Extra inputs for Devo Tx's

All I did was create a new model and add the PPM to 2 new channels, basically as a test bed to see if it works. All the values are "factory set" as it were...





The pots are on channels 6 and 7

I've tried adjusting the settings on the menu, and it needs -50 of subtrim just to centre the pot on 0. But no matter what i do with the scaler and the limits, I still cannot get that part missing from the pot, it still maximizes where it used to... in fact, it created more deadband on the other end of the pot where it was fine before. So it seems the issue is in the Arduino, because 50 tubtrim just to centre seems an awful lot to me for starters!!

I'll be opening it up today and changing the sketch to another one with that change in code you suggested, and I'll see what that does and let you know

EDIT

One thing I thought I'd try out was the entrance voltage on the pot... the thing is I use NiMH in my 7E (the Turnigy 2400mAh LSD batteries), and we are assuming that the pot is going from GND to +5V, but how can it supply +5V if the battery at that moment for the whole system is at 4.89V... so I tried with a 2 cell LiPo with the pot in centre... before, with the NiMH pack the centre of the pot marked 50 on the channel monitor, and just changing the battery over to a 2S LiPo the centre changed to 89, and just a touch put it to 100!! So in fact it made the issue even worse!!

I have the 5V for the pots and the Arduino coming from the 5V regulated switched point on the board you indicated a few pages back. I might try changing the GND and +5V points from there to the ones directly used by the gimbals and see how that does, but really I don't tant to start having to change to a LiPo as I can't find a decent one to fit in the battery bay, and I hate having a battery strapped to the back of a transmitter.

Could be a problem with the "reference voltage" the arduino takes for measuring the tension. cf www.arduino.cc/en/Reference/AnalogReference / www.arduino.cc/en/Reference/AnalogRead .
Are you sure the 5v is good ? could you measure it ? Depending on the board/Mini you have, you perhaps should input the +5v line you use to the HREF pin and set AnalogReference to EXTERNAL (After having made sure NOT to be over 5V !!!)

The arduino I am using is a Nano V3, and it is wired with the VIN pin into the switched 5V pad on the 7E's main board, which when I tested it was giving 4.7V. Supposedly, this point doesn't go over 5V no matter what battery it uses. The same pin is what's feeding the 5V into the pots.

(No arduino nor Deviation specialist here) I'd test with plugging the 5V/4.7V line to the HREF pin too, and set AnalogReference to External. Worth a try I think. Also test using a mucho Simple arrduino sketch, trying and get the 0-1024 value directly from the analog read to the console to debug. If the 0-1024 value is OK, (should be more or less linear) then move on with that.

There are a couple of things that I have found a little troublesome here with my multimeter...

First of all, the switched point on the main board does not give regulated 5V, it gives the full battery voltage. Supposedly, one gives full voltage, and the other regulated and switched, when really, they both give full voltage, only one is switched. Using 4xNiMH shouldn't be a problem with that as the voltage I am getting is within the margin for that, but on a LiPo it means that over 8V is going directly into the Arduino and the pots.

The other issue is that the pots on the gimbals are only being fed 3.24V, not 5V... now, I have found that the higher the voltage I input, the worse the problem gets, so I'm going to try with changing the GND and the Vin to the ones used by the gimbals, see what that does, and just feed the Arduino from the 5V switched from the board.

As for sketches, I've tried a few, and none of them work at all... the only one that does work is the one I posted, albeit I haven't tried adapting it yet how Deal57 said, so I'll try that later.

OK, it seems something has been damaged with all this messing around with voltages, and I'm going to change one of the pots... but the other one works perfect on range when I feed it with the Vin from the gimbal, which is 3.24V, full range of throw, centres perfect, everything!! So it seems that it just needed the right voltage going into it to work, and this is with the NiMH pack.

I'll just change the pot which is giving problems, it could be physical the issue as the board you solder to is wobbly even... let's see what happens...

OK... I've just changed the pot, and rewired everything so one pot is getting GND and Vcc from the elevator pot, and the other pot is getting them from the throttle pot, maing the installation a little cleaner in the middle...



The result is everything working absolutely perfect, and this is with the original sketch I posted on here, which is just for 2 pots. If I ever feel the need for more switches then I can just add it onto the sketch and upload it, with the changes in the wiring needed and that's it.

But a couple of words of warning people, just so you know...:

First, it didn't give the proper range until I wired them un with the original gimbal pots which have a lower voltage than 5V, so this was changing the range on the pot itself. With the change, it is now centring perfectly on 0, and the -100 to 100 range is on the full throw of each pot, just as they are supposed to work, with a tiny deadband of about 2º on either side which is good to make sure you have it at minimum and maximum. The change in value on the monitor is steady and smooth, and I've tried it out now also with a servo to see that it translates properly, which it does, giving a perfect usable pot for anything you would want it for.

The other thing I need to point out is that the regulated, switched point on the board that was photographed a few pages back is NOT regulated, only switched. It gives the full voltage of the battery that you have plugged in, so if you use a 2S battery, then you are feeding the full 8.4V into the arduino via this, and it is only being switched so it turns off when you turn the transmitter off, and the pad next to it is constant, full voltage, not switched. If you are using 4 AA NiMH batteries like I am, then this isn't a problem, but if not, make sure the Arduino you are using can take the voltage on the connection you are using!!

