Electronics > Projects, Designs, and Technical Stuff
What's the most reliable way to check if a door of a cabinet is closed?
David Hess:
--- Quote from: redgear on June 24, 2020, 01:29:22 pm ---
--- Quote from: David Hess on June 24, 2020, 01:21:56 pm ---A mechanical switch embedded in the frame of the door is the common method. My next choice would be a magnetic reed switch which is commonly used by alarm systems so inexpensive and easy to acquire.
--- End quote ---
I am thinking is go with a microswitch with a mechanical protrusion on the inner frame of the door such that the switch is pressed only if the door is fully closed. What do you think? I did think of the magnet but it is susceptible to the interference from the magnets in fitness trackers and other wearables.
--- End quote ---
That is how most microwave ovens do it. I like the idea of using two switches in series which prevents being able to defeat only one.
Common wearable magnets are never going to be close enough to the door frame matter. And even if they are, two magnetic reed switches in series could be used so both would have to be defeated.
I should add that a door interlock switch is common on anything involving UV; my UVEPROM erasers all include one.
wizard69:
Is this for commercial use in the USA or the EU? I ask because you will need to follow local regulation. Most likely that would mean a mechanically locked door switch that is tamper proof. These sorts of witches have an oddly shaped actuator to prevent finger operation of the switch along with an electrically locked catch for that actuator. One of many manufactures is: https://products.schmersal.com/en_US/category/2411/safe-switching-and-monitoring. Depending upon the scale of the risk you may need to have redundancy in place.
Your safety requirements might require the use of controlled contact relays. or series connected relays for redundancy, when it comes to control of the UV source. Oh and you may even need separate safety relays connected to the door switches.One requirement that often gets missed is that closing the doors can not restart the UV, there must be a manual reset process.
As noted it depends upon the location but in the USA you have OSHA, NEC and UL requirements to meet. This might help a bit: https://www.radtech.org/images/sustainability_pdfs/UVEBEquipmentSafetyTopics.pdf More info are here: https://www.cdc.gov/niosh/docs/73-11009/default.html
There are all sorts of access control hardware out there the trick is to avoid going broke buying it. UV exposure can be hazardous but unless the intensity is extremely high there are far more dangerous things in the work place. In any event redundancy is a key element in safety as switches do fail with great regularity.
--- Quote from: redgear on June 24, 2020, 09:23:07 am ---Hi all,
I am building a UVC Sanitizer Cabinet. Since, UVC rays are very harmful for human beings, I want to restrict the circuit to work only when the door of the cabinet is closed. What's the most reliable way to do it? I thought of using two contact switches in series such that they are ON only when the door's closed. I would like to know about other reliable techniques that I can use.
Thanks
--- End quote ---
Doctorandus_P:
A simple regular household microwave oven usually has 4 microswitches to check whether the door is closed properly.
It's not just about the switches though, but also what you do with them.
For example one of the switches in the microwave simply hard shorts (after a fuse) the transformer itself as long as the door is open.
This gives an extra safety because it's independent from the rest of the circuit. You can have some fancy magnets with HALL switches, which go to some logic board, and even if the whole logic board (Software!) goes haywire and tries to turn the oven on, then it will just blow a fuse (and probably damage the switch). Magnets with Reed switches have been used for alarm systems singe 50+ years. The advantage of magnets is that you can have bigger mechanical tolerances.
Simple circuits are usually better, because there is less that can go wrong.
But with adding more complexity, you can also build in more checks.
If there is a microcontroller involved, I would not put multiple switches in series, but route each switch to the microcontroller, and check the time between activation of the switches, and give some warning if one of the switches fails to switch.
Navigation
[0] Message Index
[*] Previous page
Go to full version