[sdiy] Pots read by microcontroller and sourcing question

[Roman Sowa]

On test bench I have no problem with 300k pots, but I know what I'm 
doing and what to expect. If I set scan rate faster than obout 0.5ms per 
pot, this becomes unusable, voltages from 300k pots cannot follow that fast.
Another thing is usual slowish scanning of few onboard pots, and another 
thing when user connects his potentiometers on whatever cable he can 
find. Sometimes people use 1m unshielded wire going thru strange places 
and then obviously scanning that will produce rubbish.
100n will not eliminate 50Hz hum, and nobody will want to solder an 
opamp by the pot if there are other products not requiring it. That's 
why I always recommend to my customers pot values 10-50k, 20k being the 
best compromise between noise immunity, secure fast scan and supply 
current waste.

I mean all those ideas are great -  capacitor or opamp by the pot, I 
used them too, this is actually what my POTAMP does, but it's not always 
possible/justified, nor it makes any difference depending on final 
installation.

BTW, it was never said how many pots you want to scan, so maybe this 
discussion, except being hopefully informative, is also pointless.

Roman


W dniu 2015-02-02 o 20:54, Richie Burnett pisze:
> http://fa.utfs.org/diy/jx3p/schemo/JX3P_schematics_06_pg-200.jpg
>
> ...shows a bunch of 100k linear pots fed through CMOS multiplexers, a
> unity-gain follower, and a "discrete" successive approximation ADC in
> Roland's PG-200 programmer.
>
> I've not had any problems using 10k linear pots feeding PIC ADC inputs
> directly, provided that you put the 100nF capacitors between the pot
> wipers and ground.  This 100nF capacitor lowers the AC impedance at the
> input of the ADC and it is this that really matters.  It provides some
> local stored charge to "stiffen" the analogue voltage so that it doesn't
> sag dramatically when the ADC's sample and hold circuit takes its sample.
>
> The total current consumption starts to add up though if you have lots
> of 10k pots across a 5V or 3V3 supply!
>
> -Richie,
>
> -----Original Message----- From: David G Dixon
> Sent: Monday, February 02, 2015 7:27 PM
> To: 'Jack Jackson' ; 'Sdiy'
> Subject: Re: [sdiy] Pots read by microcontroller and sourcing question
>
>> Hi thanks for the replies.
>>
>> I'm using an Arduino. Looks like ideal input impedance of 10kb or less
>
> If you use an opamp buffer in front of every pot, then the output impedance
> to the uC is effectively 0 ohms.  This will cost you about 10 cents per
> pot,
> so not really a bank-breaker.  I personally would never take a voltage
> directly from a pot for this sort of control, since you really can't
> guarantee linearity unless you feed the pot to a near-infinite impedance
> such as the + input of an opamp.
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 2014.0.4800 / Virus Database: 4257/9046 - Release Date: 02/02/15
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>