I'm building a moisture/temperature sensor for outdoor use in soil. It uses capacitive sensing for moisture and an NTC thermistor for the temperature. It's going to be used for vegetables so food safety is a concern.

I don't think the PCB is food safe, or safe from corrosion/degradation due to the soil and moisture would creep into the FR4 material in the PCB and cause my measurements to drift so I would like to coat the entire thing to protect it from the elements, keep my vegetables free from toxins, and keep the measurements from drifting.

So these are the requirements in order of importance for the coating:

  • Not electrically conductive
  • Food safe
  • Doesn't degrade/react in/to soil/sun light
  • Hydrophobic
  • Resistant to abrasion when insertion/removing from the soil
  • Affordable
  • Low relative permeability (electromagnetic)
  • Decent thermal conductivity
  • Not a host for bacteria

I first considered a food safe epoxy but I had difficulties finding a reputable supplier of food safe epoxies in Germany (my German sucks). I did find a supply of food safe silicone which is nice in that it has all the properties above but I could only find it in large quantities and it's quite expensive. I also considered L/HDPE which seems to have all the criteria apart from the bacterial aspect (I'm not sure about that) but it seems like melting it to have a low enough viscosity to be able to form a solid but thin coating seems hard/impossible.

My knowledge on the topic of epoxies, plastics, rubbers and chemistry in general is limited, so I'm turning to chemistry@sx for recommendations to what I can use for a coating and what process to use for coating. And maybe I've misunderstood something of the materials I've considered so far... Also I'm sorry if this is the wrong place for this question, I'm not sure where else to post.

  • 2
    $\begingroup$ It might be overkill, but PTFE (Teflon) looks like a great option to me. It is temperature and chemicals-resistant. Plus, Teflon coatings are employed in cookware. The only "bad" point is its price, but a thin coating would do the job, and not much of it would be necessary. $\endgroup$
    – The_Vinz
    Apr 29 '19 at 21:51
  • $\begingroup$ @The_Vinz have you worked with PTFE before? Can you offer any advice how it is applied? I know it's a thermo plastic, so I'd get a block and melt it the liquid phase and dip? $\endgroup$
    – Emily L.
    Apr 30 '19 at 20:56
  • $\begingroup$ I am sorry, I didn't understand that you intended to use it in DIY. I think that working with PTFE might be hard without proper equipment, but a possible experiment would be using some PTFE tape used for sealing threads and sold in any supermarket. Try to tightly coat the apparatus with it, and then try to lightly melt-solder it with a heating gun. I am not sure that it works, and probably this will require temperatures around 200°C, but maybe it's worth trying on some scrap material. $\endgroup$
    – The_Vinz
    May 1 '19 at 5:37
  • $\begingroup$ Otherwise, another good option for DIY would be using... a condom :) Latex-free condoms are made of non-toxic polymers: polyurethane, polyisoprene, nitrile or silicone. They are less resistant than PTFE, but maybe their shape and ease of reperibility might be an advantage. $\endgroup$
    – The_Vinz
    May 1 '19 at 7:18
  • $\begingroup$ @The_Vinz latex condoms don't have the necessary tensile strength to survive the insertion :/ (I tried) $\endgroup$
    – Emily L.
    May 5 '19 at 15:37

Various DIY capacitive moisture meters use a PCB for the sensor:

PCB capacitive probe

While not directly related to your question I thought I'd mention that you can buy these sensors for several dollars:

Sensor Kit

Purchasing a readymade kit and relying on the discount of mass production (along with a manufacturer's warranty) is probably easier than designing your own, and etching your own PCB.

I think how you plan to make the probe plays a part in how you plan on coating it; for example you could fit the parts in a test tube, that would meet all the requirements in your list. Such a capacitive sensor would operate similar to the touchscreen of a cellphone.

  • $\begingroup$ I have looked at the off the shelf sensors. They all lack features that I need on a system level and are not suitable for my planned network topology. I have already manufactured PCBs and written the software. None of these sensors you show come with coating so they are not safe for use as-is with plants you intend to consume and will suffer from drift as moisture creeps into the FR4. Sticking a tube into the soil and putting the sensor in there is obviously a bad idea unless you cap the tube, as water may enter from the top it also pushes the soil further away, weakening your measured signal. $\endgroup$
    – Emily L.
    Apr 30 '19 at 6:24
  • $\begingroup$ Thank you for your reply. Instead of using the above sensors exactly as-is you can order a conformal coating or if you are using them as an example then apply your own. If the test tube is a "bad idea" unless you put a stopper in it then I guess there's only one thing to do. Perhaps another answer will be forthcoming that you find more to your liking. $\endgroup$
    – Rob
    Apr 30 '19 at 6:34
  • $\begingroup$ This whole question is about choosing a conformal coating... An answer to buy another probe and then conformal coat it isn't really helpful :/ $\endgroup$
    – Emily L.
    Apr 30 '19 at 15:33
  • $\begingroup$ I explained what standard practice is, and that you can do what people whom sell many of these probes do, additional coatings beyond a decent PCB material wasn't deemed necessary by many people. In the event that you wanted something simple and inexpensive a glass tube provides certain protection and food safety; unless you are prone to breaking things. As a point of interest, and stated as unrelated, you can to buy one and choose the coated option when ordering, and get all the work done for several dollars. If your time isn't valuable then making it yourself will save a lot of money. $\endgroup$
    – Rob
    Apr 30 '19 at 16:58
  • $\begingroup$ You have given no advice about best practice about conformal coatings, full stop. PCB manufacturers make no guarantees and pay no attention to food safety of PCB soldermask or other materials, and unless the device is RoHS compliant (you're linked probes are not) the solder on the joints or on the HASL'd pads may contain lead which is definitely poisonous. People use uncoated PCBs for their soil sensors for the same reason they use PVC piping in their hydroponics/gardens, it's cheap and they don't know that it is unsafe to use if you intend to consume the plants. $\endgroup$
    – Emily L.
    Apr 30 '19 at 20:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.