This is an old project that I never got around to finishing, the PCB's have been lying around for donkeys years. I do remember testing them individually and setting the bias but they never saw action as a stereo pair.
Now, they have been built into a Dual Mono pair. As the transformers were bought at different times their characteristics are slightly different. Although both nominally 35-0-35 at 250VA, the DC voltage to the two amps is slightly different. One has 50-0-50 and the other 53-0-53.
AFAIK this will not affect them in any way other than the fact that one will clip slightly earlier than the other.
Now, they have been built into a Dual Mono pair. As the transformers were bought at different times their characteristics are slightly different. Although both nominally 35-0-35 at 250VA, the DC voltage to the two amps is slightly different. One has 50-0-50 and the other 53-0-53.
AFAIK this will not affect them in any way other than the fact that one will clip slightly earlier than the other.
Attachments
Interesting ... Looks like a classic cottage industry britamp from the 80's - any selling points from the designer on this one ?
This wasn't intended as an amplifier comparison thread. I'm simply asking will the difference in rail voltages affect the amps.
It is based on the original Hitachi Lateral FET data sheet. JLH implemented it in his 80W MOS-FET amplifier as did Maplin with their 150W MOSFET amplifier.
I beleive this comes from about the early 90s.
It's still a very competent amplifier that will beat most commercial offerings hands down.
It's nothing like the Hitachi data book circuit, it uses an op-amp as input stage and providing most of the voltage gain then the output stage provides a low amount of gain to boost the limited swing from the op-amp.
In practice, no, you won't see any difference in performance with the rail difference betwen channels and your assumption is correct.
As a building standards comment, it looks good but green/yellow coloured wire is reserved for safety earth connections only. Also you should insulate your switch terminals.
It looks like the amplifier ground is not connected to safety earth, I think you had some ground loop problems on another thread, but it's worth trying to get right for safety reasons.
In practice, no, you won't see any difference in performance with the rail difference betwen channels and your assumption is correct.
As a building standards comment, it looks good but green/yellow coloured wire is reserved for safety earth connections only. Also you should insulate your switch terminals.
It looks like the amplifier ground is not connected to safety earth, I think you had some ground loop problems on another thread, but it's worth trying to get right for safety reasons.
Ok so I don't think the power rail differences are going to matter much.
There's something quite interesting about the amplifier schematic tho that's not so run of the mill..... For starters the op stage is not operated as a follower (output is taken from the drains) which is a stark contrast to the datasheet amps but there is localized feedback and loop feedback. The way the op stage is biased is also interesting ....
If you have the article would be worth a post IMO.....
There's something quite interesting about the amplifier schematic tho that's not so run of the mill..... For starters the op stage is not operated as a follower (output is taken from the drains) which is a stark contrast to the datasheet amps but there is localized feedback and loop feedback. The way the op stage is biased is also interesting ....
If you have the article would be worth a post IMO.....
The op-amp is set up as a follower. TR1 and TR3 are current source/sink and form part of the bias network as well as feed the drivers. Output stage is CFP with gain.
The PDF article found after a casual search on the forum:
http://www.diyaudio.com/forums/solid-state/242014-anyone-recognise-these.html#post3621440
http://www.diyaudio.com/forums/solid-state/242014-anyone-recognise-these.html#post3621440
This is the other schematic http://www.diyaudio.com/forums/soli...-ab-125w-mosfet-amp-schematic-discussion.html
how to increase the power into 250 watt?
best regards,
josh
how to increase the power into 250 watt?
best regards,
josh
White Noise produced a 250W version using the double-die MOSFETs (and a few component value changes). Because the output stage has gain, adding additional discrete devices can be problematic.
I'm not trying to improve on it, I only asked a simple question which has been answered - thank you.
Well the amp only works on one channel, the other channel is there but is very heavily distorted.
Dilemma. Whether to fault find on old design or replace with F5 turbo.
Any other suggestions using 50-0-50V rails at 250VA / Channel.
Dilemma. Whether to fault find on old design or replace with F5 turbo.
Any other suggestions using 50-0-50V rails at 250VA / Channel.
I'm going to give them the once over this morning.
It could equally well be a poor track as they are home brew PCBs.
It could equally well be a poor track as they are home brew PCBs.
Well, the Mk 1 eyeball didn't reveal any bad joints nor bad tracks.
Oddly, the amplifier starts up OK but starts to distort after a second or two, getting worse over a few seconds. No white smoke and the amp is still working except that it is distorting.
Oddly, the amplifier starts up OK but starts to distort after a second or two, getting worse over a few seconds. No white smoke and the amp is still working except that it is distorting.
Is this a clue or a Red Herring ?
I can't find anything immediately wrong.
The amplifier bias was adjusted when the amps were initially built.
However, the faulty amp now has bias of only 30mA as opposed to 100mA. The bias does adjust but requires almost the full travel of VR1 to achieve 100mA.
OK the supply is very slightly different. The LH amp is using the original 50-0-50V PSU that used to power both amps, the RH amp now has its own 250VA 53-0-53V supply.
I'm aware that germanium transistors used to age, but I though that didn't happen with silicon?
I can't find anything immediately wrong.
The amplifier bias was adjusted when the amps were initially built.
However, the faulty amp now has bias of only 30mA as opposed to 100mA. The bias does adjust but requires almost the full travel of VR1 to achieve 100mA.
OK the supply is very slightly different. The LH amp is using the original 50-0-50V PSU that used to power both amps, the RH amp now has its own 250VA 53-0-53V supply.
I'm aware that germanium transistors used to age, but I though that didn't happen with silicon?
I'd try a new bias pot on the faulty channel.
Silicon doesn't age, but it could be a duff component that has failed in its "infant mortality" stage.
Silicon doesn't age, but it could be a duff component that has failed in its "infant mortality" stage.
I suppose if the bias is all wrong, the amp will be almost running in Class B, maybe even edging into Class C.
However, that doesn't explain why it is OK for the first half second or so ?
However, that doesn't explain why it is OK for the first half second or so ?
Last edited:
- Status
- Not open for further replies.