ADC read error RM48

Hi,
Can you try to increase the sampling time and try to see if it makes a difference? Currently your sampling time is 200ns. Try to increase it to 1000ns.

Hi, Mr.Charles,

Thank you for the suggestion. I will try and update the result.

regards

K.Manjunathan

hi, Mr.charles,

I have increase the sampling time to 1000ns and checked the output but the problem is not solved. The problem remains same as the sensor read outputs are not correct.

please suggest some other ideas

Thanks and regards
K.Manjunathan

Hi,
Can you elaborate a bit of what values are read? Can you show the 10 values before you disconnect one of the sensors and the 10 values after you disconnect the sensor. How much are they off?

Hi, Mr.charles.

I have attached the problem description in  a word document. 

Please check the document and if you need even more clear document I will again give you in a more elaborate document.5758.post.docx

Hi,
In your scope shot, the flow is an input to the MCU, right? Please confirm. If the flow is a input to the MCU then it must be above 0 volts. I don't understand your scale. The Y-axis shows from -16 to 16. What is the unit? Also in your second figure the scale is from -8 to 8.

Please check the actual ADC result memory and show what is the difference between the before and after. What if you instead of disconnecting one of the sensors and tie the input to either 3.3V or 0V?

Hi, Mr.Charles

I need sometimes I will clearly document it and send you within two days.

Hi, Mr.charles,

The screen shot i sent earlier was my instruments display (designed by us). Please don't consider the graph it is not fully coded yet but please consider the readings in the both the images.
In bottom u can see set parameters (input to MCU - for example take vt 240 ml -input value to MCU) and in the right side u can see monitored parameters (sensor output converted into corresponding real world values from the relevant datasheets - for eg please check vt 234 ml in vertical blocks) both input and output are almost same.

The figure 1 shows the correct output without any error. (Eg . set vt 240 ml in bottom scale and observed vt in vertical scale 234 ml) both are almost same there is negligible amount of error that's ok for me

the figure 2 shows the wrong reading when fio2 sensor is disconnected (fio2 in figure 1 is 24% - correct reading and figure 2 is 81% - wrong reading ). the fio2 sensor value is fixed in my instrument and it is 24% .

consider figure 2:
I have disconnected fio2 sensor its shows wrong reading fio2 81% and also affects vt it is 168 ml while set is 240. it affects the vt value also.

If it is not ok i will take few more screen shots and make a document for your clear understanding. If this explanation is not ok. please tell i will do it again Mr.charles.

Hi,
Which ADC input is connected to FIO2? Do you have the identical code running when you disconnect FIO2? If the ADC input which is connected to the FIO2 is tied to 3.3V or 0V does that affect the result differently.

A simple block diagram will help me. For example, you said there are 10 sensors connected to the ADC, right? One of the sensors is the FIO2, correct? You simply disconnect the FIO2 to the ADC and you see all 10 inputs with wrong readings. Is that the correct understanding?

HI,Charles

Which ADC input is connected to FIO2? Ans: channel no 9. (if there is a change in any change in voltage of channel 9 it is affecting channel 0. and if I connect the channel 9 to the ground then channel 0 is working fine.

  Do you have the identical code running when you disconnect FIO2?  Ans: yes.

If the ADC input which is connected to the FIO2 is tied to 3.3V or 0V does that affect the result differently. Ans:  I have tied the channel 9 to ground other ADC channels are working fine.

you said there are 10 sensors connected to the ADC, right? One of the sensors is the FIO2, correct? You simply disconnect the FIO2 to the ADC and you see all 10 inputs with wrong readings. Is that the correct understanding?

Ans: Yes not all the channels mostly if i disconnect channel 9. channel 0 and channel 8 are getting affected. other channels are working fine.

