Building a Monster... Class A

[Adrculda]

Hi everyone...
well after pokin around the net and lurkin around here, i came to the conclusion that i have to do something with my spare time and what better to do and try and fry myself trying to build a monster LOL

i came across the Elliott Sound Products website and a certain Project 117 caught my eye...

After reading the page several times over came to the demented idea.. I Have To Build one, but not that one must beef it UP!!!
so after a couple of hours on the computer i slowly reviewed and "improved" a couple of items:

- Power Supply
DOUBLED the caps from 8 to 16 10,000uf caps
TRIPLED the Transformer from a 2KVA to a 6KVA unit ( in case i want to add a second channel )
RECTIFIERS went from 2 x 35A units to 3 x 50A units
Also Bumped the voltage from 135V to 142V (i believe will be 140V) by using a 12v transformer instead of the 6V as well as using 5A rectifiers instead of the 1A ones shown

-Amplifier section
As of right now...
board was a 1 piece deal, but i moved the drivers onto a separate board to fit a larger heat sink and forced cooling along with slots cut into the board for breathing room for the resistors. Also number of drivers was increased from 18 to 40 !! ( see diagrams )

I WILL do other changes on th course of the build, but that's it for now...

AS OF TODAY...
dropped off the pre-driver section to have a few board made up.
the guy was amazed at first and was wondering if i was building a welder of some sort
When i told him what i was working on he gave me the look and with a tiny voice he asked if he heard right...
At this point I couldn't resist the temptation, but had to trow in that i left the flux capacitor out of the diagram since that's the top secret part of it

long story short...
$50 for 4 boards and got him to order me 2 forced cooling heat sinks, really nice ones at that!!
Its pretty much a square aluminum tube with copper honeycomb design inside and a wall thickness about 3/16 on the mounting side.
dimensions are as follows 120mm x 120mm x 400mm long, and though it was a steal for $70 each. He uses them to keep triacs cool dissapating over 3500W into the tunnel and keeping it at only 22C.

As far as the transformer goes...
The closest i found so far was a monster 8KVA toroid with 2oo/440V primaries and Tripple tapped secondaries at 25, 60, 120V on each secondary... $500 ( I think its a bit too big.. i mean i don't want to "brown out" the whole neighbour hood when i turn my amp on or i'm listening to music )

STILL TO GO:
Need to find BOM so i can start ordering some parts cuz the ones that were on the schematic are there but its incomplete information.
If anyone has any info on this would be appreciated.

Thanks
Adrian C.

 

[lineup]

Insanity Can Be Yours
1,500W / 4 Ohms Power Amplifier
Rod Elliott (ESP)
Introduction

This project is mainly in answer to those for whom no amount of power is enough.

Yes, it is like plain insanity to me, too.

Nevertheless, it is a challenge to try to see how many watts can be mastered by an audio amplifier.
Even good old Nelson Pass has made some really insane amplfiers
when it comes to power and current output to speakers.

Here is the basic schematic from Rod Elliott Project 117.
Uses many MJL21193/94 or MJ15024/25 TO-3 metal can transistors.

An externally hosted image should be here but it was not working when we last tested it.

Figure 1 - 1.5kW Power Amplifier

 

[bear]

why?

there are a number of really good class A designs on this forum, in the Pass section, and elsewhere online...

for that matter, the Krell copiers all have schematics here and they use copious numbers of output devices...

Guess I am saying that if ur going through the expense of heatsinking that much which = $$, you might as well build an amp that happens to be a good circuit to start with??

_-_-bear

PS. do you really want to hear the sound of a fan cranking when you run the amp??

 

[wg_ski]

Not to rain on anybody's parade, but this post screams of "newbie". An appropriately experienced builder would know that the auxiliary supplies don't need 5A rectifiers for example. And would already have acquired the necessary toroid(s) before proceeding. And that P117 isn't and can't be made to work class A. It would be insanity for a newbie to attempt this.

Physical construction of a project like this is 99.999% of the work. The circuit itself is nothing. The majority of the "bill of materials" is wiring harnessses, custom machined parts, heavy iron and hardware - not transistors, resistors,and capacitors. And you won't really know what you need until you seriously sit down and start mocking it up. That process may take weeks or months depending on your shop capabilities and how frustrated you get along the way. It may take several iterations and false starts before everything starts falling into place.

