New Products Flash Sale
Thread Tools
Old Jan 24, 2016, 04:46 PM
AndyKunz is online now
Find More Posts by AndyKunz
AndyKunz's Avatar
Illinois
Joined Sep 2001
32,487 Posts
Yes, the 400us pulse width is to verify the host and is one of those JR things we inherited.

The pulse for the servo is still measured from leading edge to leading edge.

Andy
AndyKunz is online now Find More Posts by AndyKunz
Site Sponsor
Reply With Quote
Sign up now
to remove ads between posts
Old Jan 24, 2016, 04:46 PM
phil_g is offline
Find More Posts by phil_g
Who let the dogs out?
phil_g's Avatar
Pontefract, Yorkshire, UK
Joined Jul 2007
1,404 Posts
Quote:
Originally Posted by AndyKunz View Post
Phil did you say that Spektrum uses non-standard or hi does? The correct timed is measured from leading edge to leading edge. That means if the line is normally low that the pulse sixth to be sent to the servo is from the time the pulse starts to go high until the pulse starts to go high again. Using this technique means that the time the pulse is high is immaterial. Back in the old days we used pulses that ranged from about 130us to about 375us. It didn't matter so long as the electronics were set up to handle pulses properly. Some RF decks needed different widths.

Andy
Thats exactly as it should be Andy - edge to edge - traditional PPM does not depend on the precise width of the PPM pulses, only the time between matching edges. But as we've discussed before - the DM9 does in fact need its PPM pulses to be 400uS or the channel timing is out by the difference between the actual PPM pulse width, and 400uS. All my project stuff has used 250uS so I have to do a specific 'Spekky version' with a 150uS fudge

Cheers
Phil
EDIT: I should add, its not a problem once you're aware... heres my Spektrum DM9 transmitter:
phil_g is offline Find More Posts by phil_g
Last edited by phil_g; Jan 24, 2016 at 06:53 PM.
Reply With Quote
Old Jan 24, 2016, 04:56 PM
AndyKunz is online now
Find More Posts by AndyKunz
AndyKunz's Avatar
Illinois
Joined Sep 2001
32,487 Posts
Actually, I think it was supposed to be rejecting you completely if you were too far away from 400. Kinda late now to fix that bug though!

Andy
AndyKunz is online now Find More Posts by AndyKunz
Site Sponsor
Reply With Quote
Old Jan 24, 2016, 05:06 PM
phil_g is offline
Find More Posts by phil_g
Who let the dogs out?
phil_g's Avatar
Pontefract, Yorkshire, UK
Joined Jul 2007
1,404 Posts
Part of my test setup is a ppm generator with all parameters variable including channel count, polarity, pulse width - the DM9 works fine with all (reasonable) widths, but the channel timings only coincide at 400uS. But as I said, its not a problem once you're aware
phil_g is offline Find More Posts by phil_g
Reply With Quote
Old Jan 24, 2016, 05:34 PM
Daedalus66 is online now
Find More Posts by Daedalus66
60 years of RC flying
Daedalus66's Avatar
Canada, ON, Ottawa
Joined Feb 2006
19,161 Posts
Quote:
Originally Posted by pmackenzie View Post
It is possible that the DM9 is doing something a bit different.



I can take a model from an XP9303 with an FrSKy module and copy it over to an X9503 using a DSMX receiver and not even have to touch the trims.



But many have said that when using the DM9 unless you set the PPM width to 400 usec the trims will be off. Other modules don't have this problem.
I can verify all of that. I just checked with a DM9 in my Taranis. With Protocol set to 300us the neutral is at 1600, at 350 it's at 1550 and at 400 it's at the desired 1500.
Daedalus66 is online now Find More Posts by Daedalus66
Reply With Quote
Old Jan 24, 2016, 07:22 PM
ivanc is offline
Find More Posts by ivanc
S.A.D. member
ivanc's Avatar
United States, TX, Round Rock
Joined Dec 2004
10,877 Posts
Quote:
Originally Posted by Daedalus66 View Post
I can verify all of that. I just checked with a DM9 in my Taranis. With Protocol set to 300us the neutral is at 1600, at 350 it's at 1550 and at 400 it's at the desired 1500.
Same here with my own PPM generator (stick encoder) - the DM8/DM9 does not measure the complete pulse (rising edge to rising edge which includes the "start" pulse) but the length of the low portion of the pulse for the DM9 and the high portion for the DM8. I can vary all aspects of the PPM and the DM8/9 will work fine with a start pulse anywhere between 250us to 450us simply ignoring it.

Ivan
ivanc is offline Find More Posts by ivanc
Reply With Quote
Old Jan 24, 2016, 07:44 PM
phil_g is offline
Find More Posts by phil_g
Who let the dogs out?
phil_g's Avatar
Pontefract, Yorkshire, UK
Joined Jul 2007
1,404 Posts
Sorry Ivan that doesnt quite tally with my findings. That would put the channel timings miles out, and in the opposite direction. If the DM9 did only use the 'low' portion of the pulse then a 1500uS neutral with 250uS pulses would give 1250uS neutrals whereas I measure 1650.
My own unsubstantiated belief is that they measure the low portion and add 400uS. This would give perfect timing on 400uS pulses and also the exact offsets we have witnessed when using other pulse widths.
Cheers
Phil
phil_g is offline Find More Posts by phil_g
Reply With Quote
Old Jan 24, 2016, 10:40 PM
ivanc is offline
Find More Posts by ivanc
S.A.D. member
ivanc's Avatar
United States, TX, Round Rock
Joined Dec 2004
10,877 Posts
Quote:
Originally Posted by phil_g View Post
Sorry Ivan that doesnt quite tally with my findings. That would put the channel timings miles out, and in the opposite direction. If the DM9 did only use the 'low' portion of the pulse then a 1500uS neutral with 250uS pulses would give 1250uS neutrals whereas I measure 1650.
My own unsubstantiated belief is that they measure the low portion and add 400uS. This would give perfect timing on 400uS pulses and also the exact offsets we have witnessed when using other pulse widths.
Cheers
Phil
Same thing said in different way. The DM8/9 ignores the start pulse and assumes it is 400us. So it considers 1100us of what it actually measures as neutral and goes from there. Whether it adds a fixed 400us value to the measured value and uses 1500us as neutral or uses the measured value only with 1100us as neutral reference is immaterial. The final result is the same - it ignores the actual length of the start pulse so for the module to generate correct servo output at the receiver the PPM fed to it must have exactly 400us start pulse.

I don't have a problem with that peculiarity of the DM8/9 as I can easily get the needed pulse duration on my designs or modify the timing circuit in the old trannies so they generate a 400us pulse.

Ivan
ivanc is offline Find More Posts by ivanc
Reply With Quote


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion HobbyKing Orange Rx DSM, any problems/complaints? big_poppa4904 Radios 26 Aug 12, 2013 04:08 AM
For Sale Blade sr dsm tx hp6dsm and blade 120 sr tx Garrett1234567 Aircraft - Electric - Micro & Indoor Airplanes (FS/W) 0 Nov 08, 2012 04:46 PM
For Sale Spektrum Telemetry TM1000's, TM1100's,AR6000 DSM2, 4 New Orange DSM 2 Receivers, daign FPV Equipment (FS/W) 8 Jun 02, 2012 03:26 PM
Sold Brand New Orange DSM 2 Satellite Receivers healthyfatboy Aircraft - General - Radio Equipment (FS/W) 9 May 06, 2012 04:17 PM
Sold Mint Spektrum DX6i, DSM 2 with 6ch Orange RX tankbuster Aircraft - General - Radio Equipment (FS/W) 2 Apr 06, 2012 08:39 AM