Simple question about frequently issuing adcconfig commands

E

Eclectica

Member
Hi all,

I am building a dual thermometer circuit using precision NTC thermistors with parallel and series resistors (thus creating in effect a potential divider with low power dissipation through the NTC). Clearly using adcconfig to set Vref-, Vref+ to external pins with preset voltages will give me increased precision over a defined range.

However, when the temperatures experienced by the NTC travel outside the associated Vref- to Vref+ boundaries, I want to be able to set Vref+ back to 5v so that I can still read a value and then using a different scaling routine, deliver sensible values outside the 'precision' range.

Now, is it ok to repeatedly swap back and forth with the reference pins by issuing adcconfig commands without risking damage to the ADCs or any internal PIC circuitry?

Sorry if this appears a simple question :rolleyes:, just me being over cautious as ever!

Regards
Eclectica
 
A

AllyCat

Senior Member
Hi,

I don't think that what you propose risks any damage to the chip, but I'm not convinced it will necessarily give any better accuracy. You need to check the base PIC data sheet for restrictions on using the A/D converter. I believe that it may be necessary (or at least recommended) to maintain around two volts (or more) between the reference + and - pins.

Cheers, Alan.
 
N

nick12ab

Senior Member
Eclectica said:
Now, is it ok to repeatedly swap back and forth with the reference pins by issuing adcconfig commands without risking damage to the ADCs or any internal PIC circuitry?
Click to expand...
The PICAXE is not mechanical and won't wear out if you repeatedly do this.

Eclectica said:
I am building a dual thermometer circuit using precision NTC thermistors with parallel and series resistors (thus creating in effect a potential divider with low power dissipation through the NTC). Clearly using adcconfig to set Vref-, Vref+ to external pins with preset voltages will give me increased precision over a defined range.
Click to expand...
Before deciding to go to the effort of doing this, are you aware of the readadc10 command?
 
P

PADJ

Member
Eclectica said:
I am building a dual thermometer circuit using precision NTC thermistors with parallel and series resistors (thus creating in effect a potential divider with low power dissipation through the NTC).
Click to expand...
I have a similar application.
Can you share the circuit, and say what temperature range and accuracy it should provide?
 
E

Eclectica

Member
Thanks for the replies and reassurance!

Yes, I'm using readadc10, and hopefully with Vref+ and Vref- set correctly (which will give a span of around 3 volts), hopefully the 0-1023 10bit range will be available over this 3 volt range. ie. 3/1024 resolution as opposed to 5/1024 => 2.93mV vs 4.88mV per ADC10 increment and hence increased resolution which will help with my simple NTC parallel series resistor circuit.

It is probably best for me to try and find the Microchip datasheet section on ADC conversion on the PIC28x2 equivalent - agreed - there is always more to learn about these cool devices after all.

I hope to post a project containing copy of the full circuit, Pebble layout and PICaxe code in due course as I guess others might find this useful too. It is also going to provide 2 x 0-10v outputs (which will interface to our BMS system) and log a total of 5 such NTC temperature readings to an openLog device.

[It is especially useful to use the huge flexibilty and ease of programming of the PICaxe in this design which will investigate heating and cooling behaviour in our horticultural environment. Ultimately, following data collection and analysis, I'll combine the appropriate readings to provide 0-10V control signals which will be used to modulate a heating setpoint under a variety of conditions and in this way save money and maintain plant health.]

Thanks for the replies chaps.;)

Cheers,
Eclectica
 
You must log in or register to reply here.
Top