Ok, I have a bipolar design for my 400W amp, so I guess I'll use that,
and I'll probably use MOSFETs for the big one. (8 x http://members.optusnet.com.au/~aussieamps/1kwamp.htm )
[Edited by Kilowatt on 12-11-2001 at 09:57 PM]
and I'll probably use MOSFETs for the big one. (8 x http://members.optusnet.com.au/~aussieamps/1kwamp.htm )
[Edited by Kilowatt on 12-11-2001 at 09:57 PM]
No, I didn't. I'm going to build 4 Leach amps instead. I still don't have the layout for the 400W one, but you can buy the PCBs from th author. I might be able to help you with a cartain 550W MOSFET amp by G. Randy Sloan. As is, it supposedly measures 0.05% THD@550W into 4 ohms, not bad for a MOSFET amp of it's design.
Hmm, >I< might build it, though.
I've got this monster 'wind tunnel' heatsink that's hardwired
to run up to five pairs of output devices per side. It was originally a
voltage regulator for a mainframe computer, I think. I've had
it in my collection for twenty years or more, waiting to get
around to doing something really crazy, and this might be
it.
Have to come up with an awfully big transformer and a chassis,
though. Wish I had access to sheet metal bending tools.
Being unable to actually >build< projects has often prevented
me from doing them.
Meanwhile, I can rig up something using one of my old Leach
boards to drive several output pairs just to see how it behaves
thermally. Not sure if the MJE15030/31 can drive five bipolar
devices into low impedance loads though, but there's one
way to find out...
I've got this monster 'wind tunnel' heatsink that's hardwired
to run up to five pairs of output devices per side. It was originally a
voltage regulator for a mainframe computer, I think. I've had
it in my collection for twenty years or more, waiting to get
around to doing something really crazy, and this might be
it.
Have to come up with an awfully big transformer and a chassis,
though. Wish I had access to sheet metal bending tools.
Being unable to actually >build< projects has often prevented
me from doing them.
Meanwhile, I can rig up something using one of my old Leach
boards to drive several output pairs just to see how it behaves
thermally. Not sure if the MJE15030/31 can drive five bipolar
devices into low impedance loads though, but there's one
way to find out...
djk:
Yeah, that's what I was thinking, too. It'll be some time before
I've got the money to buy enough parts to build a working
version with the parts you suggest, but it appears doable. I'll
have to think about how to implement the protection circuit.
It'd be an excuse to order a really large batch of output transistors and then sell what I don't use--assuming I don't
kill a batch during testing! There's going to be a lot of energy
stored up in the power supply...
Kilowatt:
Unless I get really lucky again and find a usable chassis
already built, I'll have to sit down and design my own, maybe
pay to have a sheet metal shop fabricate it.
Better find some books on sheet metal design...ah, for the
good old days at Solid State Systems, Inc., when I had access
to all those tools and scrap metal.
I hate to think what the heat sink components would cost new.
I'm not even sure who makes this kind of component--it's four
pieces that fit together with some insulated hardware so that
each quadrant is "live" and insulating the output devices is
unnecessary. Great thermal characteristics but busybusybusy
with lots of little and unique parts.
I was just real lucky at the time that a friend practically gave the
thing to me. Next weekend I'm going to check with a surplus
place in Tacoma (Electronic Dimensions) and see if Glen's got
anything like it, or has seen them on the market.
Yeah, that's what I was thinking, too. It'll be some time before
I've got the money to buy enough parts to build a working
version with the parts you suggest, but it appears doable. I'll
have to think about how to implement the protection circuit.
It'd be an excuse to order a really large batch of output transistors and then sell what I don't use--assuming I don't
kill a batch during testing! There's going to be a lot of energy
stored up in the power supply...
Kilowatt:
Unless I get really lucky again and find a usable chassis
already built, I'll have to sit down and design my own, maybe
pay to have a sheet metal shop fabricate it.
Better find some books on sheet metal design...ah, for the
good old days at Solid State Systems, Inc., when I had access
to all those tools and scrap metal.
I hate to think what the heat sink components would cost new.
I'm not even sure who makes this kind of component--it's four
pieces that fit together with some insulated hardware so that
each quadrant is "live" and insulating the output devices is
unnecessary. Great thermal characteristics but busybusybusy
with lots of little and unique parts.
I was just real lucky at the time that a friend practically gave the
thing to me. Next weekend I'm going to check with a surplus
place in Tacoma (Electronic Dimensions) and see if Glen's got
anything like it, or has seen them on the market.
Hi Kilowatt,
Good MOSFET's cost more than good bi-polar's, but are more rugged when it comes to surviving overload conditions. Most have a positive temp coefficient which means their resistance increases as the temperature goes up (bipolar's resistance drops as temperature increases). This means that MOSFET's don't go into thermal runaway like bi-polar amps can and secondary breakdown is not a problem.
I suggest you read Randy Sloan's "High Power Audio Amplifier Construction Manual". He has several decent amps (biggest one will deliver 550 W into 4 ohms) and has PCB's available for several of them. I built a pair of his 250 W MOSFET amps last year and am very pleased with their performance. Randy's designs are heavily influenced by Dougles Self (which is a good thing in my opinion).
The Hitachi Lateral MOSFET's that Randy uses in most of his designs sound every bit as good as any bipolar output stage I've heard so far.
P.S. If you want to branch out with your own designs I suggest you read Dougles Self's book "Audio Power Amplifier Design Handbook". This guy really knows what he is talking about when it comes to audio design. Sloan's book is better than Self's you are looking for "cook book" designs to build.
Best Regards,
Phil Ouellette
Good MOSFET's cost more than good bi-polar's, but are more rugged when it comes to surviving overload conditions. Most have a positive temp coefficient which means their resistance increases as the temperature goes up (bipolar's resistance drops as temperature increases). This means that MOSFET's don't go into thermal runaway like bi-polar amps can and secondary breakdown is not a problem.
I suggest you read Randy Sloan's "High Power Audio Amplifier Construction Manual". He has several decent amps (biggest one will deliver 550 W into 4 ohms) and has PCB's available for several of them. I built a pair of his 250 W MOSFET amps last year and am very pleased with their performance. Randy's designs are heavily influenced by Dougles Self (which is a good thing in my opinion).
The Hitachi Lateral MOSFET's that Randy uses in most of his designs sound every bit as good as any bipolar output stage I've heard so far.
P.S. If you want to branch out with your own designs I suggest you read Dougles Self's book "Audio Power Amplifier Design Handbook". This guy really knows what he is talking about when it comes to audio design. Sloan's book is better than Self's you are looking for "cook book" designs to build.
Best Regards,
Phil Ouellette
Ha, guess what. I AM reading G. Randy Sloan's "High Power Audio Amplifier Construction Manual". The 550W amp I've been talking about in this thread is the very one from that book. There is no layout for it, for it is meant to be derived from the 250W MOSFET amp in the book.
I was going to get Dougles Self's "Audio Power Amplifier Design Handbook" at the same time that I got "High Power Audio Amplifier Construction Manual", but they (Amazon.com) couldn't get it. I'll have to find it somewhere else.
I was going to get Dougles Self's "Audio Power Amplifier Design Handbook" at the same time that I got "High Power Audio Amplifier Construction Manual", but they (Amazon.com) couldn't get it. I'll have to find it somewhere else.
Leach SuperAmp
I'm selling Leach SuperAmp PCB's. Good design, good reputation. The boards are $25 each and they are a little bit more complicated, but the parts are not too hard to find and if you need any help I can probably manage. Check these sites to see if you are interested:
Remember, the boards are 500W into 4ohms when they are not bridged!
http://djss.hypermart.net/beast.htm
http://users.ece.gatech.edu/~mleach/superamp/
=DJSS=
(djss)(at)(mail.utexas.edu)
I'm selling Leach SuperAmp PCB's. Good design, good reputation. The boards are $25 each and they are a little bit more complicated, but the parts are not too hard to find and if you need any help I can probably manage. Check these sites to see if you are interested:
Remember, the boards are 500W into 4ohms when they are not bridged!
http://djss.hypermart.net/beast.htm
http://users.ece.gatech.edu/~mleach/superamp/
=DJSS=
(djss)(at)(mail.utexas.edu)
djk:
I'm definitely interested in getting the MJ21193/4 transistors:
I was able to buy 10 of the -3s but they wouldn't sell the others
in less than 100s. I only got the 3s because they already
had a broken rail of that part (Active Electronics).
At this point, I'm figuring on 20 of the -4s and 10 more of the 3s;
I'll let you know in a day or several. If we can get a good price
break on a quantity order, that certainly helps. From those I
can select matched sets for my existing amp.
After that, it's finding chassis, transformers and some other
parts. Alas, I definitely don't have the money for those; I just
dropped $600 in car repairs with at least that much more in
deferred repairs, so the most I can hope to do is breadboard
a channel with a lower power transformer.
I need a job, real bad.
I'm definitely interested in getting the MJ21193/4 transistors:
I was able to buy 10 of the -3s but they wouldn't sell the others
in less than 100s. I only got the 3s because they already
had a broken rail of that part (Active Electronics).
At this point, I'm figuring on 20 of the -4s and 10 more of the 3s;
I'll let you know in a day or several. If we can get a good price
break on a quantity order, that certainly helps. From those I
can select matched sets for my existing amp.
After that, it's finding chassis, transformers and some other
parts. Alas, I definitely don't have the money for those; I just
dropped $600 in car repairs with at least that much more in
deferred repairs, so the most I can hope to do is breadboard
a channel with a lower power transformer.
I need a job, real bad.
Oops, let me rephrase that:
I'll need ten more of the MJ21194s and twenty of the MJ21193s.
I bought ten of the former through Active Electronics for $2.59
apiece (shipping and taxes extra).
And I've named the amplifier; it's bodacious, so it's got to be...
...Bubba.
I'm guess that 5 pairs of output devices are good for 500 watts
into 8 ohms and I might get away with 1000 watts into 4 ohms,
but I'm not counting on it surviving 2 ohm loads without some
serious protection circuitry. I'm thinking more in terms of
swinging a lot of voltage rather than a lot of current.
I'll need ten more of the MJ21194s and twenty of the MJ21193s.
I bought ten of the former through Active Electronics for $2.59
apiece (shipping and taxes extra).
And I've named the amplifier; it's bodacious, so it's got to be...
...Bubba.
I'm guess that 5 pairs of output devices are good for 500 watts
into 8 ohms and I might get away with 1000 watts into 4 ohms,
but I'm not counting on it surviving 2 ohm loads without some
serious protection circuitry. I'm thinking more in terms of
swinging a lot of voltage rather than a lot of current.
"On that subject with anything higher than 300 watt
mosfet is better. bjt are somewhat to fragile for high power. If you use high power mosfet it's going to be nearly indestructible. " A statement that is often made, and never with any proof.The most commonly used MOSFET for this use is the 2SJ162/2SK1058 pair.About USD$7 each.Only 100W SOA at 100V and a maximum die temperature of 150*C.Consider the MJ21193/21194 pair.About USD$2.50 each.Has 200W safe area at 100V and a maximum die temperaure of 200*C.The same amount of money will buy you over FIVE times as much safe area of operation wattage.Of course if you like the sound of lateral MOSFET bass(can you say mud?).Of course the mud is from the 12V drain to source voltage drop at 7A, that means it is like a 1.7 ohm resistor.The bipolar is less than 1.4V at 8A, or .17 ohm, ONE TENTH the loss.
mosfet is better. bjt are somewhat to fragile for high power. If you use high power mosfet it's going to be nearly indestructible. " A statement that is often made, and never with any proof.The most commonly used MOSFET for this use is the 2SJ162/2SK1058 pair.About USD$7 each.Only 100W SOA at 100V and a maximum die temperature of 150*C.Consider the MJ21193/21194 pair.About USD$2.50 each.Has 200W safe area at 100V and a maximum die temperaure of 200*C.The same amount of money will buy you over FIVE times as much safe area of operation wattage.Of course if you like the sound of lateral MOSFET bass(can you say mud?).Of course the mud is from the 12V drain to source voltage drop at 7A, that means it is like a 1.7 ohm resistor.The bipolar is less than 1.4V at 8A, or .17 ohm, ONE TENTH the loss.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.