My meat curing chamber build

[dpeart]

A few months ago I started on my journey to make a raspberry pi controller for my future meat curing chamber (a side by side fridge).

I've had it running for weeks now and it seems to be working fine in my test mode. Test mode is to sample humidity/temp, handle power failures, sensor failure, send email when this happens, and log data to a SQL database. The temperature is just the temp sensor sitting above a light bulb that is controlled by the controller. This has been running for about a month without issue so it is now time to make the next step. I'm going to now actually control the fridge and humidifier instead of just driving a light bulb.

Does anyone know the minimum safe cycle time for a fridge? I.E. how quickly can I turn it on/off without damaging the compressor? I looked around on line and for home AC compressors it is ~3 times per hour. Not sure if that holds for fridge or not. Hopefully someone here knows the answer. Or for those of you that have done this already, what do you see with your chamber?

Next I need to pretty up the web interface that allows me to control the unit, plot the data and everything via php javascript. Anyone out there interested in helping with this aspect? I've got the c++ code done for the pi, and while the javascript I put together is functional, it is quite ugly

I'll post later on how it actually works to control the fridge.

thanks,
dave

P.S. - this will be a slow process for me as real life, work kids interfere with my fun, but I'll keep the thread updated.

[Chuckwagon]

Hi Dave,
Wow, it's over my head. Hang in there and someone will be along to answer your question. It sounds like you have the ideal set up. Good luck with your project.

Best Wishes,
Chuckwagon

[el Ducko]

Does anyone know the minimum safe cycle time for a fridge? I.E. how quickly can I turn it on/off without damaging the compressor? I looked around on line and for home AC compressors it is ~3 times per hour. Not sure if that holds for fridge or not.

Dave,
I have a temperature controller that I bought at a home brew beer supply store. There is a two or three degree F "dead band" built into it, so however fast heat leaks into the refrigerator is how fast it cycles.

But here's a better answer. Normal refrigerators have similar controls, so some Sunday afternoon you might drag up a chair, crack open a beer, settle back, and see if you can stay awake long enough to get a real-life answer for cycle time. This should probably put a lower limit on your "how quickly" answer. Since you'll be running quite a bit warmer for sausages, the cycle time will be longer than what you have at the normal 40 degree or so operating temperature.

Sounds like a real intriguing project. Keep us posted, please. Maybe some day I can work up the nerve to try something that ambitious.

Russ

[dpeart]

I am still working on this. Not being a web programmer this has been a big learning experience. Here is what I have at the moment.

This is the graph that displays the temperature, humidity, fridge runtime, humidifier runtime. It will let you keep track of what is happening in the chamber.



The Admin tab let's you create your batch and add product to that batch. Eventually I'll add more details to the product, like weight pre/post fermentation and drying.



I got my humidifier and will be hooking it up shortly to my fridge and tuning the PID algorithms to maintain temp/humidity. At the moment the temp readings are from just sitting there on my desk.

Anyone have a recommendation for a good analog humidity sensor? Humidity is a beast to accurately measure so I'd like an analog device to give me a sanity check that the digital ones are working.

thanks
dave

[el Ducko]

Looks like you have the web stuff working fine. Congratulations! (That's quite a feat.)

Looks like the control equipment is just sitting there. (Internal temperature tracks external temperature.) Go ahead and hook up the temperature controller and make sure it works well. Then, you can hook up the humidity portion. In automatic controls, it's best to get all the "easy to control" variables controlled first, so you'll have a steady platform on which to base the control settings for the more sensitive variables.

You'd think that temperature control ought to be easy, but there's one exception: during some seasons and some operating conditions, you may need heat for one time period (night?) and cooling for another (day?). Staying one way or the other is easily controlled, but the crossovers don't work so well. Some of the better building HVAC units do this by cooling the air too much, then heating it back up to the desired setpoint. Older systems (most of what's out there) can't, and do a lousy job in the fall and spring.

Depending on the desired operating temperature, you may have the same problem. Hopefully you've built a system which can work in cooling mode or in heating mode. This may require a manual switch-over. On the other hand, a good computer control system ought to be able to make the switch for you.

However, let's suppose that yours doesn't. In that case, you can cleverly solve your problem by moving your equipment to a location with a stable higher or lower temperature, or by looking in on the thing at least twice a day and using "hand-o-matic" (manual) switchover, or by waiting until the weather shifts from summer to winter or vice versa.

Confused? Take two bourbon and call me in the morning. I apologize. I get carried away sometimes. This is very doable, though, and with the control system you have built, it ought to work like a dream.

I'm envious. Would you mind (1) commenting on how well your light-bulb-and-refriferator hardware setup is operating? and (2) sharing your progress and lessons learned? There are a bunch of us who would like to follow in your footsteps. (I was just reading up on "Project A" last night. I've never had "real" salami.)