Sensor read functions : coding:

uint32 ADChex(int i){
gioSetBit(gioPORTB, 0, 1);
while((adcIsConversionComplete(adcREG1,adcGROUP1))==0);
ch_count = adcGetData(adcREG1, adcGROUP1,&adc_data[0]);
ch_count = ch_count;
gioSetBit(gioPORTB, 0, 0);
return adc_data[i].value;
}

void airpr()
{
ad = ADChex(4);
adc2dec = (ad/0x100)*256 + ((ad/0x10)%0x10)*16 + (ad%0x10)*1 ;
HP_Air= (1.5608*adc2dec)/819 - 1.2835;
HP_Air = HP_Air*100;

}

void o2pr(){
ad = ADChex(5);
adc2dec = (ad/0x100)*256 + ((ad/0x10)%0x10)*16 + (ad%0x10)*1 ;
Hp_O2= (1.5608*adc2dec)/819 - 1.2835;
Hp_O2 = Hp_O2*100;

}

float airflow(){ //S/N: tsi840523 - 1219052; //Updated;

float FS_ATMP;

float a,b,c,FS_AFLW,Vf_std;
int t_over;

ad1 = ADChex(1);

adc2dec = (ad1/0x100)*256 + ((ad1/0x10)%0x10)*16 + (ad1%0x10)*1 ;

FS_ATMP=(adc2dec*1000)/819;
FS_ATMP=FS_ATMP/1000;

Ta = ((-8.0295*FS_ATMP*FS_ATMP*FS_ATMP*FS_ATMP*FS_ATMP)
+(53.128*FS_ATMP*FS_ATMP*FS_ATMP*FS_ATMP)-(144.94*FS_ATMP*FS_ATMP*FS_ATMP)
+(205.1*FS_ATMP*FS_ATMP)-(185.02*FS_ATMP)+113.71+0.2);
Ta = Ta -0.5620628;

ad2 = ADChex(0);

adc2dec = (ad2/0x100)*256 + ((ad2/0x10)%0x10)*16 + (ad2%0x10)*1 ;

FS_AFLW=(adc2dec*1000)/819;
FS_AFLW=FS_AFLW/1000;

if(Ta>=21.11)
t_over=174;
else
t_over=190;

Vf_std=(FS_AFLW+0.53790998)*(pow((t_over-21.11),0.5) / pow((t_over-Ta),0.5))-0.53790;

if(Vf_std<0.25868)
{
a=-0.399304;
b= 5.215665;
c=43.420101;
}
else if(Vf_std<0.41288)
{
a=-0.53508;
b=6.92152;
c=30.5;
}
else if(Vf_std<0.58846)
{
a=-1.42355;
b=11.95656;
c= 18.38220;
}
else if(Vf_std<0.75597)
{
a=-2.82609;
b=16.49881;
c=13.54905;
}
else if(Vf_std<0.99178)
{
a=-4.32834;
b=19.67331;
c=11.86052;
}
else if(Vf_std<1.17879)
{
a=-15.24258;
b=38.45006;
c=5.19245;
}
else if(Vf_std<1.49696)
{
a=-25.08969;
b=47.14514;
c=3.91037;
}
else if(Vf_std<1.73973)
{
a=-86.78648;
b=74.69264;
c=2.54999;
}
else if(Vf_std<2.01587)
{
a=-86.78648;
b=74.69264;
c=2.54999;
}
else{}

Qa = a+b*Vf_std*Vf_std+c*Vf_std*Vf_std*Vf_std*Vf_std*Vf_std;
//Qa=0;
return Qa;
}

float o2flow(){ //S/N: tsi840523 - 1219057

float FS_OTMP;

float a,b,c,FS_OFLW,Vf_std;
int t_over;

ad1 = ADChex(3);

adc2dec = (ad1/0x100)*256 + ((ad1/0x10)%0x10)*16 + (ad1%0x10)*1 ;

FS_OTMP=(adc2dec*1000)/819;
FS_OTMP=FS_OTMP/1000;

