General > General Technical Chat
PWM'ing a 250W heater.
<< < (4/6) > >>
Circlotron:
You may like to use a transformer to run the heater at reduced voltage when it is operating to slow the rate of rise of temperature. Put the tranny upstream of the heater fan so any wasted energy gets fed into the controlled environment.
paulca:
Well, I would say this 6 hour graph shows it's working as intended.  I'd call that pretty tight temp control. :)

paulca:

--- Quote from: Circlotron on March 10, 2021, 11:01:55 pm ---You may like to use a transformer to run the heater at reduced voltage when it is operating to slow the rate of rise of temperature. Put the tranny upstream of the heater fan so any wasted energy gets fed into the controlled environment.

--- End quote ---

This was an idea I considered, although using a triac to cut the wave rather than a transformer.  The fiddly part is making a software controller one... which led to using an STC-1000 or similar.  What put me off that, is it's last generation tech.  Yes it works, yes it will be reliable, but it's so "offline" and I'm sure a "Smart" one which can be hooked up to a network data logger and accept parameters remotely... would cost $$$$ and probably force me to sign up to a cloud service.

In fact the 4 plug bar I bought tried that too.  They even closed the loop hole where you could pretend to be their update server because they used a PSK (pre-shared key) which had become "leaked" to the community.  That no longer allows you to inject your own firmware.  I had to open the thing up to flash it over serial.  Even then, they had switched the orientation of the module to make it very difficult to get to the GPIO 0 test pad on the back of the board to put it into flash mode.  Still didn't stop me.

As an aside, the commercial spyware industry is trying to make it illegal for you to modify any software installed on a device you own if the company declare it so.  We should fight against this while we fight for the "right to repair" clause.  Allowing companies free run as to what software runs in our homes when the regulations and bodies overseeing "Smart home" gadgets is pretty much non-existent.  The average consumer is completely unaware and completely unprepared for the level of tracking, profiling and behavioural analysis that is happening to them everyday.  The REAL worry however is what happens when these devices get hacked, taken over en-mass, just like 150,000 security cameras this week did.
paulca:
Whoops!

After a few beers I decided to see how far I could push it on "Tropical conditions", so I bumped the humidity up to 75% and the temp up to 25*C... Deciding it was a bad idea to leave it over night like that I dropped the temp back to 22*C.

So what happens if you take air at 25*C and 75% humidity and quickly drop it to 22*C.  It rains or at least dew forms...  first lesson.  Humidity sensors stop working when dew is present, the first one to get hit was mine which rose rapidly to 100% and stuck there, but the one in the humidifier failed the other way and that sent it into full throttle, max humidifying mode.

The project camera still functioned and you can just watch the water level drop in the humidifier on the time lapse and the water running down the walls, dripping off cables.

Around 5am the temperature sensors on one sensor board dropped out, then the pressure sensor.

I found the damage at 8:30.  It's not hard to fix the humidity, open the door.  Drying everything out is the other story.  The heaters and fan still work, so I'm closing the extractor, letting it heat up, the cycling the air out with the extractor and repeating.  I also need to be extremely careful in there, in regards touching things as there is a 240V heater sitting in a puddle of dew on the floor of a propagator lined with metalized reflective fabric.

I'm hoping the sensors and humidifier recover when I get rid of the dew.  At least it's basically distilled water anyway.

EDIT:   I have one free socket on the wifi power bar.  The humidifier is going on it.  I can't change the panel settings remotely, but I can kill it's power if the humidity goes out of spec again.
Ian.M:
It would appear you need a more robust humidity sensor that can cope with 100% humidity and condensing conditions. There are various options, including heated humidity sensors, where as the humidity approaches 100%, the sensor is heated to maintain a lower target readout, and the actual humidity calculated from the readout, sensor and ambient temperatures, however a much simpler to implement one would be to add a wet bulb thermometer (actually a wet bulb temperature sensor) consisting of a glass, epoxy or stainless steel encapsulated temperature probe, covered in cotton lamp-wick, with the tail of the wick immersed in a reservoir of distilled water.
See https://en.wikipedia.org/wiki/Wet-bulb_temperature

As to recovering your current 'swamp chamber',  I'd isolate power,  mop everything off to remove gross condensation,  then if I didn't have a dehumidifier handy, proceed as you are doing.  If I had a dehumidifier, Id simply run it and program the chamber temperature to drop slowly from say 27 deg C to the desired target temperature, at a rate calculated so that the volume of water that can be expected to condense from that volume of air at 100% RH cooled over that temperature range is significantly less than the specified max. extraction rate of the dehumidifier.
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod