I just got myself a yamaha M-4 amp which works fine. I know this is a "Close" equal to the Japanese issued B-4 in certain way but I know not all. In the B-4 there was a button you could press and the amp would run in Class A 30 watts. No such button exists for the M-4 but I would like to permanantly put the amp to use in class A.
I was wondering if anyone had any suggestions.
Thanks in advance
I was wondering if anyone had any suggestions.
Thanks in advance
To run the amp in class A at reduced output Yamaha switches the power supply voltage in half in addition to upping the bias.
This would involve buying a new power transformer since yours wouldn't have the correct windings.
Other that that, it's a good idea.
This would involve buying a new power transformer since yours wouldn't have the correct windings.
Other that that, it's a good idea.
So instead of 65v your saying 32v?
Instead of 15mv bias your saying 30 or 45 mv?
The B4 has 4 caps and 2 PT's where the M4 has 2 caps and one PT.
The amp runs as cold as ice now so I feel I could use the existing PT and just voltage divide down to 30 to 35 volts or better yet build a regulated supply which will dump off the extra voltage and then crank up the bias to be somewhere between 30mv and 45mv.
Let me know if you think I am on the right track here.
Thanks so much!!
Instead of 15mv bias your saying 30 or 45 mv?
The B4 has 4 caps and 2 PT's where the M4 has 2 caps and one PT.
The amp runs as cold as ice now so I feel I could use the existing PT and just voltage divide down to 30 to 35 volts or better yet build a regulated supply which will dump off the extra voltage and then crank up the bias to be somewhere between 30mv and 45mv.
Let me know if you think I am on the right track here.
Thanks so much!!
The B-4 also had Output Impedance control,
there are some good pics of the PCB on that site too.
This thread has a little info on how its implemented, "mrfeedback" who started that thread might be able to help you some more?
I have a pdf of the service manual for theM-4 if you need a copy.
Regards
James
An externally hosted image should be here but it was not working when we last tested it.
there are some good pics of the PCB on that site too.
This thread has a little info on how its implemented, "mrfeedback" who started that thread might be able to help you some more?
I have a pdf of the service manual for theM-4 if you need a copy.
Regards
James
Thanks to tvi the Service Manual can now be downloaded from my website: www.audio-circuit.dk
Just click on "Schematics" in the left menu and scroll down to Yamaha 😉
Just click on "Schematics" in the left menu and scroll down to Yamaha 😉
"So instead of 65v your saying 32v?"
Yes.
"Instead of 15mv bias your saying 30 or 45 mv?"
30W in class A takes 1.4A bias, or 330mV across each 0R47 emitter resistor. If you reduce the supply voltage to ±27V each channels's heatsink will dissipate about 75W.
"The amp runs as cold as ice now so I feel I could use the existing PT and just voltage divide down to 30 to 35 volts or better yet build a regulated supply which will dump off the extra voltage and then crank up the bias to be somewhere between 30mv and 45mv."
45mV would be about 190mA of bias per channel, about 23W of heat per channel. This will allow about 600mW in class A.
You will need a 20-0-20 300VA (or larger) transformer to provide the ±27V DC. I might even go for an 18-0-18 if I could find one, 75W of heat is a lot. The heatsink looks like about 0.5°C/W, so it would run about 60°C in a 23°C room.
If you just cranked the bias until the heatsinks would be about 60°C the amp would put out about 6.5W in class A. This would require about 130mV of bias. I would try this and see if I liked the sound, and could live with that much heat.
I sold these amps when they were new. The CA2010 was the same power in class AB, and had the switch for class A. I can't say that I could hear the difference.
Nelson Pass had a letter published in the Audio Amateur about class A in which he stated that one of the biggest benefits was the pseudo-regulation of the supply that comes with running class A. The M4 already has a fully regulated supply for the front end.
My experience with the Electrocompaniet amplifiers (which were biased in class A to about 2W and had fully regulated supplies for the front end) was that this seemed like a good way to go. The EC point of view was that the peak-to-average-ratio in music was such that the amplifier only need to biased in class A to about 13dB down from full power. This would be about 6.5W for the M4.
I would think long and hard before doing anything more than cranking up the bias a bit.
Yes.
"Instead of 15mv bias your saying 30 or 45 mv?"
30W in class A takes 1.4A bias, or 330mV across each 0R47 emitter resistor. If you reduce the supply voltage to ±27V each channels's heatsink will dissipate about 75W.
"The amp runs as cold as ice now so I feel I could use the existing PT and just voltage divide down to 30 to 35 volts or better yet build a regulated supply which will dump off the extra voltage and then crank up the bias to be somewhere between 30mv and 45mv."
45mV would be about 190mA of bias per channel, about 23W of heat per channel. This will allow about 600mW in class A.
You will need a 20-0-20 300VA (or larger) transformer to provide the ±27V DC. I might even go for an 18-0-18 if I could find one, 75W of heat is a lot. The heatsink looks like about 0.5°C/W, so it would run about 60°C in a 23°C room.
If you just cranked the bias until the heatsinks would be about 60°C the amp would put out about 6.5W in class A. This would require about 130mV of bias. I would try this and see if I liked the sound, and could live with that much heat.
I sold these amps when they were new. The CA2010 was the same power in class AB, and had the switch for class A. I can't say that I could hear the difference.
