What's the best cap 'TYPE' for DC filter at line input.
Load across cap is around 25 KOhm, max 25v, I want flat to 1Hz. I figure it's be around 47uf-100uf.
This will be the only cap in the amp, I hate distortions created by caps & I want to minimize the damage. I don't mind spending a bit on it.
Load across cap is around 25 KOhm, max 25v, I want flat to 1Hz. I figure it's be around 47uf-100uf.
This will be the only cap in the amp, I hate distortions created by caps & I want to minimize the damage. I don't mind spending a bit on it.
Hi Brian,
By "type" do you mean brand or dielectric type?
There's an article on cap selection here that might be of interest to you:
http://www.capacitors.com/picking_capacitors/pickcap.htm
By "type" do you mean brand or dielectric type?
There's an article on cap selection here that might be of interest to you:
http://www.capacitors.com/picking_capacitors/pickcap.htm
I vote for metal film. About the largest reasonably sized value is 10uF. That should will be sufficient to have minimal effect on LF frequency response. Personally, in this application I don.t feel too happy with any electrolytic regardless of hoe pricy or exotic.
You also say no other caps in your amp. Uh... does that mean no power suppply bypass caps? If you are looking for "sonic purity" you might want to reconsider. Unless of course you like the sound of ripple. Clever remarks aside, power supply bypass caps prtect your amp from induced humm (100-120Hz depending on geography) entereing the input section where it is then amplified. It also protects it against any HF or RF that has somewho found it's way into the rails.
You also say no other caps in your amp. Uh... does that mean no power suppply bypass caps? If you are looking for "sonic purity" you might want to reconsider. Unless of course you like the sound of ripple. Clever remarks aside, power supply bypass caps prtect your amp from induced humm (100-120Hz depending on geography) entereing the input section where it is then amplified. It also protects it against any HF or RF that has somewho found it's way into the rails.
In my experience this is always a slightly contentious issue - I know experienced engineers who claim cap type makes no significant difference, and others who won't use anything but PPS.
Perhaps the main thing is to try to understand what it is that creates the distortion you wish to avoid.
I'm afraid I don't know an enormous amount about this (and am therefore more than willing to stand corrected on any of the following points ) , but one factor may be voltage-dependent capacitance. Since an audio signal in the analogue domain is a constantly varying voltage, a cap which is particularly sensitive to changes in voltage will not be a good choice. PPS caps are often selected for use in the signal path because of their low sensitivity to voltage change, but I doubt you'll find one large enough to suit your requirement.
As far as I'm aware, electrolytics have decent performance in this area within the audio bandwidth (another consideration), so, in this case would be perfectly adequate, and I wouldn't think you'd need to spend a great deal either.
In any case, if you work out what's causing the distortion, it should just be a case of checking the manufacturers data sheets to make sure the part you choose has suitable characteristics. Then maybe just try a few parts out and see which one you like the sound of. It is, IMHO, an almost entirely subjective issue.
As far as this being the only cap in your design, if you do mean that you have no supply decoupling then I can only concurr with Sam9 that you probably should have (well I can't think of a good reason why not to).
Perhaps the main thing is to try to understand what it is that creates the distortion you wish to avoid.
I'm afraid I don't know an enormous amount about this (and am therefore more than willing to stand corrected on any of the following points ) , but one factor may be voltage-dependent capacitance. Since an audio signal in the analogue domain is a constantly varying voltage, a cap which is particularly sensitive to changes in voltage will not be a good choice. PPS caps are often selected for use in the signal path because of their low sensitivity to voltage change, but I doubt you'll find one large enough to suit your requirement.
As far as I'm aware, electrolytics have decent performance in this area within the audio bandwidth (another consideration), so, in this case would be perfectly adequate, and I wouldn't think you'd need to spend a great deal either.
In any case, if you work out what's causing the distortion, it should just be a case of checking the manufacturers data sheets to make sure the part you choose has suitable characteristics. Then maybe just try a few parts out and see which one you like the sound of. It is, IMHO, an almost entirely subjective issue.
As far as this being the only cap in your design, if you do mean that you have no supply decoupling then I can only concurr with Sam9 that you probably should have (well I can't think of a good reason why not to).
sam9 said:You also say no other caps in your amp. Uh... does that mean no power suppply bypass caps? If you are looking for "sonic purity" you might want to reconsider. Unless of course you like the sound of ripple. Clever remarks aside, power supply bypass caps prtect your amp from induced humm (100-120Hz depending on geography) entereing the input section where it is then amplified. It also protects it against any HF or RF that has somewho found it's way into the rails.
I meant in the gain & drive section. Only resistors and mosfets. Yes, the powersupply has a bank of computer grade caps.
OK. Whithout knowing you are the details of your amp I thought I'd mention that. One reads a lot of extreme sounding stuff about caps that doesn't consider what the intended function of a perticular cap is.
By the way, there is often/usually a cap involved in the VA section that manipulates open-loop feed back. Are you including one or does your amp avoid that?
By the way, there is often/usually a cap involved in the VA section that manipulates open-loop feed back. Are you including one or does your amp avoid that?
sam9 said:By the way, there is often/usually a cap involved in the VA section that manipulates open-loop feed back. Are you including one or does your amp avoid that?
Yes I'm even avoiding this cap. This is one of the areas which really bugs me in amp design. Not to also mention that MOSFETS have internal capacitance issues as well. I'm begining to think bipolar.
sam9 said:When you finalize the design, I'm sure it would be of interest to see it posted. On it's own thread, of course.
Finished the Pre-amp attenuator module. This 1 is using 9.9uf Film PPs caps (3x3.3uf in parallel). Almost finished the controler for that board, the amp is next.
I posted some of the pre-amp specs here if you are interested.
http://www.diyaudio.com/forums/showthread.php?goto=newpost&threadid=5790
In any signal path application you should look for the cap with the lowest DF and DA you can find. Film and foil types are better than metallized but much more money. DF = dissipation factor, put x in and you get x - DF out. DA is dielectric absorbtion, the cap retains some of the signal causing a 'smeared' effect. I like polystyrene as a bypass to a larger polyprop. Don't forget to bypass your power supply caps w/ a few uf of film type cap. Computer grade caps are not made for audio's needs and a bypass helps a little(or a lot!)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Design & Build
- Parts
- What's the best cap 'TYPE' for DC filter at line input.