diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Chip Amps (http://www.diyaudio.com/forums/chip-amps/)
-   -   Hi! And I built one too... (http://www.diyaudio.com/forums/chip-amps/140233-hi-i-built-one-too.html)

The RB 11th March 2009 03:47 PM

Hi! And I built one too...
 
Hi people, just a quick post to say Hi (I just joined the forum) and to show off my new homemade "gainclone" amp;
http://www.romanblack.com/gainamp.htm

maybe someone will find it interesting.

http://www.romanblack.com/gainamp/amp01.jpg

its nothing pretty compared to all the superb amps you guys have built!

I may not be here much due to my many commitments but I will try to pop in from time to time and see whats new in the exciting world of diy Audio.

Cheers! :clown:

PS. How do you get pictures to show in a post?

AndrewT 11th March 2009 04:16 PM

Hi,
nice amp and nice build, nothing to be ashamed about.

A few comments.

1.) the high pass filter or low frequency roll off is controlled by the 22uF negative feedback capacitor. Not the recommended way to do it.
You can extend the LF response by doubling or even quadrupling that cap to 47uF or 100uF and even go to 150uF.

2.) there is no DC blocking cap on the input. There is no DC detection at the output.
The DC blocking cap should define the high pass input filter.
for Zin=150k, try 330nF, or 470nF.

Consider, maybe, reducing Zin to 51k or 100k and making the DC block larger.

3.) consider swapping the locations of the 100nF and 470uF decoupling capacitors.

larryg 11th March 2009 09:39 PM

Very nice work!

The RB 13th March 2009 12:42 AM

Thanks guys for the nice words. :)

re your questions AndrewT;

1.) the high pass filter or low frequency roll off is controlled by the 22uF negative feedback capacitor. Not the recommended way to do it.
You can extend the LF response by doubling or even quadrupling that cap to 47uF or 100uF and even go to 150uF.

Yep the 22uF cap sets the low freq gain. I kept SiliconChip's value of 22uF, it measured about -3db at 15Hz which is fine. My speakers don't do much below 35Hz anyway, this amp of for my recording studio monitoring, not for theatre use (no subwoofer etc). At <3Hz (and DC etc) this cap reduces the amp's gain to 1.

2.) there is no DC blocking cap on the input. There is no DC detection at the output.
The DC blocking cap should define the high pass input filter.
for Zin=150k, try 330nF, or 470nF.
Consider, maybe, reducing Zin to 51k or 100k and making the DC block larger.

Your comment surprised me. As a pro electronics designer I always consider it good practice that anytime you can avoid passing signal through a cap then you should avoid it. This amp is fully DC coupled, zero phase shift, incredibly low THD, no caps anywhere in the signal path. That's just about the perfect amp by my reckoning.

Each amp DC offset measured at <3mV with no signal in. The amp has a DC gain of 1, so worst case it will echo the DC offset of the signal source. It is driven from my Pioneer 1bit CD and my studio multitrack recorder. Both have output DC offset <5mV.

I'm interested in your input, but I can't see why it would benefit from sticking a cap right in the front of the signal path with the associated phase shift and all the other signal cap issues (caps seem a hot topic on this forum).

3.) consider swapping the locations of the 100nF and 470uF decoupling capacitors.

Yep I too scratched my head when I saw SiliconChip's cap setup on the power fuses. Basically it forms a pi filter with the few milliohms of the 2A fast-blow fuses in the middle. The chip datasheet recommends putting the main caps close to the chip with minimal resistance tracks, so the big caps are definitely better after the fuses. As for the 0.1uF cap before the fuse? I'm guessing that since SiliconChip have had some RF instabilities in their amps before, maybe they thought it wise to take out some of the psu HF noise before the low pass filter formed by the fuses->main caps.

I'm interested in any thoughts re the caps etc.
:)

AndrewT 13th March 2009 09:12 AM

Hi RB,
you have established good reasons for doing it your way.

That's the whole point.
Informed decisions rather than blindly following/modifying because someone on the net said so.


I don't happen to agree with your decisions, but they are up to you and your priorities.


All times are GMT. The time now is 12:10 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2