ADC input using a current transformer?

[jessey]

I'm using a current transformer (one that I had kicking around) with a 600 to 5 ratio that is suppose to be good for up to 25 amps. The motor I'm using it on (under no load) uses approximately 1/4 of an amp and produces about 1.5 vac. I connected the output of the current transformer to a 1.5 Amp 1000 Volt bridge rectifier and I get a voltage of about 1.15 dc. I have it connected to an adc input on my 16f688 with a .1 mf cap to ground. When I power up the circuit the LCD shows the number very gradually creeping up to about 162 then drops very slowly to about 140, it's not consistent at all and sometimes goes up to over 200. I'm using ADCIN with the code below:



'PIC Config Fuse Definitions for 16f688
@ __config _INTRC_OSC_NOCLKOUT & _WDT_ON & _MCLRE_OFF & _CP_OFF & _BOD_ON

ANSEL = %00001000 ' enable analog AN3 (pin 3), the rest as digital
ADCON0 = %0000001 ' VREF to VDD & left justify & sets converter On
ADCON1 = %00100000 ' set conversion clock FOSC/32
CMCON0 = 7 ' turn off comparators
OSCCON.3 = 1 ' Use internal clock
OSCCON = %1110111 ' Set to use the internal oscillator
DEFINE OSC 8 ' Without this define the 688 will defalt to 4 Mhz

ADCIN 3,Ugo_Motor
LCDOut $fe, 1,"Ugo Motor = ",DEC Ugo_Motor


I've never used a current transformer before so maybe I'm using the wrong one for my small motor? I've been using the same code on another 16f688 ADC with the same settings as above and it's been working good for quite some time now. Would anyone have any suggestions that might help solve my problem?

Thanks
jessey

[mister_e]

High likely you'll need at least to load your current transformer, do a search for Burden Resistor, there's a load of discussion/appnote all over the web.

http://www.coilwinder.com/current_transformer_design_and_theory.htm
http://www.datatronics.com/pdf/current_sense_article.pdf
http://powerelectronics.com/mag/power_exploring_current_transformer/

Probably not a bad idea to buffer the signal with an op-amp as well.

Not sure of the defaut DEFINE parameter for ADCIN, but you might need to include them to reach better results, or not use ADCIN and set/read PIC regsiters manually.... which is the method i prefer, but it's me ;)

HTH

[Tom@Nottingham]

Hi

If you are using a 600/5 amp ct, this is intended to measure up to 600amps?
The output then usually passes through moving iron panel meters and heaters/relays to protect large a.c motors etc. This works fine as the meters are passing good sine wave currents and the heaters work the same no matter what the wave-form looks like. I assume you have wound many turns of your motor through the centre of your ct.

I have been trying to measure rms current using a ct but with limited success.
Since a dc output is required, the bridge I made was from 4 IN4148 diodes as these have low forward voltage of 200mv? The output from the bridge was then fed through a resistor(variable 0-100k) to a/d input and any capacitor 1 to 100mfd together with a resistor to ground etc .

This should work fine providing you are using constant smooth sine wave currents. If you have any noise, sudden load changes or are thinking of using a triac/thristor to vary fan speed etc. then strange readings appear. This is due to the bridge rectifier measuring peak volts not root mean square. Induced voltage in the secondary of a transformer is proportional to the rate
of change of current in the primary. It is not easy to measure accurately a.c current, unless using hall effect sensors and and integrating wave-form etc.
This should I think also be possible with ct output by biasing 1 output with 2 resistors to mid-point of a/d input and other output to actual a/d input,
or 2 a/d inputs (like op amp differential inputs) one for each ct output. Suitable load resistors/capacitor/ zenner diodes required. Then busy job for Pic converting to rms? or may not work? not sure.

Hope this helps

Tom