One of those days....

[Ocean]

we need to buy you a humidifier

I think the biggest issue KISS has is 2300kv cobras with bullnose pulling a lot more amps than expected (and subpar tuning), combined with the issues cleanflight has in 1.9 with vibrations... the new filters take care of a lot of that though. I want to run bulls on 2300, but I think I'm going to wait for the 30a version before I give it a go

You could try 5x4x3.

Steele ran them in the freestyle event and swears by them.

 

[Cyberdactyl]

we need to buy you a humidifier

Oh man, I suppose that is one of the very small benefits of living in the southeast from May to December, is never having to worry much of static discharge.

I am spoiled, and get irritated when have to deal with 15%RH, I can only imagine the nightmare the southwest has to deal with pretty much everyday.

 

[cranialrectosis]

Dust and sand and wind and lightning and sometimes hail three or four times per week in the summer. It's nice. It cools off the day.

You grow to expect a little zap when you turn on the lights. It's like an old friend saying good morning!

 

[narcolepticltd]

Oh man, I suppose that is one of the very small benefits of living in the southeast from May to December, is never having to worry much of static discharge.

I am spoiled, and get irritated when have to deal with 15%RH, I can only imagine the nightmare the southwest has to deal with pretty much everyday.

^^ this. I can't breathe the air here it's so thick outside, but at least I know I'm not going to pop a component when I'm soldering

 

[jhitesma]

I had a bigger static issues in Ohio than I have here in AZ. I would have thought it would be the other way around given that our humidity here is usually <15%.

There are things I've learned will almost always give me a zap though, certain metal handles on certain cases at certain stores. My truck always did for awhile, not sure why that stopped.

In my house though I almost never get zapped. Must the the same faraday cage construction my walls seem to have that mean I need two wifi routers to cover 1,100 sqft of living space...and still have dead zones :black_eyed:

 

[narcolepticltd]

You could try 5x4x3.

Steele ran them in the freestyle event and swears by them.

I'm thinking the new (hopefully for sale very soon) non-bullnose HQ5x4.5s will perform as well and be a bit more efficient, we'll see. Awdspider got him hooked on those things though Definitely good stuff.

 

[zenguerilla]

Bruce has always believed the state is out to allegedly get him.

Here is a working video of the "Cruise Missile" story.

Part1


Part 2

Everyone knows this is bull...

It's 'the singularity' that is out to get... anyone.

Go'damn computer god... you know what they say... while the cat's away...

And what better a disguise... 'the cat' or a mirror...

Don't get me wrong... I don't harbor hate... but, we are mechanical / electrical...


Some say when you look into the matrix, the matrix looks into you. But, you know, when the matrix looks into you, you also look into the matrix...

But, endure enough humiliation... and it might satisfy something... unless you are really important... and, then it's just 'hell on earth'.

"Sorry miss Jackson..." ...I just be 'pregnant' with ideas and sh*t... ssstar fishhh....

 

[jhitesma]

Well, got my new heat shrink today, so won't be using silicone tape other than for field repairs from now on. Thankfully weather stayed a little cooler (barely over 100f) the past few days so the heat shrink didn't arrive pre-shrunk

Over lunch did a few bench tests on the hex trying to see if I could figure out which motor was running hot...and what I found was perplexing. I'm starting to wonder if maybe something happened to my FCB in that mild crash at FF.

Just hand holding with minimal throttle I'm seeing what at first looked like a possible reset. All the ESC's cut out (blue lights off then back on) but nothing was running hot yet. Did a few more tests and found I could get it to do it by giving the hex a quick tip while hand holding it, the more aggressively I tipped it the more likely it was things would freak out.

Then something really odd happened. Gave it a really strong tip to the right. All 3 motors/ESC's on the right and the left rear motor/ESC suddenly cut out. Lights off (and didn't come back on) while the two front left did keep going.

So now I'm starting to think it's either a FCB issue or something between the FCB and the ESC's. The hot ESC/motor I detected last weekend may have just been a symptom of whatever is going on. (Note, this testing is being done in acro mode to minimize the number of sensors involved.)

There is a new firmware build out for the Brain, but given I've been flying on the current build for 3.5 months with no issues I don't believe it's likely to be the firmware.

