Solid State Power Amp

I'm looking at putting together a solid state power amp. So far I've looked at the AKSA amps, SKA amps and a design I found on ebay by a company called New Class-D. I've also considered the NewClassD modules, the Trueforce kits by 41hz, Coldamps & Hypex.

I'm particularly taken (so far ) with the SKA kit, its reasonably priced and gets good reviews. Has anyone any experience of these amps or any of the others Ive mentioned??

 

Can't help with any experience but would second the request for anyone who has. I've been looking at similar options and was also wondering about the LC Audio modules. I think MichaelAB who used to post a lot round here built an amp using these a few years back - there's an old thread somewhere. But I'm also thinking the SKA kit (http://www.ska-audio.com/diy/index.html) looks promising and relatively economical. (I would need at least 4 channels...)

 

I built a four channel amp with active crossover from LC Audio
using their modules. Works very well with my Manger drivers.
All reports are that the new version is even better.
LC Audio was started by Larrs Clarsen he then left and started
New Class-D. I would be tempted to go with them as Larrs always
answered my questions and also repaired the crossover for free
when one of the channels went down.
regards

 

The NewClassD board was my original favourite, I just like the idea of a complete kit, which is something SKA do.

 

I'm new here but have a SKA amp - a GB150D. It is an excellent sounding piece of kit, easy to build (with a little DIY skills), great instructions and great support from Greg Ball, the designer. There is an active forum where a range of posters from novice to expert are involved. Tweaks have been posted to this forum by Greg & SKA owners which have proved to raise the performance of this amp even more.

What attracted me to this amp in 2006 was Gregs take on what were the main issues affecting sonics in power amplifiers. His theory (which is proven by the sound of his amp) is that PS is corrupted in Class AB amp by how class AB operates. His contention is that these PS errors affect the input stage & are then amplified by the VAS stage & O/P stage.

His design has achieved a very high gain the input stage with the result that less amplification needed in the VAS stage, etc. This has resulted in a PSRR of about -115db in the audio band which is almost like having a regulated supply feeding the amp.

Anyway, I can highly recommend his amps & other products, such as the optivol which is a LDR based potentiometer replacement.

As this is my first post I'm sure suspicion will be raised that I'm a sock puppet but just see my posts on the SKA forum & DIYaudio under the same name jkeny. I have no connection to Greg or his offerings other than being a satisfied user

 

thats well known ! isn't it ?

a power amp is only as good as its ps ...or thats what I've been lead tobelieve for twenty odd years

 

Yes, Zanash, it's well known that PS is paramount to the sound of all HiFi but how many amp designers build it into their circuit to address this important issue?.

The SKA removes the high amplification factor of the VAS (found in most amp circuits) which amplifies the PS artifacts that have intruded into the the input stage and hence you have such exemplary PSRR for the amp. This makes it feasible to use a simple PS without a huge bank of capacitors to remove ripple.

It is also a damn fine sounding amp!

 

I use SKA amps and I do like them a lot. They achieve a higher PSRR while having a less complex design. Lovely I also run mine from a fully regulated supply.

I wouldn't go for LC Audio. I bought some and they blew up and I couldn't fix them. It was my fault in a way as I had a ground loop at ultra sonic frequencies, but you can't really tell that sort of thing until you switch on an test. I think they are too unstable.

I also have some Hypex amps which I have not hooked up yet, but I hear they are very stable and sound great. The fact many commercial manufacturers use these modules in their own branded products suggests they are one of the best.

So, SKA or Hypex. I know Shinobiwan likes Aussie Amps - http://www.aussieamplifiers.com/

 

I have read great reviews of the AKSA amp over on the DIYaudio forums but often wondered if it is just hype. I am interested in what you find out too!

 

I went for the SKA power amp kit and the SKpre preamp.

I can vouch for everything jkeny has said; the quality is good and the service is excellent.

I have got to be honest though and say I've had a couple of niggles; it took me ages to bias the amp due to a huge jump in volts for the first 10 mins or so. I've also got an issue with a whine on one channel thats only there when the fuses are in place. I must say that I've not done anything with this due to illness for the last two weeks so it may be something simple.

On the whole I think its a dam fine sounding amp backed-up by some seriously good support from the owner of SKA.

 

You need to make sure the diodes are really well thermally connected to the main output transistors to get a stable thermal loop. I drilled a little hole and embedded them in the heatsink with thermally conductive epoxy. Once that is done it should be easier to get a stable bias, but yes you do need to leave them playing long enough for the heatsink to fully reach thermal equilibrium before setting it finally.

That really doesn't sound good. Must be a duff component in there.

 

Thanks for the advice. I have the diodes piggy-backed on the transistors, did you try this? I've been thinking about trying bolting the diodes to the heatsink.

I dont know about the whine. It's only there when the fuses are in place thus bypassing the test resistors. Like I said, I've not done anything about finding yet except inspecting all the joints (which are fine). My 'gut feeling' is that its something to do with that huge surge in volts at start-up.

 

How did you fix the diodes to the transistors?

 

I have the BD139 & BD140 so its just a case of using a slightly longer bolt. At the moment I have a silipad inbetween in order to control the bias over a period of time (it seems to stop it going too low).