I'm about 2 months into the actaul build of four 2kW/channel PA amps. The design process took about 2 years from concept to a prototype board working on a test bench. Then the fun really began. The final chassis design and layout started in March and finally came together back in October. There were a lot of problems with fit and finish to figure out, and without a real machine shop it's a CHORE. I've just now got the power supplies and supervisor boards built, tested, and running. The front end and temp sensor circuits are under construction now. The power amp boards will be done LAST, because it is a known working amp design, just scaled up a bit from the prototype. It's an old class B design that I've used since 1990 at voltages up to +/-127V and adapted to class H by adding the QSC-style rail switches. The modules have already been mocked up and tested for proper fit and airflow. The project 117 writeup has no real mention of what's required to get a big amp like this working, fit into a chassis and reliable, but the basic steps are pretty much the same regardless of design and always different in the details. It is assumed that the builder can do all of this without any hand-holding. If you can't, the project is too big for you. If you've got access to a complete machine shop and manufacturing facilities - great - it will make life a lot eaier than it has been for me.

 

[jacco vermeulen]

Oh, in that case, you guys think it's the same idiot who keeps coming back every moon cycle ?

 

[ostripper]

wow, it took till post 4 to finally point out that there is no class
A..

For that much power 7 pairs of mosfets tacked on to the
blameless would be as far as I would go.(class AB/B)
BJT,s just don't seem to have the SOAR to be economical
at insane levels
For Insane power class d is the only way to go.
I just saw the 24 1KW power amps at the eagle's concert
(lucky me , sitting right next to them)
all class D/ no heatsinks in back (24 X QSC 1KW).
OS

 

[wg_ski]

But class D is even less newbie-friendly at this power level. That requires a working knowledge of high frequency magnetics and more specialized test equipment. You can overbuild a class B amp (relatively) easily. Switching circuits are a lot harder to overbuild and still get nanosecond rise times and low overshoots.

 

[ostripper]

That is why I mentioned the 7 pair of Vertical mosfets,
as I have not yet advanced enough to try a D.

Rod , in the article mentioned that he never built the project
and that IF someone did they better have the big cash and
know what they were doing..

A pair of Hypex UCD 700's are what I meant, not building from
scratch. A project using them is still 800-1K$
OS

 

[Nico Ras]

That is why I mentioned the 7 pair of Vertical mosfets,
as I have not yet advanced enough to try a D.

Rod , in the article mentioned that he never built the project
and that IF someone did they better have the big cash and
know what they were doing..

A pair of Hypex UCD 700's are what I meant, not building from
scratch. A project using them is still 800-1K$
OS

Think $ 800 - 1K is bad, go look here....

http://www.higherfi.com/amplist/amplist.htm

 

[d to the g]

Think $ 800 - 1K is bad, go look here....

http://www.higherfi.com/amplist/amplist.htm

I live about 20 minutes from here, and they had advertised for a tech position. When I answered the ad and enquired into the compensation, the gentlemen was incredibly rude. He told me that just being able to touch this equipment would be it's own compensation.

 

[Nico Ras]

I live about 20 minutes from here, and they had advertised for a tech position. When I answered the ad and enquired into the compensation, the gentlemen was incredibly rude. He told me that just being able to touch this equipment would be it's own compensation.

So did you get to touch anything?????

 

[Adrculda]

Yes, it is like plain insanity to me, too.

Nevertheless, it is a challenge to try to see how many watts can be mastered by an audio amplifier.
Even good old Nelson Pass has made some really insane amplfiers
when it comes to power and current output to speakers.

Here is the basic schematic from Rod Elliott Project 117.
Uses many MJL21193/94 or MJ15024/25 TO-3 metal can transistors.

An externally hosted image should be here but it was not working when we last tested it.

Figure 1 - 1.5kW Power Amplifier

why?

there are a number of really good class A designs on this forum, in the Pass section, and elsewhere online...

for that matter, the Krell copiers all have schematics here and they use copious numbers of output devices...

Guess I am saying that if ur going through the expense of heatsinking that much which = $$, you might as well build an amp that happens to be a good circuit to start with??

_-_-bear

PS. do you really want to hear the sound of a fan cranking when you run the amp??

Not to rain on anybody's parade, but this post screams of "newbie". An appropriately experienced builder would know that the auxiliary supplies don't need 5A rectifiers for example. And would already have acquired the necessary toroid(s) before proceeding. And that P117 isn't and can't be made to work class A. It would be insanity for a newbie to attempt this.

