reyh
New Member
Posts: 37
|
Post by reyh on Jan 9, 2016 9:35:07 GMT -8
I started designing a combustion analyzer cheap. This project will maybe interest some of you, and I have a few questions for those who own (Peter, you seem only). I would like to find people with whom to exchange on the subject.
For now, it remains rather theoretical. An Arduino board serves as an interface between a PC and the sensors. Opendaqcalc is a macro that can retrieve and present data in the spreadsheet libreoffice calc. I tested this with ordinary sensors, now I expect to receive the good to continue. I begins with a CO sensor and a thermocouple to the temperature of the fumes. So I have for € 150-200, add the O2 sensor will cost € 100 more and is reliable only 2 years ... it will be easy to add when the unit will be developed.
We'll have a case. The sensors do not withstand the high temperatures, they must not be directly in the flue gas, unless the thermocouple. I think to a kind of copper tube cane to suck fumes and cool, with the thermocouple included. In the box: the components and a micro fan for drawing some smoke. There, I think of two nuisances: condensation and particles. Condensation is not necessarily a problem is to see, and I have several ideas. For particles: a filter, but what kind?
Peterberg, your knowledge will be very useful to me because I have never seen such a machine, and it seems that you know by heart. Do you think that this project is relevant?
Here. It is a project started but there is still much to do. Any suggestion is welcome. Thank you.
|
|
|
Post by peterberg on Jan 9, 2016 11:45:30 GMT -8
Yes it is. Most analizers use a small vacuum pump to bring the gases to the sensors. The Testo 330-2 which I have here is equipped with a temperature sensor and a thermocouple, a CO sensor and an O2 sensor the rest of the numbers is calculated. So it can be done, you are in the position to develop a cheap DIY combustion analizer. And yes, there are filters between the probe and the cells but nothing very complicated. Important thing is you need to implement a condensation trap for the moist which inevitably will be there.
|
|
reyh
New Member
Posts: 37
|
Post by reyh on Jan 9, 2016 12:53:15 GMT -8
Thanks Peter,I am relieved that you found it feasible. By reading you I realize a pump will actually be more effective than a fan in this configuration.
For the condensation i hope the copper tube do the job, with sort of siphon to collect the water. Did you know how it work on yours?
About the filters, do you know if that's particular materials? I wonder about the chemical reactions that could distort the measurement.
|
|
|
Post by DCish on Jan 9, 2016 21:20:03 GMT -8
Reyh, I am following your exploration with great interest. I started only briefly down this path before I realized that my programming abilities are far too rudimentary to consider being the first one to do this. However, if you succeed I may have just enough skill to follow in your footsteps. I wish you luck, and eagerly await the outcome of your experiments!
|
|
|
Post by independentenergy on Jan 10, 2016 9:56:54 GMT -8
Using Arduino to create a program to optimize the operation of the bypass, the ducts of the air distribution in my stove, I also provided a display and a sensor for carbon monoxide in the environment, I have now ordered also an IR sensor to detect the flames and perhaps able to build a door with primary air regulation autonomous that adjusts according to the flame. You could make a measurement with multiple sensors to get an overview of the combustion, what I find difficult is to create a chart for the reading over time, you should use Processing ..
|
|
|
Post by Karl L on Jan 11, 2016 13:58:13 GMT -8
I may be able to help with some of this. I am an embedded systems engineer (electronic and software control systems), and I do quite a lot of instrument design. My company ( Micro-Robotics) makes and sells general purpose controllers (like the Arduino, but better ) They are a bit more expensive than the Arduino, so you probably wouldn't want to buy one, but I may be able to help with general electronics and software advice.
|
|
|
Post by independentenergy on Jan 12, 2016 5:49:45 GMT -8
I may be able to help with some of this. I am an embedded systems engineer (electronic and software control systems), and I do quite a lot of instrument design. My company ( Micro-Robotics) makes and sells general purpose controllers (like the Arduino, but better ) They are a bit more expensive than the Arduino, so you probably wouldn't want to buy one, but I may be able to help with general electronics and software advice. you are welcome
|
|
reyh
New Member
Posts: 37
|
Post by reyh on Jan 17, 2016 7:48:09 GMT -8
Indipendentenergy, For the chart, you could try to use "Opendaqcalc" macro, with openoffice. As you already have an arduino and the sensor connected on, you just have to print their values on the serial, it's easy to program. I don't find anything in english, the original page is french. If you are interested i may help you to translate. Hi Karl, your skills will be appreciated . I'm still waiting for the component, so nothing new at this time. I'll keep you informed.
|
|
derek
New Member
Posts: 14
|
Post by derek on Feb 12, 2016 15:53:18 GMT -8
Anyone use automobile O2 "lambda" sensors? These are cheap, designed to operate at over 600°C, and easy to interface with their low voltage DC output.
|
|
derek
New Member
Posts: 14
|
Post by derek on Feb 12, 2016 15:58:12 GMT -8
Also wonder if pressure/vacuum measurements from various areas of the system are interesting?
|
|
stoker
Junior Member
Posts: 61
|
Post by stoker on Feb 15, 2016 14:14:58 GMT -8
The cheap lambda sensors are only good for measuring low concentrations of oxygen. The "wideband" ones are a bit more costly, but still only around £50.
Carbon monoxide alarms can be had for less than five pounds (or Euros or dollars), with measurement typically up to 1000ppm, but with slow reaction times and I wouldn't trust them to be very accurate, and the readings might be affected by temperature and by other flue gases. Still could be useful though.
And yes, pressure measurements could be interesting.
|
|
derek
New Member
Posts: 14
|
Post by derek on Feb 18, 2016 6:43:44 GMT -8
The cheap lambda sensors are only good for measuring low concentrations of oxygen. The "wideband" ones are a bit more costly, but still only around £50. Carbon monoxide alarms can be had for less than five pounds (or Euros or dollars), with measurement typically up to 1000ppm, but with slow reaction times and I wouldn't trust them to be very accurate, and the readings might be affected by temperature and by other flue gases. Still could be useful though. And yes, pressure measurements could be interesting. Obviously the Lambda sensors are optimized for automobile exhaust and I have no idea if these ranges are comparable with rocket burn conditions. Bosch and others sell custom interface ICs for Lambda sensors with conditioned (linearized?) analog and serial-digital outputs. Lambda sensors are differential, measuring both the exhaust and surrounding outside air. As I understand it, commercial flue gas analyzers use laser spectrophotometry to identify various gas components.
|
|