How can I waterproof an electrical component?

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
7 replies · 3K views
thegreengineer
Messages
54
Reaction score
3
Good evening, it's been a while since my last thread here in PF. I will go straight to the point:
I want to measure temperature using an NTC3950 thermistor. My circuit diagram is as follows:
52602050_2283098551974676_3639846829010452480_n.jpg?_nc_cat=104&_nc_ht=scontent.fntr6-2.jpg

It's basically a voltage divider. By varying the temperature acting upon the thermistor the output voltage will change as well.
The thing is that I was given the task to measure the temperature in a water heating system, and I required on using a thermistor.
1000-immersion-rod-1-0-kw-sun-flame-original-imaffca6dyvx5g9y.jpg

To heat the water, I will use an immersion rod, like the one on the previous photo. The "tank" is basically a plastic container like the one in the photo.
protein-HDPE-bottle-manufacturers-plastic-powder-containers.jpg_350x350.jpg

Since plastic doesn't conduct heat so well, I need to put the thermistor inside the water-filled container. My issue is the following: how can I waterproof the thermistor? Can I use nail polish or silicone to waterproof? I would appreciate your advice.
 

Attachments

  • 52602050_2283098551974676_3639846829010452480_n.jpg?_nc_cat=104&_nc_ht=scontent.fntr6-2.jpg
    52602050_2283098551974676_3639846829010452480_n.jpg?_nc_cat=104&_nc_ht=scontent.fntr6-2.jpg
    12.7 KB · Views: 918
  • 1000-immersion-rod-1-0-kw-sun-flame-original-imaffca6dyvx5g9y.jpg
    1000-immersion-rod-1-0-kw-sun-flame-original-imaffca6dyvx5g9y.jpg
    18.6 KB · Views: 713
  • protein-HDPE-bottle-manufacturers-plastic-powder-containers.jpg_350x350.jpg
    protein-HDPE-bottle-manufacturers-plastic-powder-containers.jpg_350x350.jpg
    7.5 KB · Views: 695
Physics news on Phys.org
I have waterproofed temperature sensors by encasing them in silicone. Ordinary clear silicone bathroom caulk from the hardware store works well.
 
I work with hot water tanks, which use heating elements which are controlled using thermocouples. Thermocouples are naturally waterproof and typical for use in temperature control/monitoring. They can come in any length, I use either 1m or 5m long 3mm diameter thermocouples in my work. It could be an already waterproof alternative for your needs.

Thermistors are more common in freezers as they react more to cold than thermocouples do.
 
phyzguy said:
I have waterproofed temperature sensors by encasing them in silicone. Ordinary clear silicone bathroom caulk from the hardware store works well.
If the water is for consumption the silicone needs to be food safe. Bathroom caulk contains mildewcides.
 
  • Like
Likes   Reactions: phyzguy
I have used liquid electrical tape before. Realize it will insulate a little and temperature changes will be a little delayed.

71pasIUrnfL._SL1500_.jpg
 

Attachments

  • 71pasIUrnfL._SL1500_.jpg
    71pasIUrnfL._SL1500_.jpg
    30.7 KB · Views: 673
If your app is quality critical, may I suggest that you consider commercial 'spike' type probes ? Their response curve should be provided, you just have to routinely confirm your sensor's calibration at several temperatures versus a 'traceable' instrument...

Plan_B may be to craft a robust, product-compatible 'thermometer well', put your thermistor in that.
Upside, custom made to suit pot, no worries about 'creepage' and flexing.
Downside, takes rather longer to equilibrate, a response time you must establish and document. A few drops of thin silicone oil may help heat transfer...
 
Search the online catalog stores for Glass Probe Thermistor. They start around US $1.50.
Google search for Glass Probe Thermistor.
Since this is a school project, there may be some Sodium Silicate solution ("liquid glass") around there you could use for waterproofing.
Many, not all, thermistors are Epoxy coated and will last long enough for a few experiments if the coating is not damaged.
Mix some Epoxy and coat whatever thermistors you have.

Any coating will increase the response time some, but most will not affect the accuracy.

Cheers,
Tom