[J Harris]

How to convert the AKSA 100W N+ to balanced operation:

This is the condensed version of how to convert the AKSA to balanced inputs. Actually I don't know if this makes the amp "truly" balanced because I don't really know what that means - Hugh can comment about that.

What I can tell you is that in my high RFI/EMF environment (high up in midtown Manhattan), the switch from single-ended to balanced inputs made a massive noise reduction, resulting in blacker backgrounds, more detail, more palpability and presence, tighter and better bass, etc. YMMV depending on how important this is to you.

I should also mention that I listen only to vinyl LPs so I'm quite sensitive to noise... and also that from phono stage to amp, my system is now fully balanced. (Cartridge / tonearm --> phono stage is single-ended.) Click on my system link if you want to see the details.

Now, instructions for balanced input conversion. REMEMBER, THIS ASSUMES YOU HAVE ALREADY UPGRADED TO NIRVANA+. If you haven't, these instructions will not apply!

Parts:

2 X 2.7K ohm resistors (1/2 watt, metal film)
4 X 47 ohm resistors (1/2 watt, metal film)
2 X balanced input jacks (I used Neutriks from Parts Connexion)

The resistors I already had around the house. You might too.

1. De-solder the HOT and GND wires from your single-ended inputs.

2. Run 2.7K ohm resistor from the input side of C1 (that's the left as you face the top part of the board, transistors away from you) to GND (the ground pin, not star earth).

3. Carefully de-solder R4 and R5 (see page 31 of the main 100W manual for location if you're not sure). Use de-soldering braid.

4. Carefully (VERY carefully) de-solder the negative lead of C2. This is a Black Gate cap. Be careful not to stress the cap either with heat or twisting the lead! Use de-soldering braid.

5. Solder in the 2 X 47 ohm resistors at R4 and R5.

6. Attach a 5-inch (or however long you need to reach the XLR input) hook-up wire lead to the negative terminal of C2 that you just lifted from the PCB. Be very careful not to mechanically stress the cap.

6. Attach a 5-inch  (or however long you need to reach the XLR input) hook-up wire lead to Star Earth.

7. Now repeat steps 2-6 for the other PCB.

8. Wire in your XLR inputs as follows (there should be three main solder points, plus an earth tab):

- HOT pin on PCB to POSITIVE on XLR input
- GND pin on PCB to GROUND/SHIELD on XLR input (not the earth tab)
- NEGATIVE LEAD on C2 (the one you just desoldered and attached hook-up wire to) to NEGATIVE on the XLR input

Finally, attach the new hook-up wire from STAR EARTH to the grounding tab on the XLR input.

IMPORTANT! XLRs are wired differently from component to component and from cable to cable. You need to find out how the XLR outputs are wired from your preamp to know which pin (1, 2 and 3) is which! My BAT linestage instruction manual fortunately listed exactly which pins were POSITIVE, GROUND/SHIELD and NEGATIVE. You will need to know this information when you hook up your XLRs.

Doublecheck all your wiring. Make sure your wire dressing is not stressing C2.

You're now set - hook up the cables and play music!

NOTE: Re: drilling and installing XLR inputs. Everything here depends on how you have your case / enclosure set up. I have the standard AKSA enclosure from Hugh. The Neutrik XLRs need a 9/16" hole in which to sit, along with two much smaller holes for 6-32 3/8" screws and nuts. I placed my Neutriks directly under the single-ended inputs. I left the single-ended inputs in for looks, though they don't connect to anything internally now.

I used a step-drill bit to get the holes. I already had a step-drill bit up to 1/2". I searched around and finally found a hardware store that had a 7/16"-1" step drill bit, made in England. $40!! I bought it anyway. Brought it home, and found that it's got a 1/2" haft or hasp (not sure of word) that wouldn't fit my Black & Decker. Back to hardware store, found a Milwaukee drill that fits 1/2" bits. For $150.

