Ads1115 noise issue


i have an issue with the ADS115 connected to a waterlevel-sensor. The ADS Input is between 0 and 2.2V, wich equals 0 and 100%. The input voltage is completely stable in every state.

In node red I get the the output values from the "ncd-ads1115" node. The problem is, that the values vary strongly (at the second decimal). I also tested the "ads1x15-raspi" node with the same issues.

Every ground is connectetd to one point, so i have "clean" potentials. Iam using node-red 1.2.9

maybe it´s a stupid mistake by myself, but i dont have any ídea to go on.



can you test with this node:


Danke :grinning:


thank you for your reply. i also tried this node, but it dont work. iam not sure how to handle the "redplc".

Yes redplc nodes writes the values to global memory on trigger.
use an inject node with string "input" as payload and connect to input of ads1115 node. The inject node can work in cycle of 0.1s.
The values can you view in global variable tab.
Or install node-red-contrib-redplc-module.
Connect ads1115 node to output of module-update node.
On module-in node you can get the values of ads1115.
Du kannst mir auch schreiben.

I tried it but i have the same issues with the values. They start to vary around the second decimal.

How did you measure the input voltage ?

i cheked it with a multimeter.

Habe mit einem Mulitmeter gemessen (Fluke). Der Input bliebt fest und springt nicht so hin und her wie der Ausgang.
Habe mittlerweile nur den Pi mit dem ADS1115 verbunden, und alle anderen Hardwarekomponenten entfernt - Ohne Erfolg.
Mein neuer Lösungsansatz ist, dass ich einfach einen Mittelwert aus x Messungen bilde, und den dann für meine Visualisierung verwende.

Die Benutzer werden hier sauer, wenn wir nicht in English schreiben. :rage:
Aber eine private Nachricht an mich kann deutsch sein.

The normal reference voltage on ads11115 is 2.048V.
The chip scales up the input value.
Can you select max input voltage 2.048V and repeat the the test.
If the input value is to high, use resistors to downscale.

1 Like

ok, no problem...i will give my best to write in english :slight_smile: i dont know how to send you a private message.

For the test i connect 3.3V to AI0 and 0V (GND) to AI1.
in the picture below you can see what i get from the function when 3.3V is connected. I think it looks good. but i think the range is not correct now.

Auf Benutzerbild klicken und Message Button klicken.
Dann kannst du eine private Nachricht schicken.

3.3V produce a overflow. Therefore node measure give 2048mV.
Use a potentiometer or 2 resistors to divide input voltage.

1 Like

I think the private message option has to be "earned". Iam a new member, thats because i can´t send you an private message.

i will see if i can find some resistors to devide the voltage of the input.

I finally made it. i built a voltage devider out of a resistor and a poti.
the value now remains largely stable, which is completely sufficient for my purpose. Thank you for your help. :+1:

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.