Internet/WiFi Connected Stove Monitor

Thinking about designing a wireless stove monitor and doing a small run of them if anyone else is interested. The idea would be that you would be able to check on your stove from anywhere through an computer or web capable phone.

Current Features

*Multiple temperature sensors (0-500C thermocouples) and would include two of the sensors - Would be able to have more temperature sensors via a prototype area on the PCB
*Two thermistors for indoor and outdoor temperature readings (or wherever you put the sensors)
*CO detector
*Backlit LCD display
*WiFi connectivity for monitoring
*The ability to set 'safe ranges' for each sensor and would send an email and/or sound audible noise if an alarm was triggered
*Web page for viewing current values, historic graph of previous values (real time clock used to track date/time), exportation of historic data, and configuration
*Low voltage outputs for controlling external devices based on sensor values(IE a power transistor to turn on/off a relay for a fan/blower/stove shutdown) - This would be on a prototype area on the PCB for the end user to develop the circuit
*Inputs for external contact closures - This would be an option for the user to wire on the Prototype area. If there is enough interest in this may pre-wire 1-2 inputs for this

Current part costs would roughly be $150-175 for the above features. Most likely would provide a 'kit' option as well where someone could just purchase the base PCB and components and then add on only the sensors they wanted

A few features that have been requested but I haven't figured out yet...

*Monitoring coal level in hopper - I don't have a hopper so don't really have too much of an idea on the best way to sense the level. Possibly distance finding with ultrasound?
*Draft monitoring - The sensors to work in the range we need with the accuracy required look to be prohibitively expensive and would probably have issue with flue gasses degrading the sensor over time. Another option would be to use a typical dwyer water filled manometer with sensors to read the water level, probably wouldn't look all that great but could be done for reasonably low costs and would eliminate the flue gas issue Waiting to hear back on pricing on a sensor. If pricing is reasonable it is in a package that would be suitable as an add-on module.

Does anyone have any interest in such a device?
Does anyone have any suggestions for any other features or how to implement anything?
This is just in the planning stages in my head at this point but will most likely be making one for myself regardless, falls into the cool stuff that I want to build category. I'll keep this post updated as people make suggestions and feedback.

I have great interest in such a device!!!

I currently have an x10 cam focused on my stoker and another on a carbon Mono tester to see levels.

I also have the stoker plugged into an X10 module.

All this allows me to view my stove, view the carbo tester level and turn off the stove if need be all via the internet.

But this type of device would be much much better!!

I am interested in one of these. Also just a thought but could you leave a spot blank to add a electronic manometer in the future for those who might want to spend the money ? That way you could focus on all the meat of th project and leave something's for additions.

Would any sensors from automotive engines work for you? I have something called a "scanguage II" in my truck that plugs into the obdII sensor and gives you all the trucks sensors . It is sort of what you will be doing here.

Smokeyja wrote:Also just a thought but could you leave a spot blank to add a electronic manometer in the future for those who might want to spend the money ?

The board would have a area for prototyping such things. I'd be willing to develop the software to read these sensors if someone wanted to do this. The source code would also be available if anyone wanted that as well.

Smokeyja wrote:Would any sensors from automotive engines work for you? I have something called a "scanguage II" in my truck that plugs into the obdII sensor and gives you all the trucks sensors . It is sort of what you will be doing here.

Unfortunately the automotive pressure sensors I've looked at so far really wouldn't do the job. Need something that will be accurate in the 0-0.2 inch of water(0-50 Pa, 0-0.00725 PSI, 0-0.0005 BAR, 0-0.38 mmHg) pressure range. I'd love to find an automotive sensor to do this as they are generally inexpensive, easily obtainable, and pretty resilient but I haven't been able to find one that would work.

I'm interested as long as the price range stays around the OP. I don't need to know Hopper level as my Hopper empties when I shake down. I do however like the "turn on a fan or blower" feature. It'd be perfect for turning on the draft inducer.

I've considered building such an instrument myself, but for boilers. In fact I've purchased several microcontroller kits with that intent. Like many of my projects I never got it finished.

Here's what I would add to your list of desirable features.

1. Real Time clock
2. Removable data memory storage
3. Ethernet connector
4. Data stored in DOS CSV format for easy import into Excel
5. Add thermistors as temperature sensors (much less expensive)
6. Outdoor temperature sensor
7. External contact closure inputs
8. Ability to calculate delivered BTU/hr.

The cost for any instrument with these features would likely be far above what anyone would be will to pay. However it's a task that a hack of an Android OS tablet could do. Of course you don't get the sensors, but the display and web connectivity are there. I bought a 6" Cruz tablet for $64 at Big Lots a few months ago. Suggest you take an open source Android approach. Sensors will cost more than everything else.

I'm On Fire wrote:I'm interested as long as it isn't too expensive.

Here's a breakdown of the price so far to show what's driving it up...

CO sensor $6
Temp probe $15 each
IC to convert thermocouple to digital signal $15 each (This should be possible to eliminate and just use the Analog to Digital function on the microcontroller)
LCD $30 (4x20 backlit, SPI/I2C interface)
WiFi Chip/Antenna $26
microcontroller $12

Then there is the various passive components to connect, power, and run all of that, printed circuit board, Case, connectors for the sensors.

Yanche wrote:I've considered building such an instrument myself, but for boilers. In fact I've purchased several microcontroller kits with that intent. Like many of my projects I never got it finished.

Here's what I would add to your list of desirable features.

1. Real Time clock
2. Removable data memory storage
3. Ethernet connector
4. Data stored in DOS CSV format for easy import into Excel
5. Add thermistors as temperature sensors (much less expensive)
6. Outdoor temperature sensor
7. External contact closure inputs
8. Ability to calculate delivered BTU/hr.

The cost for any instrument with these features would likely be far above what anyone would be will to pay. However it's a task that a hack of an Android OS tablet could do. Of course you don't get the sensors, but the display and web connectivity are there. I bought a 6" Cruz tablet for $64 at Big Lots a few months ago. Suggest you take an open source Android approach. Sensors will cost more than everything else.

1. RTC would be utilized in order to get the historic data logging
2. Could be implemented, might be more cost effective to store on non removable storage and have a download option on the web interface. Also would add some complexity/cost of where to store the data when the card was removed to move the data to a computer. I'll look into it though.
3. Would probably stick with WiFi just to keep it simple. Have a feeling WiFi would be more desired then physical cabling so the first one of these will be WiFi only. If there is enough interest to make a bunch of these the second batch could have a wired connection instead of wireless, would probably reduce the price slightly
4. Would use a standard format that would be easy to manipulate/import be it CSV, XML or RRD (see RRDTool)
5. After briefly looking last night it didn't find any thermistors that would handle a large enough temperature range to handle the stove body and flue pipe. Also I've found thermistors horribly nonlinear in the past. If you have a part number of one that you think would work let me know and I'll take a look at it
6. Thermistor would work for outdoor/indoor temperature sensing. Would be able to implement without much added cost - Will add to features
7. Would be plenty of IO on the microcontroller for this. Out of curiosity what would you be connecting?
8. I would guess that there would be too many variables between stove to stove, house to house, etc to make a good calculation of this. I'm not a HVAC guy by any means so maybe this calculation would be easy. I just don't know

In terms of using a hacked tablet to make this work.. I can get the microcontroller, small display, and Wifi connection for a pretty low price, about $68. With the tablet I think you'd end up with an external controller anyways in order to get the inputs/outputs needed for all the sensors

RHarhen wrote:

I'm On Fire wrote:I'm interested as long as it isn't too expensive.

Here's a breakdown of the price so far to show what's driving it up...

CO sensor $6
Temp probe $15 each
IC to convert thermocouple to digital signal $15 each (This should be possible to eliminate and just use the Analog to Digital function on the microcontroller)
LCD $30 (4x20 backlit, SPI/I2C interface)
WiFi Chip/Antenna $26
microcontroller $12

Then there is the various passive components to connect, power, and run all of that, printed circuit board, Case, connectors for the sensors.

I edited my post earlier, probably too late and after you had replied. The joys of surfing the net through an Android smartphone. The price range you posted originally is perfect.

Could it be more of a modular kit? I know when I was building my Neon there was a device called a Megasquirt that controlled fuel with a turbo charged application. But it was a "pick-and-choose" device that the end user would purchase the components he'd need based off of his application. It was a means to keep costs down. The price of that unit ranged anywhere from $200 - $600 depending on what components you chose.

So, for a hand fed guy with the device you are willing to design and build wouldn't need an auto shut down or a co detector. So, having it so that the buyer could pick what he wants on it would be fantastic.

RHarhen wrote:

Smokeyja wrote:Also just a thought but could you leave a spot blank to add a electronic manometer in the future for those who might want to spend the money ?

The board would have a area for prototyping such things. I'd be willing to develop the software to read these sensors if someone wanted to do this. The source code would also be available if anyone wanted that as well.

Smokeyja wrote:Would any sensors from automotive engines work for you? I have something called a "scanguage II" in my truck that plugs into the obdII sensor and gives you all the trucks sensors . It is sort of what you will be doing here.

Unfortunately the automotive pressure sensors I've looked at so far really wouldn't do the job. Need something that will be accurate in the 0-0.2 inch of water(0-50 Pa, 0-0.00725 PSI, 0-0.0005 BAR, 0-0.38 mmHg) pressure range. I'd love to find an automotive sensor to do this as they are generally inexpensive, easily obtainable, and pretty resilient but I haven't been able to find one that would work.

I will do some searching and talk to my father about some older power plant equipment they might be wanting to get rid of.
I quite a few old Volvo gauges (VDO) usually and I will see if there is anything out there.

I'm On Fire wrote:Could it be more of a modular kit? I know when I was building my Neon there was a device called a Megasquirt that controlled fuel with a turbo charged application. But it was a "pick-and-choose" device that the end user would purchase the components he'd need based off of his application. It was a means to keep costs down. The price of that unit ranged anywhere from $200 - $600 depending on what components you chose.

So, for a hand fed guy with the device you are willing to design and build wouldn't need an auto shut down or a co detector. So, having it so that the buyer could pick what he wants on it would be fantastic.

At this point I'm trying to nail down what would be the 'basic' features of the device. Some of the stuff like the auto shutdown would be there just because the I/O is already there on the controller and having unpopulated space on the PCB really doesn't cost much/anything. If someone wanted to have a output like 'auto-shutdown' they would need to build/hook up a control circuit to the low voltage logic of the controller and make it work. The software on the controller would just be programmed in a manner where adding that 'module' would be possible. If someone wants a sensor or control that wasn't "supported" in the software I'd certainly be willing to make some one off firmware to control it.

The CO sensor itself is relatively inexpensive ($6) and I think it would be a meaningful feature at that price... Scenario in my head is you're at work and no one is at home, CO alarm goes off. Pets obviously wouldn't be able to get out of the house, but an alert email could allow you to call your local FD or neighbor and have them check it out hopefully saving yourself a really bad experience when you arrive home 8 hours later with the CO alarm still going off.

That being said, if you wanted a device as inexpensive as possible I think I would make a kit version available where you'd get the PCB with the surface mount stuff pre-soldered and whatever sensors you picked out. IE you might want to leave off the LCD display or something.

This is similar to what I bought for my stove - the controller is the same anyways.

http://www.hearth.com/econtent/index.ph ... ead/83902/

http://www.arduino.cc/

Wouldn't it be cheaper and easier to use one of these?

While a digital temp sensor with some programmable output is useful, I hope to achieve significantly more with this idea.

I've looked at some of the Arduino stuff and find the ATmega 328 to be a bit under powered for a lot of stuff. By the time you add on the required modules(USB/ethernet/etc) to get what you need you're well over a $12 chip from Microchip that's significantly faster as well. Also I'm very familiar with the PIC line from Microchip and have been very satisfied with them. They also can be programmed in C rather then Arduino's homebrewed language. I also have the programmers and hardware to deal with the PIC line already.


On a side note.. I may have found a pressure sensor that would work.. 0-2 inch of water, +-0.005 inch of water resolution and available as a through-hole DIP package so could easily be done as an option. Waiting to hear back on pricing. http://www.meas-spec.com/downloads/MS4515DO.pdf

WOW you guys just got me lost