I assume it is this one:
http://www.diyaudio.com/forums/soli...r-amplifier-design-handbook-douglas-self.html
Will do so. Thanks.
Yes, it is. There was a thread inviting comments on Samuel's comments. Should desserve much more discussion :
http://www.diyaudio.com/forums/soli...r-amplifier-design-handbook-douglas-self.html
Mike made a survey for for the prefered value of amp power, it leads him to adopt 30 W for his project. I think he mentions the possibility of a 100 V DC power supply somewhere. For a very powerful amp, for a sub for instance, he advices using more usual architecture.
Hi Marian
Here's my two pennorth. This amp may seem a little more expensive with 15 transistors but they are all low cost. The small signal ones are BC547/BC557; the protection transistors are current-limiting only (I've not optimised this) and have to carry more current than the small signal so they're BC337/BC327 but could be 547 /557 if you can't get them; the rest are BD139/BD140's and the output 2N3055/MJ2955 pair.
The circuit is essentially a Blameless architecture: differential input, with a decent current source; current mirror which is now standard; a Darlington VAS which has a low emitter resistor (because the VAS current is higher than normal at 25 mA or so); a current source for the VAS; and fully complementary output.
Frequency compensation is my recommended approach: no Miller capacitor, but local "sprog stoppers"; global compensation across the feedback resistor; and an input capacitor to keep the input stage drive impedance low, mainly, for stability.
Simulated distortion is .01% at 60W into 4 ohms and .02% at 20 kHz. It is this performance at high frequencies which demonstrates the better compensation scheme.
Distortion is quite high mainly because of the 4 ohm load. If you run this from 20 to 25V rails at 30W, it will be lower.
The VAS current is high because of the high drive needed by the output transistors. I would recommend the MJ21193/4 but if you can't get those, 2N3055's are OK but need a bit more "push". A compromise would be to use two pairs of 2N3055/MJ2955 in parallel, but that is probably not necessary if you keep the power to 30W.
John
Here's my two pennorth. This amp may seem a little more expensive with 15 transistors but they are all low cost. The small signal ones are BC547/BC557; the protection transistors are current-limiting only (I've not optimised this) and have to carry more current than the small signal so they're BC337/BC327 but could be 547 /557 if you can't get them; the rest are BD139/BD140's and the output 2N3055/MJ2955 pair.
The circuit is essentially a Blameless architecture: differential input, with a decent current source; current mirror which is now standard; a Darlington VAS which has a low emitter resistor (because the VAS current is higher than normal at 25 mA or so); a current source for the VAS; and fully complementary output.
Frequency compensation is my recommended approach: no Miller capacitor, but local "sprog stoppers"; global compensation across the feedback resistor; and an input capacitor to keep the input stage drive impedance low, mainly, for stability.
Simulated distortion is .01% at 60W into 4 ohms and .02% at 20 kHz. It is this performance at high frequencies which demonstrates the better compensation scheme.
Distortion is quite high mainly because of the 4 ohm load. If you run this from 20 to 25V rails at 30W, it will be lower.
The VAS current is high because of the high drive needed by the output transistors. I would recommend the MJ21193/4 but if you can't get those, 2N3055's are OK but need a bit more "push". A compromise would be to use two pairs of 2N3055/MJ2955 in parallel, but that is probably not necessary if you keep the power to 30W.
John
Attachments
I'm running mine on the original Dynaco transformer, which results in about 72Vdc after rectification/filtering. I haven't measured, but perhaps 40 WPC?
An externally hosted image should be here but it was not working when we last tested it.
Now that it's done, chicks, what is your satisfaction from the project - do you like the sound or is it all a bit clean and sharp?
Sounds as good as it looks.
An externally hosted image should be here but it was not working when we last tested it.
Hi Guys
Rod Elliot's circuits were designed many years ago using then-current devices. He has indicated newer devices at the request of his readers. It is the case with most circuits that for each transistor position, many different devices can be subbed in place with little performance difference - at least with Rod's circuits this is the case.
You can sub almost any other part types into his projects. Rod also sells PCBs for most of the amps. If you have Qs about subbing specific parts into any of his circuits he will answer your query.
Have fun
Kevin O'Connor
Rod Elliot's circuits were designed many years ago using then-current devices. He has indicated newer devices at the request of his readers. It is the case with most circuits that for each transistor position, many different devices can be subbed in place with little performance difference - at least with Rod's circuits this is the case.
You can sub almost any other part types into his projects. Rod also sells PCBs for most of the amps. If you have Qs about subbing specific parts into any of his circuits he will answer your query.
Have fun
Kevin O'Connor
Member
Joined 2009
Paid Member
marian - I give another vote for the DX amplifiers.
All the schematics and details are here: Greg's Web Site
If you want to save money though, and effort, buying a used stereo amp is much easier and quicker route to building one, even with cheap parts.
All the schematics and details are here: Greg's Web Site
If you want to save money though, and effort, buying a used stereo amp is much easier and quicker route to building one, even with cheap parts.
Goodness, I have been a member of this forum for a little bit more than a year, I still learn every time I open a thread, It would be so great getting some of you guy's together deciding on a design, hopefully not more than handful, otherwise we will have a flabbergasted mass of inconsistency ! So many competent people, so many ideas, can we not work and try and design the perfect woman without no VAS, no feedback at all and she is playing a sweet tune. I do enjoy reading these posts, I have build my first lateral mostfet , worked the first time, thank goodness and i do enjoy DIY. Cheers to all you magnificent people !
Cheers from the land of the Springbok !
So many competent people, so many ideas, can we not work and try and design the perfect woman without no VAS, no feedback at all and she is playing a sweet tune. I do enjoy reading these posts, I have build my first lateral mostfet , worked the first time, thank goodness and i do enjoy DIY. Cheers to all you magnificent people !Cheers from the land of the Springbok !
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.