Current project: 1kW

[Frisbee78]

Hi everybody!!

I thought I'd share the schematic of my current (and first) project to the community:
Full size (big)
Half size (still big but not as much)

It's from a French magazine published in 98 and is supposed to sound real, real good. I'll tell you when I get it done

The author called it "the monster" and gave these specs:
- Power: 1x1035W into 8 ohms (bridged) or 2x350W into 8 ohms
- S/N ratio: > 120 dB
- Bandwith @ 1dB: 1 Hz to 227 kHz
- Rasing time: 1.2 µs @ 10 kHz

It feaures 6 pairs of MJ15025/MJ15024 (12 in total) and the PSU is powered with 5 toroidal transormers adding up to 1500VA...

Now that's lethal (well for ma at least )

 

[Richard C]

Very erm.... strange. At a glance, I can't say I'd be tempted to spend time and money on this design

 

[Frisbee78]

Very erm.... strange. At a glance, I can't say I'd be tempted to spend time and money on this design

Strange? Why strange?

 

[Richard C]

It seems complicated yet when used as a stereo amp has undesirable capacitor coupled outputs, when used bridged then a crude phase splitter is employed at the input.

What controls bias? and how is NFB applied (if any)?

There are some very basic and not particualrly linear stages in this design and considering its complexity it doesn't look promising.

 

[dmfraser]

Strange design

Lets see. No feedback. No use of differential amplifiers. Will have so-so power supply rejection. THD will be high. Transient response may be better. Darlington output is low efficiency. There are a lot better, and simpler, ways to generate a kilowatt these days.

 

[Frisbee78]

It seems complicated yet when used as a stereo amp has undesirable capacitor coupled outputs, when used bridged then a crude phase splitter is employed at the input.

What controls bias? and how is NFB applied (if any)?

There are some very basic and not particualrly linear stages in this design and considering its complexity it doesn't look promising.

Got your points.
The coupling capacitor are rather large beasts of 22 000 µF... I hope that such values won't affect the output signal too much. But then again, it's more of an amp meant to be used bridged.
As for the input phase splitter, no interferences whatsoever can be seen on the graphs near the 0v line.

Sorry, but I have no idea of what control bias or NFB are... I don't speak English as my first language and I still have plenty of technical vocabulary to learn for electronics & amps...

I agree on its complexity. But the results look good, so?

Lets see. No feedback. No use of differential amplifiers. Will have so-so power supply rejection. THD will be high. Transient response may be better. Darlington output is low efficiency. There are a lot better, and simpler, ways to generate a kilowatt these days.

Yes, no feedback, but do amplifier always really need them? If the design is made to do without and manages that lack well, then I don't see the problem. Going to an extreme, I could even quote Nelson Pass himself: "In the hands of mediocre designers, feedback was often overused to cover up design sins elsewhere in the circuit...". No feedback, no lies?...
Differential amp? Sorry, I'm clueless again
Same with PSU rejection. What is it?
THD... mmm... Yes, THD... How do you measure that?
Transient response. Another unknown term... :/
Darlington output might not be the most efficient solution but... it's a solution... And if using plenty enough of transistors then why not?
I'm sure that since the article was published in 98, better, simplier and more economic designs came to light. But... when I started that project, well, erm... it was 98... (I know )

Thanks guys for your feedback.

 

[kASD]

Commentable Thoughts

Got your points.
The coupling capacitor are rather large beasts of 22 000 µF... I hope that such values won't affect the output signal too much. But then again, it's more of an amp meant to be used bridged.
As for the input phase splitter, no interferences whatsoever can be seen on the graphs near the 0v line.

Sorry, but I have no idea of what control bias or NFB are... I don't speak English as my first language and I still have plenty of technical vocabulary to learn for electronics & amps...

I agree on its complexity. But the results look good, so?



