diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Chip Amps (http://www.diyaudio.com/forums/chip-amps/)
-   -   Cap. at the output on bridged no DC (http://www.diyaudio.com/forums/chip-amps/22328-cap-output-bridged-no-dc.html)

avenger 3rd November 2003 09:55 PM

Cap. at the output on bridged no DC
 
Hi this is my first time posting...I have an idea that may look far fetched but i see alot of concern in regard to paralleling and bridging the lm3886 amps to achieve higher power, this concern because the dc present at the output and different dc levels will induce current flow between the amps, BUT what would happen if we connect a high current..snubber capacitor.. at the output before the low ohm resistors. Wouldn't this block ALL dc from flowing between the amps? I know that the capacitor will somewhat act like a filter, but can't we pick a cap. value that would not affect audio 20-20KHz? Please reply I would like to see other point of views on this.

bigparsnip 3rd November 2003 10:59 PM

If you look at teh maths for this I think you might be put off quite quickly, as say we had a 4 ohm load to drive, and you wanted the -3dB point to be a 10 Hz to stay out of the way of the audible range. To do this you would need to have a capacitor of at least 0.004F (4000uF). This cap would have to be a non polar type and be able to withstand the voltages present in a bridged amplifier (with could be as high as +/- 80V for an LM3875 chip), and then on top of all this, it would have to sound good.

So, if you can find a cap like this for a reasonable price, I would be quite supprised, but if you do, I would very much like to know where you can get them :D .

Andrew.

avenger 4th November 2003 09:20 PM

HI, I don't understand since I have seen amplifier schematics using a capacitor at the output...I just figure it could be beneficial to us , please look at the link below...

http://kitsrus.com/pdf/k97v2.pdf

theChris 5th November 2003 01:25 AM

you make an RC network, basically a highpass filter on the speaker. now to lower the allowed response, the cap needs to be big, or the speaker impedance needs to be large. for 4ohm speakers, this means a large electrolyic. for a power amp the votlage will be large, and thus the cap will be even larger. electrolytic caps also are further from the ideal cap then say film and foil, again making them even less ideal.

avenger 5th November 2003 02:27 AM

Oh certainly I understand the high pass filter, but what bugs me is that some amplifiers like the one on the link I posted above already have the electrolity cap on the output and it works ok, shouldn't that work on the 3886? That is what gets me

bigparsnip 5th November 2003 07:42 AM

Right, what I said above didn't mean that you couldn't do what you say, you could block DC on the amp output using the cap. But I think what we are trying to say here is that for the small benifit of getting DC blocking at the output, you have to spend a lot of money on the cap itself, and even then it will sound verry poor compared to no cap. So, I general it isn't worth doing this when it isn't too dificult to build a design with something like a DC servo which would probably cost less, give a better frequency response and sound better.

carlosfm 5th November 2003 11:01 AM

summarizing...
 
Any amp with a big cap on the output doesn't sound nearly as good as a Gainclone.
Why make things worst?
Even a zobel may do no favours to the sound.
:angel:

BTW big electrolythics are for PSUs, don't use them on your signal.

jan.didden 5th November 2003 11:45 AM

Quote:

Originally posted by bigparsnip
Right, what I said above didn't mean that you couldn't do what you say, you could block DC on the amp output using the cap. But I think what we are trying to say here is that for the small benifit of getting DC blocking at the output, you have to spend a lot of money on the cap itself, and even then it will sound verry poor compared to no cap. So, I general it isn't worth doing this when it isn't too dificult to build a design with something like a DC servo which would probably cost less, give a better frequency response and sound better.
Indeed. On the posted link with the LM380, there the cap is necessary because there is only one supply polarity, so the amp output will sit at half the supply. No way you can run this without a cap. But with the bridged GC, if you are carefully you can get the DC output below 100mV, so you can safely leave out the cap, saving a lot of money and getting better sound to boot. You can even do it with a bridged LM380, single supply, just make sure both bridge parts are at the same DC.

Jan Didden

avenger 5th November 2003 09:04 PM

Hi Guys, thank you for clarifying this for me, I though that it couldn't be done, but the issue is with the sound quality...I really appreciate this. As far as getting caps. I am a capacitor tech at one of the biggest capacitor manufacterer so finding or getting one would not be a problem..I rather not mention the name of the company. So if you guys need some info on capacitors ie construction...just shoot this way.

Ola carlos como estas meu....sou de portugal tambem..

carlosfm 6th November 2003 08:55 AM

Quote:

Originally posted by avenger

Ola carlos como estas meu....sou de portugal tambem..

:D
Good to see that there are portuguese guys in the states working in cap manufacturers.
I'm tired of buying caps...
Send me some caps, man.:D


All times are GMT. The time now is 03:09 PM.


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