Filter design

[fergcu]

Hi Everyone,
I was wondering if i could have a little more help please?!
I don't really know what it is that im asking for, or even if its possible, but essentially i would like to smooth a wave form out so that it is practically a flat line at its maximum amplitude. I dont know if you can see what i mean, so have included a file to give you the general gist.
I hope someone is brave enough to try and work out what the hell i am going on about, and if they have got this far, that they can help me with this!

Thanks for your time,

Ferg

 

[Nigel Goodwin]

Hi Everyone,
I was wondering if i could have a little more help please?!
I don't really know what it is that im asking for, or even if its possible, but essentially i would like to smooth a wave form out so that it is practically a flat line at its maximum amplitude. I dont know if you can see what i mean, so have included a file to give you the general gist.
I hope someone is brave enough to try and work out what the hell i am going on about, and if they have got this far, that they can help me with this!

Sorry, can't see your picture here! - and don't really understand what you're trying to do?.

It would probably be far better if you told us what you are wanting to do?, instead of asking us to try and understand the solution you think you need!.

 

[bmcculla]

It sounds like you want a full wave rectifier. You can do this with a simple opamp circuit - just google for schematics.

 

[Juglenaut]

Filter DC??

caps, bleeder, and regulate.

 

[fergcu]

ok, looking back at what i wrote it doesnt make much sense - i had just come out of a 6 hour meeting :shock: !! sorry about that. Instead, I will explain my project in full just so that you can get an idea of my intentions:
This is essentially a PIC project for university, using a 16f877. The idea of the final product is something which 'listens' to the ambient volume of the room. I'm using a piezo-electric to do this, and running it through a pre-amp (using an NE5534). This then goes into the 877 on RA1 for the ADC. The next part is where im sort of stuck. Obviously a sound wave is, as it says, a wave. Because i am wanting to sample just the maximum peak of the wave, i am getting problems when the ADC samples the slopes of the wave - instead of getting a constant reading of, say 5V, i am getting 5, 4, 3, 2, 1, 0, 1, 2, 3, 4, 5....(the full wave rectifier mentioned above definitely helps with this, so thank you bmcculla - it means i dont go into -ve values as well!)).
I would ideally prefer to sample it only when there is a peak, but this isnt possible due to the ever changing wavelengths of ambient sound! Instead of having the sound wave going into the adc, would it be possible to smooth the wave out, so that it gives a flat voltage of what ever the maximum, (or even mean voltage) of the wave being sampled is? I just need to differentiate between different volumes with a PIC.
I hope that is more clear, and if anyone has any suggestions, i would be extremely grateful.
Many thanks in advance.
Ferg
ps.. whats a bleeder juglenaut? i have a vague feeling that capacitors might be the way to go, but im not an electronics wizard - in case you hadnt guessed! :?

 

[Nigel Goodwin]

Right, that makes more sense.

What you are trying to do is build a VU meter, with the PIC taking the part of the actual display (even if the PIC doesn't actually display the result).

Have a look at this example **broken link removed**, you simply leave out Q1 and associated parts, and connect the PIC analogue input to where the base of Q1 connects. You may need to amplify, or attenuate, the voltage to make it suitable to feed the PIC input. I would also suggest you permanently connect the 10uF, so it's in PPM mode.

You should be able to find various VU meter circuits on the net!.

 

[fergcu]

Once again, Nigel The Electronics Bible has struck again. Thank you so much - this should help a huge amount. BTW, the IR controlling problem that i was having last time was helped a lot by your website. Thank you.
Ferg