New DIY controller for MES 20071117

[Fornax]

I'm in the planning phases of replacing my MES 30" 20071117 smoker's controller with a Raspberry Pi so that I can set up remote monitoring, alarms, and more complex control. It seems relatively simple... the controller is connected to the body via just four wires. When I opened up the controller, I found that they're labeled +5V, TEMP, GND, and HEAT. +5V and GND are easy enough, but I'm wondering if anyone else has done anything like this with this model before and can provide any insight into the working of TEMP and HEAT.

I'm assuming TEMP is the input from the internal temperature probe. My best guess is that it comes in as a 0-5V analog signal.

HEAT will be the output to the heater. This is the one I'm most curious about. Is it just a 0/5V on/off signal to a relay controlling the heating element? Some sort of analog thing? If the former, does it PWM to try to maintain constant temperature, or just turn it on and off periodically like a home thermostat?

I'm seeing if I have the components available to test these, so hopefully if no one else knows or has seen this information anywhere yet, I'll be able to figure it out and post what I find.

 

[dr k]

There's a guy that made a thread on making a PID controller out of a basic circuit board that plugs into the Mes wiring pin connectors the OEM Mes controller connects to. I'll try to track it down.

 

[dr k]

This is the thread from Tiros I mentioned above.
https://www.smokingmeatforums.com/t...full-pid-cellular-contro.267809/#post-1749134

 

[tallbm]

I'm in the planning phases of replacing my MES 30" 20071117 smoker's controller with a Raspberry Pi so that I can set up remote monitoring, alarms, and more complex control. It seems relatively simple... the controller is connected to the body via just four wires. When I opened up the controller, I found that they're labeled +5V, TEMP, GND, and HEAT. +5V and GND are easy enough, but I'm wondering if anyone else has done anything like this with this model before and can provide any insight into the working of TEMP and HEAT.

I'm assuming TEMP is the input from the internal temperature probe. My best guess is that it comes in as a 0-5V analog signal.

HEAT will be the output to the heater. This is the one I'm most curious about. Is it just a 0/5V on/off signal to a relay controlling the heating element? Some sort of analog thing? If the former, does it PWM to try to maintain constant temperature, or just turn it on and off periodically like a home thermostat?

I'm seeing if I have the components available to test these, so hopefully if no one else knows or has seen this information anywhere yet, I'll be able to figure it out and post what I find.

Hi there and welcome!

I have a post using the Heatermeter (Raspberry Pi device) as a controller with the MES rewire. Others have added their posts about leveraging the MES Relay and wiring etc instead of rewiring the MES and using an SSR.
Read through all pages and posts of the thread here and hopefully you can piece together the info you need :)
https://www.smokingmeatforums.com/t...d-electric-smoker-mes40-mod-in-detail.264277/

 

[Fornax]

Thanks guys. Good links! I'm going to go through them more thoroughly later to see what else I can learn. For this one I'm keeping it relatively simple. Like Tiros's mod, I'm only replacing the top controller, not anything in the main body.

I went ahead and did some probing while smoking some ribs yesterday. This was just an initial fact-finding to discover how the built-in controller uses the TEMP and HEAT wires. The TEMP wire is indeed an analog signal from the probe. Simple enough. The HEAT wire appears to be a straight on-off 5V signal to the relay, and operates like a home thermostat; on for a few minutes, then off for a while.

Next step will be to rig up an ADC so I can read the TEMP signal into the Raspberry Pi and correlate it (along with a separate probe for comparison) to the HEAT signal and see how quick the response is. I'm guessing it's going to be reasonably fast, in which case I think PID control might be overkill. We'll see!

 

[tallbm]

Thanks guys. Good links! I'm going to go through them more thoroughly later to see what else I can learn. For this one I'm keeping it relatively simple. Like Tiros's mod, I'm only replacing the top controller, not anything in the main body.

I went ahead and did some probing while smoking some ribs yesterday. This was just an initial fact-finding to discover how the built-in controller uses the TEMP and HEAT wires. The TEMP wire is indeed an analog signal from the probe. Simple enough. The HEAT wire appears to be a straight on-off 5V signal to the relay, and operates like a home thermostat; on for a few minutes, then off for a while.

Next step will be to rig up an ADC so I can read the TEMP signal into the Raspberry Pi and correlate it (along with a separate probe for comparison) to the HEAT signal and see how quick the response is. I'm guessing it's going to be reasonably fast, in which case I think PID control might be overkill. We'll see!

I can't wait to see what you figure out :)

 

[Fornax]

Ok, so this weekend I rigged up an ADC (an ADS1115) to the Raspberry Pi to measure the temperature probe voltage against the element activation signal while I smoked a pork shoulder. Here's the result:

I preheated at 250. Shortly after the temp peaked I put the meat in and lowered the set temp to 225. I didn't open the main compartment again. Just added woodchips to the tray a couple of times, which doesn't appear to have made a noticeable dent in the temperature.

Peak-to-trough temperature swings average around 15 degrees. There appears to be about a 1.5 degree difference between the turn-on and turn-off temperatures for the element. The element stayed on for 6-7 minutes at a time.

This was all done with the internal temperature probe, which I realize isn't great. I just wanted to get an idea of how the control worked so I had an idea of how to emulate the default control. The voltage-to-temperature conversion was done via a regression on a few dozen readings I made last weekend (simultaneously reading multimeter voltage on the wire and temperature on the controller display). It probably isn't super accurate, but a linear regression did fit the data quite well (R2=0.9985).

So there's definitely some momentum to the temperature after switching the element on or off. How close I can keep it to the set point is going to depend mostly on how often I'm willing to switch the element relay on and off. I don't know how many actuations the stock relay is rated for, but I can certainly do better than a cycle every 15 minutes without stressing it.

This week I'll be working on a first crack at coding a better control algo and we'll test it out next weekend, weather willing. Hopefully I'll have my new temperature probes calibrated by then so I can use one instead of the stock probe. Stay tuned.

 

[dward51]

Actually for BBQ, that 15* swing is nothing to worry about. Remember, it is 15* total, and not 15* above plus 15* below (or 30* total swing). So you are basically holding within +/- 7.5* of set point. I'll take that any day!

To smooth it out you are going to have to fiddle with the coding. You want to anticipate the element shutdown sooner, and anticipate the element turn on sooner. Looks like the fan was either on or off, with no graduated "pulsing of the element". If you could factor that in (pulsing with variable rates), you should be able to hold an even more precise band.

All said, this looks like a great job!!!

 

[daveomak]

I may not be able to understand the wording... but... I can read charts... You are on the verge of a great thing for your controlled smoker... congrats....

 

[tallbm]

Very interesting report.

You guys are much more well versed in this arena than I am but I agree that figuring out a better interval to turn on/off the element without hammering the Relay will/should get you much closer to holding near the set temp.

I look forward to seeing what you figure out :)

 

[Fornax]

I'm taking a brief break from this project until I get a mailbox mod sourced and installed. I did some thinking and figured that the long cycle times on the stock controller may be required to generate good smoke from the chips. Doing shorter cycles may therefore have an adverse effect on smoke production. I've been thinking of doing the mailbox mod "someday" anyway, since these things need to be pretty hot to generate good smoke. I figured this was as good a time as any. So this is sort of on hold until I can source a reasonably-priced non-galvanized mailbox (or other appropriate enclosure). Will update when I pick it back up.

 

[dward51]

A galvanized mailbox is fine for a mailbox mod. There is no way you will ever reach high enough temps in that mailbox to have "metal fume fever" from the galvanized coating. Plus the smoke will seal it even more after the first time you run the pellets. Galvanized metal is only a problem in higher temp environments. Just go with the el-cheapo metal Home Depot mailbox (but don't use plastic). Even the painted interior kind will work.

 

[Fornax]

Yeah, I've read through a number of the threads on the subject. Under normal conditions I'm sure it'd be fine, but I'm less sure about how a pellet flare-up would affect things. I've decided to play it safe.

 

[tallbm]

I use a wall hanging mailbox to make an all in one easily removable mailbox mod. The AMNPS fits it just right. So don't rule that option out if you can find a used one or one on craigslist or you want to pilfer one from an old abandon house :P

2j17cdd.jpg

• tallbm
• Jan 8, 2017

2zflemo.jpg

• tallbm
• Jan 8, 2017

pACE3-15285558enh-z7.jpg

• tallbm
• Jan 8, 2017

33fbwvn.jpg

• tallbm
• Jan 8, 2017

 

[Fornax]

I use a wall hanging mailbox to make an all in one easily removable mailbox mod. The AMNPS fits it just right. So don't rule that option out if you can find a used one or one on craigslist or you want to pilfer one from an old abandon house :p

I saw your thread! I actually found a plain aluminum box that's roughly the same size as that mailbox and not too horribly pricey, so mine is probably going to end up looking similar to yours. I'll probably try for a slightly longer duct, though.

 

[johnmeyer]

I'm taking a brief break from this project until I get a mailbox mod sourced and installed. I did some thinking and figured that the long cycle times on the stock controller may be required to generate good smoke from the chips. Doing shorter cycles may therefore have an adverse effect on smoke production. I've been thinking of doing the mailbox mod "someday" anyway, since these things need to be pretty hot to generate good smoke. I figured this was as good a time as any. So this is sort of on hold until I can source a reasonably-priced non-galvanized mailbox (or other appropriate enclosure). Will update when I pick it back up.

Yes, you are absolutely correct. There have been dozens of posts about this very thing in the past. Each of these posts usually begins with the OP worrying about the "large" temperature swings in the MES. They want to know if it is defective. Those with technical skills, like yours, eventually decide to develop a PID controller (or buy one) and use that to get near-perfect temperature regulation.

However, as you have correctly concluded, the OP usually realizes, as you have now done, that the temperature swings are not a design defect in the MES, but are intentional because otherwise, once the smoker is up to temp, the heating element would never stay on long enough to ignite any new chips you add.

The solution is to add the external smoking mod, as you are now doing.

But then you have to ask yourself, does my food really care if the temperature bounces around a bit? If I get perfect regulation, will my food taste any better?

What usually happens next is that someone suggests that the OP do these same tests on their kitchen oven. This is usually a big eye opener because when the person does that test, they find that the regulation on their oven isn't all that much better.

However, in both the case of the MES and also your kitchen oven, the thing you do need to worry about is the controller's ability to repeatably maintain the average temperature you set. So, if you set 225, you want a controller that will oscillate between 210 and 240, averaging out to 225. And you want it to do that same thing every time you use it.

I'll bet that if you cooked using that 210-240 smoker, and then did an identical smoke using a PID controller that kept the temp within two degrees of 225, you would not be able to tell the difference.

[edit]I forgot to provide a link to my post about an alternative enclosure to the mailbox:

Mailbox (popcorn) Mod
https://www.smokingmeatforums.com/threads/mailbox-popcorn-mod-with-a-small-twist.258297/

Advantages of the popcorn tin compared to a mailbox:

- It is made of food-grade tin, so no worry about zinc plating, plastic parts, etc.
- It is much cheaper: $14 (see my post for details) or free if you already have one
- It has more surface area so it condenses more creosote
- It is easier to clean

 

[Fornax]

Yeah, I hear you. The real reason I'm replacing the controller is to get remote control and monitoring of my smoker's progress (without shelling out all the extra for having it built in). Improving the temperature consistency is more of a "bonus" feature that came along more or less for free (as long as I would have done the mailbox (or whatever) mod at some point anyway). But yeah, I don't expect it'll be a whole new level of smoking performance. As my wife will happily tell you, any project I take on tends to get a bit over-engineered.

Thanks for the link to the popcorn tin thread! That's a great idea. I'm going to look into that too.

 

[tallbm]

Also with a PID you reduce the chance of rendering out fat on meat smokes like Bacon or Sausage. This was a major factor for me. This is why I always say going the PID route depends on what you intend to smoke. If no bacon, sausage, or sandwich meat loaves then the MES swings won't hurt anything.

My PID needs were to eliminate the temp swings to avoid issues when doing sausage and bacon as well as finally be able to hit a top temp of 275F. Max mine would hit on a good hot TX summer day with nothing inside the smoker was 270F :(