Compare Crecendo and Leach Amp

djk

R.I.P
2001-02-04 4:23 am
USA
Are we talking about the original 1984 Crescendo? It had a reputation as an oscillator. The new 2001 version may be better. I don't like plastic outputs and I don't like the sound of L-MOS for bass. I have owned several Leach amps. They are well worth the effort.
 
fcel,
Once upon a time, it was thought that TIM (Transient Intermodulation Distortion) and slew rate were the cures for the audible problems with amps that measured well, yet didn't sound all that good.
The Leach amp is a design that seeks to minimize TIM by having a high slew rate. It's a fairly complicated circuit as such things go, so the parts count is high and there's a lot of assembly and fiddle-factor involved.
I haven't heard a current version, but the last one I heard (I don't know what revision it might have been) didn't seem all that remarkable to me. However, lots of people seem to like them.

Grey
 
Well,

For me, since I'm a newbie, the Leach Amp is close to perfect, since it requires some hard work and prof. Leach has a lot of explanaitions on his website. It has, this far, been a goldmine for learning about the different parts of an amp.

I hope I'll be able to finnish the project in reasonable time.

But, it was an article about the Crecendo (just recentrly, the "upgraded" version") that got me hooked on building an amp. from the beginning, so I'm still curious if it is worth the effort to build it as well.

//magnus
 
I think one of the main features that the leach amp has, is that it is pretty much rock solid. Prof. Leach oversees the construction of at least 20 amps / year from just students. Not counting DIYers elsewhere. This gives him first hand knowledge of any problem that can arise and he has done well in fixing them in the 20+ years of the designs life.

I built one as a student under his supervision that I am very anxious to listen to. Its first test was on a pair of martin logan monolith III's. I liked the sound a lot, but the guy's krell fsb 600 had quite a bit of an edge of my amp(for 13 grand, it better). I have been unable to find employment since i graduated and have no money to buy a good pair of speakers. They are 2nd on my list after i do find a job tho!

jt
 
I take issue with Grey's perceptions about the Leach amp; it is fairly typical of modern amplifier design and even relatively simple as it has no constant current sources or current mirrors. I'd hardly call it 'fiddly' in terms of building and adjusting; there's only a bias pot.

While having a good slew rate, it also avoids slew
limiting by having an input filter. The symmetrical
design and plenty of power supply decoupling helps to give
it very high ripple rejection. Indeed, the symmetrical
design from input to output was one of the attractive
features when I originally chose to build it.

I'd call it a straightforward and even slightly conservative design that's well-documented and thoroughly
debugged over the years. You DO have to follow the instructions closely and be careful about wiring and grounding, but that's true of any high-performance amplifier.

It might benefit from a few tweaks; I've installed a 1 uF
capacitor between the driver transistor emitters to improve
the output device turnoff at high frequency, as per Self
and Sloan's books. Perhaps there's room for improvement in the bias tracking. I'm installing ver 4.5 boards to update my amplifier, pending arrival of some Black Gate and Nichicon Muse capacitors; I used metal film resistors throughout, and matched all mirror-image parts for tolerance or gain as closely as I could. Also used better
quality parts for the wire-wound emitter resistors instead
of the typical sandcast parts--lower Tc, I hope.

At some point I may substitute 2SA1302/2SC3281 transistors
for the MJ15003/4 devices; the former have much better linearity at higher current and a much higher Ft. But I'll
have to change the physical design of the amplifier to
mount the new transistors on new heatsinks, if I can find
suitable units at a good price. But I'm going to brassboard
this before I commit to a major overhaul of the chassis; I'd
like to be sure these new transistors are stable.

As for sound quality; I can't really comment as I've never
had much of a chance to compare it to other amplifiers. I'm
fairly pleased with the amplifier over the nearly twenty years I've been using it and hope the newest version boards will constitute an audible improvement.

I haven't seen any other amplifier designs that intrigue me enough to try them, as far as Class B amplifiers go. Class
A amplifiers probably have lower distortion, but the heat
dissipation and related complications make them impractical
for my use. It's a trade-off between absolute quality and
practicality.

I've also been working on a tube amplifer project, but it's stalled for lack of parts, money for parts, and a choice
of driver circuit design.
 
swede said:
Hi there,

Have any of you any idea on the main differenced between the (Elektor's) Crecendo and the Leach Amp.

I'm currently building the Leach Amp and are really curious about the pros and cons comapred to the Crecendo, which I'm probably about to build in a not too distant future.

Have a good day,
//magnus

I improved the original Crescendo a lot and published it in the swedish (now dead) magazine called Audio-Video. My amp at home has done excellent service since 1989 together (since 1996) with my Martin Logan SL3 electrostatic speakers.

You can read a short piece at my homepage about the QRO amp as I called it.

My point of view is that the Leach amp is well tested for years and is bipolar and Crescendo is a MOSFET but you can put MOSFET's on the Leach and vice versa.
 
My point of view is that the Leach amp is well tested for years and is bipolar and Crescendo is a MOSFET but you can put MOSFET's on the Leach and vice versa. (quote from Peranders)

You cannot replace MOSFETS with Bipolars or vice-versa, as drop-in replacements. Mosfets are voltage driven and Bipolar Transistors are current driven. Substantial changes to the driver stage is essential, before such substitutions are resorted to.

Note that the original Elektor Crescendo (1984) used Hitachi Lateral Mosfets and hence, the bias was set with a trimpot without any Vbe multiplier. The Millenium edition has a Vgs multiplier since the output Mosfets now chosen have a positive temperature co-efficient.

The Leach design uses diodes to sense the temperature and hold the bias stable.

The original Crescendo was known to be an oscillator. I was able to tame my two boards by moving the Zobel network to the outward side of the output coil and by hanging a few caps on the output devices. They do perform well and are quite good sonically, especially if you decrease the rails slightly and increase bias.

A LOT OF GOOD THINGS ARE SAID ABOUT THE LEACH LOW TIM AMP. BUT THIS AMPLIFIER WAS ALSO KNOWN TO OSCILLATE AND PROF.LEACH SUGGESTED THAT THE ZOBEL/BOUCHEROT NETWORK BE PLACED AFTER THE OUTPUT COIL AND DIRECTLY SOLDERED ON TO THE SPEAKER TERMINALS OF THE AMPLIFIER. This broke the positive feedback that was otherwise being induced.

I have not built the Leach amp and am not competent to comment about the its sonic qualities. However, what I have gathered is that the amp is rugged and reliable but its sound quality cannot be compared to for example, the Alephs or the Aleph-X. Indeed, the Leach Super Amp has enormous power capability and some folks like me who need the power (for PA in my case, to surpass the rather poor sound quality of the Dynacord S1200) will vote for it. Yes, I am going to build one in the near future.
 
Samuel Jayaraj said:
My point of view is that the Leach amp is well tested for years and is bipolar and Crescendo is a MOSFET but you can put MOSFET's on the Leach and vice versa. (quote from Peranders)

You cannot replace MOSFETS with Bipolars or vice-versa, as drop-in replacements. Mosfets are voltage driven and Bipolar Transistors are current driven. Substantial changes to the driver stage is essential, before such substitutions are resorted to.

OK, I was little bit unclear, of course you have to change the bias circuit and maybe the short circuit protection but still no major changes. It's probely easier to change from BJT to MOSFET's than the other way around. MOSFET's needs in the audio band much less drive power than BJT's. The big problem with MOSFET's is oscillation!
 
swede said:
Have any of you any idea on the main differenced between the (Elektor's) Crecendo and the Leach Amp.

I'm currently building the Leach Amp and are really curious about the pros and cons comapred to the Crecendo, which I'm probably about to build in a not too distant future.
//magnus

Some people wrinkle their noses for the Alexander Current Feedback amp but it's good, really. I know a person who has built it with good results. It's patented but this is no good reason for good qaulity and/or operation.

http://homepages.strath.ac.uk/~cnbp111/amp1.html
http://www.analog.com/library/applicationNotes/AdAudio/AN211.pdf
http://www.diymania.net/pass/project/CF-Power.htm
 
Hi,

all of you looking for the crescendo schematic, there must be a thread somewhere concerning a 300 or 350 watter on an italian homepage. This was almost an exact copy of the crescendo.

I still have the original article at home (in Dutch) so if anyone is interested I can scan and sent it (not that fast cause I'm not at home at the moment)

I've build about 15 of these in various forms (2,4,6 fets , 35-75V power supply). Only the first I build oscilated due to the use of wirewound resistors at the sources. All the later versions were ok even without the output inductance. (I even tried one without the source resistors and it worked too).

I don't know the leach but I know that the original crescendo sounded good and with some minor mods you can build a very good amp that is still very up to date soundwise.

william
 
Crescendo circuit

I tried to send the circuit, but the file is larger than this Forum allows.

If someone has a place that people can go download it from, I can send it there.

The original article is too long to be sent.

Anyway, the original circuit was revised, and also myself had several problems with it. It never worked properly.

I got tired and went on to better, proven projects, like those by Borbely on The Audio Amateur.



Carlos
 
GRollins said:
The Leach amp is a design that seeks to minimize TIM by having a high slew rate. It's a fairly complicated circuit as such things go, so the parts count is high and there's a lot of assembly and fiddle-factor involved.
I haven't heard a current version, but the last one I heard (I don't know what revision it might have been) didn't seem all that remarkable to me. However, lots of people seem to like them.
'Tis the trouble with bloody "highlanders", bring 'em down from the mountains and they are "lost at sea" ..... ;)

OK ..... I agree the Leach amp isn't as good as a good class A design ....... moreover I don't imagine Leach himself would claim that!

Many can't afford the size/heat/hardware of a class A amp and as class AB amps go, this has been proven to be a stable & reliabe amp, better than many commercial designs out there. Adjustment, as stated above is not really that difficult and requires no more than a multimeter.

I had one of these for years and am very fond of it. I may not listen to one now, but for entry level DIYers or those not entertaining a living-room furnace, you will likely be very happy.

cheers, mark