Author Topic: Disinfecting PPE masks using dry heat and Arduino temperature controller  (Read 1156 times)

0 Members and 1 Guest are viewing this topic.

Offline profdc9Topic starter

  • Frequent Contributor
  • **
  • Posts: 334
  • Country: us
PPE masks can be disinfected using dry heat, usually in the range of 60C-90C and 20-90 minutes.  I am designing a cabinet including temperature controller for this purpose.

The heater is a 60 watt incandescent bulb.  I chose this because both the bulb itself and fixtures are readily available.  It is switched on and off by a simple relay to control the temperature.

I am designing a temperature controller based on an Arduino Nano.  This circuit is designed to be built on perf board or even a solderless breadboard if required.  It does not use a LCD to communicate, instead it uses four LEDs which flash codes.  It also has three buttons and a connection for a piezo buzzer or speaker that is an alarm when the disinfection cycle is complete or there is an error.

My question is: is the circuit here simple enough to be built by most hobbyists?  I kept it as simple and with as few components as possible.  It is powered by USB except for the light bulb which is powered directly from mains.

I have included the schematic, the PCB I designed for it (which doesn't have to be used, it's just an example), and the traces for the PCB.  It is laid out as a single sided board so that home board etching techniques can be used to produce it.

Any comments are appreciated.
 

Offline Whales

  • Super Contributor
  • ***
  • Posts: 2122
  • Country: au
    • Halestrom
Quick aside: sometimes silicon isn't the best solution if you want things to be simple and accessible.

eBay search terms "temperature control oven":



Pureley mechanical device, turns mains on and off like a switch when a certain temperature is hit.  Attach your bulb directly to it, no other circuitry needed (other than fuse, power switch and power cord).

The 300C models are probably not what you would want.   EDIT: here we go:




« Last Edit: March 30, 2020, 02:13:17 am by Whales »
 

Offline profdc9Topic starter

  • Frequent Contributor
  • **
  • Posts: 334
  • Country: us
Are these able to keep the temperature to within a 2 C or so?  These usually aren't used at the lower end of their range.

https://www.amazon.com/uxcell-Thermostat-Adjustable-Temperature-Controller/dp/B07PFVV6QZ

I suppose they are easy enough to get and try, it's just that the LM35 is accurate to 0.5 C. 

There is also this too...

https://www.amazon.com/Temperature-Controller-Thermostat-Thermometer-Charcoal/dp/B01DTI755W

I have more confidence in that, and it has a timer which is very necessary in this case. I'm not sure that it has 2 C accuracy either being designed for such a wide range.

The thing I'm not sure about is how sensitive the masks are to damage.  It will be harder to get acceptance if the process deviates from what has been tested before.  On the other hand, not having to build a circuit and just plugging in something is very compelling.  The circuit though, can be built so that once set up, you just press the button and it does a cycle.


 

Offline Whales

  • Super Contributor
  • ***
  • Posts: 2122
  • Country: au
    • Halestrom
They won't be accurate to within 2C.  Their hysteresis will be larger than that for a start.

Why the 2degC number?  It's extremely difficult to get all of the air within a chamber within that tolerance to begin with (unless you have an internal fan circulating it).  Based off your requirements you have a wide tolerance range (60-90C).

N.B. don't assume the digital options are better or more reliable.  If you want a timer then it may be worth going for a mechanical one too, as the UI it provides is much simpler than any digital option (no manual needed) and it's easier for repairers to understand.

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Why reinvent the wheel? You can buy a ready made PID temperature controller for around 20 bucks, add a thermocouple and SSR and you are ready to go. Most have an auto-tune feature, add a light bulb or several and dial it in, holding the temperature within +/-1C should be trivial.
 
The following users thanked this post: NiHaoMike

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9321
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
If using incandescent bulbs as heaters, connect two in series or add a series resistor or diode. Dramatically increases lifetime, especially if the controller you use doesn't do phase angle control. The cheap PID module I have pulses at a rate of about 5Hz, very fast for most thermal applications but will thermal cycle an incandescent bulb a lot.

Or use wirewound resistors - they aren't that expensive. PTC heaters are another option that are self regulating, but I'm not if there are any commonly available ones in the temperature range you're looking for. You'll also want a fan to ensure good circulation.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
The cycle time on most PID controllers is widely varied, adjustable from several Hz to several 10s of seconds per cycle. Something around 1-2 Hz is probably a good cycle time for incandescent lamps.
« Last Edit: March 30, 2020, 09:15:01 am by james_s »
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Are these able to keep the temperature to within a 2 C or so?

usually in the range of 60C-90C

So which one is it?  It sounds like a terribly insensitive process.  What happens if you cook it too long and too hot?  The paper is too crisp to fit your face anymore?  You get a dozen fewer wearings out of it?

Or what about just tossing it in the dryer along with the other laundry?  I've had my share of socks/underwear elastic cooked off in a (unfamiliar, too-hot) dryer... :palm: I'm sure many hit that temperature range and duration, at least for suitable combinations of machine, cycle selection and loading.

Is a controller needed at all, why not use the Easy-Bake approach?  A lightbulb in a box with a timer.  What could be simpler?  Plus, mechanized and electronic timers are available cheaply.  The design effort would focus on mechanical and thermal design, and process -- ensure the user doesn't (or can't, if possible) load too many things, such that they might cook off and start a fire, etc...

Not to say I don't appreciate the effort.  It's a noble cause, and a good exercise.  And if nothing else, you can use it to... sous vide, I guess, among other things. :)  Accelerated life testing of electronics, perhaps?  Sure, why not...

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline Whales

  • Super Contributor
  • ***
  • Posts: 2122
  • Country: au
    • Halestrom
We're all jumping in with large-scale suggestions and problems.

profdc9: if you want feedback about your design from very specific angles (manufacturability, will it work, thoughts on part accessibility) then you will need to ask us to focus on those questions directly.  Otherwise we'll probably keep focusing on your overall aim & alternative solutions.

Offline profdc9Topic starter

  • Frequent Contributor
  • **
  • Posts: 334
  • Country: us
It is uncertain exactly what the temperature should be within the range of 60C-90C.  However, if a temperature is selected, the process should be repeatable.  So therefore if it is established that, for example, 70C is a good temperature, the device should keep the temperature within 68-72 C, so that others who use the same value will get the same result.  It is not that any temperature within 60C to 90C will work best.  So for example, I currently have written the software to allow for seven settings: 60 C, 65 C, 70 C, 75 C, 80 C, 90 C, 100 C.  Also I have seven time delays: 10 minutes, 20 minutes, 30 minutes, 45 minutes, 60 minutes, 90 minutes, 120 minutes.  It is important that the device produces repeatable results.  Once the device has been programmed with a particular temperature and time interval, one should just ideally have to load the chamber and press a button and the alarm goes off when its done.

The problem is that different masks from different manufacturers may be able to tolerate different temperatures.  For example, one mask might tolerate 90 C for 20 minutes, and so many could be disinfected quickly, while others might be more fragile and require 60 C to 120 minutes.  There will be some variation in the controls here. 

Also, I would prefer if the operations of controlling the heat and time were performed together.  The people using this will probably not want to mess with it.  They want to press a button and have it work.  The less you give them to think about, the less that can go wrong.  But if Arduinos are not available or local hobbyists are not available to build a circuit, having an oven timer and oven thermostat is definitely ok, but I am afraid that with all of the chaos going on in healthcare settings right now, the less that has to be messed with, the better. You can not underestimate the things that can go wrong in a healthcare setting especially when there is chaos, people are overworked, and there are emergencies popping up constant and tons of distractions.  There must be as little to go wrong a possible.

A fan may be required to equalize the temperature in the chamber if the chamber is large.

I chose wirewound resistors at first (for example, 8 ohm audio loads), especially because low voltage could be used, reducing the electrocution hazard.  However, I am more confident that light fixtures and incandescent bulbs are more widely available, hence my suggestion to use these as a heat source.  I think they are also safer as well, as the surface of the bulbs usually doesn't get hot enough to ignite wood or other materials.

So a higher cost electronic device, if off-the-shelf, that could control the temperature within the ranges given, and be self-timing, would be best.  So far I have found stuff for brewers and electric roasters.

So that's my reasoning behind this device.  The attached document describes the circuit and project so far.
« Last Edit: March 30, 2020, 02:01:05 pm by profdc9 »
 

Online ejeffrey

  • Super Contributor
  • ***
  • Posts: 4034
  • Country: us
If you want 2 degree accuracy, uniformity, and reproducibility you definitely need a fan.  A 12 volt PC case fan should work well and be widely available with standard mounting points.  I agree a lightbulb is the simplest, most economical, and most widely available heater for this type of application.

I wouldn't worry about Arduino availability but I would worry about it's ruggedness for how it sounds like you want to use this.  In a rushed situation you don't really want something where wires and sensors can be knocked loose or plugged in wrong or where the USB could be unplugged and it loose power.  If you are making a custom shield a lot of that can be eliminated, but then why not just put the microcontroller on the shield and eliminate the Arduino?  A small microcontroller with a temperature sensor a PC fan jack and a small SSR is a trivial design and none of those parts are in short supply.
 

Online ejeffrey

  • Super Contributor
  • ***
  • Posts: 4034
  • Country: us
Most of the cheap off the shelf temperature controllers in the 1/16 DIN format can probably do what you want as long as they user doesn't have to interact with their user interface.  If you can program them to run a preprogrammed cycle on powerup and light a "ready" led when the cycle is done that would be good. 
 

Offline profdc9Topic starter

  • Frequent Contributor
  • **
  • Posts: 334
  • Country: us
The Arduino Nano contains other parts already pre-assembled such as a regulator and USB connector and is easily programmable.  One can solder a Nano directly to a PCB if necessary or use pin header sockets which can secure the Nano into a circuit board.  I would propose the Arduino Uno as a second choice because a shield would be preferred as you note.  I imagine that this will be built on a solder perf board with point-to-point wiring in practice in most instances with a pin header socket if available.  I am using a mechanical relay rather than a solid state relay because these are more likely to be available I think.  Also, Arduino PCBs are probably available most anywhere where other devices for electronics assembly can be obtained.

I will look into the 1/16 DIN temperature controller.  The only disadvantage is that there are many models of these so it might be harder to recommend models or provide universal instructions.  Some of these appear to need some kind of external control to instruct the device to turn on and off.  For example:

https://assets.omega.com/manuals/M4636.pdf


 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf