400w high fidelity diy amp, where can I find them?

400w high fidelity diy amp, where can I find them? Pics of Leach superamp test pg.2

Hi!

the only one i know of ist the Leach Superamp. I am about to make a gerber file for that amp as thats the only one i know of.. It is 300w 8R but i would like even more, 4-500w class AB amp.

Is there any out there that can compete with the Leach in soundquality?

I have a thread about raising the power of the Leach superamp, but i am not experienced enough to do that on my own and i havent got any help in that thread so what..

So, i try to ask if there are any other designs in same class but with more power?

Best regards, Mathias
 
Last edited:
Or, maybe an amp that is bridgeable to 8R and leave 400w.. I dont know.

I am a strong believer of headroom. Thats why I need an amp with this high output.

I normally listen in quite high levels, maybe 30w ca, and with a good recording, that have trasients 10-20 times higher, I need this kind of power to listen to the recording as it is meant to be.

Please, i need no warning about my hearing, i know, high levels are dangerous bla bla..
 
Output Stage


Here you can read about what i mean, and what i believe adn many other.

"Music is capable of a 60dB (1,000,000:1) dynamic range. The transients in music are very small in energy but are approx 20dB above the RMS music level. The average RMS power of fully dynamic music can not go above -20dB of the amplifiers full power capacity without the transients clipping the rail supplies. 20dB is 100:1 so therefore a 100 Watt amplifier should not be driven above 1 Watt of RMS music level (over approx 1 minute of time) to avoid transients being driven into rail clipping. A 100 Watt amplifier can only be used at an average of 1 Watt with fully dynamic music. For this reason amplifiers less than 60 Watts should not be considered as audiophile status, but unfortunately many are."
 
.....The compound complementary arrangement has two advantages over the Darlington and Quasi complementary arrangements. Compound complementary has superior quiescent bias stability.....

Due to the effective higher Gm of the feedback pair because of the local feedback, output Z is primarily controlled by the OPT emitter resistors. This means that although the bias may seem more stable thermally, a condition sometimes referred to as Gm doubling may take place causing essentially crossover distortion to be present. In order to prevent Gm doubling, the bias is usually set on the lowish side.
 
Completely DIY, kit, or complete? Class-A, or some A/B? or D, or a mix? Biamped, triamped, or passive crosover? How many ampilfier channels? Need to know more about your particular idiocycratic pursuit. We are each unique, and so is our journey... There are hundreds of high-quality high-powered amps, so you'd have to say more about your criteria. Most amps scale up to what common drivers can handle (then you can use multiple amps) just add parallel output devices for current and bridge for voltage. The Leach you site, with 12 output devices per channel, could certainly use 14 or 16. An F5 turbo can have parallel output devices added and get into your territory.

I have over 18,000 watts for my front 2 channels, because I also believe the last watt is important. Conversely, Nelson Pass makes a very strong case that the first watt carries more information that the human brain appreciates and its qualities are more important. My commercial FET amps are bone-stock except for one coupling cap in each, strictly class-B, no magic, all 20 channels = 320 high-powered FETs. But I push incredibly inefficient sealed-enclosure isobarics. Each of my ten amps probably does average 2 watts with headroom for a 1000 watt peak.

Talking about electrical absolute watts is irrelevant without consideration of the speakers, room, etc. Very efficient speakers will impress you with their handling of peaks much easier than a big amp. If you really want to handle those transients, use a 3-way active crossover and multiple smaller amps with efficient speakers. You never really need an amp bigger than a single driver will handle.
 
Last edited:
2-channel amp, class A is freferred but not economical or possible in this instance, so class A/B is my preferred choico of class.

Passive X/O.

cyclecamper: Nice to have a friend in the believings..

Iam building a pair of efficient speakers, ca 93db but i still want that ammount of power in the amp i am going to build bcause i never know if i build another pair of speakers.


I started a thread about getting more power from the leach superamp, i am a novice so i thought it would just be to put more output devices in series as Prof. Leach did when he made the superamp. IT is a Leach amp with more Out. devices in series. Cascoded might it be called?


cyclecamper, you seem to be the guy on this forum that i need to get in contact with.

I would like to use the leach amp as it is a well known well sounding amp, but i would like it to put out 4-500w. Can you help me with this? I need someone who can tell me, this is what you have to do, this is what you should be vary about.

My idea was to use another 4 pair of output devices on the superamp.

I am even ready to pay for consultant help in making this amp work.

