DIY EC sensor

[Ranjib]

Yes, I believe both ADS1015 and ADS1115 are very reliable in this regard. They are much slower (~1kHz compared to 200-500 kHz for microcontrollers) but more accurate. The ADS1015 has 12 bits, e.g. about 1mV resolution in the largest range, and only about 1-2mV error, whereas the Pico has up to 8mV in certain places and the ESP32 has 140mV.

I have read that the ESP32‘s problem mainly arises in the full range, so if one uses a lower range and simply steps down the voltage oneself with a few resistors, this could become much better and can replace an ADS1015 for our purposes I guess.

That is awesome to know. i would always prefer an esp32 based solution , because it’s comparatively cheaper and simpler to build over ads115 . As long as it fits the reef keeping use cases, we are gold

 

[alex12eqw]

Hi Sral,
when searching for a cheap ec measurement alternative i just came across your thread. It sounds very cool and reliable. I am wondering if you have somewhere uploaded the final ciruit, so that i can build it myself. I am not really into electronics therefore it would help me alot:)

 

[Sral]

Phew, I haven't gone beyond a breadboard until now and haven't worked on this version for a long time. I could probably create a design that's simple enough and publish the code. At least some day, when I can find the time :loudly-crying-face:

I was working on a ready-to-buy version that's currently in the final polishing phases.

 

[Sral]

Alright, the ready-to-buy version seems to be up and running at robo-tank.ca
Though rob is currently unobtainium on R2R ^^

I'll work on getting some info on here and publish to Github.

 

[Sral]

Okay, looking back here now and then.

Seems like I never put any concrete circuit diagram on here to show the two versions that I have settled on. That will be the next step.

 

[Sral]

Lurking around sometimes ...

Issue is that one version is basically the RoboTank EC and I'm currently not too comfortable to publish that just like that.
The other one was never fully developed and tested, especially in a real tank. So I am not perfectly sure about the makeshift isolation working as intended.

 

[BeanAnimal]

Thank you for all of the work and open sharing of the process of problem solving.

 

[Sral]

So for explanation, I had this circuit diagram earlier:

Which worked well enough but uses loads of parts.
Improvement idea was to just use 1 capacitor and get the same AC coupling / DC isolation:

Which would work the same.

The issue that remains is this works fine in an isolates glass of test water, but in a real tank one has stray currents. Since the leads of R_TEST will basically end up being the two electrodes in the tank water, this will couple non-isolated VCC or GND lines to the tank voltage, which will probably lead to very undesired effects, measurement errors, I2C comms failing, etc.

I'd thus probably simply go for the simplest isolation circuit I can find and just slap it on, like Atlas Scientific recommends:

I'll see if I can replace the ADM3260, since that seems to cost a load of money by itself (10$).

Using isolation like that will make the whole system more resilient and hopefully somewhat simpler in design.

 

[Sral]

That one has very little power delivery though (about 20-30 mA @ 3.3V)
So it might be both better and cheaper to use two components for this, like:
- RFM 0505
- ISO 1540 DR

The RFM has up to 200mA delivery, which should be fine for an ESP type chip if one is careful.

Though that would require through hole manufacturing, which might be more expensive again ... will have to think about it.

 

[Sral]

Finished the example circuit I started for the Robo-Tank sensor quite some time ago: