TC334 ADC sampling the floating pin

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
Yubo
Level 1
Level 1
5 replies posted 5 sign-ins First like given

Hello,

I found a strange problem when testing the TC334, using the ADC routines in AURIX, the port would have a voltage of 0.3V when floating, and when I enabled adcGroupConfig.disablePostCalibration in the ADC configuration the port floating voltage changed to about 1.63V, What causes this and does this affect normal sampling?

Thank you in advance.

0 Likes
1 Solution
Kunqiao_L
Moderator
Moderator
Moderator
First comment on blog First like given 25 likes received

Hi,

Again when floating pin happens it is actual a high impedance state, the ADC internal circuit may reversely affect this pin. 

For the internal ADC hardware design question, it will take some process time to get answer from the hardware team. So please wait for a while. Thanks. 

View solution in original post

10 Replies
Kunqiao_L
Moderator
Moderator
Moderator
First comment on blog First like given 25 likes received

Hi Yubo, thanks for your inquiry. In order to help you get question solved, please provide me more information about this issue.

First of all, could you tell me which code you are testing on? which example code?

Secondly, which board and Analog port you are choosing?

Thirdly, please be aware that the pin state is whatever voltage you are feeding into that pin. If you are feeding in 3V then the pin reads the binary equivalent of 3V. If you feed in 2V then the pin reads the binary equivalent of 2V.

If you don't connect anything to the pin then it is floating. It varies time to time.

 

0 Likes
Yubo
Level 1
Level 1
5 replies posted 5 sign-ins First like given

Hi Kunqiao,

Thanks for your response!

1.Now I'm using  ADC_Single_Channel_1_KIT_TC334_LK example code to test the ADC module

2.The hardware is TriBoard-TC334-5V-with-socket, I changed the channel to AN7, and all other parameters are default, in this case, I observed on the oscilloscope that the voltage at this port is about 0.3V, and when I enabled adcGroupConfig.disablePostCalibration it changed to about 1.63V.

3.No other known voltages were connected to the port during the entire process, only an oscilloscope probe.

Because this measured voltage is still relatively stable so I don't know why a change in the configuration would cause this floating voltage and whether it is normal?

The below picture shows the Voltage in both cases:

022.bmp

0 Likes
Kunqiao_L
Moderator
Moderator
Moderator
First comment on blog First like given 25 likes received

Hi,

Have you tried the original AN0 input with and without enabling adcGroupConfig.disablePostCalibration? Do you see the same issue or not?

Also, have you connect anything on AN7 except oscilloscope probe?

And what is the ADC actual result and expected result on AN7? What is the result difference between with Post-calibration enable and disable on the aspect of the digital conversion result or error.

 

0 Likes
Yubo
Level 1
Level 1
5 replies posted 5 sign-ins First like given

Hello kunqiao,

I tested the AN0 port, and it's the same thing as the AN7, except the port levels are slightly different 0.36V and 1.5V(enable disablePostCalibration) respectively. I didn't have anything connected except the oscilloscope probes, in AN0, the values read from the software is about 280 and 1120 in 5V voltage base when connecting an oscilloscope, when I removed the oscilloscope the AD value is 314 and 1895, I can't figure out why.

0 Likes
Kunqiao_L
Moderator
Moderator
Moderator
First comment on blog First like given 25 likes received

Hi,

What I mean is that you connect AN0 or AN7 to a analog circuit as input, let's say 3.3 volt. And then check the actual ADC conversion result on both enable/disable post calibration are correct or not. Let's make sure the ADC functionality is no problem first.

Both conditions you mentioned above are your input pin on floating state. And you already saw that by connecting an oscilloscope (it is actually also a circuit loop), the ADC analog/ digital value are affected. The post calibration could also affect on this floating pin, this might related to the hardware design and I will double check with hardware team.

0 Likes
Yubo
Level 1
Level 1
5 replies posted 5 sign-ins First like given

Hi,

Yes, when I connect it to a certain input it works fine in both situations, the value can match the input, I'm just wondering if this measured voltage is normal when floating, if you could confirm I'd very appreciate it!

Thank you!

0 Likes
Kunqiao_L
Moderator
Moderator
Moderator
First comment on blog First like given 25 likes received

Hi,

Again when floating pin happens it is actual a high impedance state, the ADC internal circuit may reversely affect this pin. 

For the internal ADC hardware design question, it will take some process time to get answer from the hardware team. So please wait for a while. Thanks. 

Kunqiao_L
Moderator
Moderator
Moderator
First comment on blog First like given 25 likes received

Hi Yubo,

Some updates:

We confirmed and also observed similar levels to what the you reported, 0.45V and 1.5V with and without post-calibration enabled.

The basic principle is at the end of the startup calibration process the internal sampling capacitor is preconditioned to a level of ~Varef.  Due to the discharge of a sampling capacitor during conversion a lower residual voltage is read by ADC. The post-calibration process performs 1 step (out of a total calibration 256 steps procedure) after each conversion and the residual level might be different. For further investigation, I have to align with the R&D team to get more details on the post-calibration effect on a sampling capacitor.

Please note that It is generally recommended not to leave analog pins floating and connect them through a resistor to the ground. Please also note that post-calibration is not necessary in TC3xx applications.

 

IFX_likunqiao_1-1693361296610.png

 

 

0 Likes
Yubo
Level 1
Level 1
5 replies posted 5 sign-ins First like given

Hello Kunqiao,

This has helped me a lot, and I will look more closely at the information, thanks again!

0 Likes
Kunqiao_L
Moderator
Moderator
Moderator
First comment on blog First like given 25 likes received

Hi Yubo,

Since the internal circuit can't be exposed to the public, but the principle is like this:

with the post-calibration function enabled, the residual charge value is affected due to the internal algorithm that connects an extra correction C-network in parallel to the sampling Cap. 

This is confirmed what I mentioned in the solution, some internal circuit did reversely affect the float pin

0 Likes