I have just finished building, wiring-up and testing my new 15W per channel class A amp. you can read all about it if you follow the link below:-
Ovation sx-Amplifier
There's a .pdf file as well (about 6.3MB) that discusses the design and my listening tests.
I will make DSTHP PCB's available in a few months for this project, if there is enough interest.
This was a real fun project - nothing like the e-Amp or Ovation 250 projects (or the ACA-a for that matter!) but I've ended up getting a lot of pleasure out of it.
Anyway, happy reading!
Ovation sx-Amplifier
There's a .pdf file as well (about 6.3MB) that discusses the design and my listening tests.
I will make DSTHP PCB's available in a few months for this project, if there is enough interest.
This was a real fun project - nothing like the e-Amp or Ovation 250 projects (or the ACA-a for that matter!) but I've ended up getting a lot of pleasure out of it.
Anyway, happy reading!
Very interesting write-up article and excellent project construction, as we have come to expect, Bonsai. Thanks indeed.
As it seems to follow from your repair thread and thoughts on the MF A1, it begs the awkward question; how do you rate it subjectively against that or any of the types discussed? I put that as the preference you might have for listening to your favourite music after a hard day at the office, as this is a likely use for those who now flock to class A and the gross distortion of the old designs for their modest daily relaxation.
As it seems to follow from your repair thread and thoughts on the MF A1, it begs the awkward question; how do you rate it subjectively against that or any of the types discussed? I put that as the preference you might have for listening to your favourite music after a hard day at the office, as this is a likely use for those who now flock to class A and the gross distortion of the old designs for their modest daily relaxation.
Good question Ian. Don't know if it's answerable in simple terms because the bigger ones offer things the sx-Amp does not do, while the latter has more appeal on certain types of music that I prefer (eg top end), but here goes (don't hold me to this if I change my view with time though!)
e-Amp. Tight forward sound. Big sound stage. Lots of scale - can handle complex pieces well; great bass. Lively.
Ovation 250; lots of scale, great sound staging; distinctly 'dark' sound. Very relaxed, smooth sound.
sx-Amp. Surprising is the first thing I'd say. I was not expecting it to sound as good as it does. Great imaging that comes very close or matches to the big amps. Top end is definitely sweeter. Bass surprisingly well formed. On demanding material, cannot match the scale, but I think that's purely related to the power.
The sx is a very simple, minimalist design whereas the e-Amp and the Ovation 250 are pretty complex and there were many many months of effort in going design and completion. I remarked in the write-up that this was a fun project and very low cost for what I consider after the listening tests, a great sound.
e-Amp. Tight forward sound. Big sound stage. Lots of scale - can handle complex pieces well; great bass. Lively.
Ovation 250; lots of scale, great sound staging; distinctly 'dark' sound. Very relaxed, smooth sound.
sx-Amp. Surprising is the first thing I'd say. I was not expecting it to sound as good as it does. Great imaging that comes very close or matches to the big amps. Top end is definitely sweeter. Bass surprisingly well formed. On demanding material, cannot match the scale, but I think that's purely related to the power.
The sx is a very simple, minimalist design whereas the e-Amp and the Ovation 250 are pretty complex and there were many many months of effort in going design and completion. I remarked in the write-up that this was a fun project and very low cost for what I consider after the listening tests, a great sound.
Last edited:
Thanks for the write-up, informative although I dont agree on all your points mentioned.
Its not true that one cannot use miller compensation or the cap on the feedback, this depends entirely on the design. I have feedback caps on all my CFB amps and I see at least another 4 manufacturers use it including accuphase whithout problems. Miller compensation for me doesnt work well with CFB although this is the route taken by NAD with some of their current range of CFB amps.
Its not true that one cannot use miller compensation or the cap on the feedback, this depends entirely on the design. I have feedback caps on all my CFB amps and I see at least another 4 manufacturers use it including accuphase whithout problems. Miller compensation for me doesnt work well with CFB although this is the route taken by NAD with some of their current range of CFB amps.
Strictly speaking, yes you can get away with Miller comp if the CFA amplifer has a limited bandwidth and the loop gain is low. I did some simulations during the design phase that confirmed this. However, once the bandwidths or loop gain increase, this method of compensation is problematic, and most texts generally caution against this approach. Op-amp CFA data sheets routinely recommend adjusting the value of the feedback resistor for comp - that's not really practical in an audio power amplifer.
For a small cap across the feedback resistor, I think this is very risky with a CFA topology amp given the bandwidths on tap compared to VFA's. It might be that the benefit that is being observed is because the feedback summing node capacitance to ground is being cancelled (compensated) by the feedback cap rather than wholesale loop compensation - but under this case I would expect the value to be very small (single digit pF), and the benefit in reduced peaking on square wave response.
The Alexander Amp has an appendix that shows the complete derivation fo the comp scheme I used, and seems to offer the best compromise - single cap does the whole comp job with the front and filter doing the bandwidth limiting.
BTW I have put a copy of the Alexander Amp paper up here:- http://hifisonix.com/wordpress/wp-content/uploads/2012/09/The-Alexander-Amplifier.pdf for anyone wanting a copy. Hope ADI 'don't mind - seems there are quite a few copies floating around on the web.
For a small cap across the feedback resistor, I think this is very risky with a CFA topology amp given the bandwidths on tap compared to VFA's. It might be that the benefit that is being observed is because the feedback summing node capacitance to ground is being cancelled (compensated) by the feedback cap rather than wholesale loop compensation - but under this case I would expect the value to be very small (single digit pF), and the benefit in reduced peaking on square wave response.
The Alexander Amp has an appendix that shows the complete derivation fo the comp scheme I used, and seems to offer the best compromise - single cap does the whole comp job with the front and filter doing the bandwidth limiting.
BTW I have put a copy of the Alexander Amp paper up here:- http://hifisonix.com/wordpress/wp-content/uploads/2012/09/The-Alexander-Amplifier.pdf for anyone wanting a copy. Hope ADI 'don't mind - seems there are quite a few copies floating around on the web.
Last edited:
I have used upto 22pf as feedback cap and at least 2 other manufacturers are on 39pf so single digit is not quite right. If I remember correctly the Cyrus range also use around 39pf.
There are better ways to cancel the summing node capacitance although more complex but I find very little is discussed here at DIY on CFA amps although in the last 10 years I know of 3 major commercial concerns who are now only producing CFA amps accross their full range.
Some of the best amps Ive heard are indeed CFA with class A output aka Accuphase so Im not surprised youre happy with the 15 watter..
There are better ways to cancel the summing node capacitance although more complex but I find very little is discussed here at DIY on CFA amps although in the last 10 years I know of 3 major commercial concerns who are now only producing CFA amps accross their full range.
Some of the best amps Ive heard are indeed CFA with class A output aka Accuphase so Im not surprised youre happy with the 15 watter..
Bonsai, just glanced through the 37 page article. Paused and gazed at the amp and psu schematics. Read through the listening session pages.
Very well researched, executed and documented project. As you have always demonstrated, up to your very high standards once again. Well done and keep it up.
Just a few questions: Why not choose lower noise devices for the input section, viz., 2SC1845 and 2SA992? Can 2SC5200 and 2SA1943 devices be used for the output with little impact on sound quality?
Will 200uV noise be heard on 94db full-rangers?
Finally, would you like to give us a short comparison between the SX-Amp and your fine version of the TSSA, especially in terms of resolving low level detail and overall musicality? Special mention of bass response, if possible. Thanks.
(P.S: Just as I began to gather JFETs for the TSSA, here comes another great little amp to spice the party!)
Very well researched, executed and documented project. As you have always demonstrated, up to your very high standards once again. Well done and keep it up.
Just a few questions: Why not choose lower noise devices for the input section, viz., 2SC1845 and 2SA992? Can 2SC5200 and 2SA1943 devices be used for the output with little impact on sound quality?
Will 200uV noise be heard on 94db full-rangers?
Finally, would you like to give us a short comparison between the SX-Amp and your fine version of the TSSA, especially in terms of resolving low level detail and overall musicality? Special mention of bass response, if possible. Thanks.
(P.S: Just as I began to gather JFETs for the TSSA, here comes another great little amp to spice the party!)
Last edited:
Samuel, I'm not th TSSA designer ;-)
Re the input devices, I think any reasonably fast small signal devices would do, but given he low gain 23 dB gain, I don't think noise is a problem.
There is no noise (hum or other) with my ear up against my speakers and they are rated at 90 dB.
Thanks for he comments BTW - appreciated.
Re the input devices, I think any reasonably fast small signal devices would do, but given he low gain 23 dB gain, I don't think noise is a problem.
There is no noise (hum or other) with my ear up against my speakers and they are rated at 90 dB.
Thanks for he comments BTW - appreciated.
good morning mr.bonsai. mr.bonsai, I want to assemble ovation sx-amplifier.help
guided me to assemble. I’m drawing lay-out amplifier schematic and layout components. I assemble amplifier-amplifier sx ovation for his own use and not for sale .. email me: sofianirawan@yahoo.com
guided me to assemble. I’m drawing lay-out amplifier schematic and layout components. I assemble amplifier-amplifier sx ovation for his own use and not for sale .. email me: sofianirawan@yahoo.com
Dear sx-Amplifier friends, apologies for the delays. I have had a lot of difficulty setting up my web shop. Paypal have been good, but getting the correct infomation to make it work from the bank has been a problem. I hope that today I have the last pieces of the puzzle in place, and i can go live in the next few days.
Again, from my side, apologies for these delays.
Again, from my side, apologies for these delays.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.