Hi All, I’m wondering if this is the right place to ask. Can anyone help me build a level sensor for a grain silo? I would like to use the waterproof ultrasonic sensor (A01NYUB) and an ESP32 or something similar, ideally run off a small solar panel and transmit the level to wifi so I can read it from my phone, I’m not sure on the most cheap and efficient way to do this. Cheers for any help.
1 Like
Hi Bo
I would suggest something like a Radar type reflective sensor. But I don’t quite know what sort of reflecting surface would be presented by the grain. Lots of little curves I imagine. Also as the Silo fills the surface would probably not be level like water would be.
Maybe some sort of pressure sensor in the bottom of the Silo might be the go. It could be positioned in what might be the deepest part of the pile like under the feed chute or whatever.
Food for thought but I am afraid I have no practical experience with this sort of thing.
I assume that this Silo is the tall cylindrical container. I think that open air covered piles of grain with or without bags are also referred to as “Silos”.
Cheers Bob
1 Like
Hi Bob, thanks for the input. You’re right about the grain not being level, however this wouldn’t be an issue I don’t think as it would at least give me a rough idea as a percentage for example, Silo ‘A’ = 60% full etc. they hold 80T each, looking for someone to sketch me a wiring diagram basically and some lines of code, I can fill in the blanks. Cheers.
1 Like
Hi Bo
I think the other important quality is the reflective properties of the grain surface. This will largely determine what type of sensor is used. You might have to burn a few dollars to find out by experiment. Unless of course you find someone with a spare grain silo at their disposal.
Cheers Bob
1 Like
My first thought was a camera system with some sort of AI to interpret the images and convert to a level percentage. But may not suit your budget.
Alternatively sensors positioned at various levels to activate when covered by grain, light level is my thought here. 25% 50% 75% 95% etc. Pretty simple and no need for micro; just wiring to a LED or lamp or something.
Anyway, many ways of detecting the level, depends on what you want to check.
Cheers
Jim
1 Like
Thanks Jim, going for as simple as possible as there are 8 silos all up and will be more one day, I’d like to go on my phone and see which one needs filling first.
2 Likes
An interesting project 
.
Just as a matter of interest. What are the silo dimensions?
Are the grains the same or varying?
How much “grain dust” is present during filling?
How accurate do you require the level measurement?
2 Likes
Hey @Bo282250,
Welcome to the forum!
Some great ideas coming through already, thanks to everyone who’s contributed so far.
From my background working with mining electronics, I know how much the operating environment can affect reliability and sensor choice. To help narrow down the most cost‑effective and practical solutions, it would be useful if you could share a bit more detail on:
- Conditions inside the silo – e.g. temperature range, dust levels during filling, humidity
- Required measurement accuracy – rough “empty/half/full” or something more precise (e.g. ±5%)
- Wi‑Fi coverage at the silo site
- Target budget per unit, keeping in mind you have 8 silos to monitor
- Your preferred build involvement – comfortable with soldering, flashing firmware, etc., or looking for a more plug‑and‑play option?
This information will help members recommend the most suitable sensor type (ultrasonic, radar, pressure, etc.), power setup (mains, battery, or solar), and data logging or display approach.
One thing to consider is that certain distance sensors can struggle in dusty environments, so knowing the dust levels will be important for making the right choice.
Once we have these details, we’ll have a much better chance of putting together some practical, cost‑effective designs.
2 Likes
Hi Bo
Just for interest I googled “grain silo level sensor” and got some interesting results.
Especially this one
Would be worth a look. Saves a lot of guess work.
Cheers Bob
2 Likes
Hey Gary, silo dimensions are roughly 5m diameter and 8 m tall.
Grain will consist of barley, wheat, oats, lupens and pellet mixture. The environment inside the silo will be quite dusty when it’s being filled so that will be a factor.
It doesn’t need to be super accurate, within ± 5 % would be fine. just a guide to see which one is lowest.
2 Likes
Hi Ryan,temps inside could be anything from 0 - 50 celcius and am unsure on humidity.
Within ± 5% would be fine, the distance to the wifi is 160m but a clear line of sight, nothing in the way.
I’m comfortable with basic soldering and simple dc circuits, I have flashed arduinos and ESP32 in the past but wouldn’t call myself experienced.
1 Like
Thanks Bo,
I’m tending towards Radar considering your dimensions, accuracy and dusty environment.
Another option if possible is external load cells but that’s a costly install. Alternatively placing load cell based platforms in the inside base of the silo may allow a calculated mass “calculation” that would need to be “calculated” based on the grain type… complicated math 
It would be interesting to see if any of the low cost radars work, eg RD-03D which has an 8m range, and would not be affected (much) by dust.
2 Likes
There are a lot of different ideas here.
From my perspective the commercial one linked by @Robert93820 is the only one that is known to work. Camera, Radar, Pressure, Light are all untested and may not work long term. We are only guessing as to their viability.
If this was my project, I would start with what I said before, light sensors at say 1 Metre levels connected to a micro that provides a web page with the level on it; accessible by WiFi and a browser on a phone or tablet or laptop. With the dust and light conditions maybe light sensors would not work.
Any other sensor option would require testing too. But the common point in any design would be a micro to send data via WiFi.
The best option I see would be to contact the company linked by @Robert93820 and see if they have or can provide a cheap enough solution for your situation. Most likely someone else has done this before, grain silo’s have been around for a long time and the level if grain in the silo is a high priority when filling them.
Cheers
Jim
1 Like
I’m not sure the commercial option linked by @Robert93820 would have the 5% accuracy? It looks like an Full/Not Full solution?
Neither would light sensors every 1m (gives ~12.5%). But is cheap. Would require regular cleaning and additional cabling.
Radar is the most widely used solution commercially but is not cheap.
If this were my project I’d not use WiFi - too power hungry, and poor range, but rather opt for a low power (IoT) solution.
I’ll knock up a quick test to try out my theory
1 Like
Hi Gary
I did not go into this particular device I linked in depth but you are correct I think. Full, empty and any other critical point in between which if requiring lots of points could get a bit unwieldy and expensive I agree. But surely Full, Empty and a suitable warning point at say 60% would suffice. This bit about 5% is a bit far out. The slope on the grain pile under the chute would probably be more than that
Cheers Bob
1 Like
Hi Bob,
Agree 5% is quite “precise”. I initially though capacitive/reflective/IR at various positions but to get 5% would need every 40cm and at least 17 of them
Regards,
Gary.
2 Likes
Hi All
An add on
The company that markets the above linked device has all sorts of level sensors.
Here is the link to the parent page
Cheers Bob
PS: There are others also, this is just one.
2 Likes
Hi Bob,
I saw those and a few others, all around the $1k-3k price tag. if Bo is happy with that then there’s no reason to proceed with the “concept”.
Personally I think a “mount on top solar wireless version” should be able to be done for a lot less
G.
1 Like
Didn’t get as far as the price. Might be worth experimenting.
Cheers Bob
1 Like
Hi Bo and all!
Welcome to the forum!
I cant imagine an infra-red ToF sensor would go astray when the dust settles.
Something like this mounted off-center to get a rough volumetric estimate (will depend on the shape of the grain), and a quick measurement/calculation could be made.
If the measurements don’t make sense, either don’t publish or flag the measurement.
If you wanted to get fancy, you could use a ToF imager or LiDAR to get a better picture of the profile of the grain, for a more accurate calculation.
Hi Gary,
Unfortunately these radars have some inbuilt filters to stop them being used for ‘general’ projects. Humans are the only thing they can reliably detect.
Liam
1 Like