So a big outlay in drilling, and only now I hear from my friend that Milwaukee ain't what it used to be. Plus he owns a full set of Greenlee punches. That would be the MUCH EASIER way to go to drill the holes... and actually cheaper. Much cheaper. Though I think you can never own enough tools.

Final note: you may find that the gain increases when you switch to balanced. For me it's not been a problem with the sound - the increase is slight and I don't notice additional distortion. (I have not measured the output.) But Hugh suggests that if it is too high, then you could replace R11  (again see page 31 of the original 100W manual for exact position) with another 2.7K resistor.

These instructions make this mod seem much more complicated than it actually is. It's a piece of cake. Just make sure not to stress C2, and make sure that you wire your XLRs properly for your system.

Good luck, and please report your results! I'm LOVING it. Even better since they've been mounted in the enclosure and sealed right up.

Many thanks to Hugh for talking me through all this in the previous thread, Sintz for encouraging me to take the plunge, and my friend Andrew for holding all the delicate parts while I was drilling and soldering! (Sunglasses and leather gloves on as metal bits sprayed all over the place, of course.)

Best,
Patrick (J Harris)

[ShinOBIWAN]

Excellent post Patrick.

You wouldn't have any pictures of the mods would you? A picture speaks a thousand words and all that?

[AKSA]

Patrick,

Many thanks - you have done the AKSA community a great service and I appreciate it!  

If the gain is still a little too high, you could bring it back with more degeneration on the diff pair.  You are using 47R just now;  you could increase this to 68R to pull back gain.

I'm really pleased the balanced AKSA sounds as good if not better.  This is a revelation, and not something I actually expected.  It's interesting how a thought experiment always differs from the real thing.......

Cheers,

Hugh

[J Harris]

Hi ShinOBIwan... yes I wanted to post some pictures but my digital camera only really works well in the daytime... plus I've bolted the cases cover back  

anyway... if I get a chance in the next few days I'll post some pics.

Hugh, the gain really isn't a problem. As for balanced sounding better, it could be more about system synergy than anything else. I know one thing - those balanced-to-single-ended adapters I was using, despite being expensive (from Purist Audio), were picking up a lot of the noise. Just getting them out of the chain may have helped the whole system a lot.

That said, this was a fun and definitely rewarding project. I'd love to hear from anyone else who tries it. It's easily reversible too (aside from drilling your case, of course).

Patrick

[J Harris]

http://www.audiocircle.com/index.php?action=gallery;area=browse;album=698

[ShinOBIWAN]

Thankyou!

[J Harris]

Can we return to the subject of this thread - i.e. converting the 100W N+ to balanced inputs?

If there is something particularly dangerous about this - to my equipment - I would like to know about it.

However from the previous 2 posters' comments it would seem that the only danger would come from operating the amp without a load, shorting the speaker outputs, disconnecting equipment while it's still on, etc., none of which I ever do. I also follow a strict start-up / turn-off order of components which I think every safe audiophile probably follows (ON: source to amp, one by one; OFF: amp back to source, one by one).

Hugh also emphasizes and triple-emphasizes the need for all these safety measures -- especially for the 100W N+ -- in the instruction manuals, and also in his posts here.

It would be good to return the debate about amplifierguru's right to post here, his goals, his manners etc. to the 'Controversy' thread.

Patrick

[AKSA]

Hi Patrick,

At your suggestion, I have moved the guru posts rather inexpertly (first time I've ever done this actually) to two other threads titled 'Controversy' to confine actual discussion of the balanced conversion to the appropriate thread.

Ampguru makes the point that if the input to the - (inverting) input of a balanced AKSA is broken, that is, no signal is connected, then the amp will revert to unity gain.  This will increase feedback factor 38 times, increasing global negative feedback and setting the amp into oscillation.  Yes, this is indeed what happens as 99% of amps of this kind are not stable at unity gain - in fact a lot of IC opamps are not either.

