Hello guys
I have made my own chip based power surroud amp for my pc and it's working great. The only problem I have is DC offset. Short of installing capacitors in the signal going into the amp, is there any way to change the DC offset via windows? I'm using an audigy 2zs, incidentally the onboard should has next to no DC offset 😡
The amount of offset, after amplification is about .87 volts! It's not damaged any speakers and the sound is still ok but I really dont want it. If i have to stick 5 caps in the signal lines, which would you reccomend and what size should I use for maximum transparency?
Cheers
Dan
I have made my own chip based power surroud amp for my pc and it's working great. The only problem I have is DC offset. Short of installing capacitors in the signal going into the amp, is there any way to change the DC offset via windows? I'm using an audigy 2zs, incidentally the onboard should has next to no DC offset 😡
The amount of offset, after amplification is about .87 volts! It's not damaged any speakers and the sound is still ok but I really dont want it. If i have to stick 5 caps in the signal lines, which would you reccomend and what size should I use for maximum transparency?
Cheers
Dan
Hi,
have you got a cap between ground and the second resistor in the -ve feedback section? I'd suggest putting in a 47uF at this point, offset should reduce to next to nothing. What is the offset with input shorted?
I also use an audigy II ZS and have no problems.
BTW this should really be in the chipamp forum, I'll report the post and ask the mods to move it 🙂
Tony.
have you got a cap between ground and the second resistor in the -ve feedback section? I'd suggest putting in a 47uF at this point, offset should reduce to next to nothing. What is the offset with input shorted?
I also use an audigy II ZS and have no problems.
BTW this should really be in the chipamp forum, I'll report the post and ask the mods to move it 🙂
Tony.
Please move thread to chipamps forum 🙂 woops... that wasn't supposed to happen... seems that the report post is not supposed to be used for this anyway, just read the fine print 😉 Hopefully a kind mod will notice.
Last edited:
Worth confirming with a meter. My Benchmark DAC isn't supposed to have DC offset either but the 40 mV measured was enough to put 1.5 Vdc on the speaker outputs of an Audiosector gainclone kit. On the plus side it created an opportunity to whip up a transformer-coupled tube pre with independent floating secondaries. This particular chipamp seems to really like a low DC source impedance. 🙂
Thanks for the reply, yeah if the thread needs moving then no probs =]
The problem doesn't lie within the amp, its the output of the soundcard. When the soundcard isnt connected to the amp, i only get 70mv after amplification DC offset, which I can definately live with.
http://www.dogbreath.de/Chipamps/ThreeResAmp/ThreeResAmp.html
Thats the amp I've made, adapted for 5 channels.
The reason why I know the onboard has next to no offset is because theres (nearly) no offset at the speakers when everything is powered up.
thanks for your replies
The problem doesn't lie within the amp, its the output of the soundcard. When the soundcard isnt connected to the amp, i only get 70mv after amplification DC offset, which I can definately live with.
http://www.dogbreath.de/Chipamps/ThreeResAmp/ThreeResAmp.html
Thats the amp I've made, adapted for 5 channels.
The reason why I know the onboard has next to no offset is because theres (nearly) no offset at the speakers when everything is powered up.
thanks for your replies
The problem DOES lie with the amp. The cause is the lack of DC gain capacitor on Ri as Wintermute has already mentioned.
The soundcard is already AC coupled, adding more caps to the input of the amp will only make it worse.
The soundcard is already AC coupled, adding more caps to the input of the amp will only make it worse.
Ok. Could you explain to me a few things.
If the problem is with the amp, why does the problem not appear when the onboard soundcard is connected? What does a DC gain cap do, and how does it do it? Where exactly would I need to install these caps (in between rf and ri to ground)? What type of cap should I use and what capacitance will I need? Will it affect the sound quality much?
Sorry for the barrage of questions
If the problem is with the amp, why does the problem not appear when the onboard soundcard is connected? What does a DC gain cap do, and how does it do it? Where exactly would I need to install these caps (in between rf and ri to ground)? What type of cap should I use and what capacitance will I need? Will it affect the sound quality much?
Sorry for the barrage of questions
Last edited:
It reduces the gain of the amp to DC signals to unity so offets etc are not amplified. Put a 100uF or 220uF cap between the ground side of Ri (+ve on cap) and ground (-ve on cap), ensure you break the link that is currently there or you will achieve nothing as it will be shorted out.
electrolytic? and just to clarify -
----[ ri ]-----(GND)
to
----[ ri ]-----|(----(GND)
What effect will different sizes of caps have? will this affect audio output at all?
----[ ri ]-----(GND)
to
----[ ri ]-----|(----(GND)
What effect will different sizes of caps have? will this affect audio output at all?
A properly operating DC-coupled amp is not a problem, DC on the input is a problem. My offset is about 5 mV with no blocking caps. Examine the situation and pick your poison.
Great drawing! Yes that's spot on. Some people claim it affects the sound, others do not.
Yes different sizes WILL affect the audio output 🙂 Basically the smaller the cap value the earlier the low freq rolloff will start.
formula for calculating this for an LM3886 (from the datasheet) is: fc = 1/(2PiRi Ci) so for your amp and a 47uf cap fc = ~5Hz [1/(2*Pi*680*0.000047)] with a 10uF cap the fc would be ~ 23Hz and with a 220uF it would be approx 1Hz.
Using something as low as 10uF can be useful to ensure your speakers don't get killed by extremely low freq signals.
As Richieboy said, some people claim a cap here affects the sound (other than what I have said above) and some say it doesn't. I'd personally think that a cap in the feedback circuit (if it does have an affect other than as stated above) is probably going to have less of an effect on the sound than one in the direct signal path (ie a coupling cap on the input).
I have a 47uF cap between Ri and ground on my LM3886 P2P gainclone with 680 ohm Ri and get a dc offset of 1.4mv with shorted input 🙂
Tony.
formula for calculating this for an LM3886 (from the datasheet) is: fc = 1/(2PiRi Ci) so for your amp and a 47uf cap fc = ~5Hz [1/(2*Pi*680*0.000047)] with a 10uF cap the fc would be ~ 23Hz and with a 220uF it would be approx 1Hz.
Using something as low as 10uF can be useful to ensure your speakers don't get killed by extremely low freq signals.
As Richieboy said, some people claim a cap here affects the sound (other than what I have said above) and some say it doesn't. I'd personally think that a cap in the feedback circuit (if it does have an affect other than as stated above) is probably going to have less of an effect on the sound than one in the direct signal path (ie a coupling cap on the input).
I have a 47uF cap between Ri and ground on my LM3886 P2P gainclone with 680 ohm Ri and get a dc offset of 1.4mv with shorted input 🙂
Tony.
- Status
- Not open for further replies.