'AK Mods' SC Class A 15W Amplifier

[Deleted]

Hi Sean
How did the monoblocks go driving those difficult 4 ohms speakers of yours in Stereo?

Kind Regards
Alex

[Deleted]

Half way there !

Shaun won't read this for a few days, but mine is now running with one of Will's very nice PCBs
The Output devices are spaced further apart with Will's PCB, so I had to remove the left heat sink and drill another mounting hole.
I then found that as Will's PCBs are a little longer than the original S.C. PCBs, I had to relocate it's JLH a bit further back.
As I already had some 2SA1943 and 2SC5200 Toshiba output devices, I am using them instead of the NJL's in the original PCBs.
I tried it initially with a 1K resistor at the input, and although it is presently using AD744 in the Offset Corrector, DC Out into no load was only 0.9mV.
Unfortunately it looks though this time I will need to change the BC549C 120 ohm emitter resistor to 130 ohms like the Northern Hemisphere guys have done , as I could only adjust the balance to within 8mV, and this slowly increased to 10mV as it warmed up.(That can wait for now though.)
Tomorrow, I hope to do the same for the other PCB after first fitting a 130 ohm resistor instead of the 120 ohm.
My original SF11 diodes must have had slightly lower forward voltage drops than the higher voltage rated SF12, which is typical of many higher rated voltage diodes in a series.
Regards
Alex

P.S.
It is VERY reassuring to see the front panel green LED come on after several seconds delay !!!
At present I am still using the 2 x S.C. PSU PCBs in the other rack case.

[jeffc]

Hi Alex,

Very nice work with the boards , they do look good don't they. However, where's the naked-foil resistors

cheers..jeffc

[Deleted]

Apr 30, 2013 5:37:19 GMT jeffc said:

Hi Alex,

Very nice work with the boards , they do look good don't they. However, where's the naked-foil resistors

cheers..jeffc

Hi Jeff
I presently remain unconvinced about the need, rather than personal taste, for naked foil resistors in the 15W Class A where the gain isn't attenuated, but agree that they could be worthwhile in the gain setting feedback path of the Class A preamplifier ,where any small improvement in S/N is worthwhile, and perhaps also the 220 ohm input resistors and the 2 LS352 200 ohm emitter resistors.
I will be getting at least the gain setting FB resistors when the GB is organised.
BTW, how is yours driving those panels now that you have 6dB more available from a lower source impedance output, instead of 10 ohms in series?
Kind Regards
Alex,

[Deleted]

Nice one Alex Will's boards are good to work with, 130 ohms seems to be optimal for us all then.
Regards,
Alan

[Deleted]

Apr 30, 2013 5:54:16 GMT @sandyk said:

Apr 30, 2013 5:37:19 GMT jeffc said:

Hi Alex,

Very nice work with the boards , they do look good don't they. However, where's the naked-foil resistors

cheers..jeffc

Hi Jeff
I presently remain unconvinced about the need, rather than personal taste, for naked foil resistors in the 15W Class A where the gain isn't attenuated, but agree that they could be worthwhile in the gain setting feedback path of the Class A preamplifier ,where any small improvement in S/N is worthwhile, and perhaps also the 220 ohm input resistors and the 2 LS352 200 ohm emitter resistors.
I will be getting at least the gain setting FB resistors when the GB is organised.
BTW, how is yours driving those panels now that you have 6dB more available from a lower source impedance output, instead of 10 ohms in series?
Kind Regards
Alex,

Hi Alex

what'what'what

Unconvinced

Yep I think that you may be right on that but I’m still planning to have a fiddle. I may try some Tants on the IP and FB loop.

Just out of curiosity.

And also something for the front end emitter resistors as you’ve rightly pointed out.

I’d also like to try some of those multiple personality Silmics at some point.

Just some thoughts ATM.

Glad that you have Will’s boards up and running.

Take care

[Deleted]

IMPORTANT - NEW TESTING PROCEDURES.
As both the Class A HA/Preamp and 15W Class A amplifier have separate JLH PSUS for the front end, I would recommend that before fully testing the amplifier,that you initially ONLY connect the JLH PSUs. Fit a 1K resistor or thereabouts, across the Input of the PCB, and a 10 to 100 ohm resistor across the Output terminals. Then check that the voltages are similar to those shown in the schematics. The offset correctors can be plugged in too.
Check that the DC across the output resistor is only a mV or so . As well, you can also check that the bias voltage is below 1.6V .
This can be measured between the 2 x 100 ohm resistors that connect to the BC327 and BC337.Adjust this voltage as low as you can go initially
Do NOT connect the main power unless these voltages seem correct.It may also be a good idea with the 15W Class A to temporarily fit 1.8 ohm PW5 resistors in series with the main power leads, until you are confident that all is well with DC at the output, and that you can properly adjust the bias voltage.
Then adjust the bias to around 100mV as measured across a pair of the parallel 1 ohm resistors.
If all is O.K. with DC out etc. then the series resistors may be removed, and bias increased to say 220 mV when cold.Again check for a mV or so DC Out.
Alternatively, if all is O.K. you can increase the bias to read just under 1.8V across the 1.8 ohm PW5 resistors.
Regards
Alex
P.S.
When testing with just the JLH connected to the front end, and the input terminated with a 1K resistor, it is also possible to set the front end balance close to 0mV .
Doing this as well, will tell you that the whole front end, including the VAS stage and the bias transistor setting is working correctly.

[Deleted]

Hi Alex

All good advice.

BTW I found that I could do a quick and dirty test on the front end without having the OP devices powered up at all.

I just had the boards on the test bench and hooked up the front end JLH supply, did a quick measure round and made sure that the front end balance was working in the way that it should.

Maybe UN necessary but I found it easy to do it that way first and then mount the amp boards to the heat sinks and hook up the main supplies.

I did start out with my bias set much lower just to power the whole thing up without putting toooo much strain on things. Then just upped the bias when I was happy that all was well.

However

Your method looks better so I’d go with that.

Take care

[Deleted]

Another point to watch out for.

With the 15W Class A, it is even more important to make sure that the front end using the JLH works correctly as described.
In the event of a higher than acceptable DC out, the Speaker Protector will ground both sides of your speakers, BUT, this may blow up an Output device in the process.
Whether or not it does will depend on the current limiting in your PSU, but you may still damage a BC327 or BC337. When fault finding the output section, don't forget to disconnect the leads from the Output terminals to the LS Protector.

Regards
Alex

[Deleted]

Dave Kendrick has asked me to post this on his behalf. It's a shame that due to a few silly disagreements that Dave was not able to post this for himself.
Perhaps Mike could review this situation in order to reduce present tensions?
Alex

I had a good 4 hour listen tonight to a wide variety of music and I am well pleased. It would be nice if Alex posted a precis of this on RG by way of a vote of thanks to Will and Shaun for agreeing to include me in the GB and then proffering their help when Sean hit problems.
I am very happy with the SQ on all the music that I listened to. Getting it and then connecting it up was a bit of a journey but well worth the wait * happy .To put the cherry on the cake my wife walked in during my listening session and said, 'That sounds good, it's pure, just like she is singing in the room" !!!!! This from someone with no real appreciation of music in the accepted sense * winking , being perfectly happy listening to her impeccable Michael Bubble through her Bose dock and mp3 player - praise indeed and something with which I would not disagree.
Separation between the channels, inky silences and solid bass all the way down were very noticeable.
Mainly for Alex, the difference between standard Red Book and Best Audiophile tracks was much more noticeable than I have heard before.
The source was my HTPC playing FLAC files (via Sean's RC source selector) to Sean's MF-X10 clone buffer amp to the Class As and on to my tri-wired Mission 774 speakers with added super tweeters.
I also had a brief listen through my Senn HD 280 pro cans through Sean's HA10 HP amp (placed immediately after the Class As in the above layout) and again I was very impressed with what I heard.

[Deleted]

Will's PSU PCB.

I have now changed over from the original Silicon Chip designed + and -20V PSUs, of which I was using a pair, instead of the original single PSU. The S.C. design used series pass transistors to increase the output from the LM317T and LM337T voltage regulators, as well as an off board 35A Bridge Rectifier.
Initially the sound was underwhelming, but after a couple of hours to help form the 6,800uF electros, and most likely with further help due to the big 8A Schottky Diodes, it is now clearly outperforming the original S.C. designed PSU. "Dire Straits- Love Over Gold" is the best yet that I have heard it from the old Oppo 981 DVD-A/SACD player.
Alex

[Deleted]

Hi Alex

Yes those big caps do take a little time to settle in but well worth the wait.

I’ve not tried the original PSU but I’ve been more than happy with my etched version of Will’s PSU and the overall SQ of your amp.

Good work

Take care

[Deleted]

May 7, 2013 10:17:19 GMT @shaun said:

Hi Alex

Yes those big caps do take a little time to settle in but well worth the wait.

Hi Shaun
Unfortunately, there are always questions in your mind like "would Rubycons /whatever sound a little better"
Kind Regards
Alex

[pagan]

May 7, 2013 10:27:10 GMT @sandyk said:

May 7, 2013 10:17:19 GMT @shaun said:

Hi Alex

Yes those big caps do take a little time to settle in but well worth the wait.

