take a look at the JJ 6L6GC as one choice as well as Philips/Sylvania ECG 7581A (they sound quite good in Citation V and several other amps I have tried imho) and are a drop in for 6L6GC. Another obvious choice would be one of the several repro KT66 floating around out there.
6CA7 are an option if the transformer filament winding has sufficient reserve to run them safely, but not without the precautions mentioned above.
I would take a look at the JJ 6L6GC as one choice as well as Philips/Sylvania ECG 7581A (they sound quite good in Citation V and several other amps I have tried imho) and are a drop in for 6L6GC. Another obvious choice would be one of the several repro KT66 floating around out there.
hello
Thanks for your posts
one of my Sovtek 6L6's is flashing to the music, i.e the electrons dim and brighten in time to bass and drum beats etc. the other three dont do it and it stops when i turn the sound down. Ive only just noticed this last night so not sure how long its been doing it. Is the tube going to fail or is this natural?
Regards
Jack.
Thanks for your posts
one of my Sovtek 6L6's is flashing to the music, i.e the electrons dim and brighten in time to bass and drum beats etc. the other three dont do it and it stops when i turn the sound down. Ive only just noticed this last night so not sure how long its been doing it. Is the tube going to fail or is this natural?
Regards
Jack.
Hi Jack,
That tube might doesn't sound healthy. Replace that channel's outputs.
I have had very good luck with the Electroharmonix 6L6EH. It looks very close to some 7581A's I have (another excellent choice). You may see 7581A also branded with KT-66, they are very close to the best of my knowledge. I hear JJ is another good brand, I haven't tried them.
-Chris
That tube might doesn't sound healthy. Replace that channel's outputs.
I have had very good luck with the Electroharmonix 6L6EH. It looks very close to some 7581A's I have (another excellent choice). You may see 7581A also branded with KT-66, they are very close to the best of my knowledge. I hear JJ is another good brand, I haven't tried them.
-Chris
My 6CA7 mod IS SAFE
Kevin,
I have had a MC34B purchased new for three years. For the past two years I have logged hundreds of hours with a set of 2nd generation, vintage good Mullard 6CA7's. I have the quiescent cathode current set at 35 mA per valve by adjustment of G1 bias available as a separate control for each output valve. I too was concerned about the extra filament current needs of these tubes and watched power transformer heating carefully. As it is set up my amp has been left running for as long as 24 hours with nothing but safe mild temperature rise. As tube amps go I consider this one to be a cool runner. Note that I did specify to leave the tube cover off to help the transformer enclosure cool.
A wattmeter check shows exactly the same consumption as the nameplate rating, so I must have ended up lower on my cathode current than stock with the 6L6's to maintain the same draw on the power xfmer.
With my Mullard 6CA7's and cathode current settings of 35 mA this amp now outputs 28 wartts rms per channel, both channels driven and sounds fantastic. It put out a measured 18 watts/ch with the SINO 6L6's out of the box.
I would not make a recommendation unless I had tried it myself with tested results. This works very well on this amp and I feel confident to recommend it.
That said, try any other brand 6CA7, particularly the newly manufactured junk and take your chances.
kevinkr said:
Kevin,
I have had a MC34B purchased new for three years. For the past two years I have logged hundreds of hours with a set of 2nd generation, vintage good Mullard 6CA7's. I have the quiescent cathode current set at 35 mA per valve by adjustment of G1 bias available as a separate control for each output valve. I too was concerned about the extra filament current needs of these tubes and watched power transformer heating carefully. As it is set up my amp has been left running for as long as 24 hours with nothing but safe mild temperature rise. As tube amps go I consider this one to be a cool runner. Note that I did specify to leave the tube cover off to help the transformer enclosure cool.
A wattmeter check shows exactly the same consumption as the nameplate rating, so I must have ended up lower on my cathode current than stock with the 6L6's to maintain the same draw on the power xfmer.
With my Mullard 6CA7's and cathode current settings of 35 mA this amp now outputs 28 wartts rms per channel, both channels driven and sounds fantastic. It put out a measured 18 watts/ch with the SINO 6L6's out of the box.
I would not make a recommendation unless I had tried it myself with tested results. This works very well on this amp and I feel confident to recommend it.
That said, try any other brand 6CA7, particularly the newly manufactured junk and take your chances.
I purcased a quad of brand new Sovtek 6L6WXT's to try in the MC34B.
Two of the four tubes failed in less than 50 hours. Garbage!!!! Am I impressed? NO! The sound was not as good as I get from the Mullard 6CA7's either.
I bought a quad of brand new fat bottle ElectroHARMonix 6CA7's. Two had glowing grid two segments that ran a bright yellow orange. Garbage!!!!
I am still trying to obtain one good set of ElectroHARMonix gold grid 300B's. In the first set one was fine and one does a lightshow every turn on from partial Grid cathode short. So far the one sent to replace it has the same problem. Garbage!!!!
Two of the four tubes failed in less than 50 hours. Garbage!!!! Am I impressed? NO! The sound was not as good as I get from the Mullard 6CA7's either.
