Atmel AVR + Motorola MPX Sensors approach - Notes and Discussions to the project

[Bob's57]

[MrHijet]

First of all:

The great green thing is a commercial development platform. This is not needed when I completed the whole little digital display. It will be a much easier and smaller amount of parts when I am getting forward.

First of all I will build a little board with an LCD display with 2 (or three) sensors. Only thing it will do is showing the pressure rates.
If that works fine, I am nearly done with the hardware part and the rest will be software.

My approach will an USB or RS232 (serial) interface to a computer, so that the calculation can be done at the PC/Mac/Unix-System by utilizing a calculation tool like Excel or OpenOffice Calc.

I have just limited time to play on this and I have to build my first flow bench to do real life testings with the water gauges AND the digital ones.

When I have the first milestone done, I will offer the layout and hardware plans at bruce plans forum. I hope there will be at least one or two additional people who could help me out in the future, but first of all, let me get to the first step and then we will see.

Cheers,

Daniel

[Tony]

Looking good Daniel.

If you can make a direct pressure reading "front end" that would be extremely useful for many automotive applications. Excellent for both data logging and diagnostic applications, as well as working with a flow bench.

[MrHijet]

Since some people ask what my approach could or can do, I will try to make a little overview:

Technically my approach is NOT a fully flow computer. In my first steps I will just offer a multi channel pressure viewer.
This means:
You have up to 2 or maybe 4 differential pressure sensors which will be displayed on an LCD output.
Not more, not less.

If thats working fine at some environments, I will enhance the system.
My Ideas are following ( and external impressions and ideas are very welcome):

1) Adding Temperature sensors. ( Very easy to implement in hardware and programming )
I already played around with a very tiny and laser calibrated temperature sensor. They are a bit expensive (aprox. 10-15 us$), but they
would fit virtually anywhere at the flowbench and it would be easy to add this info to the LCD display.
If there is a need of more than one temp sensor ? - No problem ... I assume that I could easily add up to 8 or more without major hardware changes.

1b) Humidity Sensor ? ( unsure of the hardware implementation work )
Do you guys thinks a humidity sensor would make sense ? I know buying one and having it sticked somewhere near the flow bench can be
fine, but eventually its still good to have it included ?

2) Computer Connection (Easy hardware, medium work on programming)
Both USB or serial connections are possible. I tend to the serial connection, because today USB to Serial adapters cost aprox. 5-10 US$ and include all kind of software drivers. So I don't have to worry about implementing a USB device and afterall there are driver issues in lets say Vista or MacOSX. Using a classic serial port is cheap on the cost side and on the programming (plus old computer always have COM-ports, but eventually no USB).

The computer connection could be used in the first row for sending the measured pressures (and temperatures + humidity(?) ) to the computer. The data then can be stored into Excel / Openoffice Calc or just printed on a big screen. I will at least build a receiver software, which views the incoming data and store it in an Excel/OpenOffice Calc compatible format. ( I am programming typically in Python, so it can run on Windows, MacOSX and Linux/Unix).

3) Calculating the cfms ( Huge work in programming, additional smaller things in hardware)
I am still not sure if it will be possible to calculate the CFM directly inside the microcontroller. Doing the calculations inside means not just number crunchings, but also far more interaction with the user.

Up to this point the programming is very easy to implement: 1. Measure Stuff -> Send to Display (+ serial).

At the point where I calculate, I have build in a little user interface, so that the user can choose (and set/create !) the orifice sizes. This also means additional buttons or a bi-directional communication via the computer connection.
Thats why I am not sure when or if I will get the this step. We will have to see how much time I will find to get into this.

4. Controlling the vac motors ( Don't ask my how much work that would be )
Since I already bought some nice power controls, I am thinking about letting the control of the vac motors been done by a microcontroller. I will investigate into this eventually later this year, but it's not impossible. The motor controls just need potentiometers which I could implement digitally. Since I already have pressure sensors the programming of such a digital control is no rocket science.
This would allow to just set a preferred pressure and the rest will be done by the microcontroller.

[MrHijet]

What you guys can expect is absolutely points 1 - 2. This will allow a fully automated measurement, as long as you have a computer.
Point 3 would be great, but I need much more time for this.
Point 4 is definetly far future, so don't expect too much on this.

Tony thought about using it with batteries. Thats a nice idea. I already played around with a 8-16V DC flexible power source, but eventually I should implement the power source which can even work in car environment. I assume that would be the most usable area when its away from the flow bench.

Any more questions ? Please let me know.

Cheers,

Daniel

[slracer]

Hi Daniel, Thanks for the long response. You and Tony are helping me learn more than I ever thought possible. I have some thoughts on your system which I will throw out for thinking. These are just MY thoughts and I am NOT an experienced user, so read with caution!

0) I would like to see 3 sensors so you could do as on the PTS system and read depression, flow, and have the third for a probe. I am now using water gauges and am planning on adding a 3rd manometer along side my depression manometer to do that.

1) I have a pitot type bench so temperature is required for my flow calculations, but it sounds like an orifice bench does not as they are "ratiometric" (if heat is not added between the 2 measuring points). I don't really understand Orifice Benches, so ???

1b) Ed (86rocco) had a comment that he had included a humidity correction but it was so small that he deleted it. It may not be worth the effort for hardware / software. Check here:


2) If the data is automatically input to a computer in spreadsheet readable format, that is fine!

3) Being able to read CFM's sounds like a great idea (at first), then I think that on a water gauge you are reading either % of flow for an orifice bench (which you then calculate or approximate in your head before moving on to the next test point) or read a depression (which you then calculate ....). When I have a complete set of results, I look at the data and resulting flow curves and take the time to think about it. It may be more useful in a probe application, but I haven't done that so again???

Your work looks terrific! I am enjoying reading your posts and your "diary". Very educational for me! -- Doug

[MrHijet]

I had to invest a bit ... just ordered a new JTAG/ICE mk2 programmer, because I invested too much time into figuring out why sometimes the old programmer works and sometimes not.
Additionally I ordered a 10 Mhz digital osziloscope and a 500 Mhz, 34 pin logic analyzer.

Hope the stuff arrives next week, so that I can get forward again.

Daniel

[MrHijet]

New JTAG/ICE mk2 programmer arrived.
Tomorrow my new logic analyzer should be arrived too, so I just have to wait for the digital oscilloscope

Since I have some free days next week, I hope I can get forward in the project.

Daniel

[MrHijet]

Now I am fully equipped

My 2 channel oscilloscope, 34 port logic analyzer and my microcontroller programmer arrived and are working fine.

I already found an issue with my SPI clocking (Thats a serial connection between the microcontroller and the 16bit ADC). I am pretty sure I am getting forward now

Daniel

[MrHijet]

It's a long time that I haven't updated my project.

I had to do some other (real life) projects, so this project got stuck a bit.

I am currently rebuilding the setup. I will use a LTC1290 instead of the MX7705/AD7705 analog-digital converter.

This will allow me to utilize 8 channels of analog input connection with a resolution of 12 bit. Using the new LTC1290 the microcontroller be a bit more versatile and we can easily add more sensors or switches, if there is a need for it.

I will continue working on the measuring and data collection first. When there are the first field experience with bruce's PTS digital sensor project, we can decide again which Motorola MPX sensors would make the most sense.

Cheers,

Daniel

[MrHijet]

It is a just a sad update, but I haven't had time to work on this project. I will continue this month, as I have completed several other things, which had a higher priority.

Cheers,

Daniel