On 2025/02/21 at 6:47 PM, stormjp said:

Hi Muttley

 

Thank you for this - Claude and I are still struggling with each other, maybe because of my very Afrikaans-like English.

 

I've gone through the schematics and cannot figure out the reason for DC-DC voltage regulator, as both the INA219 and the DF Robot can accept 24V input? What voltage setting do you use on the 'out' on the DC-DC voltage regulator then?

The 24V power supply I got have two DC outputs, so I'll use a buck converter to power the ESP32 from there, but can I not connect the INA219 and DF Robot directly to the 24V power supply?

I'm a little confused

And also, can you cut the cable and tube on the DF Robot shorter, or should it remain at the full 5m length?

Hey...

Welcome to the world of tinkering I guess

I use the DC-DC voltage regulator to smooth out any potential voltage jumps - you could probably get away with not using one and connecting the INA and sensor directly but I do think it helps smooth out the data - this is what was also recommended on the home assistant community forum. 

The Voltage regullator is putting out 24V - perfect for both the INA219 (Max 26V) and the throw in sensor (12 - 36V, so 24V is the sweet spot I think )

I would deff use a buck converter to power the ESP32.
From what I've read, you can certainly cut the cable, it shouldn't make any difference. 

I hope this helps
 

Nice one.You guys are way too technical for me.I just smack the side of my 5000l jojo and I can tell where the level is but of course not as accurately as you .

1 hour ago, Nicholas Strachan said:

Nice one.You guys are way too technical for me.I just smack the side of my 5000l jojo and I can tell where the level is but of course not as accurately as you .

The problem with your technique is that you have to smack yourself against the head if the level control doesn't work.

12 hours ago, Muttley said:

Hey...

Welcome to the world of tinkering I guess

I use the DC-DC voltage regulator to smooth out any potential voltage jumps - you could probably get away with not using one and connecting the INA and sensor directly but I do think it helps smooth out the data - this is what was also recommended on the home assistant community forum. 

The Voltage regullator is putting out 24V - perfect for both the INA219 (Max 26V) and the throw in sensor (12 - 36V, so 24V is the sweet spot I think )

I would deff use a buck converter to power the ESP32.
From what I've read, you can certainly cut the cable, it shouldn't make any difference. 

I hope this helps
 

Awesome, I'll have to resolder a bit and hopefully we're good to go then.

I'll wire it all like yours and just use an additional buck converter down to 5V for the ESP32

The trusty 3D printer is busy doing its thing for an enclosure, so once it is all done, Ill give an update.

Edited February 23, 20251 yr by stormjp

Whilst the 3D Printer is printing away, I managed to fry an S2 Mini and the only DC-DC converter I have. I honestly do not know how the dc-dc got fried, but I ordered some new ones and will update once it arrives and everything is up and running. 

1 hour ago, stormjp said:

Whilst the 3D Printer is printing away, I managed to fry an S2 Mini and the only DC-DC converter I have. I honestly do not know how the dc-dc got fried, but I ordered some new ones and will update once it arrives and everything is up and running. 

That looks awesome... sorry about the fried components!

I just saw that I didn't reply to your message on the home assistant forum - sorry about that 🤓

Here's the link to the original home assistant forum on this topic: https://community.home-assistant.io/t/esphome-water-level-sensor/126504/582

Edited February 23, 20251 yr by Muttley

Okay, so I think I am winning.

Everything is installed and working flawlessly. I have noticed that the installer of my tank system was very very conservative with the float and when the float stops the tank from filling, we are only indicating about 1980lt. So, I'll address that at some stage. The enclosure might need a little tweaking and reprinting too, but for the interim it'll do. Otherwise, I'm happy with how everything turned out. I'll monitor it for a couple of weeks and then start building one for the other tank too.

Thanks for all the help - the inner nerd in me would not have been able to figure this out.

4 hours ago, stormjp said:

Okay, so I think I am winning.

Everything is installed and working flawlessly. I have noticed that the installer of my tank system was very very conservative with the float and when the float stops the tank from filling, we are only indicating about 1980lt. So, I'll address that at some stage. The enclosure might need a little tweaking and reprinting too, but for the interim it'll do. Otherwise, I'm happy with how everything turned out. I'll monitor it for a couple of weeks and then start building one for the other tank too.

Thanks for all the help - the inner nerd in me would not have been able to figure this out.

Nicely done....😍

When the community WhatsApp group starts moaning about there being no water, I occassionally go up and take a measurement with the tape measure - mine shows an accuracy of about 98% which works for me...

The nerd in me also added in this to keep an eye on the raw readings (anything between 0.92 & 0.93 in my case = 100% capacity)

 

Keep us posted on how it's working

Edited February 27, 20251 yr by Muttley

Hi everyone,

I finally got time to change the INA219 sensor and things are looking very promising. I'm just waiting a bit for more data and then I'll report back.

Thanks for the help!

Update:

So the issue was indeed the INA219 sensor from DFRobot. When I changed the INA219 with the suggested generic sensor, everything worked, even the tank volume started updating!

Thanks especially to @Muttley for your help!

I'm going to experiment on the "days of water left" and the "Tank Refill Rate"

This is amazing!

11 hours ago, Robbie3337 said:

Update:

So the issue was indeed the INA219 sensor from DFRobot. When I changed the INA219 with the suggested generic sensor, everything worked, even the tank volume started updating!

Thanks especially to @Muttley for your help!

I'm going to experiment on the "days of water left" and the "Tank Refill Rate"

This is amazing!

So glad to hear... :)

Sometimes I do find that DF Robot sensors are just weird - had similar issues with one of their ADC but anyway.

Your suspicions about the DFRobot sensor were correct. Thanks for that! I'll be doing the other 2 JoJo tanks as well that's on out farm.

Hi all,

I'm having a bit of an issue here. My tank volume stopped calculating correctly and just showed it was full (I didn't fix the calculation yet, so it shows 10284L when full). Upon investigating, I discovered that the MV has suddenly changed after months of reading the same values. We did not change the tank or anything like that. The last "loss of data" is a result of unplugging our whole network to protect our devices from lightning. After restoring power to our network equipment, the voltage had jumped quite significantly. I did not cut the power to any smart device, just a simple network disconnect and reconnect after a few hours.

Does anyone maybe know what the issue could be? I can, of course, just recalibrate, but I'm really curious to find out the cause of this change.

I'd appreciate any kind of insight on this.

5 minutes ago, Robbie3337 said:

Hi all,

I'm having a bit of an issue here. My tank volume stopped calculating correctly and just showed it was full (I didn't fix the calculation yet, so it shows 10284L when full). Upon investigating, I discovered that the MV has suddenly changed after months of reading the same values. We did not change the tank or anything like that. The last "loss of data" is a result of unplugging our whole network to protect our devices from lightning. After restoring power to our network equipment, the voltage had jumped quite significantly. I did not cut the power to any smart device, just a simple network disconnect and reconnect after a few hours.

Does anyone maybe know what the issue could be? I can, of course, just recalibrate, but I'm really curious to find out the cause of this change.

I'd appreciate any kind of insight on this.

I'm not sure what's causing the issue but I'm wondering if the voltage that's powering the sensor is providing stable voltages?
I installed a Meanwell power supply and then a DC - DC Buck converter to try keep the power as stable as possible.

Perhaps this could help?

Thanks for the rapid response Muttley!

I bought the same power supply and DC - DC voltage buck converter as per your original post. I always ran it on 24V, but I changed it to 18V now without any success.

Maybe I should just accept the new MV ranges and recalibrate?

43 minutes ago, Robbie3337 said:

Thanks for the rapid response Muttley!

I bought the same power supply and DC - DC voltage buck converter as per your original post. I always ran it on 24V, but I changed it to 18V now without any success.

Maybe I should just accept the new MV ranges and recalibrate?

Now that you mention that, I think I also had to do that once..

I'd re-calibrate based on the new MV Value and keep an eye on it.

Good luck :)

Ok great, thanks a lot Muttley! I appreciate your help :)