Yes, no feedback, but do amplifier always really need them? If the design is made to do without and manages that lack well, then I don't see the problem. Going to an extreme, I could even quote Nelson Pass himself: "In the hands of mediocre designers, feedback was often overused to cover up design sins elsewhere in the circuit...". No feedback, no lies?...
Differential amp? Sorry, I'm clueless again
Same with PSU rejection. What is it?
THD... mmm... Yes, THD... How do you measure that?
Transient response. Another unknown term... :/
Darlington output might not be the most efficient solution but... it's a solution... And if using plenty enough of transistors then why not?
I'm sure that since the article was published in 98, better, simplier and more economic designs came to light. But... when I started that project, well, erm... it was 98... (I know )

Thanks guys for your feedback.

MY Friend AmpmaN Requests u to donot attempt this amplifier because u donot have enough level of understanding of electronics related to amplifiers and u also seems that u dont have relavant knowledge in amplification systems.

Some Facts about this amplifier.
No Negative Feedback.
Poor power supply rejection ratio.
Nil common-mode rejection ratio.
No differential pair at input.
slew rate is inadequate as compared to its power rating.
Thermal compensation and biasing is inadequate.
Higher components of Total Harmonic Distortion at output.

Regards
AmpMan.

 

[Circlotron]

Re: Commentable Thoughts

Thermal compensation and biasing is inadequate.

Ampman, you just hung yourself. In another thread (I will look for it if you like) you made the claim that you never use thermal compensation and yet here you say that another circuit's thermal compensation is inadequate.

I was going to say more but I have decided not to.

Edit -> Changed my mind. http://www.diyaudio.com/forums/showthread.php?postid=304119#post304119

 

[kASD]

Re: Re: Commentable Thoughts

Ampman, you just hung yourself. In another thread (I will look for it if you like) you made the claim that you never use thermal compensation and yet here you say that another circuit's thermal compensation is inadequate.

I was going to say more but I have decided not to.

Edit -> Changed my mind. http://www.diyaudio.com/forums/showthread.php?postid=304119#post304119

MY Friend Circlotron
Yes I still made that claim that I never use Thermal compensation but:
That is in the case of MOSFETS not in the case of BJTs , so understand the logic.

Regards

AmpMaN

 

[Richard C]

Yes, no feedback, but do amplifier always really need them? If the design is made to do without and manages that lack well, then I don't see the problem. Going to an extreme, I could even quote Nelson Pass himself: "In the hands of mediocre designers, feedback was often overused to cover up design sins elsewhere in the circuit...". No feedback, no lies?...

If an amplifier is intended not to use feedback then open-loop linearity is paramount as there is no mechanism to reduce distortion.

This amp appears to be designed to encourage distortion rather than reduce it.

1000W with >50W distortion...

Don't do it

 

[bigparsnip]

Re: Re: Commentable Thoughts

Edit -> Changed my mind. http://www.diyaudio.com/forums/showthread.php?postid=304119#post304119

Isn't that in relation to MOSFET's not bipolars as this amp is?

And, just my $0.02 worth, but there are a lot of much better amps out there than this one, and unless you have spent lots of money on parts (which could probably be used in another design anyway) and boards for this one I would sujest going looking for something a bit diffrent (try aussieamps for a start).

 

[kASD]

Re: Re: Re: Commentable Thoughts

Isn't that in relation to MOSFET's not bipolars as this amp is?

And, just my $0.02 worth, but there are a lot of much better amps out there than this one, and unless you have spent lots of money on parts (which could probably be used in another design anyway) and boards for this one I would sujest going looking for something a bit diffrent (try aussieamps for a start).

Ampman ThanX Bigparsnip For Clarifying the statement to CIRCLOTRON!

Regards
AmPmAn

 

[Ultima Thule]

Re: Re: Commentable Thoughts

Ampman, you just hung yourself. In another thread (I will look for it if you like) you made the claim that you never use thermal compensation and yet here you say that another circuit's thermal compensation is inadequate.

I was going to say more but I have decided not to.

Edit -> Changed my mind. http://www.diyaudio.com/forums/showthread.php?postid=304119#post304119

Well, he have "known" this for 7 years!!

http://www.diyaudio.com/forums/showthread.php?postid=315142#post315142

 

[Frisbee78]

Someone, kill me now! (nearly there)

donot attempt this amplifier because u donot have enough level of understanding of electronics related to amplifiers and u also seems that u dont have relavant knowledge in amplification systems

Don't we have to start somewhere? OK this not my own design, but does that mean I can't build someone else's???

