Hi,
The simple answer is no.
The more complex answer is perhaps with high impedance speakers,
but come across a pair of low impedance speakers and the probability
of self-destruction will be much higher.
Having had similar circuits in a bass amplifier running off 50V rails,
3 different versions of similar circuits have self expired. Might be OK
for hi-fi but IMO an unreliable circuit for heavy duty use @ 50V rails.
/sreten.
The simple answer is no.
The more complex answer is perhaps with high impedance speakers,
but come across a pair of low impedance speakers and the probability
of self-destruction will be much higher.
Having had similar circuits in a bass amplifier running off 50V rails,
3 different versions of similar circuits have self expired. Might be OK
for hi-fi but IMO an unreliable circuit for heavy duty use @ 50V rails.
/sreten.
100-150W? Not even by accident!
For an 8 ohm nominal speaker with reasonable reactance margin, perhaps 50-60W would be the maximum safe power. It should also be noted that the amp will lose about 7V off the power rails at maximum power, so efficiency will be lower than expected, and good cooling will be needed even for this safe power level. The design is, as seems usual for kits, a stripped down version of the original Hitach MOS application schematic. I always wonder how one justifies the 0.2$ saving in a proper tail current source and current mirror transistor (instead of diode)...
For an 8 ohm nominal speaker with reasonable reactance margin, perhaps 50-60W would be the maximum safe power. It should also be noted that the amp will lose about 7V off the power rails at maximum power, so efficiency will be lower than expected, and good cooling will be needed even for this safe power level. The design is, as seems usual for kits, a stripped down version of the original Hitach MOS application schematic. I always wonder how one justifies the 0.2$ saving in a proper tail current source and current mirror transistor (instead of diode)...
Oh well I may as well add my bit.
I would not even run them at the specified rails. I havn't done a SOAR v Load analysis, but I just know it would fail for a typical 8 ohm speaker. Still for a home situation it could survive, but definately not for extended high power use.
I would not even modify this cct for greater power because it actually aint that good. You are better off starting a new cct.
Cheers
Quasi
I would not even run them at the specified rails. I havn't done a SOAR v Load analysis, but I just know it would fail for a typical 8 ohm speaker. Still for a home situation it could survive, but definately not for extended high power use.
I would not even modify this cct for greater power because it actually aint that good. You are better off starting a new cct.
Cheers
Quasi
Back in ther day, Hafler made a model like that. With ±52V it could do 60W/8R and 90W/4R. It will require very large heatsinks to do it.
The Hitachi L-FETs have 1 ohms RDS at room, 1,7 ohms over temperature. At near 100W/4R you are at maximum drain current (7A) and RDS is maximum (1,7), and you will drop about 12V across the device from drain to source.
Hafler originally used two pair of outputs at ±63V for their 100W/8R design, later increasing to three pair. Driver transistors were added to handle the extra gate capacitance.
The Hitachi L-FETs have 1 ohms RDS at room, 1,7 ohms over temperature. At near 100W/4R you are at maximum drain current (7A) and RDS is maximum (1,7), and you will drop about 12V across the device from drain to source.
Hafler originally used two pair of outputs at ±63V for their 100W/8R design, later increasing to three pair. Driver transistors were added to handle the extra gate capacitance.
I don't think so, the poor T4/5/3 TO92 devices are already dissipating ~370mw. This means they are already too hot to touch. Further increasing the voltage will make them smoke.
Replacing the mosfets with IRF type will make you need to replace the trimpot r13 with a vbe/vgs multiplier as the IRFs need thermal compensation.
What makes you believe you can take some simple amp and increase power output by simply increasing voltage supply ? There is a reason why high power amps are more complex...
Quasi's amps are about the minimum complexity necessary for reliable high power amps.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=43331
Is your subwoofer 4 or 8 ohms ? What is your targeted power output ?
I doubt that 4 pairs of TO220 mosfets will allow more than ~200watts.
What is your targeted voltage supply ? Are you experienced enough to handle lethal supply voltages ?
Mike
Replacing the mosfets with IRF type will make you need to replace the trimpot r13 with a vbe/vgs multiplier as the IRFs need thermal compensation.
What makes you believe you can take some simple amp and increase power output by simply increasing voltage supply ? There is a reason why high power amps are more complex...
Quasi's amps are about the minimum complexity necessary for reliable high power amps.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=43331
Is your subwoofer 4 or 8 ohms ? What is your targeted power output ?
I doubt that 4 pairs of TO220 mosfets will allow more than ~200watts.
What is your targeted voltage supply ? Are you experienced enough to handle lethal supply voltages ?
Mike
mike, i'm young but i'm a diyer from more than 7 years...
i'm only an electronic student, i didn't have the preparation of a engineer but i 'have someting called passion(i also learned to play with english only for this great site...), a bit of experience(passing through numerous fails and "booms") i have build +/- 10 amps some low power and also medium high power... i'm not a total newbie in this range...I KNOW WATH I DO(i ever be carefull with electrics!!!)
however i dont reach to understand HOW I MISS in all'quasi's amps
, the quasis negative rail appears to me in my dreams... BUT I'M MORE STRONG THAN AN AMPLIFIER, AND I WILL TRY TO BUILD IT SINCE MY LAST DAY!!!
i have two subs ready for kick... an isobaric push-pull from coral-peerless of 4+4ohms speakers and a 12'' 8ohms RCF woofer in a 56 lit reflexenclosure...
i'm only an electronic student, i didn't have the preparation of a engineer but i 'have someting called passion(i also learned to play with english only for this great site...), a bit of experience(passing through numerous fails and "booms")
i have build +/- 10 amps some low power and also medium high power... i'm not a total newbie in this range...I KNOW WATH I DO(i ever be carefull with electrics!!!)however i dont reach to understand HOW I MISS in all'quasi's amps
, the quasis negative rail appears to me in my dreams... BUT I'M MORE STRONG THAN AN AMPLIFIER, AND I WILL TRY TO BUILD IT SINCE MY LAST DAY!!!i have two subs ready for kick... an isobaric push-pull from coral-peerless of 4+4ohms speakers and a 12'' 8ohms RCF woofer in a 56 lit reflexenclosure...
Flavio88 said:mike, i'm young but i'm a diyer from more than 7 years...
i'm only an electronic student, i didn't have the preparation of a engineer but i 'have someting called passion(i also learned to play with english only for this great site...), a bit of experience(passing through numerous fails and "booms")
Hello Flavio,
Dont get dissapointed with Mike's serious enough statements, You are passing through a phase which is known as "Social Engineering Runaway" and which is very common in young and passionate guys like you....
Mike donot apply your Class-A biasing to Flavio's passion.....
best regards,
K a n w a r
It is possible to change T3,4,5 to 2N3440/5415 types, re-bias every stage, and use the MOSFETs as drivers to run three pair of MJ15024/25 outputs.
With ±63V from a 45-0-45 1KVA transformer and a pair of 22,000µF 75V caps I estimate it would do 300W per channel at 4 ohms.
The outputs would be biased what Crown calls AB+B, with about 0.34V on the bases of the outputs at idle with 3R3 source resistors to drive the outputs. This would be 100mA of bias in the driver stage.
With ±63V from a 45-0-45 1KVA transformer and a pair of 22,000µF 75V caps I estimate it would do 300W per channel at 4 ohms.
The outputs would be biased what Crown calls AB+B, with about 0.34V on the bases of the outputs at idle with 3R3 source resistors to drive the outputs. This would be 100mA of bias in the driver stage.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.