Which begs the question: what meat product do you plan to tackle first? (...and can I come over and help sample the product???)

[dpeart]

Initially I just plan to support either heat or cool, not switch over. Being here in TX that will only prohibit working over a small period of time where the weather switches.

I do plan to automate the controls to handle switching between the two, which is why I have an internal and external temperature reading. I can use the relationship between the temp and my target temp to determine whether to run the fridge or turn on the heat. At least that is the plan. I'll probably have to store two different PID values, one for each "mode", and load them appropriately to get it to converge well, but that is doable after I tune the two modes.

I plan to move the unit from my office and into the fridge to start tuning the PID controllers next.

dave

[el Ducko]

It oughta work great, Dave.

I'm in Texas, too. Now that the temperature is dipping below 100, it's almost time to... Uh... Looks like you'll be in refrigeration mode quite a while longer. My keg-o-rator sounds like it's working overtime, and here it is, way past time to brew a batch of my famous OktoberPest pilsner.

[dpeart]

OK did my first temperature control test. I'm surprised at how easily it maintained 70 degrees. For my smoker I had to tweak the crap out of the PID values. Not sure why this was so easy, maybe the temp just changes so slowly it worked.

Next it humidity control, but I have to wait for some new SSRs. I bought two 3-32VDC control SSRs (cheapos from China) and they don't even turn on at 5V. So I bought some name brand ones, hopefully they will work better.

Oh yeah, the humidity percentage axis on the right isn't correct. The humidity is aligned to the temp scale on the left. I need to fix that also.

Is the up and down in the humidity expected? I kind of expected it to just drop, not go up and down like that.

anyway here is the pict.

[el Ducko]

Humidity control can be a real bear. Supposing the humidifier is designed for a room, and you use it in an ice chest. You can overwhelm the ice chest when it comes on, then underwhelm (if there is such a word) when it cycles off. Or vice versa, suppose the humidifier is small and the other equipment is large. You'll never "get there."

So, to answer your question, I'll cite el Ducko's Law: "Some do; Some don't." If it's cycling badly, maybe there's some way you can cut down on the amount of humid air introduced to your box. If it's not cycling at all and you can't reach setpoint, you'll need more moist air.

Fascinating, isn't it? (...in a frustrating sort of way.)

[ssorllih]

Only my uneducated opinion but it seems to me that the capacity of all of the mechanical systems is far too large for the chambers that we build. I have considered using a complete refrigerator as a source for chilled air and building a duct work with supply and returns to a curing chamber. The air flow to the curing chamber could be controlled by fans in the ducts.

[dpeart]

for clarification. I was not doing anything to try and control the humidity. I was just watching it change as the fridge ran. I was surprised to see the spike in humidity while the fridge was still running. that's all.

dave

[el Ducko]

Ah! Okay.

Can we explain that by saying that, when you cool down a volume of air, it becomes less able to hold as much moisture (relative humidity increases)? At the dew point temperature, the air is saturated, and droplets of water start to condense (clouds form). Note that we didn't change the amount of water in the air, just the temperature (or the ability to hold a greater quantity of water, if you added it, but you didn't).

Clear as mud? As temperature cycles, relative humidity (ability to hold water) cycles with it.

Ross is right about overcapacity and how to handle air flow. Don't let him fool you- - he has the best education you can get. He's done this stuff hands-on, for real.

[dpeart]

It appears that when the fridge is not running, the humidity increases. You can see this in the plot below. The temperature does not change much at all, so I don't think that is the issue. The humidity just creaps up. Then when the fridge turns back on, the humidity drops. Or is the humidity sensitive enought to change 3% for a .3 degree F change in temp?

Anyway just another pretty picture really.

[el Ducko]

Creepy. Maybe it's a Halloween problem. (Addams Family theme plays in background.)

...couple of wild guesses:
(1) Is there an air leak or air flow?
(2) There is a possibility that, when the refrigerator is running, moisture condenses out on the coils or freezes. Then, when it shuts, the moisture evaporates. It wouldn't take much to make a noticeable swing.
(3) "It's on account a yo' boogaloo situation." (Wolfman Jack, late '50's.)

I'll go for either #2 or #3.

Ross...? ...got some more insight?

[ssorllih]

When you run a cooling coil it is usually operating below the dew point. It takes about a 1000 BTU's to condense a pound of water and you can't cool the air below the dew point. So the net result is the RH will fluctuate but the temperature will remain quite steady.
I was trying to use ice to cool my chamber and was constantly pulling water vapor from the air and condensing it on the ice. When I sometime build a chamber I will treat it like a single room in a house with central heat and AC and control the conditions in the room very gently.