Monday May 4th

Humidity is at 40% today. Carrot that went in yesterday, nearly gone, was washed and wet when it went in. Not sure of external humidity – may need to monitor this to see if there’s any connection.



The ideal humidity level should be around 70%. During testing I was getting around 22%. When I added the mealworms and a piece of carrot it was at 30%.

The humidifiers that are widely available seem too big for my set up and would be unmanageable for such a small volume. Foggers used in vivariums would also produce too much humidity and require an open tank.

Water gel crystals, used for keeping plants moist and humidors humid, might be a solution. I’m thinking of putting some in a small tub with a few holes in to give a slow release of moisture. I’ll order some.

In position

This is the farm in position in my office. Mealworms should arrive tomorrow.

I’ve checked temperature with a calibrated thermometer and sensor is correct. Also checked temp all over and varies  no more than 0.2C. These mealworms better be happy. And productive.


More testing

Been away to Scotland drinking whisky, mmm, now there’s a hobby.

I’ve now removed the vent flaps and servo. It holds temperature really well.  Having tested for some time I’m ready to order first batch.


A longer test today. Heat cycle works fine and first few vent cycles were ok but during venting, sensor data was corrupt. After approx 5 vent cycles, servo stalled. Quickly reset arduino and then servo performed well straight away.

I was reading up on the motor control shield for another project and it detailed the need for three suppressor capacitors on cheap dc motors. The servo I’m using is the Tower Pro SG90 – I think that’s cheap.

I’m now thinking of removing the vent flaps and replacing them with some forward facing ducts. This would do away with the need for the servo. I’ll think some more

Ready to go

I’ve tested it now with longer vent cycle and it works. Front sensor keeps throwing errors so will replace but it’s only for reference. I’m going to test for longer periods over next few days and then order first stock.

And here’s another video

And here’s a picture

Possibly fixed

Ok. So I replaced the damaged DHT22 and wired a separate 0v for the 5v electronics. I tested without venting cycle as this was when pop happened. It functioned apart from no humidity reading from front sensor (reads 1.0 constantly), must have been damaged when other melted, temperature appeared in line with new sensor. I tried venting cycle and immediately servo was struggling so I switched off. The 5v supply is supposed to be 800mA and it is powering two DHT22, the relay module and the servo. I looked into peak power for sg90 and thought I’d be ok with 800mA supply. I had a spare power module so decided to wire it in at the terminal block on the farm, powering just the servo (connected 0v to other 0v from control box). Tested it and servo worked fine at speed.

I’m guessing it’s one of two things. Either the supply module isn’t sufficient for servo and other bits together or the wire I’ve used to connect control box to farm isn’t up to the job and is limiting the current supplied.

Whatever it is, the new setup seems ok for now.