Taking the arms off for travel did include two screws that both hold the arms and the FCB stand-offs. Which has me wondering if maybe I overtightened those causing the FCB mount screws to tighten as well transmitting more vibrations to the FCB. Seems unlikely but possible.

It could be I damaged the gyro in the MPU-9250 - but again seems unlikely as it was a very mild crash and I've had MPU's survive MUCH harder crashes in the past.

I suppose it could be damage to signal wires on the ESC's - but they look good. One did look a bit beat up the other day but I replaced it. Or it could be damage to the PDB since the signals run through there...I'll have to do some major disassembly to inspect that.

Very odd situation indeed. But at lest I seem to be closing in on potential causes.

Just very odd that two different builds with very little in common both developed similar issues. Would be nice if my LC meter from China got here quicker so I could take a closer look at the motors...though the DYS with notably different resistances on the quad sure seems like the issue there.

 

[cranialrectosis]

Hope it isn't the pdb.

 

[jhitesma]

So do I...but it wasn't really designed to have the arms taken on and off without being disassembled like I did. So I could have scratched some of the solder mask and be causing a signal to short out somewhere...I have to look back at my build thread though I thought the PDB on this one was designed so the traces were safe....but it may have been after I discovered the design flaw in the quad PDB and this one may still have that potential. Would explain quite a bit. But would also be fairly easy to fix.

 

[jhitesma]

Looking at my build photos it looks like the two center arms signal traces on the PDB do run under the arms - so there is a chance that something has abraded and could be messing those up. But I'm still not sure.

With the top and FPV gear removed it's still doing the same thing for me. I thought for a bit that maybe the reset button on the arduino that makes up my RX was getting bumped - but confirmed even with it completely safe it's doing the same thing. And the power lights on
the RX never even dim. So it doesn't appear to be an RX cutting out issue.

I also hooked up my watt meter and saw no abnormalities in the power draw when the issue happens. No sudden spikes or dropouts - at least none that I was able to observe.

I really need to get my log parsing tools setup so I can review the flight logs and see what the FC is doing.

There is one other thing both this and the quad have in common I realized. They're both using pololu regs. Those should be pretty robust...but I have broken those little inductors they use in RX's before in crashes. So I'm starting to wonder if maybe the pololu's took some damage. They're well protected...but maybe just the shock from ground impact. Guess we'll see once the teardown gets a bit further along since the reg is really buried in there.

Whatever it is it does seem to be getting worse, I have't been able to run it long enough for a motor or ESC to even get warm again but it keeps doing what appears to be a reset. I also no longer suspect a change to the FC mounting affecting it as I realized that the two mount screws I redid are full hard points, the screws with o-rings on the board itself I haven't touched.

Hmmmm....

 

[jhitesma]

Found and fixed two potential issues...seems to be back to it's old self stable as can be with motors/ESC's running cool again after a few minutes of hovering and a couple of flips in the front yard.

Found one signal wire from the PDB to an ESC that had rubbed through it's insulation on the bottom against a screw. Thought I had checked them all but guess I missed this one. I checked the screw for continuity to anything and it seems isolated...but better safe than sorry replaced that wire.

Second was with how motors 5&6 were wired. The BrainFPV board only has 4 primary motor connections, any others have to be run though one of the aux connectors. I'm running the PPM+UART+Output config which puts my RX, a serial connection and motors 5/6 all on the main port. I have Pins 1/2/3 for GND/+V/PPM going to a 3 pin female connector that mates with my RX. Pin 4 is unused, Pins 5/6 are motor outputs 5/6 and are run to another 3 pin female connector, then pins 7/8 are TX/RX for a serial port that I have run to a 3 pin female header where I can plugin my bluetooth or if I get a new openLRS RX I could put the telemetry connection through there.

The wires for motors 5/6 come off the PDB and go to a short bit of IDE drive cable with a male pin header on the end that mates with the 5/6 connector. It's a little confusing as I got motors 5/6 mixed up so the connection is green to yellow and yellow to green...but that wasn't the problem I found. The problem was apparently I never shrunk the heat shrink over that male header. I think what was happening was the heat shrink had worked it was back and exposed some of the solder connection which was hitting something (bottom of the FC, or the other motor signal pins on the PDB) and shorting them out.

