I have a prototype setup, where an ultrasonic humidifier is controlled from a microcontroller (Teensy 3.2) through a power MOSFET to humidify the air passing through a metal container partially filled with water in which the humidifier is submerged.

The humidifier itself is unbranded and I cannot find a datasheet, but it's identical to this one. The humidifier has a metal "rod" going out from the housing and extends up above the membrane, and this rod has to be submerged in water for the humidifier to turn on.

My best guess is that the rod is one electrode in some kind of conductive level switch used to only excite the membrane, when it is fully submerged, to avoid damage. Near the membrane there is another metal "ring", which (maybe?) constitutes the other electrode of the switch.

The system was first tested in a plastic container, where everything worked as expected, but when moved to the metal enclosure (needed because of the water temperature) the humidifier turns on and off at irregular intervals even when fully submerged. This behaviour is easily replicated when moved between the containers.

I have included a simplified schematic of the setup here: enter image description here

My best guess is that when placed in the metal container, some kind of leakage current influences the internal readout from the level sensor, making it unstable. I tried grounding the metal container properly which seemed to help a little (longer time between the humidifier turns off), but it did not solve the issue.

PSU1 and PSU2 are two separate switch mode power supplies operated from the same 230VAC power outlet.

Do I need some kind of galvanic isolation of the humidifier, or what would you suggest the next step should be? Any help is much appreciated!

  • \$\begingroup\$ Im curious but if the bucket is placed on an insulating mat and ungrounded is the behavior unchanged here? Have you tried buckets of significantly different size or dimension? \$\endgroup\$ Commented Sep 8, 2020 at 11:17
  • \$\begingroup\$ What makes you think the humidifier itself will be able to function at a water temperature where a plastic container (possibly inside an outer metal one) is not workable? Note that questions on the usage of undocumented products (or modules scavenged out of them) are off topic here, for this to survive you're going to have to provide some substantial documentation of this device, and some meaningful results of your investigations of it. \$\endgroup\$ Commented Sep 8, 2020 at 13:42


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.