Hi Shaun
Unfortunately, there are always questions in your mind like "would Rubycons /whatever sound a little better"
Kind Regards
Alex

Alex
That's mean....

now Shaun's going to go "what if" on different caps

[Deleted]

My only experience of that was trying out some Mundorf resevoir caps in my integrated amp (of equal v and z of original standard Elnas). IMHO they made no difference what so ever. A later mod of using higher capacitance along with larger bridge diodes did, however, make a very noticeable improvement, mainly in the weight and control dept.

So I'll be staying quite standard for this build. They are 10k each though

[Deleted]

May 7, 2013 12:39:01 GMT pagan said:

May 7, 2013 10:27:10 GMT @sandyk said:

Hi Shaun
Unfortunately, there are always questions in your mind like "would Rubycons /whatever sound a little better"
Kind Regards
Alex

Alex
That's mean....

now Shaun's going to go "what if" on different caps

Now then now then

What’s all this cap malarkey?

Ok then

I already have Rubicon’s in the main regs chosen for well err because I already had them but to be honest I’ve never found caps that far back on the reg have made much difference SQ wise.

But

it’s still worth having a little quality there just for the sake of longevity if nothing else.

Further up the circuit

Well that’s a different matter IMHO.

From my little experiments with Alex’s Pre/HA that those PSU de coupling caps do make some difference to the overall sound. I have Carafine for the last JLH cap and Silmic for the main boards which now that they have had time to settle (80 hours) do sound really good IMHO.

I liked that scheme so much that I’ve done the same thing with Alex’s PA but it’s early days on that which is why I’ve not mentioned it previously.

I’ve also replaced the MF on the IP with Tans and also used them for the FB loop along with some of those nice red 100PPM MF that I used in the Pre.

I’ll keep you posted on how that goes.

Take care

[sean]

I was going to post up on here Dave's comments on his new amps, but I see that Alex has beat me to it. In reply to your earlier query Alex, which I forgot to reply to at the time, the amps drove my 4 Ohm speakers fairly well - they did not quite have the dynamic response and 'kick' of much more powerful class AB designs of amp (Naim NAP300/ Avondale NCC200), but in all other aspects they sounded very nice indeed. A smooth presentation, very typical of class A amps.

When he came to collectthem and have a quick listen at mine, Dave commented that with some of his CDs he could clearly hear things that he had never heard before; the improvement varied between some different CDs we tried. Overall he is really happy with the amps, as you can tell.

Now that Dave's happy with the amps and he hasn't found any issues with them at all, i'm pretty much finished with the project now. It's been interesting; a bit frustrating at first when it didn't work i'll admit, no blame to Will or Alex or anyone else of course, this was just one of those things. Dave would like to extend his thanks to everyone involved, as would I; thanks to Alex, Will, Shaun, and everyone else that was involved with pulling the project together and/or offered their help and expertise when Dave's amps weren't workign properly, be it through these forums or by email.

Cheers all!

-Sean

[jeffc]

BTW, how is yours driving those panels now that you have 6dB more available from a lower source impedance output, instead of 10 ohms in series?

Hi Alex,

Thanks for the gentle push .... but I've only just now got to disconnecting the amp/PSU boxes and take a peak. It seems I wasn't spared, with the 10 ohm GND lift resistors on both boards burned like happened with Alan's. Could have been during use with my panels or with those cheap 4 ohm drivers tested to investigate Sean's problems before the PCB error was identified. Whatever, boards will need to be lifted to replace these and attend to GND track cutting etc.

Might hold back on reassembly as there's no urgency, although much curiosity, and as I'm keen to hear what Mr Not-So-Content Cap-Resistor roller aka Shaun, learns in his latest adventure. I already have Elna Cerafines at PSU and JLH output positions. I have 100uf 50V Silmic IIs lying about too but having questioned colourations that these imparted in 2 previous attempts to use them in PSUs (including at the output of JLHs), which resulted in me swapping back to what was there before, I'm reluctant to revisit them. What I just couldn't warm to was their effect on transients, seeming to soften their leading edge and then over exaggerate them, and their added richness/colouration to lower-mids. But that said, I've only tried them twice and could have been unlucky with where/how I used them. Some original ARS-type Silmic's I've used in PSUs and as output caps in HAs I've really liked, but not sure which type you're playing with Shaun.

And Shaun,
Got these resistor replacers on your 'to try' list..... The resistor you have when you're not having a resistor.
www.buildanamp.com/Assembled-Resistor-Replacer-BuiltRR.htm

cheers.. jeffc

[Deleted]

Hi Jeff

Before you get in too deep, it would be a good idea to check for DC out to ensure that the PCB error didn't lead to a driver device failure.
Use the procedure that I detailed recently on this page . (May 3). IF you have a spare RCA plug and a 1K resistor laying around, you could fit the resistor inside the RCA plug and simply plug it into the appropriate RCA input socket for testing that channel.
Kind Regards
Alex

[Deleted]

May 14, 2013 11:53:10 GMT jeffc said:

BTW, how is yours driving those panels now that you have 6dB more available from a lower source impedance output, instead of 10 ohms in series?

Hi Alex,

Thanks for the gentle push .... but I've only just now got to disconnecting the amp/PSU boxes and take a peak. It seems I wasn't spared, with the 10 ohm GND lift resistors on both boards burned like happened with Alan's. Could have been during use with my panels or with those cheap 4 ohm drivers tested to investigate Sean's problems before the PCB error was identified. Whatever, boards will need to be lifted to replace these and attend to GND track cutting etc.

Might hold back on reassembly as there's no urgency, although much curiosity, and as I'm keen to hear what Mr Not-So-Content Cap-Resistor roller aka Shaun, learns in his latest adventure. I already have Elna Cerafines at PSU and JLH output positions. I have 100uf 50V Silmic IIs lying about too but having questioned colourations that these imparted in 2 previous attempts to use them in PSUs (including at the output of JLHs), which resulted in me swapping back to what was there before, I'm reluctant to revisit them. What I just couldn't warm to was their effect on transients, seeming to soften their leading edge and then over exaggerate them, and their added richness/colouration to lower-mids. But that said, I've only tried them twice and could have been unlucky with where/how I used them. Some original ARS-type Silmic's I've used in PSUs and as output caps in HAs I've really liked, but not sure which type you're playing with Shaun.

And Shaun,
Got these resistor replacers on your 'to try' list..... The resistor you have when you're not having a resistor.
www.buildanamp.com/Assembled-Resistor-Replacer-BuiltRR.htm

cheers.. jeffc

Hi Jeff

MR not so content.

Mmm well I have had Alex’s PRE/HA for two years now and have been more than happy with its SQ.

But I’ve since built another and well what’s the point in building a replica? So I used build no 2 (the mule) to do a if just to satisfy my own curiosity.

OK let me take a wild guess.

With the Silmic ones that you tried you liked em pretty much off the bat but with the Silmic II you put them in and thought my god these sound f**G bad (I know I did).

And you’d be right

Unlike the Silmic 1 they need a long burn in period and I found that they needed at least 60 hours of crap sound and thinking I’ll just rip em out before they settle down. On the main amp board to my ears they sounded a little dry and errrr boring but did a better job on the JLH.

I used Silmic on the last 100uf cap on the JLH and have Carafine II for PSU de coupling on the main board.

Now the Carafines have been fun they sounded not too bad if a little warm when I put them in and then ‘’as if by magic’’ the bass totally disappeared without trace. In my set up they seem to start at the top and work down with the bass last to make an appearance.

So a ‘’dark night of the soul’’ bad sound wise needs to be put up with to get the final goodness.

With the same caps in Alex’s class A the effect has not been quite as obvious as it was with the pre.

I think that Leo mentions in one of his posts that caps can be like marmite either love them or hate them so it could be worth having a think about the sound you want and then doing a little reading around.

All I can say is that to my ears/in my set up/IMHO they sound superb but it took time to get there.

And with Alex’s PA and the few tweaks i've done well I’m

Nuff said.

Take care

[jeffc]

Hi Alex,

Thanks for the advice.... nothing disconnected yet.... so I'll do the checks as suggested for damaged transistors before lifting the boards for surgery.

With the same caps in Alex’s class A the effect has not been quite as obvious as it was with the pre.

Hi Shaun,

That's good for all to know so no one here using other capacitor types feels as though there missing out too much in the way of potential with this amp. I thought long about capacitor types to use in this build and .... who knows... ended up using a mix of Nichicon HE and FG, Elna Cerafine, and Nover LA for-audio 10,000uf in the PSU... for an 'all flavours' kind of sound.