Pulled out the embossing heat gun and gently shrank the heat shrink. Then put the RX back on with a rubberband to hold it in place temporarially and it passed bench tests for the first time in almost a week.

Oh - I also did some tweaking on the DIY RX. Since I now have a lot of nice clear heat shrink again I took off the silicone tape I had on the RX - but doing so I pulled off the status LED by mistake. So re-soldered that and then put some heat shrink around the RX nice and clean.

I'm thinking that the motor 5/6 header connection was where the problem was. It was right over the connections on the PDB for all the other motors (I've been meaning to go back and put some liquid electrical tape over those once the build is "finished"...too bad my builds are never "finished") and when the whole thing would freak out except for two motors it was motors 5/6 that didn't freak out. So they were behaving differently than the others. I may just go ahead and get rid of the connection there and just solder the pigtail from the brain's connector right to the PDB. Eliminate a source of potential issues entirely.

Need to re-assemble the top and FPV gear though and hopefully give a good test flight this weekend before I fully declare it fixed though!

 

[cranialrectosis]

Good deal. It always feels great to fix one of these little gotcha problems!

 

[PHugger]

Good catch! So many people just give up and toss the whole thing. I remember a DEC repair technician who used to fixes our VAX machines. When something didn't work he would just replace all of the boards until it started working. I remember thinking - really? Basic debugging skills are lost on many people. Thanks for sharing both your successes and failures with us.


Best regards,
PCH

 

[Cyberdactyl]

. . . .The wires for motors 5/6 come off the PDB and go to a short bit of IDE drive cable with a male pin header on the end that mates with the 5/6 connector. It's a little confusing as I got motors 5/6 mixed up so the connection is green to yellow and yellow to green...but that wasn't the problem I found. The problem was apparently I never shrunk the heat shrink over that male header. I think what was happening was the heat shrink had worked it was back and exposed some of the solder connection which was hitting something (bottom of the FC, or the other motor signal pins on the PDB) and shorting them out. . . .

I've found you really need to check shrink after it's been on a multi after a few weeks. I use clear shrink as my only method to hold my ESCs to my hex. Over time the shrink needs a quick hit with the heat gun to tighten everything up again. The one on the left below is getting a little sloppy. I think micro vibrations never stop working at the material, especially if the shrink is doing a job beyond merely wrapping.

 

[jhitesma]

This was 100% my fault. As soon as I got the board off and saw that connection I remembered that I never shrunk those connections in the first place. The brain board arrived while my parents were visiting back in March - it was a busy visit and I didn't think I'd have time to get it installed...but really wanted to take my dad for a FPV flight. So I rushed the installation. Didn't have my heat gun handy and just made a mental note to go back and shrink them figuring they were "good enough" for a few quick test flights. Except with all the craziness of the visit that mental note just have fallen off my mental fridge and I never went back and finished cleaning it up.

Just glad it didn't act up at Flite Fest!

 

[jhitesma]

Got her hat back on and even gave her a spiffy new skirt to clean up her lines. First test pack flew great - looks like the gremlins have been successfully banished....for now.

 

[cranialrectosis]

How do you power your FPV cam and do you use a filter?

I need to move my lipo under the copter but that's a full takedown. Still it needs to happen to get this OSD working.

 

[jhitesma]

I just tap it right off one of the vbat+ pads twitchity provided around the PDB. I should add a filter but so far my video has been clean enough I haven't been in a big hurry. I have all the parts to build one...but I'm in a bigger hurry to add my current/voltage sensor and hook that up for the OSD. I just haven't figure out where/how I want to mount it yet even though it's been sitting on my desk taunting me for almost 3 months

I left my power lead long enough I can run the lipo on top or bottom. But usually put it on the bottom since I tend to fly this on a 2200 and a 2200 doesn't fit well on top with the camera mount and VTX connector up there.

I did find I've damaged the top plate where I had the GPS mounted - too many crashes and the CF is starting to delaminate around that one hole. So I may dab a bit of epoxy on it to hep reinforce it and work out a better GPS mounting method. I really should add some kind of landing gear as well...but landing on the lipo gives me incentive to make nice soft landings. With springy legs I tend to get lazy about my landings

Need to decorate that plain white skirt tonight. It's bugging me

 

[Cyberdactyl]

I like the idea. Almost no weight. Successful racers could lease the surface to sponsers. *winks at FGA*