I've played around a little with this approach. The Lynx 3.0 is a popular example here on this forum. It's main topological difference from the PA300 is the use of a pair of common base transistors to do the DC level shifting of the op amp output, where the PA300 uses a resistor divider.
The main problem with using an op amp for the input stage is that you can't get all the voltage gain out of it because of its limited supply rails. That means you have to add some sort of additional voltage gain stage, and suddenly you have more gain than you know what to do with.
It's not hard to design circuits of this type with 120 - 140 dB of loop gain. A traditional dominant pole compensation approach, with gain declining at 20 dB per decade above the dominant pole, means you have 6 or 7 decades before you reach unity loop gain. If you put the pole at 1 Hz, you don't get there until the MHz regions, and its really hard to get away with that without first running into other poles caused by transistor capacitances in other parts of the circuit.
You can make these circuits work, but they are notoriously hard to make stable, and the compensation can get pretty complex. One word of advice: don't try to change op amps unless you're ready to redesign the entire compensation scheme.
All of this said, I also have to say that a well built version of an op amp based design would probably beat the Melody 400. I am not overly impressed with that design. I can't help but think there are other choices that might be better than anything suggested so far. I'd take a look at Leach's Super Amp, to start with.
regards,
Mike
one does not need to go the Super Amp route if one uses modern output devices that have a much more robust HV SOAR.
The Lo Tim version manages 311W into 4r0 with a 3pair output stage and all the data is on the Forum for a 5pair output stage that will easily achieve 400W if that is what is needed.
The Lo Tim version manages 311W into 4r0 with a 3pair output stage and all the data is on the Forum for a 5pair output stage that will easily achieve 400W if that is what is needed.
Ermac -
I think I agree w/ Andrew that the SuperAmp is a poor suggestion on my part, given that you've said you want something optimized for 50 V rails and a 2 ohm load.
I'm not aware of any existing DIY design that's really optimized in that way, but you should probably start with something designed for 50 - 60 V rails, and then
(1) make sure that 50-0-50 supply can provided the currents necessary to drive 2 ohm loads,
(2) add parallel output devices if necessary to keep them in their safe operating region with 2 ohm loads,
(3) make sure the driver stage has enough grunt to drive the extra output devices,
(4) make sure everything is still stable, look particularly for parasitic oscillations at the peaks of voltage swings.
All of this can get rather involved. Are you looking for something where you can buy a PCB for a tested design, or are you happy with modifying or extending a design to better meet your specs?
regards,
Mike
I think I agree w/ Andrew that the SuperAmp is a poor suggestion on my part, given that you've said you want something optimized for 50 V rails and a 2 ohm load.
I'm not aware of any existing DIY design that's really optimized in that way, but you should probably start with something designed for 50 - 60 V rails, and then
(1) make sure that 50-0-50 supply can provided the currents necessary to drive 2 ohm loads,
(2) add parallel output devices if necessary to keep them in their safe operating region with 2 ohm loads,
(3) make sure the driver stage has enough grunt to drive the extra output devices,
(4) make sure everything is still stable, look particularly for parasitic oscillations at the peaks of voltage swings.
All of this can get rather involved. Are you looking for something where you can buy a PCB for a tested design, or are you happy with modifying or extending a design to better meet your specs?
regards,
Mike
Maybe Carlos's DHR Turbo is a suitable alternative.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=124931&pagenumber=1
..Todd
http://www.diyaudio.com/forums/showthread.php?s=&threadid=124931&pagenumber=1
..Todd
Honestly I'm looking for a low distortion and powerfull amplifier which can be built with these components that I have:
35V-0-35V transformer (8A);
2SC5200/2SA1943;
TIP42/TIP41;
2SC669/2SB649;
BD139/BD140;
2N5551/2N5401;
The amp could be able to drive 2 ohms speakers and also possesses an output short circuit protection.
I won't buy a PCB but I'll build one.
Now please suggest me the best choice!
Thank you!
35V-0-35V transformer (8A);
2SC5200/2SA1943;
TIP42/TIP41;
2SC669/2SB649;
BD139/BD140;
2N5551/2N5401;
The amp could be able to drive 2 ohms speakers and also possesses an output short circuit protection.
I won't buy a PCB but I'll build one.
Now please suggest me the best choice!
Thank you!
Ermac -
I have no recommendations that use only those devices. To get that, you may have to design it yourself.
Carlos' Turbo amp that Taj mentioned is a definite candidate.
There was a design that popped up on this forum a few months ago, called Patchwork, designed by MJL21193. It uses the C5200/A1943 output devices, although it only uses 2 pair, and you might need to add some more. See here for what I think is the latest schematic: http://www.diyaudio.com/forums/showthread.php?postid=1760403#post1760403
The FrugalAmp by OStripper is also a candidate, see the thread at http://www.diyaudio.com/forums/showthread.php?s=&threadid=133306&highlight=
The bottom line is that there are lots to choose from. Poke around in the Amplifiers -> Solid State section of this forum, and you'll find quite a few. None of them can be called the best of the lot, although opinions abound.
regards,
Mike
I have no recommendations that use only those devices. To get that, you may have to design it yourself.
Carlos' Turbo amp that Taj mentioned is a definite candidate.
There was a design that popped up on this forum a few months ago, called Patchwork, designed by MJL21193. It uses the C5200/A1943 output devices, although it only uses 2 pair, and you might need to add some more. See here for what I think is the latest schematic: http://www.diyaudio.com/forums/showthread.php?postid=1760403#post1760403
The FrugalAmp by OStripper is also a candidate, see the thread at http://www.diyaudio.com/forums/showthread.php?s=&threadid=133306&highlight=
The bottom line is that there are lots to choose from. Poke around in the Amplifiers -> Solid State section of this forum, and you'll find quite a few. None of them can be called the best of the lot, although opinions abound.
regards,
Mike
Hi,
the 35+35 8A transformer = 560VA.
That is good for 300W to 400W of output power.
This could be two 4ohm amplifiers or one 2ohm amplifier.
The 1943/5200 will need many pairs to drive 2ohm speakers and this in turn will require at least an mje15034/5 as the driver pair.
An option would be 2pair of drivers or use another output device as the driver pair.
If you want to use the b649/d669 as drivers then 4ohm can be driven with 1pair and 2ohm with 2pair.
the 35+35 8A transformer = 560VA.
That is good for 300W to 400W of output power.
This could be two 4ohm amplifiers or one 2ohm amplifier.
The 1943/5200 will need many pairs to drive 2ohm speakers and this in turn will require at least an mje15034/5 as the driver pair.
An option would be 2pair of drivers or use another output device as the driver pair.
If you want to use the b649/d669 as drivers then 4ohm can be driven with 1pair and 2ohm with 2pair.
first, you cannot get 150W into 8r0 from +-52Vdc PSU.
second, you should be able to get 200W into 4r0 from +-52Vdc PSU. But, only if you optimise the design of both the amplifier and the PSU to achieve this target.
third, one pair of genuine 1943/5200 plastic package devices will struggle to give reliable 200W maximum power due to the temperature and SOAR constraints of the package.
second, you should be able to get 200W into 4r0 from +-52Vdc PSU. But, only if you optimise the design of both the amplifier and the PSU to achieve this target.
third, one pair of genuine 1943/5200 plastic package devices will struggle to give reliable 200W maximum power due to the temperature and SOAR constraints of the package.
Ermac, you may need to buy a few more transistors but use the same transformer; check this http://www.diyaudio.com/forums/showthread.php?postid=1863821#post1863821
Hi guys!
I just copy this schematics, and can some one check if there are any errors please.
I just copy this schematics, and can some one check if there are any errors please.
An externally hosted image should be here but it was not working when we last tested it.
The take-off point for the top differential pair doesn't look right...
Low Ft devices.
Keep in mind that this was designed for "slow" devices. Use modern higher Ft equivalents and "oscillator city"
. Input stage / VAS are conventional and would most likely work. If you do use low Cob High Ft devices for the VAS , keep in mind to adjust Cdom(120pf - whoa !!).
OS
Keep in mind that this was designed for "slow" devices. Use modern higher Ft equivalents and "oscillator city"
. Input stage / VAS are conventional and would most likely work. If you do use low Cob High Ft devices for the VAS , keep in mind to adjust Cdom(120pf - whoa !!).OS
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.