Fun, Learning, Friendship and Mutual Respect START  HERE


Unregistered
Go Back   HeliFreak > R/C Electronics Support > MSH Brain FBL Unit


MSH Brain FBL Unit MSH Brain Support Forum


Like Tree155Likes
Reply
 
LinkBack Thread Tools Display Modes
Old 02-21-2021, 04:25 PM   #1 (permalink)
Registered Users
 
Posts: 1,128
 

Join Date: Jul 2016
Default Rx Fail Safe (150) without Hold or enough Fades?

Hey all, and especially BrainDev,

today I had an "oh shit" moment when the motor cut-off during flight. Luckily motor got back and gained enough headspeed before it hit the ground. Being back home I looked at the logs and now I am completely puzzled and unsure about the reliability of my setup.

Setup is:
Brain2, on most recent firmware (3.3.039)
TX Spektrum DX8 G2
RX SPM4651T SLRX2
Remote RX SPM9645

This is the diagnostics & events (better quality pics, full log file as well config mtr attached):





As you can see at the time of the motor cut, the SLRX2 rx had a couple of consecutive fades, but the remote had only a minor number of fades. The Brain did not log one single frame lost at all, and overall the number of fades is also low and very acceptable.

Interestingly, if I look at the DX8 telemetry log, there is one single frame lost (only 1!) and no hold at all.



Now, I wonder, why did the brain cut the throttle? The RX-TX signal did seem quite happy? From the data, everything was fine as the remote was doing its job perfectly well.

I have the impression, interpreting the provided telemetry data, that the lost frame on the main SRXL2 rx got somehow interpreted wrong and led to the short throttle cut.

Not sure what to do now, thb. I don't feel confident sending it again like this, but also not sure what do change / fix / improve.

Thanks for your help
Attached Thumbnails
Click image for larger version

Name:	Kraken580_Brain_Diag.png
Views:	541
Size:	464.5 KB
ID:	905971   Click image for larger version

Name:	Kraken580_Brain_Events.png
Views:	555
Size:	180.2 KB
ID:	905973   Click image for larger version

Name:	Kraken580_DX8.png
Views:	530
Size:	68.1 KB
ID:	905975  
Attached Files
File Type: csv Kraken580_210221_3.csv (1.67 MB, 21 views)
File Type: mtr 210221.mtr (16.6 KB, 23 views)
cyclo likes this.
__________________
Michael
Kraken 700 & 580; Oxy 4 MAX; Oxy 3 TE18

Last edited by mulchi; 02-22-2021 at 03:37 PM.. Reason: Typo
mulchi is online now        Reply With Quote Quick reply to this message
Sponsored Links
Advertisement
 
Old 02-21-2021, 05:58 PM   #2 (permalink)
Registered Users
 
Posts: 12,444
 

Join Date: Feb 2012
Location: Northern, VA
Default

It does look like the rcvr fades did spike, although not sure where that point is where the fades reach a level that it goes into fail safe. I do know that even at 50hz, you may not see the level that it reaches before that fail safe triggers. Glad you saved it.
__________________
-Ben
Oxy 2/3/4/5; XL 550/700; Protos 380/700 EVO; RAW; Jeti DS12 Carbon

EGODRIFT Team, Lynx/Oxy Field Rep
jrman83 is offline        Reply With Quote Quick reply to this message
Old 02-21-2021, 06:50 PM   #3 (permalink)
Registered Users
 
Posts: 5,753
 

Join Date: Jan 2010
Default

Quote:
Now, I wonder, why did the brain cut the throttle?
The flight controller cut the throttle because the flight controller went into fail safe.
Also, there are several Throttle and Elevator output limits reached alerts and the one last Elevator output limit reached at the same instant the flight controller went into Fail Safe: What was happening at the point?

Both the flight controller and the transmitter recorded a burst of fades a the same time.
Having simultaneous fades on Receiver A (the SPM4651T) and the Sat1 will cause a Frame Loss (and it recorded it).
The question is what caused both receivers to suffer a burst of fades at the same time.
That could have been voltage, placement of the receivers on the airframe or a problem on the transmitter.

I guess a good question would be "Does a single Frame Loss cause the flight controller to go into fail safe?". I would say "Yes, it looks like it does".
Maybe you should add another sat.
__________________
"Those who can make you believe absurdities can make you commit atrocities." (Voltaire)
Watch out: There are a lot of absurdities being thrown around lately.
+ > RIP: John Lewis
ticedoff8 is online now        Reply With Quote Quick reply to this message
Old 02-21-2021, 07:05 PM   #4 (permalink)
Registered Users
 
Posts: 12,444
 

Join Date: Feb 2012
Location: Northern, VA
Default

What ESC and are you using a buffer pack?
__________________
-Ben
Oxy 2/3/4/5; XL 550/700; Protos 380/700 EVO; RAW; Jeti DS12 Carbon

EGODRIFT Team, Lynx/Oxy Field Rep
jrman83 is offline        Reply With Quote Quick reply to this message
Old 02-21-2021, 07:32 PM   #5 (permalink)
Registered Users
 

Join Date: Oct 2010
Default

Is it a 12s setup or 10s setup on the 580?
__________________
Agile 5.5, Logo 550sx, X5, Goblin 500, Oxy 3, Trex 450proV2, 250DFC #309
nflanagin is online now        Reply With Quote Quick reply to this message
Old 02-21-2021, 07:58 PM   #6 (permalink)
Registered Users
 

Join Date: Nov 2005
Default

It looks to me that the receiver shut down the throttle after ~ 50 fades/frame losses as it went into fail safe.

Brain dev would have to comment as to what the Brain does if a receiver goes FS but the satellite is still (apparently) giving valid data.

Jim Davey
__________________
teamdavey
teamdavey is offline        Reply With Quote Quick reply to this message
Old 02-22-2021, 02:08 AM   #7 (permalink)
Registered Users
 
Posts: 1,128
Thread Starter Thread Starter
 

Join Date: Jul 2016
Default

Thanks for your replies guys.

It is a 10S system using Scorpion 200A ESC with a 1st Buffer-Pack 25F as failsafe.

@ticedoff8: The interesting part is, that only one RX had lots of fades. The other RX (Sat1) had only very few fades not nearly enough for a fail safe.

45 consecutive lost frames (across all RX) are needed to cause a single hold. As per TX log (none logged at Brain), at most one frame was lost in total. RX lost (>45), but Sat1 only lost 5, which is totally acceptable.

Therefore both together should not have caused a fail safe condition, as this is exactly what the RX redundancy is all about (and which normally works flawlessly with other RX/Sat combinations).

Regarding what I was doing, probably practicing pyro flips or something similar. The output limits reached just indicate that I probably should tune down my rates (or increase pitch / head speed?).

The main RX is installed as usual on a Kraken (on top with the 3d printed piece) and it sometimes get shielded by the body (not unusual). Therefore I have installed the RX to give good coverage during this period.
Attached Thumbnails
Click image for larger version

Name:	IMG_20200627_202648.jpg
Views:	52
Size:	969.1 KB
ID:	906073  
__________________
Michael
Kraken 700 & 580; Oxy 4 MAX; Oxy 3 TE18
mulchi is online now        Reply With Quote Quick reply to this message
Old 02-22-2021, 10:24 AM   #8 (permalink)
Registered Users
 
Posts: 3,208
 

Join Date: May 2018
Location: Mosquito Farm
Default

The FBL calculates its own values for Fades/Holds based on what it sees. The calculation was updated a few releases back to be consistent with how Spektrum calculates them.

Kinda makes me think that the ‘Hold’ is triggered by:
(Fades RX1 + Fades RX2)

instead of:
MINIMUM(Fades RX1, Fades RX2)
__________________
Jeti DS12 / Ikon2 / Oxy2/4/5 / SAB Mini Comet / Fireball / G380 / Gaui X5v2 / Kraken 580
JimJW is offline        Reply With Quote Quick reply to this message
Old 02-22-2021, 11:58 AM   #9 (permalink)
Registered Users
 
Posts: 1,128
Thread Starter Thread Starter
 

Join Date: Jul 2016
Default

I did an experiment on the bench. I disconnected the SRXL2 RX to force a "disconnect" and then reconnect after a while. What happend was quite interesting, and is actually similar to what happened in flight:

After the disconnect everything worked fine, as expected, I had full control over the model the whole time while the SRXL2 RX was disconnected AND the motor was running.

