er9x development

[LTMNO]

Both great valid options. I use the pot for tuning of PID's during flight. But maybe an 'alt screen' would be good as Jaoa mentioned.

I do find it a pain that when I am in the house that the volume is loud as my wife can here it great me and then she knows I am messing around with my toys. Its like the hand in the cookie jar.

On a different note, now that you are thinking about options for the TX....
Would it be possible to have the ability to change the Headings/Titles of the Telemetry Items. For Quad/APM/IOBoard transmission via FrySky and APM information, the reuse of existing headings/titles are being used... Like T1 and T2 are used for GPS Sat count... etc... I will provide a list if this is possible.. or an option to switch the Telemetry Headings for use with FrySky and APM(ARDUCOPTER).

Thanks...

[MikeB]

That could end up using a lot of flash.
One way of doing this is to have a "language option", but that would require another hex file (or even one per actual language).

How many items involved needing different lables?

Looks like volume control on a per model basis is needed.

Mike.

[LTMNO]

Hi Mike,
Here is the List or Mappings thus far...
Values that are running and known to be correct:

Accel - X, Y, Z = Accelerometer values, -+ values

Temperature 1 = Flight controller internal temperature, Celsius

Temperature 2 = Visible satellite count and GPS Fix style, 2-3 digits where LAST digit is fix status 0-1 No fix, 2 2D Fix, 3 3D Fix. FRONT digits are amount of satellites. For example "63" = 6 satellites, 3D FIx or "103" is 10 satellites, 3D fix or "32" = 3 satellites, 2D fix

Heading = Your current compass heading

RPM = Armed/Disarmed status and approximate throttle value when armed. 0 = Disarmed/Motors off, Values between 1020 - 1960 are shown when motors are armed and throttle position. Number represents pulse width so for example around 1500 is middle throttle. This value is NOT exactly same what Mission Planner might show for your throttle channel as we need to use special scaling made by FrSky but it's close to it. Accuracy is -+30 uS.

Altitude = Altitude detected by your flight controller (works on tabletop tests, not confirmed on real flight yet. Anyone??)

Lat/Lon = GPS Latitude/Longitude details with N/S, W/E indicators

Time = Sec/Min/Hours since last reboot of IOBoard

Things that needs to be fixed/confirmed are:

Voltage = Battery voltage. It seems to work but startis drifting after awhile.

Current = It's seen by IOBoard but not forwarded correctly

Fuel Level = Seen by IOBoard but not confirmed yet

Speed = Seen by IOBoard but not confirmed yet

Here is the thread on JDrones...
http://diydrones.com/profiles/blogs/add ... 1#comments

Perhaps the ones that would be best to test with would be Armed/Disarmed for RPM and Temp2 for GPS Fix and Sat Count.

[MikeB]

There is enough flash, and probably EEPROM, for something like this on the '128 build, and ersky9x as well.
On stock with a M64 it will be much tighter to fit.

Mike.

[LTMNO]

Looks like I have some more
Modding to do to my radio. I will follow the footsteps of Joao. Upgrade to 128 it is.

[jhsa]

loool.. prepare yourself for a nice ride..

[LTMNO]

I guess I need special solder. Off to the electronics store.

[MikeB]

Only change to a '128 if you really want to. I wouldn't want to be the cause of a broken board!

Do you use the custom telemetry screen to display all your data?

There are some unused ID's in the FrSky hub protocol. Could the data be changed to use some of these? I think it is highly unlikely that FrSky will be using them since they are moving to using the S-PORT and the protocol for that uses IDs that are all greater than 255.

Ideas are slowly ticking over! I may come up with something simple to handle this!

Mike.

[Iksbob]

Will the RX or TX module reject data IDs that aren't explicitly part of the hub protocol? If the FCB is parsing all the sensor values, there's no hub to worry about on the aircraft end of the connection... Just the RX. And since we're talking about er9X on the radio end of the connection, there's no need to stick to known values if the telemetry data link will carry them.

[LTMNO]

Well, I am always up for the challenge. I have a week or so for some parts to come before I can fly again anyway. I can get the m128 on Monday PM.
I really wasn't sure how the Screens where be presented... The language idea seems to be a good one. Just duplicate the English and make some changes?
I completely use the Custom Telemetry Screens but then again the main screen shows the Battery information for A1 and A2.


This image has been reused from JDrones Site....

But what makes you think just now you are responsible for all work done to the Radio... If it wasn't for you, I would have done nothing to my radio... its all this darn good stuff you keep building for us to consume.

Keep up the great work!

Pino.

[MikeB]

Iksbob: The serial is just a byte stream, neither the Rx or the Tx looks at the data. We should be able to drop in any other IDs we like.

LTMNO: Are you running the standard (FrSky) version of er9x? Those screens (not yours I believe) look to come from a modified version of the Ardupilot version.
One problem with re0using the T2 value for something else occurs if you select IMPerial units. As a temperature it will be converted from C to F.

Maybe I've been making this more complicated than it needs to be. A possible simple solution is to add an entry in the telemetry values list (e.g. Scnt for Satellite count) and simply get this to pick up the data from the T2 field. Quite easy, it's all there in a table to select!

Mike.

[MikeB]

Volume controlled by a POT seems to be working. The way it currently works is:
Assign a POT to control the volume in the model setup.
The "welcome" message and the model name are announced using the volume setting in the RADIO SETUP.
While you are on the main screen, the POT will control the volume.
If you go to any other screen, the volume reverts to the volume setting in the RADIO SETUP.
Returning to the main screen the volume goes back to the POT setting.
I haven't included the rotary encoder as a volume source, that will take more code.

Mike.