I bought a quad of brand new fat bottle ElectroHARMonix 6CA7's. Two had glowing grid two segments that ran a bright yellow orange. Garbage!!!!
I am still trying to obtain one good set of ElectroHARMonix gold grid 300B's. In the first set one was fine and one does a lightshow every turn on from partial Grid cathode short. So far the one sent to replace it has the same problem. Garbage!!!!
Unfortunately Rob, we have opposite experiences.
Every Electroharmonix I use reinforces my good opinion of them. They are not the same as Sovteks. I am using a set of 6CA7EH in my Eico. Really excellent sound.
The 6L6GC family are known for vast reserves of power - as long as the circuit is designed for it. I use Electroharmonix in Fender Twins. Most other tube brands fail. I have only ever had one noisy 12AX7EH, everything else has been good.
300B's I can't vouch for. I'm a P-P man myself.
Don't know if that helps Kevin.
-Chris
Every Electroharmonix I use reinforces my good opinion of them. They are not the same as Sovteks. I am using a set of 6CA7EH in my Eico. Really excellent sound.
The 6L6GC family are known for vast reserves of power - as long as the circuit is designed for it. I use Electroharmonix in Fender Twins. Most other tube brands fail. I have only ever had one noisy 12AX7EH, everything else has been good.
300B's I can't vouch for. I'm a P-P man myself.
Don't know if that helps Kevin.
-Chris
My MC34-B experience
I was given one of these because it had stopped working, by an audiophile with many amps I found the fuse blown and worried about what had provoked this. But, it turned out there was nothing by way of a hard failure to blow the spare fuse, and I monitored cautiously while advancing the Variac supply, and monitoring the real input power with a Valhalla meter (indispensable for this kind of work!).
The fuse is a 2A fast blow, and I would conjecture that it opened by virtue of simply the number of power on-off cycles. The online schematic I found was inaccurate and showed cathode biasing of the 6L6 output tubes, although it was clear that this unit had fixed bias as the others discussed in this thread. I adjusted the bias for a fairly rich 45mA, although I may back that off a bit. Things do run quite warm, and I removed the cover as a matter of course. Fortunately I don't have kids running around.
As it has quite a lot of gain (almost 38dB) for normal line-level sources, I also experimented with reducing this a bit by returning the 100k feedback resistor from the 8 ohm output tap directly to the 6N2 cathodes, and clipping out the partial electrolytic bypass caps. With resistive loads the square wave performance showed no onset of ringing or other instability, although with real loudspeaker loads this needs to be checked carefully. Things fall apart at high levels and very low frequencies with square waves, but this is to be expected. But I like to eliminate capacitors if they are not needed.
But I got ahead of myself on the troubleshooting front here. When I began testing the amplifier one of the channels was out. I had noticed the rather absurd positioning of a toggle switch on the side of the unit and thought it looked like an accident begging to happen --- I suspect a substantial population of these out there with the handle sheared off. But this appeared intact, and I worked it a bit to see if that was the problem.
Well, the channel didn't start working, but I eventually traced the problem to the switch: it was not "making" for the right channel common connection. Since I did not plan to use the front panel 1/4" jack anyway, I wound up removing the switch, the jack, and substantially re-routing a number of connections (among them the left channel connection between the 8 ohm tap and the 100k feedback resistor, which had been bundled and tightly tie-wrapped to the mains power wires* to and from the front panel power switch!).
After this all was well. I listened to some orchestral selections via my AKG K1000 phones, connected to the 8 ohm taps and with the 8 ohm resistive loads left in place to keep the amplifier happy. It was quite a decent showing. Considering that I'm not out-of-pocket on this at all I may invest in some different tubes, as I am sure these are the stock ones from the manufacturer.
It's also tempting to fiddle with the topology, possibly introducing something along the lines of Cordell's recent amp as published in Linear Audio, since the input 6N2 has the two sections available for differential operation, and the output bias supply would probably support some current for current sources. I'm not sure why the manufacturer felt it necessary to use the paralleled connection, other than possibly to reduce voltage noise a bit, which I suppose could make sense given the rather high gain. I may also introduce an adjustment for the second stage balance, to null even-order. From looking at the drive signals to the 6L6 grids, it appears as if the two second-stage plate resistors are close to, but not precisely, values that result in equal magnitudes. Since an adjustment to make them precisely equal will itself not guarantee a minimum in even-order distortion, as it doesn't include the contribution from output tube mismatch, the best policy is probably just to adjust a trimmer while observing the distortion.
Brad
*at least hot and neutral were tightly twisted together, so the B field coupling was fairly small; and of course the source impedance is low.
I was given one of these because it had stopped working, by an audiophile with many amps
I found the fuse blown and worried about what had provoked this. But, it turned out there was nothing by way of a hard failure to blow the spare fuse, and I monitored cautiously while advancing the Variac supply, and monitoring the real input power with a Valhalla meter (indispensable for this kind of work!).The fuse is a 2A fast blow, and I would conjecture that it opened by virtue of simply the number of power on-off cycles. The online schematic I found was inaccurate and showed cathode biasing of the 6L6 output tubes, although it was clear that this unit had fixed bias as the others discussed in this thread. I adjusted the bias for a fairly rich 45mA, although I may back that off a bit. Things do run quite warm, and I removed the cover as a matter of course. Fortunately I don't have kids running around.
