diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Chip Amps (http://www.diyaudio.com/forums/chip-amps/)
-   -   Gainclone with DC-offset problems (http://www.diyaudio.com/forums/chip-amps/34930-gainclone-dc-offset-problems.html)

seve 28th May 2004 12:58 PM

Gainclone with DC-offset problems
 
Hi,

I have just finished my gainclone. I have build it around briangt's layout but on a veroboard. My plan is to use it with a preamp based on the AD825.

I have encountered some problems. When I measure the gainclone without the preamp and any sources I get a stable DC at 5mv and 30 mv. The problem is when I connect a source either directly to the gainclone or through the preamp I get alot of DC. The DC goes between 200mv to above 1V and it does not seatle down. Also if I connect the source through my preamp then the Volume control is influencing the DC-ofset.

BTW the preamp has a DC-ofset at 0-1mv.

Could it be either the missing input cap or the missing zobel?

Please, if someone can help me figure out what the problem is then I will be very happy.

Regards
Seve

peranders 28th May 2004 01:37 PM

Can you show us how you have connected? Schematics? have you measured the output with an oscilloscope, any oscillations?

Do you have DC current path from the input (bias current) through the pot? If yes, not good if the pot has a high value. Less than a uA times 50kohms = 50 mV => 30X => 1.5 V out.....

seve 28th May 2004 02:34 PM

Hi peranders

I have used the schematic that briangt has based his pcb on. Also I have made the layout like the pcb but on a veroboard and then made the connections underneath the veroboard. Most places I have just used the resistors "legs" to connect.I have almost wire it like the pcb traces from the briangt pcb. I have keept the signal and power ground separate and then connected them.

Sadly I don't have an oscilloscope!

I will see if I can find a Camera to take some pics of it

Seve

seve 28th May 2004 02:38 PM

Quote:

Originally posted by peranders
Do you have DC current path from the input (bias current) through the pot? If yes, not good if the pot has a high value. Less than a uA times 50kohms = 50 mV => 30X => 1.5 V out..... [/B]
I forgot to answer this. Sorry this is a little too technical for me. I don't have a pot in my gainclone but only in my preamp and the pot is a 10K

Seve

peranders 28th May 2004 02:48 PM

OK, but it helps if you have a schematic. I advice also to check the datasheet because there you have all basic information.

My idea about your truobles is the "input bias currents" which is currents for the input transistors. Those can vary rather much an aren't particulary either.

I missed to read some parts of your text. Brian uses 220 k as feedback and the non-inverting (+) input i grounded. I suspect that you IC has rather high !input offset voltage" and maybe also high "input bioas current". If that is the case you must trim the offset. Haven't got the time to tell you how but someone else here knows.

seve 28th May 2004 05:49 PM

2 Attachment(s)
Quote:

Originally posted by peranders
OK, but it helps if you have a schematic.
I have found a schematic of the version I have used. I have borrowed it from the thread about Briangt's non-inverted pcb I also think Briangt is the one who has made it.

seve 28th May 2004 06:17 PM

I just tried placing a pot (10K) at the input of the gainclone that actually helped a lot. Now I have DC-offset varying from 20-50mv, but it is still influenced by my pot meaning when I turn up the volume on the pot the DC also increases!

Though still strange that I cannot use it with my preamp or with out volume control just connecting the cd-player and using the internal volume control on the cd-player!

seve

peranders 28th May 2004 06:45 PM

Have you connected signal ground and power ground together? If not, do it :nod:

seve 28th May 2004 06:53 PM

Quote:

Originally posted by peranders
Have you connected signal ground and power ground together? If not, do it :nod:

Yes they are connected.

And thanks for trying to help me out.

I will try connecting the peamp and gainclone again later tonight and see if a wire or something was loose.

seve

jan.didden 28th May 2004 08:28 PM

Without the pot there is already an unbalance in the input resistance to grnd, but it is apparently not too bad. When you add a pot, the input R goes down (pot + 220// 22k), that upsets the offset. The fact that it moves if you turn the pot gives it away: the varying pot varies the R to gnd through which the input bias current goes (in // with the 22k of course). so the offset varies with the pot. You can completely solve it with a cap between the pot wiper and the input (or between the input and the preamp/CD connection).

For a 10k pot which has a lowest equivalent R of 2.5 k, the cap should be for 10Hz lower 3dB point, about 1/(2*pie*2.5*10e3*10), let's say 6uF or make it 10uF. Should be non-polarized.

Jan Didden


All times are GMT. The time now is 12:45 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