mosfet class a/ab amp

Anyone tried the JLH Class A/AB?

Just wondering if anyone has tried this in the last 7 years or so?
The amp uses VP080, VN1210, 2SK133 and 2SJ48 mosfets. These are difficult, but seemingly not impossible to get hold of.
However:
1) Finding them on the actual manufacturer's website is nigh impossible - hence limited information.
2) There seem to be no spice models for them.
3) Perhaps there are newer, better alternatives to these mosfets now available - and also with the spice models or data suitable for LTSpice?

It would be nice if the manufacturers stated they were suitable for audio use too.

Comments/suggestions gratefully accepted!

Cheers!
 
Jlh

Whole load of JLH here The Class-A Amplifier Site -- even one where they tried to make something that resembles a Krell KSA -- only with a chassis made of wood.

Personal opinion, I know - but after designing amplification for over 2 decades you get to know what works and what is a bit ..........er..........well...... of a minger.

Linsley Hood stuff is very basic - noisy - with crummy damping factors, insignificant slew rates - offering pretty below average performance ,(even for back then) - we tended to use them as training tools -- mainly because they are cheap and the parts list is short.

Dont get me wrong - great for enty level, have a go type of thing but
If you are seeking quality - I'd scour the world for better.
 

Mooly

Administrator
Paid Member
2007-09-15 8:14 am
Just wondering if anyone has tried this in the last 7 years or so?
The amp uses VP080, VN1210, 2SK133 and 2SJ48 mosfets. These are difficult, but seemingly not impossible to get hold of.
However:
1) Finding them on the actual manufacturer's website is nigh impossible - hence limited information.
2) There seem to be no spice models for them.
3) Perhaps there are newer, better alternatives to these mosfets now available - and also with the spice models or data suitable for LTSpice?

It would be nice if the manufacturers stated they were suitable for audio use too.

Comments/suggestions gratefully accepted!

Cheers!

I can recall this amp... never built it though. It used FET's for the drivers and VAS I seem to remember. The outputs are lateral FETs and there are several new devices that would work well.
Check out Exicon/Profusion

Lateral Mosfet

JLH's amps all had one thing in common... they really made music.

I surprised PhaseLockedLoopy finds them poor performers. Perhaps he could recommend one of his own designs instead, or some other equally good performer.
 
Oooooh lets not go into my projects just yet -- I'm going to be starting a thread on that later as it just so happens I am developing Pure Class A bipolar.

But...As I say -- its down to personal preference -I find them all a bit simplistic - but one could expand out a bit and examine other designers such as Carver - or Hafler

...here's the DH250 FET

...as an example of what I mean...one of the biggest improvments that can be made to any amplifier is how power is provided and delivered - in the JLH link in my previous post - they try and use common, cheap plastic package regulators ....hmmmm ... not my preference I have to say.

....here's mine for comparison
 

Attachments

  • hafler.jpg
    hafler.jpg
    171 KB · Views: 540
  • psu.jpg
    psu.jpg
    125.6 KB · Views: 519
Last edited:
PhaseLockLoopy, your regular does not include Emitter resistors for the Pass Transistors. I think it is important to share the current load.

About 47-100uF caps from Base of Q1, Q4 to Ground would improve performance and C11,12 could be of higher value. C11, 12 will run out on sustained music peaks.

A supply, whether regulated or unregulated, which allows for relatively uncompressed transients is desirable, provided the amp's output stage is designed to handle the peak currents.
 
Hi guys

If I've walked into a JLH appreciation society - I apologise if my personal tastes dont fit - I just thought some alternatives exist that could be improvements

Hi Andrew -- both really -- as I said for me personally - its all down to the cheapskateyness they seemed to extol.

Hi Samuel - absolutely - if I were delivering to much larger amplifiers - But under drive conditions down to 2 Ohm inductive loads, with the rest of the amplifier on the end that, peak issues havnt proved to be the case. I may improve that further when I do the bigger 300W version. There is enough stable current in there to sink a battleship. I may post the whole design this weekend - I'm still testing the bench prototype. You'll find it is consistent at that max power.
 
But under drive conditions down to 2 Ohm inductive loads, with the rest of the amplifier on the end that, peak issues havnt proved to be the case. I may improve that further when I do the bigger 300W version. There is enough stable current in there to sink a battleship.
I think you underestimate the peak current requirements of fast transients.
 