Now all I need is for the knob that is missing to finally arrive, and I can be happy with this transmitter as is, because I think putting more on it now for me is just overkill. I wanted a good way to control flaps with variable angle, as well as be able to adjust things like butterfly and such in flight, and maybe also for things like gimbals, and I now have it!!

If you still have it open do me a favor and measure the voltage on the arduino's 5V pin. I supplied 5V directly to the 5V pin (not vin, which can accept 12v), if you applied it to vin your arduino had to buck that down and you would of been trying to run to low.


Also measure the voltage on the PPM output (D13 unless you changed it) PPM is, by standard, 5V, infact STM32 /devo is a 3v3 system but ppm is 5V so devo must accept it. Glad you got it working and if it's back together I'd just use it as it's working now but if you want to dig in, it's strange [to me] you're getting full range with 3.3v to the pot.

I was getting less range the higher the voltage was... so I thought if I lowered it I might get the reverse effect, and the voltage from the other pots on the gimbals were just what the system needed to work spot on, not even a touch of trim needed or anything.

I'm afraid I can't measure the voltage now as It's all closed up, and also the Arduino itself is encased in the last of the heatshrink I have of that size...

one sec, hold that thought, I have an idea...

No no, I insist, dont take anything apart to measure voltages for me if it's working now! Go young padwon, fly your planes!

OK, what I did was plug the jack into the DSC port and measure the voltage between the jack tip which is the PPM out from the Arduino (D5 on mine), and the negative of the multimeter to the black wire of the battery JST plug, and I got a tension of 3.62V coming through there.

the power I have going into the VIN port of the Arduino, but that was direct from the switched battery pad as I said, so it was receiving what the battery was giving it. The same was originally feeding the pots, so the voltage in the pots depended on what battery was used and how drained it was. Now the VIN is still being used and works fine with the NiMH batteries I'm using, and the pots are fed from the same Vcc and GND as the gimbal pots

Epitaph wrote: OK, what I did was plug the jack into the DSC port and measure the voltage between the jack tip which is the PPM out from the Arduino (D5 on mine), and the negative of the multimeter to the black wire of the battery JST plug, and I got a tension of 3.62V coming through there.

That's very strange as ppm is a 5V signal. Im wondering if all your problems stem from that? The signal is so weak the timers can't count the rises / falls...

I don't know, but before I was getting more voltage, and problems, and now I have less voltage and no problems... it's going perfect with those new figures. Maybe my transmitter was put together by Mulder himself!!

I'll be keeping an eye on it whilst the battery level fluctuates and all that, but as far as I'm concerned, if it works, why touch it?

Also, I tried with a number of other sketches from on here just changing the pin values to what I had wired up, and none of them worked, only the simple, 2 pot sketch. And it's with the old code, not the new change that Deal57 posted, I didn't need that... and with the values on the original code I posted.

Don't ask me why it works, it might not even work on someone else's transmitter the same, but it works on mine and goes great!! Who am I to complain??

OK, one thing I did try, I've just tested it with a 2S battery, and again problems... the PPM output via the DSC jack is 4.81V, but the pots are off-centre, and going from -100 to 56 in range, not to 100. Put the NiMH back in and all working perfect again.

Could there be a problem on my transmitter that the solder point everyone is using for 5V on mine isn't working isn't working as it should and is just feeding the full battery voltage unregulated? I mean, it doesn't bother me if I have it working with the setup I'm using, as the battery I'm using is regulating the voltage, but if it's the case, it might explain something.

I guess the issue is with the input voltage into the Arduino, if it's being fed from the NiMH batteries, it gives a lower PPM out than if it's fed off a 2S battery, and this in turn changes the voltage needed on the pots. Seeing as I had planned to keep it on NiMH as they fit better and the ones I have are pretty good (and I'm not bothered if it takes a while to charge) then I might as well just keep it like that. The only thing I want to keep an eye on is that the Arduino continues to work ok when the battery drops low (not that I let it drop low anyway)...

I'm going to fully charge up my NiMH batteries to see if that gives any difference.

I guess the BEST way to fix this would be to use a 5V step-up/stepdown converter like (for example) this one www.pololu.com/product/2119 , to make sure your arduino is fed a stabilized 5V (and the pots too).

But I tried with the pots being fed from the battery before along with the arduino which was 4.96v and that''s where all the trouble started!!

And besides, fix what? It's working great now!!

The regulator might be a solution should I change to a LiPo, but I honestle can't see that happening; I just hate having a huge LiPo strapped to the back... what's the point on a small transmitter? It's like a battleship anchor on a rowboat!!

Nice work and this looks great!
If this stuff was easy, anyone could do it!

I've never been tempted to use 2S batteries in my Devo7e. If you put it behind a 5v regulator, maybe. I'm happy with my NIMH rechargables. The PPM voltage I've measured on the Devo7e was about 3.6v, and it seems to work with all the Arduino or PC tools I've used. IIUC the PPM signal is differential, so as long as the receiver can differentiate between high and low, it should be able to decode the signals, so voltage isn't critical -- however I'd never put more than 5v onto something connected to my TX.

So what have we learned?

1. The Arduino sketch works great
2. The Devo7e is a pretty awesome platform when the Arduino is doing it's job
3. The Devo7e 5v power is elusive, need to determine an "ideal" Vcc point; what worked for some may not work for others
4. (Corollary) We need to identify the best way to power the Arduino at the right voltage