There exists no gerber files for it, i am slowly getting one done, but, i dont know what more has to be added than just 4 pairs of devices..
 

mcd99

Member
2008-03-21 10:13 pm
UK
You could bridge a pair of leach (non superamp) amps. That would give you some decent power. I have done this running a peerless xls10 in sealed enclosure with EQing down to 17hz. You could hear the poor xls bottoming out but the leaches weren't struggling. The price being paid bridging amps is an increase in distortion.

You idea of adding output transistors to a superamp is definitely possible.
 
You could bridge a pair of leach (non superamp) amps. That would give you some decent power. I have done this running a peerless xls10 in sealed enclosure with EQing down to 17hz. You could hear the poor xls bottoming out but the leaches weren't struggling. The price being paid bridging amps is an increase in distortion.

You idea of adding output transistors to a superamp is definitely possible.

Well then i could bridge a pair of superamps too, right? Then, then were talking the power i want. But this is for a pair of speakers, not subs so the hihger distortion is not wanted. How much higher will it be when bridged, are there some kind of general rule that when bridging you get x amount more distortion.

What happens with slew rate when bridged?
 

mcd99

Member
2008-03-21 10:13 pm
UK
slew rate doubles as does distortion but remember each amp sees effectively half the speakers impedance. Eg for an 8 ohm speaker each amp sees 4 ohms. So you would use the distortion figures for the amps driving 4 ohm loads.

Bridged leach amps would give you 480W into 8 ohms.
 
This is the slewrate and squarewave/sine response of the Leach Superamp, It is taken from the original article; Build a double barrelled amp in Audio magazine 1980.


Slewrate:

[IMGDEAD]http://s24.postimg.org/vqnhigir9/slewrate.png[/IMGDEAD]

Squarewave at 50hz, 10khx and [email protected] 8/R
:
[IMGDEAD]http://s24.postimg.org/sc5b2bncl/10khz_50hz_20khz.png[/IMGDEAD]

Square and sinewave with a 2uf load, not many amps that even can do that..:
[IMGDEAD]http://s14.postimg.org/oarg023w1/sqwave10khz_20khz.png[/IMGDEAD]


You people understand why I like this amp? :) But with 4 more pair of output devices and hopefully the same testresults, then i have the only amp i´ll ever need.

The Superamp at Prof. Leachs page is a updated version if this amp, with a more compact layout.

I intend to use the suggested and preferred devices thah Prof. Leach has suggested.


Can you help me reach that goal?
 
Last edited:
Yeah, i know that i "don´t" need it, but i still want it. The speakers are capable of over 120 db, and i would like an amp that can push them to that level.

AndrewT, can you please help me with how i should go on with putting 4 pairs of extra out devices on the superamp? You are one of the most knowledgeable on this forum, and i would really like to have your help. If you find it interesting, of course.

There must be more people that would be interested in a real high output amp, i suppose.

Best regards, Mathias
 
If you don´t have the skills I think it´s unwise to experiment with your own solutions. .......
Agreed.
I will go further:
If one doesn't have the skills to recognise one's own limitations then better to find another hobby.
With ill informed or uninformed guesses with mains electric then it's unwise to experiment with your own solutions.
 
I will do this:

I will build the superamp. Then, i learn me something, under the time i biuld it, i will read out my 3 books i have on the subject. Cordells, Selfs and Sloans book.

I think i know how to do it, just put more devices in parallel with the already existing devices. But, at the moment i dont know how to calculate the other things that will ned to be adjusted and replaced/newly put in. SOA and so on, that i can. But the other bit, will come with time.

It is the part with the resistors i dont know how to change. I belive that i can do it, with a little more reading and experience, which oi will get under the time i build the superamp.

I am not interested in buying modules, i know how to connect cables.. that wont learn me a single bit...
was interested in that until i realised that it is just to connect some cables, dvs 30 minutes.

This has been an on and off interest for me for the last 3 years. Then in the beginnign it was all chipamps. Now, i am interested in building a real amplifier. This is a real amplifier that fulfills all my requrements, but i still want mroe power. When i have built it, then i will know how to redo it with more devices.

Thanks for your sparse help so far. You are right. I still think you could have telled me what i should read on to get hold of the info on paralleling moree devices and how to calculate the resistors value, and how the bias should be changed.

Can you please provide me with what i should read on up regarding paralleling more output devices and how that changes the bias? Last time I ask for help in this thread, hopefully someone will answer it..