A question about inaccurate sampling with thermistor (involving WIFI module)

Sub forums for various specialist XMOS applications. e.g. USB audio, motor control and robotics.
New User
Posts: 3
Joined: Wed Nov 28, 2018 4:24 am

A question about inaccurate sampling with thermistor (involving WIFI module)

Post by 928928 »

A temperature measuring device, with WIFI module, its principle is that the power supply board supplies 5V. After LDO, it will supply 3.3V to WIFI module and system. The sampling is done with a 1.2v voltage regulator, and use precision 10K and thermistor to divide and value.
The phenomenon is: If the WIFI module is disconnected, the temperature sampled by the AD is very accurate, but the WIFI module is powered on, the sampling temperature will be high, and the rise is not a little bit higher, rising to the ratio. The normal temperature is stable at around 7 degrees.
Possible reasons
The first one: program error, AD sampling is not allowed, I take the WIFI module, the device with high temperature, replace the thermistor with a fixed resistance value, the thermistor is 25 °C under the table 10K, so I took it. Precision 10K replaced the thermistor, the result is 25 ° C, which means that there is no problem with AD sampling and procedures, then how to explain this phenomenon, the power supply and WIFI module can not be high 7 degrees, using a multimeter to measure the high temperature of heat Sensitive resistance, it is indeed the resistance of the high temperature, is my experiment has a mistake or in addition to the temperature can affect the thermistor and other can affect the thermistor, please enlighten me
The second one: I suspect that the LDO drive capability is not enough. The LDO I chose is 500mA. The actual power consumption of the whole board is only 260mA. The peak value is about 350. I use the regulator source to output the 3.3V system and the WIFI power supply at the back end of the LDO. The temperature is normal, but I use the voltage regulator to deliver 5V at the front end of the LDO. After the LDO is depressurized to 3.3, the temperature rises.
The third one: cut off the power of the WIFI module, the temperature is normal immediately, the power supply temperature of the WIFI module is immediately increased, using AD sampling.
The connection between the above three points makes me not very clear. I found a lot of passages about[url]thermistorhttp://www.apogeeweb.net/article/171.html[/url] in the Internet, but I really don't know where it is. The main reason is that I don't understand the first point. I feel that the experiment is loophole and I don't know where the vulnerability is.
So here, if you know the answers, could you please share them to me?
Thanks a lot.
Attach the temperature sampling chart:

User avatar
Respected Member
Posts: 286
Joined: Wed May 31, 2017 6:55 pm

Post by CousinItt »

Hi, you haven't said where your thermistor is located. If it's off the board and connected by wires, it could be interference from the wifi module, but I don't think this is particularly likely. If your thermistor is on your board, it's probably being warmed up by the heat being produced locally - about 1.3 W to 1.75 W. It will heat up more when your wifi module is in use. Have you tried thermally isolating the thermistor - e.g. putting it on thin wires a good distance from other components?