[arm-allstar] FOB 48KHz Aliases and spurs

[David McGough]

A simple RC 2nd order passive filter is probably adequate and should
provide 12db/octave rolloff. The main drawback with a passive filter is 
insertion loss, which is why active designs that provide gain to 
compensate, combined with higher order filters, are popular.

Here is a website with some on-line design tools:

2nd order filter design:
http://sim.okawa-denshi.jp/en/CRCRtool.php

Many other general purpose tools:
http://sim.okawa-denshi.jp/en/Fkeisan.htm


If anyone is interested, a 192KHz/24bit sound card (something like:
http://www.ebay.com/itm/131767155488 ) could be used to build a
high-resolution audio spectrum analyzer to "test" the lower sample rate
(48KHz) USB FOB. Some typical software for this type of application is
found here: http://www.qsl.net/dl4yhf/spectra1.html

Anyhow, if you're using a mic-input on your transmitter, and even most 
high-speed MODEM inputs, the transmitter frequency response is well below 
20KHz, and DAC Aliases are not of concern.  HOWEVER, if you were to place 
an unfiltered FOB directly behind an old GE Mastr II exciter, you 
might have spurs thru the entire ham band and into the business band!!! 
So, the key here is to know your radio hardware--or, add a filter and 
eliminate all concerns.


73, David KB4FXC



On Fri, 8 Apr 2016, Ken via arm-allstar wrote:

> The response cutoff issue on the FOBS is probably a non-issue for most.
> Check it with a scope if you think it is a problem and put a non active RC
> filter whihc should be fine for 48Khz especially with the rolloff of the
> radio, etc. 
> 73 Doug
> WA3DSP
> 
> I agree that the 48 kHz would only be a problem in specific cases, where the
> modulation was introduced directly into the frequency control circuit.
> For the vast majority - using standard audio inputs - it's no problem.
> I would point out that a simple RC filter has 20 dB per decade roll-off.
> With (for example) a 4.8 kHz corner frequency, the response would be down
> only 20 dB at 48 kHz.  That is nowhere near enough for the case where you
> are injecting directly.
> 
> Regards
> Ken
>  
> 
> 
> 
> _______________________________________________
> 
> arm-allstar mailing list
> arm-allstar at hamvoip.org
> http://lists.hamvoip.org/cgi-bin/mailman/listinfo/arm-allstar
> 
> Visit the BBB and RPi2 web page - http://hamvoip.org
>