Therefore, the danger of this conversion is that if for any reason the inverting input is disconnected, the amp will go unstable and likely take out the speakers (particularly the tweeter) and the output stage.

If both inputs, inverting and NON-inverting, are disconnected, no problem at all.  Nothing to amplifier, so no problem.

If you use XLR connectors, and all the soldering inside is correctly done, there is not much risk anything nasty will happen.  But it remains a possibility, and so you should not disconnect or play with the wires while the amp is playing.  It's only an unbalanced signal which might damage things.

It is possible to modify the input stage so that a disconnected input signal cannot result in unity gain.  I will figure this out in time, but as I earlier described some time back the best option from a safety standpoint is the use of a balanced line transformer, such as a Sowter or Jensen.   There are also special chips which can convert balanced input to single ended, but I wanted to avoid this as it does add considerable additional hardware to the pcb.  A transformer is the preferred approach of the music industry, as you probably realise, but it's expensive and there are other disadvantages such as slightly reduced bandwidth (which could work for your actually as you live in an EMI prone area).

Nelson Pass and Norman Thagard discuss this precise method of balanced operation in one of their articles on the Pass website, and they don't even refer to the disconnected inverting input conundrum.

The chances of your having any problems with your existing setup are very small, particularly with the precautions you take.  Since the upside is so strong, I'd leave it as is and we can talk about this as the design evolves.  I'd be happy to work on this in the background.

Cheers,

Hugh

[J Harris]

Hi Hugh,

Thanks very much for the clarification. I've plenty of confidence in my solder connections and my Neutrik locking XLR inputs and Kimber XLR cable plugs... not much chance of disconnection there. And my case cover is securely fastened on.

Seems to me that guy is a bit of a scaremonger. Or something.

Patrick

[amplifierguru]

Hi L Harris,

Terribly sorry if you think I'm scaremongering - I simply don't like to see people's DIY effort go up in flames, possibly taking speakers with it - through oversight. I was designing commercial /professional active balanced input mixing desks back in the '70's. I see accidents waiting to happen.

Also disconnecting both inputs to an amp will still revert it to unity gain, and if not UG stable, will more than likely oscillate with or without input, unless this amp is somehow different?

It's up to Hugh to offer up a solution.

Cheers,
Greg

[J Harris]

Sorry ampguru - I didn't mean to offend. It's just that Hugh has been an ultra-careful, lead-my-by-the-hand asset in building his kits... I've never encountered so many warnings in all my kitbuilding life!

Of course my previous kits were tube SETs and phonostages - much more robust beasts in general, so this was probably just part of the baptism in solid-state kit-building.

All that said, I find it hard to believe that Hugh would have recommended anything with potential instability in the absence of user/builder carelessness. It just doesn't seem to be in his personality. He does encourage experimentation, but this mod was purely of his own device.

That said, I think anyone playing around on a kit building board probably does have a pretty good idea of what they're getting into. It's hardly like buying an off the shelf amp! I've modded, reverse-modded, upgraded my various kits -- sometimes merely on apparently informed forum member instructions (with notes from previous builders like mine above), and have never had a serious mishap. I observe all basic electrical safety procedures in testing (esp with the tube amps!), discharge capacitors, and so on and so forth.

Probably, admittedly, beginners' luck... and the less experience you have had, provided you play by the basic rules, the less chance that something would have gone wrong at some point in your history. I'm pretty sure though that if and when something does go really wrong in a kit (something other than a channel not functioning or something), the damage will be confined to the component itself - not my other expensive gear, or the person handling the component.

there's also the possibility that we Americans are more casual about electricity in general, what with half of your voltage coming out of the mains... most of us had some unpleasant childhood experience relating to a piece of cutlery in the outlet or something and "got the idea" that it wasn't a fun thing to do... rather than being seriously injured or killed. NOT to downplay the significance of 120 volts to the wrong body part... just pointing out a cultural difference that might be of some use in clarifying the current situation.

Take care ampguru

best
Patrick