However, when connecting the SRXL2 RX back, the motor would STOP and start again immediately (and there is a 150 event logged).

I recorded the experiment (sorry for vertical)
https://photos.google.com/share/AF1Q...83SFJKV3BXZF9n

FYI, when I disconnect the satellite and reconnect after a few seconds nothing like this happens.

I have the strong impression, that after the SRXL2 RX comes back online (e.g. after being shielded for consecutive 45 frames), the Brain either gets a signal from that RX that a fail safe occurred and stop-re-starts the motor , or it somehow generates this signal on its own.

It is definitely reproducible, which is probably good.

Here is the log of the experiment:
Attached Thumbnails
Click image for larger version

Name:	no_force.png
Views:	27
Size:	300.2 KB
ID:	906111   Click image for larger version

Name:	Forced_Disconnect_Reconnect.png
Views:	465
Size:	301.6 KB
ID:	906113  
Attached Files
File Type: csv Kraken580_220221_8_forced_disconnect_bench.csv (170.4 KB, 9 views)
File Type: csv Kraken580_220221_10_normal_bench.csv (411.5 KB, 10 views)
cyclo likes this.
__________________
Michael
Kraken 700 & 580; Oxy 4 MAX; Oxy 3 TE18
mulchi is online now        Reply With Quote Quick reply to this message
Old 02-23-2021, 08:30 AM   #10 (permalink)
Registered Users
 
Posts: 1,128
Thread Starter Thread Starter
 

Join Date: Jul 2016
Default

Out of curiosity I did the same experiment today on my Oxy4max which sports an old and trusted setup (SPM4649T RX + 1 satellite).

The results is as expected (in a positive way) and very different from my test with the SRXL2 RX:

- No motor cut-re-engage after re-connecting main RX
- No event 150 in the events log
- No total frame lost, only fades from main RX when disconnected
- No "fake" fades from satellite while main RX is disconnected

The (boring) logfile of that tests looks like this, and this is precisely what I would expect to see also with the new SRXL2:



I think I am done now with testing and am looking forward to a fix
Attached Thumbnails
Click image for larger version

Name:	Oxy4max_disconnect.png
Views:	449
Size:	276.9 KB
ID:	906179  
Attached Files
File Type: csv Oxy4_210223_16.csv (261.8 KB, 10 views)
cyclo likes this.
__________________
Michael
Kraken 700 & 580; Oxy 4 MAX; Oxy 3 TE18

Last edited by mulchi; 02-23-2021 at 11:58 AM..
mulchi is online now        Reply With Quote Quick reply to this message
Old 02-23-2021, 11:31 AM   #11 (permalink)
Registered Users
 
Posts: 5,753
 

Join Date: Jan 2010
Default

Quote:
@ticedoff8: The interesting part is, that only one RX had lots of fades. The other RX (Sat1) had only very few fades not nearly enough for a fail safe.
45 consecutive lost frames (across all RX) are needed to cause a single hold. As per TX log (none logged at Brain), at most one frame was lost in total. RX lost (>45), but Sat1 only lost 5, which is totally acceptable.
My understanding is that with DSMX, all it takes is to have all the receivers experience a fade at the same time. I'm pretty sure "count to 45 and trigger a frame loss" was the standard for DSM2.

There appears to be no doubt the receivers (somehow) triggered a frame loss.
And it appears the 1 frame loss put the flight controller into fail-safe.
And a fail-safe causes the flight controller to cut the throttle.

I know there was a discussion about how the "Fade", "Frame Loss" "Hold" counts were calculated in this forum last year. You may want to search through here for that and see if there is something helpful in it.
__________________
"Those who can make you believe absurdities can make you commit atrocities." (Voltaire)
Watch out: There are a lot of absurdities being thrown around lately.
+ > RIP: John Lewis
ticedoff8 is online now        Reply With Quote Quick reply to this message
Old 02-23-2021, 11:55 AM   #12 (permalink)
Registered Users
 
Posts: 379
 

Join Date: Jul 2005
Default

Just for the sake of clarity the SPM4649t is an SRXL receiver and not SRXL2 like the SPM4651t. I am pretty sure you know, maybe edit your last post for clarity.

Also, have you tried different firmware versions in the FBL?
__________________
VE7FM
TheSteve is online now        Reply With Quote Quick reply to this message
Old 02-23-2021, 12:01 PM   #13 (permalink)
Registered Users
 
Posts: 1,128
Thread Starter Thread Starter
 

Join Date: Jul 2016
Default

Quote:
Originally Posted by TheSteve View Post
Just for the sake of clarity the SPM4649t is an SRXL receiver and not SRXL2 like the SPM4651t. I am pretty sure you know, maybe edit your last post for clarity.
Absolutely, thanks for mentioning it. Already fixed.

Quote:
Originally Posted by TheSteve View Post
Also, have you tried different firmware versions in the FBL?
Haven't done that yet. I updated the firmware of the 4651T to the latest public beta fw and repeated the test (same result).
__________________
Michael
Kraken 700 & 580; Oxy 4 MAX; Oxy 3 TE18
mulchi is online now        Reply With Quote Quick reply to this message
Old 02-23-2021, 01:12 PM   #14 (permalink)
Registered Users
 
Posts: 2,312
 

Join Date: Jun 2016
Default

Hmm, so some issues still with SRXL2? Might be worth emailing your findings to [email protected] in case Rik (BrainDev) isn't aware yet?

Sent from my SM-G981B using Tapatalk
cyclo and mulchi like this.
AndyO is offline        Reply With Quote Quick reply to this message
Old 02-23-2021, 02:51 PM   #15 (permalink)
Registered Users
 
Posts: 314
 

Join Date: May 2014
Location: France
Default

Hi guys, I had a very similar issue in November, similar setup (4651 + 2Sats, DX8G2). I posted some data on MSH FB page and somebody suggested to check static issues (belt driven tail) and I was lacking continuity. After having added a wire from the tail gear box to the main plate it never happened again. This was my FB post:

"Today during one of my flights I had a short loss of power (less than 1s). Never happened before on that machine; see graph below at 196 seconds. On the events file I found the event 150 (loss of Tx signal, fail safe), see picture.
On the logs you can see a rapid increase of RX fades (from 10 to 54) just before the signal loss, recovered very quickly...BEC voltage looks OK, no decrease in that period. Any idea of what happened?
Config: BRAIN2 + 2SATs + SPM4651T, last FW everywhere (Brain and SPM4651). HW130A ESC, Spektrum DX8G2. I have flown that config at least 30 times recently, no issue whatsoever (I check the events every time)...No other alarms on the other flights of the same session today. TX Fades always increase progressively during the flight, no more than 20/30 per flight..."

Attached Thumbnails
Click image for larger version

Name:	123207081_807182846787969_1662989001256409562_n.jpg
Views:	558
Size:	36.0 KB
ID:	906291   Click image for larger version

Name:	123233679_807182896787964_85331036515139470_o.jpg
Views:	562
Size:	76.9 KB
ID:	906293  
mulchi likes this.
__________________
Goblin 500 sport / MKS HV9767 & tail KST servo/ HW120A V4 / EGODRIFT 4020HS-1050 / DH blades 511 & 87
Goblin Black Thunder Sport 700 / BK 7001 & MKS HBL669 / HW130AV4 / XNOVA 4530-500 / RTech ULT700 & 106. Both with BRAIN2BT, SPM4651T & DX8G2
lucasouthoffrance is offline        Reply With Quote Quick reply to this message
Old 02-23-2021, 03:57 PM   #16 (permalink)
Registered Users
 