I think you underestimate the peak current requirements of fast transients.

Hey Andrew

Actually you might be right there :) ......I havenet done any transient yet
I been testing for max output -- but because of the performance of the MJs I assumed I had 60Amps peak (12.6 continuous) per rail available - which is huge.

...can we take this to another thread later today or tomorrow? I'll post my project up (all of it)

...I'm just a bit concerned for poor P-Robertson,here, coming back to his thread thinking "WTF" .....LOL
 
Last edited:
Dont get me wrong - great for enty level, have a go type of thing but
If you are seeking quality - I'd scour the world for better.

Actually I was just following up on the OP question of +7 years ago. This was one of JLH's last designs. I think that a kit was offered at the time but never heard what the sound quality was like.
I am thinking of building it to see but would like to check out the design on LTSpice as a first step - hence my query on the FETs and their .models.
 
Check out Exicon/Profusion

JLH's amps all had one thing in common... they really made music.

Thanks for the reply. Yes the 10W LJH Class A always seems to get good reviews - in any one of it's many guises.

Had heard that the Exicon MOSFets were not always well received but at least they're TO3. No small signal on the Profusion site though.
Do Exicon have they're own web site? Have been trying to find it.

RS indicate many types of MOSFet as available but they don't appear on the manufacturer's site. Tends to be confussing.

Would be nice to get a reference to a manufacturer's site for small and large signal MOSFets that are considered good for audio and have Spice models....
 
Hi guys

If I've walked into a JLH appreciation society - I apologise if my personal tastes dont fit - I just thought some alternatives exist that could be improvements

I don't think it's so much of a JLH appreciatation society but rather quite a number of people appreciated JLH's contribution to amp design from 1969 onwards.

AFAIK, the original JLH class A is still considered very good even after all these years.
Ultimately, I want to have a 'reference amp' to compare other designs to, possibly my own one day. I think mosfets should give superior sonic results because of their linearity but then this is a hot topic so we'll stay away from it. Just my thoughts - hence the query on his mosfet amp design.
 
Linsley Hood stuff is very basic - noisy - with crummy damping factors, insignificant slew rates - offering pretty below average performance ,(even for back then) - we tended to use them as training tools -- mainly because they are cheap and the parts list is short.

Then you'll LOVE this, JLH=TTL. Everything you asked for above in spades!

My beef with JLH is overbias in the center, not enough bias at extremes...
Entirely because base current is the only thing gets regulated or steered.
Turn a blind eye to quiescent output currents, and hope for the best.

On the other hand, that same blind eye makes paralleling them easy.
I can't think of any other advantage...

JLH is fixable, but thats not what I'm attempting to show here.
 

Attachments

  • 74.gif
    74.gif
    54.6 KB · Views: 258
Last edited:
Then you'll LOVE this, JLH=TTL. Everything you asked for above in spades!

My beef with JLH is overbias in the center, not enough bias at extremes...
Entirely because base current is the only thing gets regulated or steered.
Turn a blind eye to quiescent emitter currents, and hope for the best.

On the other hand, that same blind eye makes paralleling them easy.
I can't think of any other advantage...

JLH is fixable, but thats not what I'm showing you here.

eeeeeeek :yikes:

ROFL
 
Sorry to be bashing JLH. Its not all that bad, and as I said, fixable...

Merely replacing the bipolar output devices with MOSFETs and some
appropriate resistors from gate to source corrects a lot. Square law
vs Square law helps restore the quiescent current profile such that
the little Iq current is in the middle, and big current at the extremes.
And temperature co-efficients now bend in a much safer direction.

The biggest new prob introduced by MOSFETs is matching threshold
voltages. And to include a compliment of similar, but opposite threshold
up front, assuming you want the DC to cancel.

Nelson Pass' PLH is probably the circuit you are looking for.
 
Last edited:
If you wanna fix JLH, but stay all bipolar, thats do-able too. You gotta do
something bout making the output currents behave by a rule. Rather than
merely spray and pray a drive current to one base or the other...

http://www.diyaudio.com/forums/soli...-buffer-power-output-stage-9.html#post2056889

See schematic of post #87 in the above thread...

Its a good circuit. Just don't expect my "explanation" of it to make
any sense. Cause I was simultaneously talking bout something else.
My mind was scattered all over that day.
 
Last edited: