How to connect FMA co-pilot

[1alonso]

Hi
I got my OSD, GPS, Head Tracker, camera, etc.
Also I have my FMA copilot ... Vova told me I need two FMA copilots for the artificial horizon to work... but I dont see any information or much information in how to connect these two copilots or a copilot and a z sensor?

How to program ... etc. Just information please


Thanks

[Alex Villa]

Hi
I got my OSD, GPS, Head Tracker, camera, etc.
Also I have my FMA copilot ... Vova told me I need two FMA copilots for the artificial horizon to work... but I dont see any information or much information in how to connect these two copilots or a copilot and a z sensor?

How to program ... etc. Just information please


Thanks
In the first page of the forum you can find all that information... just one page back.

http://www.rangevideo.com/forum/showthread.php?t=153
http://www.rangevideo.com/forum/showthread.php?t=147
http://www.rangevideo.com/index.php?main_page=product_info&cPath=49&products_id=135
http://rvosd.rangevideo.com/

[1alonso]

I have the FMA... but dont have the cable.. is there any guide of how to make this cable or maybe it can be purchased somewhere?
Thanks again

[Alex Villa]

Cable: http://www.fmadirect.com/detail.htm?item=1496&section=20
XY head: http://www.fmadirect.com/detail.htm?item=1778&section=20
Z head: http://www.fmadirect.com/detail.htm?item=1888&section=29

[elpimous]

Hello Alex,
I've soldered the resistor in a wrong place, on Z sensor !!! Why....don't know (silly boy i think !):(

in the wrong side, and the one nearest to center...poor of me !:p
I've de-soldered a light blue small resistor (500 ohm..I hope....don't know !)

my question : can I solder a standard 500 ohm resistor in place of those smallest pieces (hard to solder!) to repair my error....Cause 1: i'D prefer ! 2: don't know how to find them !!!!

Hope you'll be able to help me..........or i'll have to buy another one Z sensor !!

Big thanks Alex.
Vincent.

[Alex Villa]

Nope need to be another SMD resistor. Those 1K resistors have the label "102" on it, you can find it on all computer hardware, like CDROMS, Hard drives, motherboards, cards... IT si a widely used value so you will find it for sure. This thread can be really usefull http://www.rangevideo.com/forum/showthread.php?t=147

Pics please.

[elpimous]

Hello Alex
here is the location of my error !!
http://img259.imageshack.us/img259/7140/zsensor.th.jpg (http://img259.imageshack.us/my.php?image=zsensor.jpg)

in the red circle ! It was e light blue smd resistor, but don't know if it's e 500 or 1k resistor..or another thing ! IS IT A 500 OHM ?????:confused:

Thanks for asking me "SMD resistor" !! i'll find in old motherboards !.
Vincent.

[Alex Villa]

http://www.rangevideo.com/forum/showpost.php?p=718&postcount=2

[chiloschista]

Hi Alex, hi Vova,

1)I connected an FMA XY sensor to Z axis. It works properly but I must mount it inverted.

I connected an FMA Z sensor (with 102 resistor). It works properly, but I also have to mount it inverted: up face down.

Is this right?

2)I set CDP4 gain to 30 to reduce servo travel. I use V tail with external mixer, between RVOSD and servos. So I can use normal elevator and rudder setup. I use ailerons too (with differential), but they are connected directly to the receiver, with neutral failsafe setup, so they don't work on autopilot. The model turns very well with rudder also.
Why sometimes, on autopilot, tail servos goes to end travel without control?
This is dangerous, they have to travel less because of the double travel necessary for v-tail.

Thanks and best regards

chilo

[Alex Villa]

Hi Alex, hi Vova,

1)I connected an FMA XY sensor to Z axis. It works properly but I must mount it inverted.

I connected an FMA Z sensor (with 102 resistor). It works properly, but I also have to mount it inverted: up face down.

Is this right?

2)I set CDP4 gain to 30 to reduce servo travel. I use V tail with external mixer, between RVOSD and servos. So I can use normal elevator and rudder setup. I use ailerons too (with differential), but they are connected directly to the receiver, with neutral failsafe setup, so they don't work on autopilot. The model turns very well with rudder also.
Why sometimes, on autopilot, tail servos goes to end travel without control?
This is dangerous, they have to travel less because of the double travel necessary for v-tail.

Thanks and best regards

chilo
-Regarding FMA sensors positions, post a pic please.

-Autopilot might command the servos to go to the extreme (+-0.54mS from the neutral point) if the desired attitude is to much far from the actual attitude. Untill now I didnt saw the need for this to be user configurable.

[chiloschista]

Here is the picture.

Servo maximum travel configurable by the user is a must, if you consider that RVOSD overwrite transmitter program.
Have a programmable transmitter means to use it, otherwise we can go back to stone age (I think I still have a yellow Graupner somewhere, no it's not a tip-tip radio :mad:, it's already proportional :o).
If you program a servo travel limit on the tx, this has to be programmable in the RVOSD too.
Considering that on such a model usually there aren't high end servos, medium quality servos, when over travel, would probabily break, that means probabily a crash :eek:.
Objectionable opinion, of coarse ;)

Best regards.

chilo

[Alex Villa]

lol, every servo should be able to handle +-0.54mS from mid point.
I find weird that you have to place the Z sensor backwards, plese chek this carefully

[chiloschista]

... about servos, it is not a servo problem, but a mechanical problem with rudder's travel.
If I can't program a travel limit I must reset all in a mechanical way (stone age :rolleyes:, with all that electronics on board it would seems a little embarassing...).
I remeber you that I use v-tail, so rudder travel has to add to elevator travel. For example rudders all right and all up, means that the left tail is full up (some 3D travel :D). It should go to maximum that it can, but it is difficult to program this in a mechanical way, so: happy to have a computerized TX.
Would this be a big problem to insert such a string in the RVOSD program?
Anyway, if you plan a firmware upgrade for elevon (I read about this somewhere ;)) I think you have to put a string for maximum servo's travel. Otherwise some people will likely have problems.

Another personal disputable opinion...

... about z sensor: I tryed and I tryed...
Any suggestion about a new way to try it?

Thanks.

chilo

[Alex Villa]

... about servos, it is not a servo problem, but a mechanical problem with rudder's travel.
If I can't program a travel limit I must reset all in a mechanical way (stone age :rolleyes:, with all that electronics on board it would seems a little embarassing...).
I remeber you that I use v-tail, so rudder travel has to add to elevator travel. For example rudders all right and all up, means that the left tail is full up (some 3D travel :D). It should go to maximum that it can, but it is difficult to program this in a mechanical way, so: happy to have a computerized TX.
Would this be a big problem to insert such a string in the RVOSD program?
Anyway, if you plan a firmware upgrade for elevon (I read about this somewhere ;)) I think you have to put a string for maximum servo's travel. Otherwise some people will likely have problems.

Another personal disputable opinion...

... about z sensor: I tryed and I tryed...
Any suggestion about a new way to try it?

Thanks.

chilo
Its not that difficult to add that, after writing more than 22K lines of code this is like a piece of cake. So sorry you feel on the stone age with RVOSD. To those that never saw a program made on assembly might look like a single string, but new menu item needs to be done, several modifications needs to be added to those routines that handle the menus, the autopilot and the attitude stabilizator.

About your Z sensor placement, use the debug screen, and look at the value "ABS thermal", testing with unobstructed view of the sky, it should give you the maximum value when it is upside, and the minimum when it is upside down. This is not a new way to try it. I guess you readed about that in the instructions post?

[1alonso]

Sorry to take you from your dicussion. I would like to come back to my original question... does anyone know where I can get the cables for XY and Z FMA sensors to RVOSD? I dont have them and dont know how they go.

Thanks to all

[Alex Villa]

Sorry to take you from your dicussion. I would like to come back to my original question... does anyone know where I can get the cables for XY and Z FMA sensors to RVOSD? I dont have them and dont know how they go.

Thanks to all
Well, I posted the links on this thread, post #4. Be kind to read it please.

[Stephane67]

Does anyone know if we can replace the FMA Flat Cable by any other cable ?
Is there any risk of problems by using another cable (self-made) ?

Br Stéphane

[1alonso]

Thanks Alex... Iam sorry... since I saw other discussions I didnt pay well attention to it.

Thanks again

[AndreK]

Servo maximum travel configurable by the user is a must,


After todays "crash" - (lucky, hard landing) - I must agree !

-I was testing autopilot takeovers at dive/inverted position at altitude of 200-300meter.
-After few tests, I the plane were "falling" to the right, as if I lost a part of the right wing.
-I had to try as good as I could to get close to "home", the lift on right wind was worse while at low speed, and controls was bad at high speed.
-When altitude was only 80, I had to ditch the video-googles, and try to find a spot (was too low for good video-feed 200meters from where I was standing)
- I got down, hard, without too much damage.

A very quick analysis, proved that RVOSD was guilty:
Right aileron servo torn the plastic part that is mounted on the aileron away from the aileron !
It was clearly torn away, as if the servo went *way* to far.
Reason: simple, - Usually I the aileron travels only 60% - but when autopilot took over from a crazy position, it must have yanked the aileron-servo-rod-mount clear off !.

Solution - please make one routine that can "learn" maximum aileron/elevator/rudder movements. (just let the user do some stick-banging while recording maxumun/minimum output, and limit autopilot to that)

I would prefer not to have to use the solution that were used in "old days", and modern digital servos are obviously too powerful to fool around with :)
On the positive side: Autopilot worked great - if fact, I did not took such decision to let it try, but I think it might even had a better chance og getting closer to me than I did. (but I was unsure what caused that behavior, so I did not let it try in the heat of the moment)

[Alex Villa]

More learning routines = more chance users can make wrong things. Before push the autopilot to the limits, you should test servo travel at ground. Yes, I forgot to state this, but I guess is common sense?

[AndreK]

servo travel is usually limited by RC-controller programming, the controllers original setting is 100% , and it may go to 150 in each direction - I assume RVOSD did output 150% - while I was flying with a maximum throw of 60%

Do you say everybody should first program the transmitter to go 150% , then adjust the horns/links, then reduce it ? - it's like in the 80's. :)

[Alex Villa]

nope, but engage autopilot after have the AHI calibration comlplete. you can tilt your plane to look at the AP servo corrections.

However, the autopilot doesnt know what the maximum servo travels are, I think to be able to adjust this there is two posibilities.
- Set a default higher throw, and learn from user inputs while the plane is in the air, or in a calibration stage.
- Make new menu items to allow the user to store the maximum throws.

Wich one would you like to have?

[AndreK]

Learning in the air have one drawback, of user starts, and climbs and turns left until desired altitude, the autopilot - if needed - will not have "learned" to allow the servo go to the right og down.

May I suggest - a menu-item saying "learn maximum servo throw"
user selects this item, and sees 6 numbers aileron,rudder,elevator min+max - then user moves sticks in all directions while each value is "learned" - the function may exit automatically after x seconds. - This needs to be done on the ground only.

Of course, "Make new menu items to allow the user to store the maximum throws." would do the job as well.

thanks for this great product !

[chiloschista]

@AndreK: sorry for your crash.

@Alex: both possibilities are good:

1) it make life easyer, above all if you retouch tx programming and/or if you change plane
2) it is preferable if anything goes wrong; you have the possibility to see how RVOSD is working, better understand what is it making, and retouch it.

Personally I prefer the second, but both are valid, so up to your programming experience.

About post 14:
-z sensor wash faulty (bad soldering of resistor 102, I guess), so I use an xy sensor that works good (but upside down...)
-Alex I don't feel at the stone age with RVODS. RVOSD is a great product and I am very satisfied with it. Of coarse it is not plug and play. It is a growing up device, and it needs some calibration etc. This is the most exciting side of the matter.

Keep up the good work!... and continue to be patient with us :D

chilo

[aaronredbaron]

If you mount the XY sensor on the bottom of the plane, is it possible to do your pitch calibration with the airplane inverted? I am trying to decide where to put the sensor, someone in this thread mentioned that if you put it on the bottom, your hand blocks the calibration. What is the best solution, and if you put it on the bottom, is there a problem with calibration?

[aaronredbaron]

oh, by the way... I was able to modify a thunder power 3s balance connector to work with the FMA sensors... Goodbye FMA ribbon cable!!!

[AndreK]

oh, by the way... I was able to modify a thunder power 3s balance connector to work with the FMA sensors... Goodbye FMA ribbon cable!!!

Why do they use those crappy cables anyway ? - the output is amplified, so I cannot imagine any electrical reason.
I shortened mine.

[aaronredbaron]

So, what is the verdict on mounting the sensor on the bottom? Can we initialize the pitch axis with the plane inverted? or is it just better to put the sensor on a pod on the top, and if so, does the vertical stabilizer affect it? (maybe it needs to be off center?) I finally seem to have a good GPS lock, and I have the sensors, I'm looking into my options on where to put the sensors...

[Alex Villa]

I place the XY head in a pod because I want the sensors to have the best FOV possible, with the head in the belly some of the FOV is blocked by the plane.