Posts: 1,128
Thread Starter Thread Starter
 

Join Date: Jul 2016
Default

Hey Luca,

Sounds similar. Maybe you could give it a try on the bench and make a similar experiment like I did? World be interesting to see if it dies the same.
__________________
Michael
Kraken 700 & 580; Oxy 4 MAX; Oxy 3 TE18
mulchi is online now        Reply With Quote Quick reply to this message
Old 02-23-2021, 05:10 PM   #17 (permalink)
Registered Users
 
Posts: 3,208
 

Join Date: May 2018
Location: Mosquito Farm
Default

Quote:
Originally Posted by mulchi View Post
Hey Luca,

Sounds similar. Maybe you could give it a try on the bench and make a similar experiment like I did? World be interesting to see if it dies the same.
For whatever it’s worth, I can try it out as well. Might be the weekend at this point.

I had a similar experience last summer on my first flights with a SPM4651T. It was different versions of the Rx and FBL firmware (obviously).

It seems as though there is some ‘incompatibilities’ between the Ikon / SRXL2 RXs. (Without pointing fingers either way)
mulchi likes this.
__________________
Jeti DS12 / Ikon2 / Oxy2/4/5 / SAB Mini Comet / Fireball / G380 / Gaui X5v2 / Kraken 580
JimJW is offline        Reply With Quote Quick reply to this message
Old 02-24-2021, 05:40 AM   #18 (permalink)
Registered Users
 
Posts: 1,128
Thread Starter Thread Starter
 

Join Date: Jul 2016
Default

As suggested I tested an older FW version (019) and this is the result:



I did three disconnect-reconnect cycles (visible as +45 increase in RX fades). All of them were recorded in the events log with a 150 event. The first one did not lead to a motor cut, but the second and third did.

No fake fades were recorded for Sat1.
Attached Thumbnails
Click image for larger version

Name:	Kraken580_disconnect_019.png
Views:	425
Size:	279.6 KB
ID:	906351  
cyclo likes this.
__________________
Michael
Kraken 700 & 580; Oxy 4 MAX; Oxy 3 TE18
mulchi is online now        Reply With Quote Quick reply to this message
Old 02-24-2021, 02:34 PM   #19 (permalink)
Registered Users
 
Posts: 379
 

Join Date: Jul 2005
Default

Thanks for trying another firmware version. I would expect that a signal fade is not quite the same as disconnecting and reconnecting an SRXL2 receiver. However if a SRXL2 receiver was to restart in flight(for whatever reason) I wouldn't expect the FBL to go into fail safe at any point if a valid signal was being received by a satellite the entire time.
__________________
VE7FM
TheSteve is online now        Reply With Quote Quick reply to this message
Old 02-24-2021, 03:46 PM   #20 (permalink)
Registered Users
 
Posts: 3,208
 

Join Date: May 2018
Location: Mosquito Farm
Default

mulchi, maybe try shielding the antenna of the SPM4651 with some tinfoil to block the signal. Does it have the same effect? (i.e. With no reboot on the RX)
__________________
Jeti DS12 / Ikon2 / Oxy2/4/5 / SAB Mini Comet / Fireball / G380 / Gaui X5v2 / Kraken 580

Last edited by JimJW; 02-24-2021 at 03:57 PM..
JimJW is offline        Reply With Quote Quick reply to this message
Reply




Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the HeliFreak forums, you must first register.
Please enter your desired user name, your REAL and WORKING email address and other required details in the form below.
User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself. Use a real email address or you will not be granted access to the site. Thank you.
Email Address:
Location
Where do you live? ie: Country, State, City or General Geographic Location please.
Name and Lastname
Enter name and last name here. (This information is not shown to the general public. Optional)
Helicopter #1
Enter Helicopter #1 type and equipment.
Helicopter #2
Enter Helicopter #2 type and equipment.
Helicopter #3
Enter Helicopter #3 type and equipment.
Helicopter #4
Enter Helicopter #4 type and equipment.

Log-in


Thread Tools
Display Modes

Posting Rules
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On




Copyright © Website Acquisitions Inc. All rights reserved.
vBulletin Security provided by vBSecurity v2.2.2 (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.

SEO by vBSEO 3.6.1