Simple Killer Amp Constructor Thread

[Wavebourn]

I lived in Randburg, worked in Sandton, but did not have opportunity to meet people who love to DIY!

 

[gbyleveldt]

Yup, those are a bit scarce around here 😀

 

[jacco vermeulen]

Mmm, same here actually.

(Heita da, ons kyk hier vandag saam toch. Naweek ook )

 

[jkeny]

George,
Don't know if anybody asked you this but did the sound change after changing the grounding scheme & decreasing the crosstalk - should have done but was it noticable?

John

 

[WillemFick]

Your dad is a Math teacher at Hans Moore? Your surname rings a bell...

That's me! Your surname sound familiar too...

 

[gbyleveldt]

Ah, small world then! You also building one of these? They come highly recommended, by me for one 😀 😀

If you into DIY, why don't you pop in to our local forum
http://www.avforums.za.net

Cheers

Gert

 

[gpapag]

Hi John (jkeny)

Don't know if anybody asked you this but did the sound change after changing the grounding scheme & decreasing the crosstalk - should have done but was it noticable?

As you know, instead of following Greg’s recommendation on earthing(PSU earthing point be connected to chassis ground through a 2R7 with two antiparallel diodes), I went initially for another earthing scheme (direct wire instead of 2R7//diodes).
Although with this direct earting the amplifier behaved OK on dummy load tests with sinusoidal and square waves, on real situation with long cables, my amplifier was hf oscillating (in a controlled manner though and only after a certain output level).
Earthing PSUs according to Greg’s recommendation, stopped oscillations.
The sound of the crypto-oscillating amplifier (very quiet, detailed, but lacking dynamics and fatiquing in the long run), had nothing to do with the sound of the non-oscillating amplifier (still quiet, detailed, but dynamic and non- fatiguing).
Regards
George

 

[HiFiddle]

Greg's just released his new SKpre -

http://www.audiocircle.com/circles/...ies=148;sesc=1f722030caab1d31b397944c596c64cf

and a new power supply for the GB150's -

http://www.audiocircle.com/circles/...ies=148;sesc=1f722030caab1d31b397944c596c64cf

 

[jkeny]

Hey George,
Based on your measurements I changed my grounding scheme (I was running to ground like you with no Resistor or diodes) - The amp sounds better now - I was worried after I had done my zener mods that the amp had become a little hiFi'ish but this has solved the issue & brought back that silken sound again without any tendency towards skitishness, if you now what I mean.

Thanks george - very useful measurements

regards
John

 

[gpapag]

Hi John
With the direct earthing, your amplifier might have gone into controlled oscillation like mine, thus the "skitishness". Anyway, after going Greg's way (2R7//1N400X) things became rock stable again. I will post some oscilloscope screenshots from 20KHz square wave tests with the 2R7//1N400X earthing. Loading data info are embeded into the pictures

 

[gpapag]

Higher output

 

[gpapag]

A bit higher

 

[gpapag]

Well above clipping

 

[gpapag]

Now 50Vpp output, but paralleling capacitors at the 6 Ohm dummy load
0.1uF

 

[gpapag]

0.47uF

 

[gpapag]

1.4uF

 

[gpapag]

And 2.2uF

 

[gpapag]

This is a very wide bandwidth amplifier.
My manual measurements with my 1MHz oscillator
show the -3dB points to be at ~3Hz and ~330KHz.
Roll-off slope is 6dB/octave beyond these points.
I verified it accidentally with an oscillating preamplifier. It was outputing a perfect 1800mVpp 5MHz sinusoid. Amplifier managed to output 680mV.
This is -8.45dB against the theoretically predicted -9dB.
Not bad.


Note:For attached measurement, power output was 1Wrms into 6 Ohm

 

[gpapag]

And for the fans of automatic data collection, here is the 3Hz to 20KHz response (ARTA software X100 averaging, Soundblaster Live card, Amplifier output 0.1Wrms into 6 Ohm).


Regards
George

 

[jkeny]

Hey George,
Only difference with my setup is I used 10R 1Watt resistor instead of 2R7 (it was all I had) - I wonder if this makes any diff?

Regards
John