All,

I have a MPC-20 (date code K4, REV-E) flashed with the MSTD configuration bundled with PowerVision 2.8.10366 and I have noticed the following,

The MPC-20 analog inputs, when set to 0-5VDC are seemingly reading off a bit, the analog inputs track fine when supplied a 4-20mA signal. The 0-5VDC inputs seem to track to the low side when the raw count is converted using the calibration set-points and linear interpolation.


Test replication:

The MPC-20 is on a programmable regulated 24VDC power supply. set to 24VDC. There is a common ground between the power supply and the test equipment.

Set Analog Input 7 to "0-5V Oil Pressure". Leave the calibration settings at 0.0 VDC = 0 PSI (0 kPa), 5.0 VDC = 200 PSI (1378.95145863 kPa).

Looking at the front display (page the MPC-20 goes to when booted up)

VDC vs Oil Pressure (my test results)

0.0V 0 PSI
0.5V 19 PSI
1.0V 39 PSI
1.5V 59 PSI
2.0V 79 PSI
2.5V 98 PSI
3.0V 118 PSI
3.5V 138 PSI
4.0V 158 PSI
4.5V 178 PSI
5.0V 197 PSI
5.12V 198 PSI
6.00V 198 PSI

The process calibrators we are using to supply the signals are accurate/precise within +/- 0.004 VDC at this particular voltage band.


I have tried two different MPC-20's, and for each MPC-20 I used two different Fluke 87 III multimeters with different lead probes (with cal stickers current), and two different oscilloscopes with two different sets of lead probes to measure the voltage being supplied to the MPC-20 analog input to see if the process calibrators were off and needed calibration. Per each MPC-20 I also tried using two different process calibration units to supply the 0-5VDC signal to the MPC-20, as well. I also tried a 10k Ohm, 10 turn linear potentiometer supplied with 5VDC from a regulated power supply to see if I was running into an impedance issue with the analog input, it did not seem to make a difference.


So the questions I am looking for an answer on are as follows.

1. Is there a tolerance spec on the MPC-20 0-5VDC analog inputs?

If so, if FWM can supply that information it would be appreciated.

2. One thing I noticed, I cannot seem to get the digital gauge to ever read 200 PSI, in PowerVision I look under the Programming Tab -> Analog Calibration -> 0-5V Oil Pressure Y2 and it's factory defaulted for 1378.95145863 kPa ~ 200 PSI. I also checked the Smart Text Gauge on this page and it's max is set to 200 PSI.

Any indication as to what might be happening, and a suggested solution, would be appreciated. I have thought of editing the Library -> Pre Defined Curves -> 0-5V Oil Pressure -> Expression from,

CurrentIOValue * ((_0Minus5V_Oil_Pressure_Y2 )/(5-_0Minus5V_Oil_Pressure_X1)) + (0 - (((_0Minus5V_Oil_Pressure_Y2)/(5-_0Minus5V_Oil_Pressure_X1) ) * _0Minus5V_Oil_Pressure_X1))

to,

(CurrentIOValue + (CurrentIOValue * 0.01)) * ((_0Minus5V_Oil_Pressure_Y2 )/(5-_0Minus5V_Oil_Pressure_X1)) + (0 - (((_0Minus5V_Oil_Pressure_Y2)/(5-_0Minus5V_Oil_Pressure_X1) ) * _0Minus5V_Oil_Pressure_X1))

Or something similar, but it doesn't make theoretical sense to me to do so (I am guessing CurrentIOValue should vary from 0-1023 counts, 10 bit analog input), so I would prefer not to do that and get to the root cause.

Trying to figure out if it's me, or if something is actually off.

Let me know.

Hello Ocelot,

We have found that most of the inaccuracy displayed is due to the text gauge rounding and conversion from the base unit, kPa, to PSI. The values used in calculations and used to determine warnings/shutdowns, start and stop control are more accurate than what is displayed.

If you want to see the actual input reading, you can modify the config to show the readings.

In the Pre Defined Curves > 0-5V Oil Pressure,

• change the Polynomial Order to First Order
• coefficient for x = 1
• remove the Expression

I would recommend changing the Oil Pressure smart text gauge to show at least 2 decimals and edit the Smart Gauge for all of the items in the list to have at least 2 decimals, more if you want to be more accurate.

Set one of your analog inputs to use the 0-5V oil pressure and Pressure Units to kPa.

Maybe this will help you see a more accurate reading.

Stalley,

Calculation rounding vs. precision, that makes sense.

I will try a couple of things with the smart gauges and see how it responds.

Thanks.