Why are the measured ADC values lower than expected in F28379D Controller Card?

, , ,
1

Hi,

i’m using the RT Box 1 to make some HIL tests with my F28379D Boards.

First I tested my setup with the F28379D Launch Pad together with the LaunchPad Interface and everything works fine. -> I am giving you this information because it can be ruled out that it is a software problem.Then i changed the setup to the F28379D controller card with the RT-Box’s ControllerCard Interface. I use the power supply from the Interface and have a Data-Connection via USB between my Laptop and the controller card (mini USB not micro USB).

The problem now is that the readings from the ADCs are too low. In 12 bit mode 3.3V means 4095bit but i’m only measuring about 2935bit. If i give 3.3/2 V to the controller, i can measure 1466bit - not 4095/2.

I think this could be a problem in the power supply but I haven’t been able to come up with a solution so far. Maybe some jumpers on the controller card wich i have to open or to short…?

Is there anyone how can help me with this? Thanks!

Ben

2

Hi Ben,

The 180-pin ControlCard Interface Board actually has a signal conditioning circuit between the RT Box Analog Outputs and the ADC inputs. The DC gain of the circuit is 0.65, which means your observations are a bit more reasonable. For the LaunchPad Interface Board there is not an equivalent signal conditioning circuit.

The circuit is shown in the Control Card Interface Board Manual, and I’ve copied the relevant section below.

61
61600×509 80.5 KB

Regards,

Bryan

3

Hi Bryan,

many thanks for your your answer. To be honest, i’ve read this line in the Manual but i didnt understand that because i can measure the right value via the BNC connectors with an oscilloscope.

Does the gain mean, that i have to multiply my output signal with 1/gain? To be more precisely: If i would like to have a value of 1V on my ADC Pin, should i send a value of 1/0.65 = 1.538… out of my box?

Thanks a lot!

Ben

4

I’ve tested the above-mentiioned presumption.I create a DC-Signal of 1.65 V (half of the max. voltage of the controller) and multiplied that before sending it from the RT box to my controller with a value of 1 / 0.65.

271
271600×205 23.4 KB

The result is that I can measure a value of 2260 bits in CCS -> that is a little more than half of 4095 bits (it should be 2048 bits).So, i think i did not understand what i have to manipulate to get the exact value of the voltage into my ADCs.Could you be a little more precisly please?Thanks Ben

5

You might be confused on which voltage reference you are using for the ADC. The Vref can be 3.3V or 3.0V depending on the SW2 position. Your results seem correct when considering a 3.0V voltage reference. For example, 1.65/3.0*4095 = 2252. There’s still a very small error (0.3%) that you can calibrate out.

> i can measure the right value via the BNC connectors with an oscilloscope.

The BNC connections are directly connector to the RT Box analog output. The op-amp circuit is connected between the RT Box analog output and the ADC pins.

6

Perfekt! I will test it but it seems very obvious that this is the point.

Thanks Bryan