As it has quite a lot of gain (almost 38dB) for normal line-level sources, I also experimented with reducing this a bit by returning the 100k feedback resistor from the 8 ohm output tap directly to the 6N2 cathodes, and clipping out the partial electrolytic bypass caps. With resistive loads the square wave performance showed no onset of ringing or other instability, although with real loudspeaker loads this needs to be checked carefully. Things fall apart at high levels and very low frequencies with square waves, but this is to be expected. But I like to eliminate capacitors if they are not needed.
But I got ahead of myself on the troubleshooting front here. When I began testing the amplifier one of the channels was out. I had noticed the rather absurd positioning of a toggle switch on the side of the unit and thought it looked like an accident begging to happen --- I suspect a substantial population of these out there with the handle sheared off. But this appeared intact, and I worked it a bit to see if that was the problem.
Well, the channel didn't start working, but I eventually traced the problem to the switch: it was not "making" for the right channel common connection. Since I did not plan to use the front panel 1/4" jack anyway, I wound up removing the switch, the jack, and substantially re-routing a number of connections (among them the left channel connection between the 8 ohm tap and the 100k feedback resistor, which had been bundled and tightly tie-wrapped to the mains power wires* to and from the front panel power switch!).
After this all was well. I listened to some orchestral selections via my AKG K1000 phones, connected to the 8 ohm taps and with the 8 ohm resistive loads left in place to keep the amplifier happy. It was quite a decent showing. Considering that I'm not out-of-pocket on this at all I may invest in some different tubes, as I am sure these are the stock ones from the manufacturer.
It's also tempting to fiddle with the topology, possibly introducing something along the lines of Cordell's recent amp as published in Linear Audio, since the input 6N2 has the two sections available for differential operation, and the output bias supply would probably support some current for current sources. I'm not sure why the manufacturer felt it necessary to use the paralleled connection, other than possibly to reduce voltage noise a bit, which I suppose could make sense given the rather high gain. I may also introduce an adjustment for the second stage balance, to null even-order. From looking at the drive signals to the 6L6 grids, it appears as if the two second-stage plate resistors are close to, but not precisely, values that result in equal magnitudes. Since an adjustment to make them precisely equal will itself not guarantee a minimum in even-order distortion, as it doesn't include the contribution from output tube mismatch, the best policy is probably just to adjust a trimmer while observing the distortion.
Brad
*at least hot and neutral were tightly twisted together, so the B field coupling was fairly small; and of course the source impedance is low.
Last edited:
Brad;
the schematic is designed by "cut and paste", and sections of 6N2P are paralleled because 2 tubes look prettier than one. Don't look for technical explanations; the whole Mingda business is to build amps that look impressive, while topologies are mediocre. But I have to confess, his output transformers are quite good, unlike the rest of Chinese comrades uses.
the schematic is designed by "cut and paste", and sections of 6N2P are paralleled because 2 tubes look prettier than one. Don't look for technical explanations; the whole Mingda business is to build amps that look impressive, while topologies are mediocre. But I have to confess, his output transformers are quite good, unlike the rest of Chinese comrades uses.
Well the price was right
And I am happy to hear they are decent output transformers. The bandwidth was pretty reasonable, I think about 60kHz -3dB with a quick look, with 8 ohm resistive loads.
The HV supply is pretty wimpy in terms of filtering, the equivalent of 110uF -- although that is about 7.4 joules. There's a little room for something larger.
In any case it is a good matrix for playing around with alternative designs. I just looked at the winding resistance of the bias supply and it's comfortably low enough to support several milliamps additionally.
I can suggest you to implement something like this:
pentode LTP I used because of parallel feedback across output tubes. Without it you can go with a double triode in a single bulb, something like 6N6P. The same number of bulbs.
Also, screen grid supply is better to make stabilized, like single Zener shunt regulator with MOSFET source follower.
pentode LTP I used because of parallel feedback across output tubes. Without it you can go with a double triode in a single bulb, something like 6N6P. The same number of bulbs.
Also, screen grid supply is better to make stabilized, like single Zener shunt regulator with MOSFET source follower.
Attachments
E-mail them, ask the question, and mention that you bumped the topic with their name on DIYAudio forum. Or to their dealers who sold you the amp. I am pretty sure they will answer.
But you can experiment: when 6N2 is inserted in driver instead of 6N1, and 6N1 inserted instead of 6N2 in preamp, you will get less power and more distortions. Swap and listen.
Edit: here is another thread about this amp: http://www.diyaudio.com/forums/tubes-valves/221783-valve-upgrade.html
If your amp is wired for 6N2 and 6N1, but not for 12AX7 and 12AU7, swapping 6N1 and 6N2 you won't damage it. But you can't use 6N1 and 6N2 in the amp where filaments were wired for 12AX7 and 12AU7.
Last edited: