Jump to content
jheissjr

Making a 3 Axis Thermopile Sensor Module

Recommended Posts

Under Mr. RC-Cam's suggestion I have started this topic in its own thread...

The topic is making your own thermopile sensor module instead of buying an FMA module. The module would be like the FMA module except it will use three axes instead of two.

The module will be used to make an artificial horizon in real time on a ground PC through an RF telemetry link.

My main goal is to have the software self adjusting to the non-constant IR levels encountered during flight. To accomplish this self adjusting, I was wondering which would be the bests choice.

1) Have 3 voltage outputs like the FMA does, one for each axis (3 total outputs)

2) Have 6 output voltages, one for each sensor (6 total outputs).

3) Or maybe it does not matter

Thanks to Mr. RC-Cam, we have the link to the FMA schematic to use for reference:

http://www.uavs.net/schematic.jpg

Edited by jheissjr

Share this post


Link to post
Share on other sites

Mr. RC-Cam suggested number 1 to me earlier:

(1) Fewest required ADC inputs, inherent common mode rejection and ratiometric operation.

(2) 2X as many ADC inputs and instrumentation OpAmps. No common mode rejection.

I would definitely have to agree. Only my question is, would it still be possible to have the software self adjusting to the varying IR levels without having the six individual sensor voltages?

Share this post


Link to post
Share on other sites
Only my question is, would it still be possible to have the software self adjusting to the varying IR levels without having the six individual sensor voltages?

Cyber-flyer explained how to solve for that in the MAHI thread. It is sqrt(X^2 + Y^2 + Z^2). No need to measure six inputs -- three Co-Pilot configured thermopile pairs are fine.

Share this post


Link to post
Share on other sites

I read all 18 pages of the MAHI topic but I stilled didn't see that. :rolleyes: Sorry.

In the FMA schematic, is that a 500 ohm feedback resistor. If it is, wouldn't the gain equal 500/1.1M which is very very small?

Share this post


Link to post
Share on other sites
I read all 18 pages of the MAHI topic but I stilled didn't see that.

Buried in there are some good tidbits. Just don't blink while reading them all. :)

In the FMA schematic, is that a 500 ohm feedback resistor?

Oops, Looks like they goofed up the drawing. The 1M and 500 resistors need to be swapped.

Share this post


Link to post
Share on other sites

I thought that was a mistake :)

Along the same topic...

I know the Seagull system has an artificial horizon in their software. How do you think they are doing this without using thermopiles.

Share this post


Link to post
Share on other sites

I'm a out of touch with what their current system can do. I had no idea that Seagull had a "real" artificial horizon. Do you have a link to the feature's details?

If I had to guess, I would say that they are deriving data from the G-force sensors and then offering it as simulated horizon info just for the fun of it. But, hopefully I am wrong.

Share this post


Link to post
Share on other sites

I only see a sample screenshot there. It says "Artificial Horizon Simulation" which sort of sounds odd to me. I can't find any other mention of the feature.

Share this post


Link to post
Share on other sites

Mr. RC-Cam, I know you are using a 10 bit ADC for the MAHI and I was wondering if this was adequate or you if you would suggest using a higher bit ADC such as a 12 bit ADC if the option was available.

Share this post


Link to post
Share on other sites

Mr Rc-Cam,

is there any chance a sensor module built with 4 Thermopiles could replace the sensors of a Futaba AP-1 system? Ap-1/2 work by detecting light instead of heat

Do you know the resting voltage for level flight and max/min voltage that AP-1's sensors output?

Dimitris

Share this post


Link to post
Share on other sites

Mr. RC-Cam,

thank you for the quick answer. From what I've read, the Co-Pilots work the same way but on a diffrent principle (heat vs light). I'll try to measure the AP-1s sensor output and make a Thermopile sensor module for it.

By the way do you have an artificial horizon PCB left (PAL version preferably?)

Regards,

Dimitris, Greece

Share this post


Link to post
Share on other sites
I'll try to measure the AP-1s sensor output and make a Thermopile sensor module for it.

If the sensor module outputs a linear analog voltage then it should not be too difficult.

By the way do you have an artificial horizon PCB left (PAL version preferably?)

Yes, I do. The MAHI thread has the pricing details to the parts and built board. Just PM me if you are interested in getting one.

Share this post


Link to post
Share on other sites

What is the voltage output range of the FMA sensor? I am asking because I don't know if I can hook the sensor directly to an ADC input pin or if the range has to be scaled down in order to use the ADC's 4.096 reference voltage.

Edited by jheissjr

Share this post


Link to post
Share on other sites

The output range of the Paparazzi based sensor is about .5 to 4.5 volts. I don't think you would need the extra accuracy of the reference voltage for this sensor. I was planning to just input it into one of a PIC's 10 bit A/D converters using 5 volt supply voltage as reference.

Share this post


Link to post
Share on other sites

Thanks for the info. Also now that I think about it, you can also look in the Op amps data sheet to see what the maximum output is given Vdd.

Share this post


Link to post
Share on other sites

Mr. RC-Cam

For the MAHI, did you connect the FMA's X and Y outputs directly to the PIC's ADC pins? Did you put in a cap between the ADC pin and ground and/or a resistor in series to form a low pass filter (if so what did you set the cut off point as?)

Share this post


Link to post
Share on other sites
For the MAHI, did you connect the FMA's X and Y outputs directly to the PIC's ADC pins?

That is correct, no X/Y input filtering was used on MAHI.

Share this post


Link to post
Share on other sites

Forgot to ask :rolleyes: what did you use for the ADC's voltage reference. I assume it is Vdd which would give a range of 0v to 5v but just wanted to make sure.

Edited by jheissjr

Share this post


Link to post
Share on other sites

Thanks, I was just trying to get an idea of the resolution I will have using a 12 bit ADC (5/(2^12) = .0012v) compared to MAHI's 10 bit resolution (.003v).

Edited by jheissjr

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×