Sorry my mistake on that, but as far as the rectifiers go...
the third one is for backup since i seen plenty of rectifiers bridges go bad under severe loads due to uneven heat dispersion.
Also most people on here should know that its not only the output transistors that count for output, but also the power supply has to be robust enough to supply the voltage and current when there's a demand for it, specially in low impedance where high current is required!

Physical construction of a project like this is 99.999% of the work. The circuit itself is nothing. The majority of the "bill of materials" is wiring harnessses, custom machined parts, heavy iron and hardware - not transistors, resistors,and capacitors. And you won't really know what you need until you seriously sit down and start mocking it up. That process may take weeks or months depending on your shop capabilities and how frustrated you get along the way. It may take several iterations and false starts before everything starts falling into place.

I'm about 2 months into the actaul build of four 2kW/channel PA amps. The design process took about 2 years from concept to a prototype board working on a test bench. Then the fun really began. The final chassis design and layout started in March and finally came together back in October. There were a lot of problems with fit and finish to figure out, and without a real machine shop it's a CHORE. I've just now got the power supplies and supervisor boards built, tested, and running. The front end and temp sensor circuits are under construction now. The power amp boards will be done LAST, because it is a known working amp design, just scaled up a bit from the prototype. It's an old class B design that I've used since 1990 at voltages up to +/-127V and adapted to class H by adding the QSC-style rail switches. The modules have already been mocked up and tested for proper fit and airflow. The project 117 writeup has no real mention of what's required to get a big amp like this working, fit into a chassis and reliable, but the basic steps are pretty much the same regardless of design and always different in the details. It is assumed that the builder can do all of this without any hand-holding. If you can't, the project is too big for you. If you've got access to a complete machine shop and manufacturing facilities - great - it will make life a lot eaier than it has been for me.

It shouldn't be an issue, i have made several enclosures for some class D car audio monsters i have built years back.
Also this will be mostly to drive the low end at my gigs ( 4 x 18 EV EVX18B's loaded into folded horn cabinets ) and they require the good old brute force approach sadly.

You want to use a 6KVA transformer where? Why bother with a transformer and just run it straight off your mains feed into your property.

Hahah ... sure why not
it would save me some serious cash on that transformer
but i think i will try and NOT burn down my house...

Oh, in that case, you guys think it's the same idiot who keeps coming back every moon cycle ?

Idiot???
why cuz i'm running a DJ business and like to have and amp that will drive low resistance loads all day and be stil reliable after a harmonic beating, because some of the music that ppl hand you to play at these events, quality wise its like running your finger nails on a chalk board, and been a couple of times where not even my compressor nor my De-Esser could not clean up the signal good enough !!

wow, it took till post 4 to finally point out that there is no class
A..

For that much power 7 pairs of mosfets tacked on to the
blameless would be as far as I would go.(class AB/B)
BJT,s just don't seem to have the SOAR to be economical
at insane levels
For Insane power class d is the only way to go.
I just saw the 24 1KW power amps at the eagle's concert
(lucky me , sitting right next to them)
all class D/ no heatsinks in back (24 X QSC 1KW).
OS

Well
sad to say but i have redesigned the output section with 20 SPT's (10NPN and 10 PNP ).

 

[Steve Dunlap]

I live about 20 minutes from here, and they had advertised for a tech position. When I answered the ad and enquired into the compensation, the gentlemen was incredibly rude. He told me that just being able to touch this equipment would be it's own compensation.

I can't believe they would let a tech get close to that stuff. You know the owner would NEVER settle for anything but a factory repair.

I have touched and heard at least half of that stuff. I just feel truly blessed.

 

[bear]

Check the design with a lower power/current version, biased into class A and see how it looks clipped...
you might not like it?

You did not say you wanted a PA amp for driving Bass Bins...

Hauling those big xfmrs around will be heavy... glad ur carrying them not me.

I'd sooner go for a lightweight class D commercial amp one per speaker than haul a big monster.

If it was for a fixed location, different story.

I don't see any reason he can't change the bias circuit and bias the sucker into class A...

However, maybe you want to look at how QSC and some others did PA amp design - they work gud for that application. You could buy a fried QSC (or similar) and take those guts, beef up the driver transistors, up the biass and see how that flies?

_-_-bear

PS. not a thing wrong with 6kva power transformers... I used 4kva worth (that's the limitation of what I was able to test them to...) on my 2 x 180w/8ohm Symphony No.1 amplifier... of course you'll need a soft start and a place to plug in a 6kva PS with a load on it... maybe you want to run off 240vac??

PPS. the sweetest bass amp in the world will have zero effect on the "chalkboard" bands sound... "GIGO" rules!!

 

[bear]

I live about 20 minutes from here, and they had advertised for a tech position. When I answered the ad and enquired into the compensation, the gentlemen was incredibly rude. He told me that just being able to touch this equipment would be it's own compensation.

You should have said "oh, that's fine... you are so right!" "When can I come over?"

Then when you got there, instructed him on the proper use of a 200watt American Beauty Iron!

_-_-bear

 

[Adrculda]

Check the design with a lower power/current version, biased into class A and see how it looks clipped...
you might not like it?

You did not say you wanted a PA amp for driving Bass Bins...

Hauling those big xfmrs around will be heavy... glad ur carrying them not me.

I'd sooner go for a lightweight class D commercial amp one per speaker than haul a big monster.

If it was for a fixed location, different story.

I don't see any reason he can't change the bias circuit and bias the sucker into class A...

However, maybe you want to look at how QSC and some others did PA amp design - they work gud for that application. You could buy a fried QSC (or similar) and take those guts, beef up the driver transistors, up the biass and see how that flies?

_-_-bear

PS. not a thing wrong with 6kva power transformers... I used 4kva worth (that's the limitation of what I was able to test them to...) on my 2 x 180w/8ohm Symphony No.1 amplifier... of course you'll need a soft start and a place to plug in a 6kva PS with a load on it... maybe you want to run off 240vac??

PPS. the sweetest bass amp in the world will have zero effect on the "chalkboard" bands sound... "GIGO" rules!!

it will be a 240VAC input (dual phase) and also will be using a soft start
right now trying to decide if i want to to keep the power supply internal or external to less crowd the case, either way im well aware it will be over 55Lbs for the power supply alone and anoather 10-12 Lbs for the actual amp.

The only thing that stops me from making the power suplly external is the cabling an connector that will be able to handle 40+ Amps and 135VDC. Also ampliefiers are have a high current/voltage demand.
the only way i see feasible for this to work is to have anoather set of caps inside the case, which really wouldnt hurt, but only add to the extra expense of building the monster.

 

[lgreen]

good luck

Don't let them get you down.

Jacco was kidding about a rare species of animal found in the scifi realm. He is never serious...except he is.

People here will help you, EXCEPT if your questions indicate you are way over your head they will mock you try to get you to give up....for your own good.

So lets see what you got, and good luck.

 

[Adrculda]

Re: good luck

Don't let them get you down.

Jacco was kidding about a rare species of animal found in the scifi realm. He is never serious...except he is.

People here will help you, EXCEPT if your questions indicate you are way over your head they will mock you try to get you to give up....for your own good.

So lets see what you got, and good luck.

Thanks...

But im the "rare" head strong type that when i got my mind made up nothing really gets in my way.

 

[GK]

Well sad to say but i have redesigned the output section with 20 SPT's (10NPN and 10 PNP ).

Only 10 pairs? A 1500W / 4ohms amplifer biased into class A will dissipate about 4000W at idle. Good luck dissipating ~200W per transistor

Cheers,
Glen

 

[OMNIFEX]

It shouldn't be an issue, i have made several enclosures for some class D car audio monsters i have built years back.
Also this will be mostly to drive the low end at my gigs ( 4 x 18 EV EVX18B's loaded into folded horn cabinets ) and they require the good old brute force approach sadly.

Hi Adrculda

I use EVX 180Bs in twelve double eighteen enclosures sitting on six Crown MA 5000vzs each getting 240 volts

it will be a 240VAC input (dual phase) and also will be using a soft start
right now trying to decide if i want to to keep the power supply internal or external to less crowd the case, either way im well aware it will be over 55Lbs for the power supply alone and anoather 10-12 Lbs for the actual amp.

The only thing that stops me from making the power suplly external is the cabling an connector that will be able to handle 40+ Amps and 135VDC. Also ampliefiers are have a high current/voltage demand.
the only way i see feasible for this to work is to have anoather set of caps inside the case, which really wouldnt hurt, but only add to the extra expense of building the monster.

Might I suggest 6-Gauge SO cable? I use that as speaker wire. It can withstand up to 600 Volts.

I use the spade lugs connections on the rear of the 5000s with neutrik speakons to connect to the subwoofers. Using cable with that type of density can easily handle 40 amps continuous.

Good luck on your design and, please post some photographs of the work in progress.

Cheers!