Jump to content
Sign in to follow this  
Dimitris76

CPD4 and MAHI OSD

Recommended Posts

Thomas,

I am having your MAHI Hardware Revision B board and I need to use it with the CPD4 on a new unstable airframe.

On your datasheet you mention that this might be possible by not soldering (or unsoldering) the U3 Vref IC, but it wasn't tried by the time the manual was writen... Did you ever try it since then?

According to the OSD board's schematic, connecting the sensor in parallel will send the CPD4's reference voltage directly to pin 5 (RA3) of the PIC. Is this reference voltage compatible (3.3V) and does it work if the two devices (CPD4 and MAHI) have common ground?

Finally, do I have to also remove C2 or does it stay in place?

Thank you in advance,

Dimitris

Share this post


Link to post
Share on other sites

I never took the time to try it. However, it seems to me that sharing the reference voltage would be possible.

Here's what I would do.

(1) Use your voltmeter and confirm that the CPD4's reference voltage is 3.0 to 3.3VDC. This is pin 2 on the sensor's cable.

(2) Connect the CPD4 to MAHI, but leave off the Vref signal (disable the wire on MAHI J1 pin-2). That is, only wire J1-1, J1-3, and J1-4. This will use U3 as the MAHI Vref.

As an alternative, don't install the U3 Vreg IC. Be sure to connect J1-2 to the CPD4. This will use CPD4's internal supply for the Vref on MAHI.

Share this post


Link to post
Share on other sites

So, just to make sure I got it right, I have two options:

CDP4_MAHI.jpg

Option 1: CPD4's internal supply supplies reference voltage to sensor and U3 to MAHI. Obviously U3 stays in it's place.

Option 2: I unsolder U3 from MAHI board and CPD4 supplies voltage to both sensor and MAHI.

I looks better having common reference voltage for both devices (per Option 2) but option 1 seems to be much easier and I don't risk damaging the MAHI by unsoldering the component. Which one do you think has more chances working?

Also any thoughts how to parallel connect the sensor on the ribbon cable? I don't have a crimping tool for these 2mm pitch connectors...

Dimitris

Edited by Dimitris76

Share this post


Link to post
Share on other sites

Drawings look fine. I forgot to mention that C2 should remain in place if the Vreg is removed.

Which one do you think has more chances working?

I expect that they both will be the same in performance. Be sure to power the MAHI and the CPD4 with the same battery so that they both see on/off power at the same time.

Also any thoughts how to parallel connect the sensor on the ribbon cable?

That part is up to your imagination. I'd probably hard wire to them using a 3-wire servo cable or something similar. But, do what seems best for your project.

Share this post


Link to post
Share on other sites

Thomas,

does MAHI need a signal at it's video input in order to produce an image (the ticks and markers)?

I hooked it up on my transmitter (it's proven that it functions properly) and all I get is a black screen. I didn't have in hand the right connector to match my camera's harness - I'll buy one tomorrow - and I don't want to splice through it... So no camera was connected during this short test.

In addition I have powered the MAHI board with just 5volts. I know that in the manual you ask for 5.5 to 10.5 volts but that's all I have in this range and I figured that since you mention that at 4.9V the LED starts blinking, it would be ok.

What do you think the problem might be?

Dimitris

Edited by Dimitris76

Share this post


Link to post
Share on other sites
does MAHI need a signal at it's video input in order to produce an image (the ticks and markers)?

MAHI is a video overlay; it needs a valid video signal at its input. You must hook up the camera or other suitable video source.

There is also a PAL/NTSC jumper at location OP1. Be sure to set it for the correct video mode.

Share this post


Link to post
Share on other sites
Is 5v enough?

It might be too low. It was designed to work with 5.5V to 10.5V (2-cell LiPO).

If the battery voltage is too low you may experience inaccurately reported voltages. The "VDC: BAD!" message may appear as well. Other than that, it would operate at 5VDC.

Share this post


Link to post
Share on other sites

Sweet!

I don't mind about the voltage indication. What I am mostly worried is that U3 will not be able to supply the much needed 3.3 Vref... or that it might not work altogether.

By the way, when I checked CPD4's votage output it was EXACTLY 3.30volts!

Dimitris

Edited by Dimitris76

Share this post


Link to post
Share on other sites

Success!

Is there any way to use the sensor in the diagonal X position instead of the traditional + one?

With an external circuit between the sensor and MAHI maybe?

Dimitris

Share this post


Link to post
Share on other sites

It might be possible to create some sort of external mixer to allow MAHI to use X operation, but I don't have anything to share. The simplest workaround is to use the thermopile sensor in the traditional + orientation.

In case someone else wants to try sharing the sensor, please describe your chosen solution.

Plus, once you've had MAHI up in the air, please post some flight videos. :)

Share this post


Link to post
Share on other sites

Thomas,

unfortunately it's impossible to install the sensor pack in a + orientation on this aircraft.

It's neither a low wing nor a high wing one - the wings comes out exactly at the middle along the fuselage's height. If I place it for example on top or bottom of the left wing, fuselage is blocking the right thermopile - if I place it on the fuselage, the tail's vertical fin is blocking the rear thermopile. Under the fuselage is not an option either since it has no undercarriage....

I had your board for quite some time with this certain project in mind, but I haven't noticed this detail and now I am stuck... :(

I am wondering what does it take to make a "simple" mixer circuit with some logic gates - after all it's just voltage variations in comparison to Vreg right?

Do you have something else to suggest?

Dimitris

Share this post


Link to post
Share on other sites
I am wondering what does it take to make a "simple" mixer circuit with some logic gates - after all it's just voltage variations in comparison to Vreg right?
These are analog signals (0-3.3V), so a hardware external mixer would involve OpAmps rather than logic gates.

Do you have something else to suggest?

If it was me, I would first try it out with the CPD4 module mounted above the fuselage/wing on a short pylon tower to get it above the model. Despite the fear of obstructing its view, it still might work fine for you in such a location.

If that did not work out well, then perhaps unsolder the four thermopile sensors from the module and extend them as needed with shielded cable. This would allow placing them anywhere on the model. You should not need to remove them all -- just relocate the ones that cannot see a clear horizon.

Share this post


Link to post
Share on other sites

Thomas,

Tell me one thing though. Does this thing has it's sensors in pairs and outputs 2 voltages (pitch and roll) proportionally to the heat differece that reads across the two opposite thermopyles.

For example: assuming that Vreg is 3.3V do we get between Pin 3 (pitch) and Pin 1 (Ground), 1.65volts at level flight, less than 1.65 volts when the aircraft dives and more than 1.65V when the aircraft climbes up???

Dimitris

Edited by Dimitris76

Share this post


Link to post
Share on other sites

That is essentially how it works. There is some discussion about the Co-Pilot sensor buried in the original MAHI thread.

Share this post


Link to post
Share on other sites

I think I found the electronic solution to the problem!

If V1 is the first sensor pair's output, V2 the second pair's and Vref the reference voltage, then in diagonal mode the circuit should be able to modify the output so:

V1'= (V1+V2)/2 and V2'=(Vref+V1-V2)/2

where V1' and V2' should be the voltage pairs on the diagonal mode.

Now I just have to build the circuit.... I am guessing that a summer with a voltage divider will do the job for the first equation and a summer, a differential amp and a voltage divider will do the second one...

What do you think?

Dimitris

Edited by Dimitris76

Share this post


Link to post
Share on other sites

Doing it in the firmware would be sweet. However, if you want something now, then you will have to solder up those opamps. :)

Share this post


Link to post
Share on other sites

I have copied in principle the summing amplifier circuit that you can see here:

http://www.elfa.se/pdf/73/730/07301682.pdf

on page 9.

Here is my circuit

MAHIopamps-1.jpg

on this schematic I used the single LM741 OpAmp ICs from Eagle's library. What I want to find is a double opamp IC (easy) with a maximum Vout range as close to Vcc (harder...), so that I can power it directly from Vref and skip the +5V input to this extra circuit board. The reason I am using +5V is that these OpAmps have a Vout=Vcc-1.5, so I am guessing that powered with the reference voltage you can never actually get 3.3V at their output.

What do you think about the circuit? Do you have any specific component to suggest? - they are so many opamp ICs out there that I am lost..

Dimitris

Edited by Dimitris76

Share this post


Link to post
Share on other sites

http://www.elfa.se/pdf/73/731/07314321.pdf

what about this one?

They claim up to 4.95 volts at the output when powered with Vcc+ 5V and Vcc- 0V. Am I right to thinking that powered with Vcc+@3.3V I'll get up to 3.267 volts? Are these characteristics linear?

3.267 volts might not be the sensor packs max output at full deviation but it would be enough to move the marker all the way to the screen's side... right?

Dimitris

Edited by Dimitris76

Share this post


Link to post
Share on other sites

I'm not familiar with that IC. Nearly any single supply OpAmp that is spec'd as "rail-to-rail" should be fine.

The project is too long ago for me to recall the exact details, but the sensor's pitch and roll outputs certainly don't swing all the way to 3.3V. Your voltmeter will tell you all you need to know about such things.

Share this post


Link to post
Share on other sites

Thomas,

I just want to let you know that the circuit works PERFECTLY!

I used the TS912 (rail to rail CMOS dual OpAmp) from ST. The only difference from the above posted schematic is that I am using 11.1 volts (straight from my 3S) instead of 5V for the Vcc+. The reason is that even though I don't need more than 3.3V at the final output to MAHI, the added voltages at the OpAmp's output (Vref+V1-V2 and V1+V2) can be more than 5v before the voltage dividers.

The whole circuit is on a double sided PCB that measures approx. 15x15mm - I could make it even smaller but I had enough hard time soldering those tiny SMT resistors. I am just waiting the connectors from you to tidy it up!

I have only one problem now - I can't change the P CAL and R CAL values at MAHI's setup menu.... I can change all the other parameters (gain, filter, expo, AHI, voltage, message, etc) except these two. Any ideas?

Dimitris

Edited by Dimitris76

Share this post


Link to post
Share on other sites

That is fantastic.

Those two values are the calibration constants from the zeroing function. You don't directly change them. So, it sounds like everything is set to go.

How about photos of the cool mixer PCB? When do we get to see videos with MAHI in action?:)

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
Sign in to follow this  

×