On the Silmic 2's, I did live with them for quite a while both in my MF X10-V3 tube buffer replacing 4 x 100uf Panasonic FMs and in the output positions of a +/-V JLH supplying a CS4397 DAC opamp stage employing a LME49720 decoupled locally with tantalums that replaced Panasonic FMs that to me didn't work particularly well in that position. I suppose the fact is, that we all known too well, all caps will influence the sound to some extent, and just how much will depend on the type and the application. However, caps like the various Elna for-audio types are designed specifically to profoundly influence/enhance certain aspects of the sound in fairly consistent ways, irrespective of their application, with Silmic 2's probably being the top of the pile when it comes to how much the presentation is altered. So certainly a good cap to experiment with to hear differences, but as both you and I have found, put one in the wrong place and you can't live with it there, find the right place (this being one I bet where its influence is least exaggerated), and sure they can probably add some genuine niceness to the presentation......

While I have the boards out I'll make sure leads of resistors targeted for fancy types aren't too bent over for easier extraction from the top..... just in case I feel compelled to do some in situ tinkering once it's assembled again.....

cheers.. jeffc

[Deleted]

Hi Jeff

Yes all good points

Those sort of fine adjustment things are really down to personal taste and expectations. I still have a few caps to spin onto the boards so it's just a where I’m at right now. Yes getting the right part in the good spot is the thing so some work to do on that still. IMHO caps can have influence on SQ for good or ill but finding the right ones is all part of the4 fun and a great way to stock that parts box still further.

On Alex's advice I’ve avoided the ultra expensive Charcroft types and plumped for Tants.

Any effect?

Yup even quieter to my ears.

what I would say is it's always going to be best to build the HA/PRE PA as standard and live with it for a good while before any light tweaking as it really does pay to get to know them both over time.

One thing that I will mention is the superb Bass quality that I’ve always had from both of Alex's amps.

I posted a track by Underworld called ''Rivers of Bass'' which has a really low bass errr vibration on it. Now that was either non-existent or muddled on anything else that I’ve tried.

Here it is

rockgrotto.proboards.com/thread/3934/listening?page=70#


But with Alex's the Bass just goes so low, that note just comes up through the floorboards with my PMCs.

Awesome.

So nice to have a chat about an often ignored or argument inducing subject and nice to feel the cool breeze of subjectivism still blowing through.

I'll PM you for a chat

Take care

[jeffc]

So nice to have a chat about an often ignored or argument inducing subject and nice to feel the cool breeze of subjectivism still blowing through.

Hi Shaun,

No arguments with what you’re hearing in your cap/resistor rolling experiments, just needed to add that Silmic 2’s are quite unique and need extreme care in application to not be disappointed.

Like you I’ve never been too concerned about professing to hear or try and describe changes in the way music presents itself to my ears with even very minor tweaks to equipment/files etc. No reputation, nothing to defend....    

For example, on a whim, and because they were staring at me from my equipment rack after I disassembled the Class A amp/PSU last night, I stuck some Herbie’s Gabon wood balls in whatever special silicon cups under my CDP, had a brief listen, went away to inspect the Class A for damage, came back later for another listen and then sat through a doz tracks or so on a couple of CDs with a smile on my face thinking WTF, this has really changed how sounds are launching from my panels and being presented to my ears. Wood balls under CDP profoundly changing how sound is presented.....white jacket time......  

Seriously though, it’s great to have experimenters/innovators like you and Alex without “too many” prejudices and preconceptions, except for Mr anti-tubes and anti-Class D , leading by example. I’m always all ears for what I might tinker with when a get a few spare moments and do plenty I’m too embarrassed to discuss publically but which keep me entertained and smiling....

cheers.. jeffc

[Deleted]

Hi Jeff
When you have corrected your Class A problems, why don't you take those comparison CDs that I sent you, and pop them in your CD spinner while it is sitting on the carpet, and compare it with sitting on Herby's balls You may be able to hear differences between Herby's Balls and the carpet.
To make it more interesting, try the same tracks into both your mains polluter and the 15W Class A, and see if you can as easily hear the same differences with the mains polluter.

Kind Regards
Alex

[jeffc]

Laugh all you like.....

Repeat "No reputation, nothing to defend...."

You've gotta a better idea than most that I'll try/build essentially anything 'for fun and enjoyment' and 'because of curiosity it might sound good' as long as its cheap enough to convince myself that this audio electronics/speaker building lark is still a hobby and not an OCD obsession. Point in case, I've just been dabbling with PCB etching (HCl and H₂0₂ freebee's from work) and while I'm no expert yet, results so far not to bad except for the beginners mistake of making them mirror imaged, with first cab off the rank being 8 x Schumann resonator PCBs, just waiting on some 5V regs and NE555 timers to arrive and I'm in production for home made Christmas presents.    
I'll send you one, might help you sleep better, might even "alter" perceived sound reproduction from your stereo system.

cheers.. Jeff