Thread Tools
Oct 10, 2016, 07:39 PM
Registered User
Yea sure is a good step forward. This will completely remove any jitter .
The CRC check will help filter any invalid data too.

One issue you might face is.. multiple, or sequential packets been rejected to invalid CRC checks.
Sign up now
to remove ads between posts
Oct 10, 2016, 07:50 PM
Woohoo!
RS2K's Avatar
Quote:
Originally Posted by Ub3r
Yea sure is a good step forward. This will completely remove any jitter .
The CRC check will help filter any invalid data too.
A lot of the jitter actually comes from the FC's calculations. You'd be amazed at how accurate PWM can really be when using good grounds and twisted pair. The real winner here is no need for calibration and guaranteed resolution no matter how noisy the line is (within reason of course).

Quote:
Originally Posted by Ub3r
One issue you might face is.. multiple, or sequential packets been rejected to invalid CRC checks.
I did a lot of testing with this in April (when developing MSv2) and that will happen at higher speeds when using DMA. There are ways around it though which it sounds like Felix has figured out.
Oct 10, 2016, 08:35 PM
Quad Enthusiast
With all of us blheli_s people out here I really hope there's a way to use this. It sounds very, very nice. Great job!
Oct 11, 2016, 12:21 AM
Quadaholic
--Oz--'s Avatar
Sweet! Now we need zero desync esc's
Oct 11, 2016, 12:35 AM
Woohoo!
RS2K's Avatar
Quote:
Originally Posted by --Oz--
Sweet! Now we need zero desync esc's
Working on it.
Oct 11, 2016, 12:35 AM
Registered User
Hi Felix,

Flew 6 packs with DShot on 2 different builds with Kiss FC and Kiss 24a. Felt really nice in the air. Motors came down colder than I am used to on my noisy Skitzo build.

Two times I had what I think was a failsafe but I'm not sure. I've never had a failsafe at this location and it was only 250 feet or closer. It happened once on each quad.

Here's the DVR with blackbox overlay:

Video with Blackbox Overlay (0 min 12 sec)
or https://dl.dropboxusercontent.com/u/...20Overlay.webm

Here's the Log (Note it's the 3rd flight in this log): https://dl.dropboxusercontent.com/u/...2/LOG00089.TXT

Here's the Raw DVR Video (Sync +0.485): https://dl.dropboxusercontent.com/u/.../Raw%20DVR.mp4

I don't have that much experience reading blackbox logs but it looks like at 0:58.649 in the log RC Command becomes stuck which is what I would guess a failsafe looks like. I do see the FC starting to command more power to motor #3 during this period until it maxes it out. RC Command then comes back at 0:59.406 and you can see me try and recover on the sticks but there is no response from motors.

Does this look like a failsafe event or some sort of issue with motor/ESC #3?

This happened on another build with different motors that I don't have a logger attached to.

Edit: After looking at it some more I'm not too sure RC Command actually sticks. I may have not been moving the sticks.
Oct 11, 2016, 12:45 AM
Registered User
Boris B's Avatar
Quote:
Originally Posted by RS2K
Great job! Digital is the future for sure! I've just finished repurposing the MSv2 signal to support your protocol on RaceFlight and RaceFlight One. We're ready for testing once the ESC firmware is ready.
Quote:
Originally Posted by RS2K
A lot of the jitter actually comes from the FC's calculations. You'd be amazed at how accurate PWM can really be when using good grounds and twisted pair. The real winner here is no need for calibration and guaranteed resolution no matter how noisy the line is (within reason of course).



I did a lot of testing with this in April (when developing MSv2) and that will happen at higher speeds when using DMA. There are ways around it though which it sounds like Felix has figured out.
You know what. I actually invented a wheel 1000years ago. Really everybody believes me.

Full rewrite was done too. And multishot is a new protocol too.

Come on man you are really starting to getting really funny.
Last edited by Boris B; Oct 11, 2016 at 12:58 AM.
Oct 11, 2016, 12:46 AM
Registered User
Thread OP
Quote:
Originally Posted by lcurole
Hi Felix,

Flew 6 packs with DShot on 2 different builds with Kiss FC and Kiss 24a. Felt really nice in the air. Motors came down colder than I am used to on my noisy Skitzo build.

Two times I had what I think was a failsafe but I'm not sure. I've never had a failsafe at this location and it was only 250 feet or closer. It happened once on each quad.

Here's the DVR with blackbox overlay:

https://youtu.be/89LrFPOPGQI or https://dl.dropboxusercontent.com/u/...20Overlay.webm

Here's the Log (Note it's the 3rd flight in this log): https://dl.dropboxusercontent.com/u/...2/LOG00089.TXT

Here's the Raw DVR Video (Sync +0.485): https://dl.dropboxusercontent.com/u/.../Raw%20DVR.mp4

I don't have that much experience reading blackbox logs but it looks like at 0:58.649 in the log RC Command becomes stuck which is what I would guess a failsafe looks like. I do see the FC starting to command more power to motor #3 during this period until it maxes it out. RC Command then comes back at 0:59.406 and you can see me try and recover on the sticks but there is no response from motors.

Does this look like a failsafe event or some sort of issue with motor/ESC #3?

This happened on another build with different motors that I don't have a logger attached to.

Edit: After looking at it some more I'm not too sure RC Command actually sticks. I may have not been moving the sticks.

hi, first thanks !

im not sure what happend ... but i will make a version that sends the failed CRC's to the FC so we can log it.. just to make sute thats not the problem.

in my tests i had no crc problems.


regards

felix
Oct 11, 2016, 01:05 AM
Woohoo!
RS2K's Avatar
Quote:
Originally Posted by Boris B
You know what. I actually invented a wheel 1000years ago. Really everybody believes me.

Full rewrite was done too. And multishot is a new protocol too.

Come on man you are really starting to getting really funny.
Oct 11, 2016, 01:12 AM
Registered User
Boris B's Avatar
Quote:
Originally Posted by RS2K
Oct 11, 2016, 01:14 AM
Woohoo!
RS2K's Avatar
Quote:
Originally Posted by Boris B
Relax, the next beer's on me.

Oct 11, 2016, 01:33 AM
Closed Account
Quote:
Originally Posted by RS2K
Relax, the next beer's on me
Prestons arrogance has really rubbed off on you. I guess you guys also vote trump?
At least boris knows collaboration and doesn't have interest in profiting from hear say marketing.
Oct 11, 2016, 01:33 AM
cyrilfpv
lostcontrol's Avatar
What about not going OT????
Oct 11, 2016, 01:34 AM
Registered User
Quote:
Originally Posted by juuuut
hi, first thanks !

im not sure what happend ... but i will make a version that sends the failed CRC's to the FC so we can log it.. just to make sute thats not the problem.

in my tests i had no crc problems.


regards

felix
Thanks Felix!

I reverted the one build without logger to RC29-2 and 1.03h to see if I have any failsafes with it. I need to get me one of those carrier cables like Alex has so I can switch out my logger easily.

Would it also be possible to include the failsafe state in the blackbox data?
Oct 11, 2016, 01:46 AM
Woohoo!
RS2K's Avatar
Quote:
Originally Posted by Corporalbadger
Prestons arrogance has really rubbed off on you. I guess you guys also vote trump?
At least boris knows collaboration and doesn't have interest in profiting from hear say marketing.
Let's please not turn this into a troll thread. I think Felix is doing a good thing here which is why I'm happy to include support into RFC and RF1.


Quick Reply
Message:

Thread Tools