[sdiy] More on Xpander modes was Re: HP from LP ?

[David Moylan]

Neil, a good point.  Also, in this particular filter implementation the 
caps are 220pF.  I wonder how much stray capacitance could affect 
things.  It would only take a few pF to start racking up percentage 
points.

I'm providing non inverting buffers on my board (don't need to invert 
the stages since IR3109 does that already).  I did so because I thought 
it might make the board easier to "shoehorn" into other circuits and I 
needed a buffer close by to the first stage cap, which I moved on to my 
PCB so that it could easily be switched in/out.  So I just used a Quad 
opamp and provided 4.  Since I had the on board buffers I think I 
soldered the flying leads directly on the caps rather than post IR3109 
buffers.  That would make them more susceptible to stray capacitance 
from the leads, which additionally aren't currently as short as they 
could be.

So I have a few things to experiment with:

1. Increase cap values to largest value that won't adversely affect 
filter range, and probably tighten tolerance on those caps as well. 
Largest value I've seen in a schematic is 470pF (thanks, Florian!). 
Wonder if it could go to 1000p.

2. Move the flying leads to IR3109 buffer outs and shorten as much as 
possible.

I have to be away for a few days suddenly so video will probably be 
pushed until the weekend.



On 08/01/2017 01:15 AM, Neil Johnson wrote:
> "David Moylan" wrote:
> 
> 
>     This is due to the fact that achieving HP responses relies on
>     nulling out terms in the transfer function via mixing the different
>     poles.  If you're not exact you don't get a full null and the
>     attenuation of low frequencies suffers (looks more like a shelf
>     instead of continuing roll off of low frequencies and usually has a
>     little dip in the middle).  I'm using 1% tolerance resistors, but am
>     about to sub in some 33.2k for 33k to get even closer to the
>     theoretically perfect values.
> 
>     Even if you had perfect resistors I would imagine any mismatching
>     between the OTA stages would also cause the attenuation to suffer,
>     but I haven't modeled that.
> 
> 
> The integrator capacitor tolerance will likely swamp any improvements in 
> resistor tolerance.  For example the 2164 is quoted as having worst gain 
> matching of 0.2dB which is just over 2%, yet capacitors will likely be 
> 5% - you can get C0G down to 1% if you pay for them. So 1% resistors are 
> already better than anything else in the signal path.
> 
> Neil