[gohsthb]

Maybe make the volume controlled by a gvar? Then we could set it to a put or a trim switch, or whatever. I've been out of town so I haven't been following this, maybe that was already suggested.
-Gohst

Sent from my LG-P999 using Tapatalk 2

[jhsa]

So, what could control the volume then?
A pot. trim switch, rotary encoder, GVAR.... any other?

João

[LTMNO]

How about throttle stick while your flying. It gets louder as the copter/plane/car gets louder. Just like those high tech cars we all drive. (Just kidding btw)

[MikeB]

If we allow a GVAR, then this allows any GVAR source to control the volume - trim, rotary encoder, stick, pot or any channel output.

Throttle stick would then be a possibility

I'll look into the possibility.

Mike.

[LTMNO]

@Mike. I will take screenshots of my radio later today and answer your question. Just have to head out.

Edit: Hi Mike, I am actually running openTX with FrySky Prot.

The other screenshots is most likely running OpenTx too as the screenshots are identical to mine.... just model name different.
I jumped ship on ER9x after I couldn't set the AttoPilot Volt/Amp Sensor correctly for proper reading of Batt.
I am ready to jump back to ER9x if more functionality is available. I do like the Battery Gauge and FP's display on the OpenTx.

here are my screenshots...

[Iksbob]

Sounds like we could maybe use a simplified class of output "channel" mixes for various in-radio functions? Like GVARs, but they're value destinations rather than sources. Audio volume, haptic strength, backlight brightness (just PWM on the appropriate pin... could be used for other devices, like a dedicated alarm LED blinker).
Beyond just the audio volume, it could be applied to modulation of alarm tones that I posted about a little while ago. Given some kind of mix/destination mapping, the user could assign varying attributes (GVARs, channel outputs, telemetry, etc) to a tone/click type audio alert. The click could be a simple synthesized beep of (typically) very short duration. Assignable attributes could include volume, tone (frequency of the individual beeps), duration (length of individual beeps), rate of repetition (anywhere from a single blip when an alarm occurs to a buzzing, on up to a solid tone depending on how the duration and rate of repetition interact).
Given the flash and RAM space situation, I'm just throwing it out there as a idea, maybe it will inspire something down the line.

[Iksbob]

Sounds like we could maybe use a simplified class of output "channel" mixes for various in-radio functions? Like GVARs, but they're value destinations rather than sources. Audio volume, haptic strength, backlight brightness (just PWM on the appropriate pin... could be used for other devices, like a dedicated alarm LED blinker).
Beyond just the audio volume, it could be applied to modulation of alarm tones that I posted about a little while ago. Given some kind of mix/destination mapping, the user could assign varying attributes (GVARs, channel outputs, telemetry, etc) to a tone/click type audio alert. The click could be a simple synthesized beep of (typically) very short duration. Assignable attributes could include volume, tone (frequency of the individual beeps), duration (length of individual beeps), rate of repetition (anywhere from a single blip when an alarm occurs to a buzzing, on up to a solid tone depending on how the duration and rate of repetition interact).
Given the flash and RAM space situation, I'm just throwing it out there as a idea, maybe it will inspire something down the line.

[MikeB]

LTMNO: What is the problem with the "AttoPilot Volt/Amp Sensor?
A battery gauge is interesting, but if, like me, you use NiMh, then the battery volatge holds steady for a long time, so a gauge doesn't really mean a lot if it is based on the battery voltage.

Mike.

[jhsa]

that battery gauge comes from opentx.. you can define the low and High voltage..
I think minimum difference is 1V..

[LTMNO]

LTMNO: What is the problem with the "AttoPilot Volt/Amp Sensor?
A battery gauge is interesting, but if, like me, you use NiMh, then the battery volatge holds steady for a long time, so a gauge doesn't really mean a lot if it is based on the battery voltage.

Mike.

Hi Mike, the problem was the input level of the Amps in ER9x... last I remember and the reason for jumping ship was the limitation of supporting the 90A AttoPilot Sensor, even with offsets... I had to use OpenTx which had more range in OffSets, if I recall.

Not sure if that was fixed yet in ER9x?

51.8V
89.4A

[MikeB]

Are these values are on the A1 and A2 inputs?

Mike.

[LTMNO]

Sorry. These are the values of the sensor limits. Max. And what I would need to set a1 and a2 at in telemetry in er9x. At the time that was not possible. Was past the limits. Not sure if you raised them.

https://www.sparkfun.com/products/9028

[MikeB]

At this point I haven't changed any of the scaling on A1 and A2.
I have some plans, but not enough time!

Mike.

Edit: Checking the code I see I've started. The plan is to drop the FrSky module alarms, and re-use the EEPROM bytes to store extra scaling information.

[LTMNO]

no worries, there are bigger fish to fry..

[jhsa]

Mike, just noticed something..
When deleting a mix, I press the center button of the encoder to display the popup menu, and when I scroll down to delete the weight in the mix also changes.. It doesn't really matter as it will be deleted anyway, but if I just copy or move a mix.. the weight will be changed as well.. Now that shouldn't happen
I still don't have the latest fw installed so I don't know if this issue was corrected.. just letting you know..

João

[MikeB]

Interesting, I'll have a look when I get a moment!

Mike.

[jhsa]

sorry that I don't have the latest version on the radio yet and it's not a big help, but have been flying and I need time to check all is ok after flashing.. will probably do it today.. This summer has been quite good here in Germany and that is quite rare I've been told so I'm taking my chance to put some more flight time on my models and 9x's..

João

[MikeB]

I can reproduce the problem, and I think I have a fix for it. Quite a "buried" bug to find!

Mike.