Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
All-In-One Mini Q-Control Unit
using SPI with max6675 devices really makes it simple to add more could go over 10 easily.... HOWEVER, there would be a lot of coding required and changes to many webpages to fully support all the extra probes.

but, if you were just going to use MQTT and post the probe values to a server, then the work would be a lot less. (not using the local webpages/server for all those features)

I don't see a lot of people talking about the limit of 3 food probes..... would you want to put one in each steak Smile
hhmmm, 10...yeah that's the ticket. LOL

so let's see, a list of some of my lesser traits  Dodgy  OCD, anal retentive, oh and control freak, yea that should do it   Big Grin

I know that the master competition smoker guys out there probably wouldn't even consider using a temp control device as they have learned the way of the 'smoke'. But it may be possible that there are padawans that want to have more detailed info during their cooks to ensure success.  Or like soon to be me, have a very large smoker with different types of meat during a cook and want to know what each type of meat is doing.

My soon to be smoker is going to be rather tall, at least 4 racks (and considering 5).  So I'd want to have more than one pit temp sensor, like one on the top rack and as well as one on the bottom rack(possibly one on the middle rack).  And if the smoker is fully loaded, I'd have 4 pork butts, 2 full packer briskets, 20# spatchcock turkey, 9 racks of ribs (sitting on edge), and some sausage, fatties, and vegies.  Therefore I'd want to have a probe for each meat type.

Then my anal self might want to know the incoming air temp, exhaust (sorry I'm a car guy so...) temp, air flow speed, and maybe even have an environment sensor (temp, humidity, barometer).  I know crazy huh   Confused Huh Exclamation  Blush 

Good Grief Charley Brown

This doesn't mean I'll actually do this, but is nice to have the flexibility if needed  Wink
(03-23-2017, 06:51 PM)NailBuster Wrote: HOWEVER,  there would be a lot of coding required and changes to many webpages to fully support all the extra probes.

True, it may initially.  Please do not read the following and take offense to it, I am just having a friendly discussion in hopes to better understand and possibly contribute to the project.  I do understand that many hours (months) and lots of dedication has been spent on this and other similar projects.

I haven't looked at any source code as yet so I am just guessing here, that the code is expecting a fixed number of inputs (probes).  And I do understand that it is easy to code for this especially with just four inputs to deal with.  But isn't that limiting?  Yes there may not be much demand for more probes, but maybe that is because the majority of the temp controllers on the market only offer up to four probes, so people are just accepting of that fact.

I would think that the number of probes could be a variable input, have the system sense the number of probes in use and adjust the outputs (graphs) accordingly.

I would say that in the user config settings, have a drop down option to tell the system which probe is being used for what; e.g. main pit temp, other pit temp, meat temp, other/etc. again providing flexibility (no hardcoding what probe 1,2,3,4 are to be used for).

In the software you don't have to change any of the calculations nor output controls as you would still use just the main pit temp and cook setpoint temp for controlling the damper/fan and use the meat temps for providing details on when the meat will reach it's setpoint and the other temps would just be informational either graphed or in a table or something.

I have found in my years in IT, that not hardcoding items allows the system to have more flexibility and less headaches down the road.  It may be abit more work upfront, but you'll have a much more robust system in the end.

Just my two cents  Angel

(03-23-2017, 06:51 PM)NailBuster Wrote: but, if you were just going to use MQTT and post the probe values to a server, then the work would be a lot less.  (not using the local webpages/server for all those features)

OK, so one more comment, and believe me when I say this, I am not standing on a soap box.  I'm just providing my thoughts in hopes for good discussion.

So I do have a bit of history with web security and securing systems for the big boys, which is why I am not a fan of cloud services.  Yes I know they are the cool new thing, but just look at the recent DOS on DNS, they made use of IOT devices.  Yes, in certain situations it does make sense to use cloud, and yes we are not holding nuclear codes in our smokers.  Once I have, umm Confused  I mean a hacker has access to an IOT, then they are on your network and can then do all kinds of nefarious things.  There are those of us security geeks who are uber paranoid (because we can) and have isolated most of our network devices from each other (which then becomes a pain in the arse to communicate across the LAN). Uhg

I am not recommending military security here (load know they are full of holes lol), but suggesting to only provide MQTT as an option for those who want to make use of it. I personally would rather connect to a local host, the temp controller, to get the web services.  So I am not sure if the Arduino platform can provide a robust web server or not (I'm about to go do some more reading on this) which may be why HM uses RPi.

Please understand that I truly love your design for a temp controller, breadboarding together already built components makes building a DIY controller much more accessible to those who are timid about soldering electronics (like me).
whew, just finished reading Arduino Ethernet Shield Web Server Tutorial, I need to lay my head down now.
always good to get other viewpoints and such in a conversation....I'm an easy-going guy, so I take no offense with any ideas/suggestions...

I've been involved in 'huge' software projects that span the globe with thousands of sites/terminals....and I know that 'future-proofing' is very important, in fact I really like thinking about the 'what-ifs' in software design. HOWEVER, with the PID (atmega temperature controller) running on the atmega you have to realize one thing....32KB of prog space and 2KB of ram... Yup, this single webpage on the forum is almost 70KB, that's twice what can be stored on atmega!!! So with these micro procs you really have to hard-code limits and worry about each and every BIT, not just BYTE!! If it was using a different proc then you could easily support 99 probes..... just think of this... it has a 64 BYTE serial buffer... so really your serial com messages should fit in that... it limits your ability to be flexible like other software enviornments....

As far as IOT, i agree security is going to be a 'big' thing moving forward. If I had a choice though between running on a PI or an esp8266...I'd choose the esp8266. Also, you could easily run MQTT server on a local-lan PI (or any pc) if you don't want to use a cloud. I'm still a believer in the cloud and keeping the qcontrol as 'dumb' as possible, but with more powerful esp8266/webservers, options are always good for different scenarios....

I would avoid learning too much about the ATMEGA, and focus on more 'modern' procs....the esp8266 is a pretty powerful chip even without the WIFI enabled. I was even thinking about using two of them in this project (one handling web, one handling the PID).... the esp32 is still a bit too new, but that looks like a nice proc as well.
Good deal David, glad to know. The more I dig into this stuff the more I feel like a noob, and I really don't intend upset anyone, I'm just trying to learn.

Please let me know if you want to move this to a new thread.

So ESP8266, OK. Why the Wemos, I see there are other brands that offer the same design (at least I think so). I'm not only asking to understand but also to figure out what parts I should consider buying so I can start playing around with this stuff. I've downloaded the Arduino IDE and pulled down your source code as well as HM's. I've got lots of learning ahead of me Wink
Just to follow up, NO I am NOT going to do any programming, you definitely don't want me doing that, it would be UGLY! and it's not my thing LOL. But assembling parts on a breadboard, let me at it!!! I'd like to find parts that are readily available even if they are a couple $ more rather than wait on for those slow boats (I too have had not so good experiences).

wholly smoke batman, check this link for 8 glorious connections
The WeMos is a real nice 'package' for the esp8266..... there are a few other esp8266 boards like the NodeMCU. They all are using the same esp8266 chip so it really is just whatever one fits your needs.

wow that 8-port max board is only has one max-chip on it. It really is just multi-plexing one chip with 8 header connections. I'm not sure how good or bad that is? It wouldn't be a simple task to support as it has a special library and can only read one probe every 0.125 seconds.....strange indeed....
yeah I see that now, only one max chip, so likely not good to use.

I would like to rig up (for testing) some components based on your design. What ESP8266 board do you recommend that I can easily get here in the states (not slow boat). I'm ok if the part is a couple $ more.

Same ? for the arduino board. mini pro?
max31855 board?
I'll also need to get one of the breadboards kits (recommendations) for hooking everything up.
What other parts will I need?

To let you know, I'll also be getting some arduino stuff to build an outdoor temp/humidity dial, as a learning project (already been researching that). Ever since a couple weeks ago when started down the road for an all TC controller, I have gotten super excited to learn more about the arduino platform. I had already been playing with the RPi, but only because it supported linux which I am familiar with and could build apps around that.
I would stick with the wemos d1 for the nodemcu. That or the nodemcu....

as far as where to order, I don't know any USA seller as I've only ordered from ebay/ali (china).... I'm sure you can search for those boards and see if there are local sellers from ebay....

Depending on your timeline, I'd stick with max6675 (I don't know when I'll be supporting the max31855 board yet)....

As far as parts, I'll start a new thread with a 'shopping-list' suggestions with some examples....

Getting into the arduino world is great fun! You will soon be trying to automate all sorts of things around your house.... RPi are OK, but they are power-hungry and overkill for many simple iOT devices around the home. The wemos has some simple shields for the esp8266 for temp/humidity. If you don't need wifi then the arduino-clone 5v nanos are great procs as well....

Forum Jump:

Users browsing this thread: 2 Guest(s)