1 Meg Pot on ADC input?

S

SMC

Member
Hey guys-
I have a bunch of sweet (hermetically sealed) high quality 1 Meg pots that were given to me. I know you are supposed to use 10k pots on ADC inputs....but the 1M pots seem to do fine in my test circuit. I understand that the current available thru the 1M's for the ADC reading (charge cap etc) is low and the reading might take longer with the 1M pot...but besides this is there a problem using the 1M's?
Will it damage the picaxe?
My circuit has 0-60 second and 0-60 minute timing cycles adjustable with the pots and seems to work perfectly and accurately.
So- can I use the 1M's or should I order the 10K's?

THANKS!
 
hippy

hippy

Ex-Staff (retired)
No it won't damage the PICAXE.

As you say, the current into the internal ADC cap ( Sample & Hold ) will be slower but if you can live with that you should be fine. Multiple READADC of the same pin before taking a final reading may help there, also switching to a READADC of 0V in between can help; zap the cap to zero, switch to a pot, keep reading until stabilised.

The pots / circuit may be more sensitive to external noise pickup.

I'd use them, see how it goes, swap them out later for others if there is a problem.
 
S

SMC

Member
Thanks Hippy-
Great help as usual!
Can you think of a way to "torture test" the noise thing? Like run it next to a motor or something?
 
hippy

hippy

Ex-Staff (retired)
One simple test may be to hang a wire off the ADC pin with pot in place and tap/squeeze it while seeing what happens on a repeating READADC. Ideally report only on changes so you'll get an idea of how big the problem is.

Of course that's a bit abstract and doesn't prove or maybe mean a lot. Perhaps also compare doing the same with a 10K pot ?
 
B

BeanieBots

Moderator
Please note that when using such a large value input resistance that your ADC reading will be significantly effected by whether or not the programming cable is fitted. The enhanced download circuit will help to overcome the effect but will not completely eliminate it.
If your serial line drives at greater than 5v, you can add a series diode to completely remove the problem but downloads might be 'temperamental' from a 5v serial line.

Instability can be improved by adding 100nF to the ADC input but response time will be worse.
 
You must log in or register to reply here.
Top