No Negative Feedback.
Poor power supply rejection ratio.
Nil common-mode rejection ratio.
No differential pair at input.
slew rate is inadequate as compared to its power rating.
Thermal compensation and biasing is inadequate.
Higher components of Total Harmonic Distortion at output.

If an amplifier is intended not to use feedback then open-loop linearity is paramount as there is no mechanism to reduce distortion.
This amp appears to be designed to encourage distortion rather than reduce it.
1000W with >50W distortion...
Don't do it

That hurt more than an head on collision with a freight train!
The thing is that I have nearly completed it. In fact, I just need to put everything in a box and wire it all up!

If only I had known about this forum back in 98 (did it even exist then?) I probably would have turned away from it and considered another design.
But dropping the project at 90% completion and after $100s & countless hours of labour is simply not an option.

The author presented this amp like the best piece of kit that can ever be built and I took his word for it. After all, he is a professional electronic designer, unlike me who does it as a pass time.
I don't think I'll ever know enough to produce my own design but I am definitely willing to learn enough to avoid bad designs in future... (assuming that this project doesn't disgust me from audio DIY for good...)

 

[kASD]

Re: Re: Re: Commentable Thoughts

Well, he have "known" this for 7 years!!

http://www.diyaudio.com/forums/showthread.php?postid=315142#post315142

USE LOGICABLE THINKING!

Regards
Ampman

 

[bigparsnip]

I wouldn't be too sad, just look at it like this, if it does work you're ok. If it doesn't sound good, it's always an excuse to start on another project (a good excuse is always good if you have to convince the missus to let you play).

But, I l'll be interested to hear from you how it sounds once you have it all finished.

 

[kASD]

Re: Someone, kill me now! (nearly there)

Don't we have to start somewhere? OK this not my own design, but does that mean I can't build someone else's???


That hurt more than an head on collision with a freight train!
The thing is that I have nearly completed it. In fact, I just need to put everything in a box and wire it all up!

If only I had known about this forum back in 98 (did it even exist then?) I probably would have turned away from it and considered another design.
But dropping the project at 90% completion and after $100s & countless hours of labour is simply not an option.

The author presented this amp like the best piece of kit that can ever be built and I took his word for it. After all, he is a professional electronic designer, unlike me who does it as a pass time.
I don't think I'll ever know enough to produce my own design but I am definitely willing to learn enough to avoid bad designs in future... (assuming that this project doesn't disgust me from audio DIY for good...)

Dont woory friend , Ampman will help u in resolving the amplifier situation.

Regards
Ampman

 

[richie00boy]

Yeah, good luck and I hope it works out

I have to agree with everybody else though, some aspects of the circuit just beggars belief, particularly the o/p stage which will have atrocious turn-off time.

I also think you made a major mistake even thinking about building such a high-powered amp for your first project, let alone one so overly complex and of strange topology as this one.

Just make sure you can rip all the parts out of it to make a proper good and more sensible amp when you have built it

 

[pjacobi]

I understand the very front (phase splitter, duh) and the very end (standard EF).
Also the hiside and lowside current sources T11/T14 and the parts between as a rather baroque way to generate the bias voltage.
T7-T10 is a VAS buffer, I assume.
Looks likeT5/T6 do the actual voltage gain, but how much? Care to tell us R7/R8 values.
This leaves the question of T2/T2' - and what's in IC1? Can this be a NFB path? R4/R5 a summing network? Sorry, I'm somewhat clueless here.

IMHO no-NFB and 1kW are strange mix, also no-NFB and 120dB S/N (whatever is considered N here).

Regards,
Peter Jacobi

 

[Frisbee78]

Here are the values of the components.
I didn't want to publish those without the author's conssent, but I doubt I'll be able to contact him nor do I think anyone will attempt the building of this design anyway...