[MikeC]

It took nuch longer than I anticipated, but due to other distractions rather than kit difficulty. It is in a pine box until I get something better to mount it in, but it worked perfectly first time. I also need to put in decent output terminals and internal wiring, but I wanted to prove the amp first.

Sound: First off, it sounded very "hollow", so I left it running for one CD and went outside. About an hour later, it was sounding much better. It continues to improve, and has about 15 hours on it now. I took it into a local hi-fi shop, and the response was how musical it sounded. I agree, and briefly connected my Proceed Amp 2 back into the system. The sound appeared "flat" in comparison, but I will compare in more detail later on.

I do have two questions which someone may be able to answer - Hugh?
1. I measure DC offset at -14 & -16mV in each channel (initially -7 & -8mV). Is there a rule of thumb for increasing the value of R3 to minimise this, ie x ohms per millivolt. I know this is around the point where there should be no benefit, but I am pretty anal.
2. When swithching off (and on to a lesser extent), there is a pronounced outward movement of the speaker drivers. This has been noted in another recent post as an increase in DC offset, but I havent measured. Is there any way to prevent or minimise this? I understand that there should be no ill effect with a passive crossover in line as the series tweeter cap will prevent DC being passed to the tweeter, but there is no such protection with active designs. Am I just worrying for no reason?

Thanks

Mike

[AKSA]

Mike,

Thanks for your post!  You wrote:

1. I measure DC offset at -14 & -16mV in each channel (initially -7 & -8mV). Is there a rule of thumb for increasing the value of R3 to minimise this, ie x ohms per millivolt. I know this is around the point where there should be no benefit, but I am pretty anal.

The offset moves up and down several millivolts with the mains voltage supply, and is not worth getting to zero because within the hour it will have moved again!  If you do a comprehensive analysis of offsets for each hour of the day over 24 hours on a work day, you might get the mean.  Most speakers function just fine with 150mV of offset on them;  there's just no sonic effect at all.

4K7 change on R3 is roughly 10mV of offset change for the stock AKSA, about half that resistance for the Nirvana.  Increase R3 to raise offset, decrease it to lower offset, both wrt earth.

2. When swithching off (and on to a lesser extent), there is a pronounced outward movement of the speaker drivers. This has been noted in another recent post as an increase in DC offset, but I havent measured. Is there any way to prevent or minimise this? I understand that there should be no ill effect with a passive crossover in line as the series tweeter cap will prevent DC being passed to the tweeter, but there is no such protection with active designs. Am I just worrying for no reason?

This has in fact been noted in this forum several times, and yes, it's 12V at switch off for around half a second.  There is not one single speaker in the marketplace (even a Lowther!) which this will damage because the dissipation is around 12W for half a second, and anyone using a 12W continuous rated speaker with his AKSA is due to blow it up, because it needs at least a 35W rms rating for durability.

There are ways of eliminating this pulse at switch on (very small, on the edge of audibility) and switch off (infrasonic, in fact, visible on the cone but not audible).  They are quite complex, and range from relays on the output stage holding everything at ground until stabilization to constant current sources on the diff pair and voltage amp.  The relay trick works well enough, but costs, takes up lots of board space, and quite a few extra components.  The CCS solution costs sonics, and the amp does not sound anywhere near as impressive.  I have managed to quieten this pulse to inaudibility over the years, and know that it will never damage any driver to which it is connected;  this is a fair compromise and indeed, if you put it out of your mind you will never be troubled by it.

Congratulations for completing your AKSA, Mike!  It will steadily improve over the next week.  Why not leave it on all the time?

Cheers,

Hugh

[andyr]

It took nuch longer than I anticipated, ... but I wanted to prove the amp first.  ...

O ye of little faith!!

Now, I hope you now "have faith"!!  Please, do yourself a favour -appropriate for having spent a) your hard-earned cash and b) that more precious commodity ... your time - and get yourself a metal case.  I would think RS Components operates in South Africa, alternatively, there must be electronic parts suppliers where you can buy a 19" rack case.

If you're using toroids for power transformers a 2RU case should be fine ... alternatively, Hugh has just had some very fine AKSA cases made up, which I am sure he'll be delighted to send you for a few shekels.

Secondly, why turn your amp off?  I leave mine on all the time except when I go away for a few days.  All electronics take a number of hours to "warm up" and reach a their "sweet spot" so if they're always on, you always get good sound!!  You might be worried that the power supply caps only have X,000 hours life but this life, just like an electric light bulb, is also affected by the no. of times they are switched on and off.  So, basically (according to Hugh), it won't shorten the life of your amp by leaving it on all the time!!

Happy listening and welcome to the AKSAphile community.

Andy

[MikeC]

Me, no faith???

Well maybe a little uncertainty. The important thing was to get the amp built: I would still be playing around now if I hadn't done it this way. And, an amp this good deserves better than an RS components box. I'm thinking 1.2mm stainless steel or aluminium.

Thanks for the very quick replies.

Mike