To = ((-8.0295*FS_OTMP*FS_OTMP*FS_OTMP*FS_OTMP*FS_OTMP)
+(53.128*FS_OTMP*FS_OTMP*FS_OTMP*FS_OTMP)-(144.94*FS_OTMP*FS_OTMP*FS_OTMP)
+(205.1*FS_OTMP*FS_OTMP)-(185.02*FS_OTMP)+113.71+0.2);
To = To -0.65631;

ad2 = ADChex(2);

adc2dec = (ad2/0x100)*256 + ((ad2/0x10)%0x10)*16 + (ad2%0x10)*1 ;
FS_OFLW=(adc2dec*1000)/819;
FS_OFLW=FS_OFLW/1000;

if(To>=21.11)
t_over=174;
else
t_over=190;

Vf_std=(FS_OFLW+0.5379)*(pow((t_over-21.11),0.5) / pow((t_over-To),0.5))-0.5379;

if(Vf_std<0.25822)
{
a=-0.39532;
b=5.18076;
c=43.4201;
}
else if(Vf_std<0.41298)
{
a=-0.53274;
b=6.89655;
c=30.5;
}
else if(Vf_std<0.58855)
{
a=-1.39784;
b=11.86447;
c=18.38220;
}
else if(Vf_std<0.75635)
{
a=-3.05654;
b=16.8552;
c=13.54905;
}
else if(Vf_std<0.99039)
{
a=-4.37619;
b=19.84090;
c=11.86052;
}
else if(Vf_std<1.17898)
{
a=-14.58173;
b=38.110671;
c=5.19245;
}
else if(Vf_std<1.49598)
{
a=-26.46435;
b=47.71259;
c=3.91037;
}
else if(Vf_std<1.73836)
{
a=-84.49235;
b=74.06126;
c=2.54999;
}
else if(Vf_std<2.01595)
{
a=-84.49235;
b=74.06126;
c=2.54999;
}
else{}

Qo = a+b*Vf_std*Vf_std+c*Vf_std*Vf_std*Vf_std*Vf_std*Vf_std;
return Qo;
}

float Paw()

{

ad = ADChex(7);
adc2dec = (ad/0x100)*256 + ((ad/0x10)%0x10)*16 + (ad%0x10)*1 ;
LP_AW=(adc2dec*1000)/819;
LP_AW=LP_AW/1000;
LP_AW =((-18.903*LP_AW*LP_AW*LP_AW*LP_AW)+(232.19*LP_AW*LP_AW*LP_AW)
-(1063.6*LP_AW*LP_AW)+(2266.1*LP_AW)-1913.4);
return LP_AW;
}

float Bellow()
{
ad = ADChex(8);
adc2dec = (ad/0x100)*256 + ((ad/0x10)%0x10)*16 + (ad%0x10)*1 ;
LP_BW=(adc2dec*1000)/819;
LP_BW=LP_BW/1000;
/*LP_BW =((-18.903*LP_BW*LP_BW*LP_BW*LP_BW)+(232.19*LP_BW*LP_BW*LP_BW)
-(1063.6*LP_BW*LP_BW)+(2266.1*LP_BW)-1913.4);*/
//LP_BW = (LP_BW - 0.5*5/5*0.018)+4;
LP_BW = (((111.11*LP_BW) - 277.78)-3);

return LP_BW;
}

float fiO2(){
ad = ADChex(9);
adc2dec = (ad/0x100)*256 + ((ad/0x10)%0x10)*16 + (ad%0x10)*1 ;
O2con=(adc2dec*1000)/819;
O2con=O2con/1000;
O2con = (22.012*O2con)-0.1532;
return O2con;
}

float vtee(){
ad = ADChex(6);
adc2dec = (ad/0x100)*256 + ((ad/0x10)%0x10)*16 + (ad%0x10)*1 ;
vte=(adc2dec*1000)/819;
vte=vte/1000;
vte = -(23.352*vte*vte) +(214.7*vte) -362.83;
return vte;
}

halcogen configuration already attached: 

Simple block diagram:

Hi,Mr.charles,

Yes, i am using customized board. Yes, this might be the reason for the errors. I will try all the comments and update the outcomes. Thank you for the support you have provided so far.

Hi, Mr.charles ,

This EMI and EMC problem are creating lot of problem with the results in my customized board. i am checking with ti HDK