Nelson Pass had a letter published in the Audio Amateur about class A in which he stated that one of the biggest benefits was the pseudo-regulation of the supply that comes with running class A. The M4 already has a fully regulated supply for the front end.
My experience with the Electrocompaniet amplifiers (which were biased in class A to about 2W and had fully regulated supplies for the front end) was that this seemed like a good way to go. The EC point of view was that the peak-to-average-ratio in music was such that the amplifier only need to biased in class A to about 13dB down from full power. This would be about 6.5W for the M4.
I would think long and hard before doing anything more than cranking up the bias a bit.
the Zo control is of interest to me, though i wouldn't want to venture into -Zo territory. looks like current feedback. ElectroVoice used a similar scheme on a tube amp.
"looks like current feedback"
It is, they had a whole line of speakers that used this ActiveServoTechnology (AST).
Many of their amplifiers had the (switchable) capacity to run these.
In the end they went to dedicated amplifiers built into the subs.
The Crown Delta Omega 2000 also had a Zo control, if you turned it too far it would make violent thumping noises. It didn't stay in their line-up for too long.
It is, they had a whole line of speakers that used this ActiveServoTechnology (AST).
Many of their amplifiers had the (switchable) capacity to run these.
In the end they went to dedicated amplifiers built into the subs.
The Crown Delta Omega 2000 also had a Zo control, if you turned it too far it would make violent thumping noises. It didn't stay in their line-up for too long.
overbias (biasing a class B into class A) can cause some distortion in the crossover area due more to lowered output impedance within the crossover region than to transistor nonlinearities.
DJK
So I understand you clearly say with the 65v for now and turn the bias up a little correct? Or turn the voltage down with the voltage pot and turn the bias up which is what I think you saying?
Also would you be kind enough to give the equation on the amount of current per mV?
Thanks
So I understand you clearly say with the 65v for now and turn the bias up a little correct? Or turn the voltage down with the voltage pot and turn the bias up which is what I think you saying?
Also would you be kind enough to give the equation on the amount of current per mV?
Thanks
DJK also when you say cranking the bias up what do you think would be safe point?
Take no issue with the advise. I take on my own responsibilities as I paid next to nothing for this amp and have many others.
Take no issue with the advise. I take on my own responsibilities as I paid next to nothing for this amp and have many others.
there isn't a voltage pot. but if you want to lower the rail voltages without getting another tranny, you could do it with a pair of series pass regulators. if they were adjustable, you would then have voltage pots.
"overbias (biasing a class B into class A) can cause some distortion in the crossover area due more to lowered output impedance within the crossover region than to transistor nonlinearities."
I know.
There is an optimum bias point for distortion. As you increase bias the distortion will go down, dip, and then go up as you further increase the bias (when measured with an IM anlyzer). As I mentioned before, Yamaha made several amps around this time with class A switches that could be flipped in a matter of seconds, and it didn't make even a small difference (as long as you didn't run it out of power in class A).
Without an analyzer I would probably set the bias to 20mV (about 40mA per device) and be done with it.
"you could do it with a pair of series pass regulators"
Which would then add another 75W per channel of dissipation to the sinks (at ±30V rails), and they would run at the boiling point of water (for a short time).
"when you say cranking the bias up what do you think would be safe point?"
As mentioned before:
"If you just cranked the bias until the heatsinks would be about 60°C the amp would put out about 6.5W in class A. This would require about 130mV of bias. I would try this and see if I liked the sound, and could live with that much heat."
I don't think the heatsink will stand any more (and the output transistors life will probably be reduced).
I know.
There is an optimum bias point for distortion. As you increase bias the distortion will go down, dip, and then go up as you further increase the bias (when measured with an IM anlyzer). As I mentioned before, Yamaha made several amps around this time with class A switches that could be flipped in a matter of seconds, and it didn't make even a small difference (as long as you didn't run it out of power in class A).
Without an analyzer I would probably set the bias to 20mV (about 40mA per device) and be done with it.
"you could do it with a pair of series pass regulators"
Which would then add another 75W per channel of dissipation to the sinks (at ±30V rails), and they would run at the boiling point of water (for a short time).
"when you say cranking the bias up what do you think would be safe point?"
As mentioned before:
"If you just cranked the bias until the heatsinks would be about 60°C the amp would put out about 6.5W in class A. This would require about 130mV of bias. I would try this and see if I liked the sound, and could live with that much heat."
I don't think the heatsink will stand any more (and the output transistors life will probably be reduced).
"there is a voltage pot and it says voltage adjust right under it"
Look at the schematic, it's for the front end and only supplies a few mA.
Look at the schematic, it's for the front end and only supplies a few mA.
"i don't have a service manual for it..."
ACD posted the link, 14MB.
"on the amp board or the power supply board? "
On the front end card, one per channel.
ACD posted the link, 14MB.
"on the amp board or the power supply board? "
On the front end card, one per channel.
The heatsinks of the M4 (B4) do not look very encouraging for Class A operation.
A suggestion might be to open up the bottom plate of the M4 chassis, similar to the openings under the heatsink as in the B4 model, and mount a small desktop processor fan under each heatsink wing.
A suggestion might be to open up the bottom plate of the M4 chassis, similar to the openings under the heatsink as in the B4 model, and mount a small desktop processor fan under each heatsink wing.
- Status
- Not open for further replies.