Can't set correct PPM frame pulse width for Spektrum DM9

[GoogleCodeExporter]

Tx: Taranis
VERS: opentx-x9d-v1.0.00
EEPR: 215

What steps will reproduce the problem?
1. Go to MODEL SETUP
2. Scroll down to External PPM
3. Change Module to PPM
4. Change PPM frame pulse width
5. Get frustrated when width can't be set to 375u.

In order to get the correct neutral servo positions, the Spektrum DM9 would 
needs a pulse width setting of ~370u-380u. At the moment the pulse with can 
only be changed in steps of 50u. 400u is a bit too high and 350u a bit too low.

This can be mixed with sub-trim, but if a model has been mechanically trimmed, 
it's a shame to start adding sub-trim.

I'd expect the pulse width to be changeable in smaller steps. Steps of 10u or 
5u might be a good compromise. Steps of 1u is maybe overkill.

Original issue reported on code.google.com by jcansd...@gmail.com on 11 Sep 2013 at 8:06

[GoogleCodeExporter]

use ppm center adjustable option...

Original comment by romolo.m...@gmail.com on 11 Sep 2013 at 8:10

[GoogleCodeExporter]

PPM center can be set up to the microsecond using PPMCA option

PPM center is not intended to be adjusted using PPM interframe.

Original comment by romolo.m...@gmail.com on 11 Sep 2013 at 8:13

• Changed state: Invalid

[GoogleCodeExporter]

Where is the PPM center option? I couldn't find it on the 'MODEL SETUP' menu.

Original comment by jcansd...@gmail.com on 11 Sep 2013 at 9:41

[GoogleCodeExporter]

In limits (last column) by default is 1500

But please open an issue also with horizon hobby.
There is absolutely no reason for having the center conditioned by interpulse 
pause...
That's a very bad implementation of a PPM decoder.

Original comment by romolo.m...@gmail.com on 11 Sep 2013 at 9:54

[GoogleCodeExporter]

Oh, you're suggesting changing the center by the same amount for every servo? 
This feels like more of a hack than having finer control over the PPM frame 
pulse width. What's the point of being able to change the pulse width if we 
can't set it to what the module expects?

I realise it's probably a bad implementation of a PPM decoder, but there are a 
lot of these modules in the wild already. If people want to move over to OpenTx 
from a Spectrum radio, this is the module they will probably use unless they 
want to risk the Orange one (which has very mixed reviews). AFAIK it works okay 
in JR radios.

I'm sure Horizon will be more interested in selling their new DX9 radio than 
fixing the DM9 so that people can easily use it in the Taranis. ;)

Original comment by jcansd...@gmail.com on 11 Sep 2013 at 10:39

[GoogleCodeExporter]

Yes it is a hack , but so would be implementing strange pulse lengths... The 
PPM pulse length is supposed to be insignificant in the first place! A module 
should not depend on it at all, and even less change servo positions according 
to it!
All users so far seem to be happy with positive modulation and 400us PPM pulse. 

A model that is ported from another radio to the Taranis will likely require 
mechanical readjustment anyway due to different center points and throws 
between radios and firmwares. 

Original comment by bernet.a...@gmail.com on 12 Sep 2013 at 1:49

[GoogleCodeExporter]

I've investigated this further and a 400us PPM pulse with the DM9 does give a 
servo centre position of 1.5ms. The dependency appears to be with my DX8, which 
is showing a centre position of 1.52ms! :( I've asked on the DX8 forum about 
this.

The reason I'm interested is because I've been wondering about writing a DX8 -> 
OpenTx model conversion tool. I don't know how feasible this might be.

Original comment by jcansd...@gmail.com on 12 Sep 2013 at 9:01

[GoogleCodeExporter]

Reply on RCGroups from spectrum66:
http://www.rcgroups.com/forums/showpost.php?p=26092142&postcount=10019

That's simple:
servo center standard pulse width is just a convention:
some producer use 1500, some 1520, some other 1620

That's why in Limits opentx offers the possibility to set the center per 
channel.

Then there is the issue with DM9 module.
That module has a programming issue: it only measure pulse width considering 
interpause fixed at 400uSec instead of measuring time between two leading 
edges. 

So if you want 1520 with a DM8 you need to set interpause at 400uSec and in 
channel limit set PPM center at 1520, that's all.

That said, you cannot ask opentx to change actual interpause or center pulse 
timing defaults as another radio performs differently... 
We cannot force a lot of users to readjust centers just to conform to a single 
producer standard.

Romolo

Original comment by jcansd...@gmail.com on 12 Sep 2013 at 9:31

[GoogleCodeExporter]

LOL, it was you. Thanks for your help and explanations! :)

Original comment by jcansd...@gmail.com on 12 Sep 2013 at 9:33

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

Original comment by bson...@gmail.com on 29 Nov 2013 at 8:58

• Changed title: Cant set correct PPM frame pulse width for Spektrum DM9

[GoogleCodeExporter]

Original comment by bson...@gmail.com on 29 Nov 2013 at 8:58

• Changed title: Can't set correct PPM frame pulse width for Spektrum DM9