Post by thread:12 Bit ADC for LOAD CELL APP

I am prototyping a load cell application based on PIC14000. (internal single slope 16 bit A/D) Works well so far. I have not done the calibration routines yet, though... The temperature drift is significant, so even if y don«t need exact level I will have to recaculate values against values measured from internal bandgap reference. /Morgan At 19:25 1998-02-13 +0200, you wrote: {Quote hidden}>Hello , > >I am looking for a 12bit ADC for PIC based load cell application. >I am currently trying to use National's ADC12030 with PIC16C74. Please >tell me If you know cheaper solution for same resolution. > >Thanks. > >Kursat > > / Morgan Olsson, MORGANS REGLERTEKNIK, SE-277 35 KIVIK, Sweden \ \ spam_OUTmrtTakeThisOuTiname.com, ph: +46 (0)414 70741; fax +46 (0)414 70331 /

>I am prototyping a load cell application based on PIC14000. >(internal single slope 16 bit A/D) > Works well so far. >I have not done the calibration routines yet, though... >The temperature drift is significant, so even if y don«t need exact level I >will have to recaculate values against values measured from internal >bandgap reference. Can you use the internal bandgap reference as the bridge drive? That should make the measurement ratiometric, eliminating most of the temperature effects. Of course, that won't help if the drift is due to the tempco of the integration cap in the single slope converter. newell

At 17:36 1998-02-14 -0600, you wrote: >>I am prototyping a load cell application based on PIC14000. >>(internal single slope 16 bit A/D) >> Works well so far. >>I have not done the calibration routines yet, though... >>The temperature drift is significant, so even if y don«t need exact level I >>will have to recaculate values against values measured from internal >>bandgap reference. > >Can you use the internal bandgap reference as the bridge drive? That >should make the measurement ratiometric, eliminating most of the >temperature effects. No... Yes... It is possible to let the internal MUX output connect to a pin as well as the analog comparator, so you can select the reference and output to that pin, if you there have a cap it can hold the reference voltage while you are measuring another input. You have to buffer it and increase voltage, of course. Since I have a very stable power supply I use that directly to load cell. >Of course, that won't help if the drift is due to the tempco of the >integration cap in the single slope converter. Best integration cap is a polycarbonate. The temp drift is dependent of drift in internal current generator, and since I use internal oscillator I have that drift too. The best way is to use soft routine to calculate the voltage using the Reflo and Refhi voltages internally derived from bandgap. There are app notes on that. I have not yet implemented that. To optimize it further we can use a resistive divider to measure load cell supply, and divide the load cell output with that. /Morgan >newell / Morgan Olsson, MORGANS REGLERTEKNIK, SE-277 35 KIVIK, Sweden \ \ .....mrtKILLspam@spam@iname.com, ph: +46 (0)414 70741; fax +46 (0)414 70331 /

Did you consider using Analog Devices AD 7730 ? No hassle, includes excitation circuit, input amp, AD converter, programmable digital filter, serial interface to microprocessor..... and only for about U$ 10 in 1000 quantities..... you can't build it with discrete parts for that money.... I have some (free) samples I got when I was in Europe last month, now I'm working on a prototype for weight scale applications - but I think I'm going to use an Atmel AT89C52 micro (sorry, no PIC), together with a serial eeprom for storing calibration parameters, a MAX232 chip for serial communication with a PC, and a LCD display.... Albert A. Smulders InSAD - Encarnaci—n, Paraguay insadKILLspamitacom.com.py