Knockoff is the New Black

[jhitesma]

Have you updated to the latest baseflight build recently? Just saw this on the MW forums:

http://www.multiwii.com/forum/viewtopic.php?f=22&t=2387&start=2650#p55560

(I'd post it here but it's from TimeCop and not really family friendly...but the gist of it is that there was a bug causing hexes to lean forward that was fixed in a new build a few weeks ago.)

 

[FinalGlideAus]

Ahh Timecop, about as subtle as a rasp file on the eyeball .

That might be it Jhitesma, thanks for chiming in. It must have been a recent bug as there plenty of naze's flying hex's without dramas.

 

[califrag]

Hrmmmm goood spot... I will double check the receiver stuff and the firmware.

When I reverted back to baseflight from cleanflight, I just did the Online Firmware update, it pulled the version from 8/4.

I went ahead and flashed the firmware again, doing a 'full chip erase' this time.

I enabled just a few features: PPM, VBAT, and my telemetry stuff, and set the mixer to HEX6X. It still flips forward. I'll go about seeing how to compile the latest code.

 

[FinalGlideAus]

Don't forget about updating baseflight too.

 

[califrag]

Don't forget about updating baseflight too.

Just to clarify, you mean Baseflight Configurator?

I checked the version I had installed in Chrome it was at 0.50, I deleted it and reinstalled it, and it's still at 0.50, so I think I've got the latest version there.

I tried making a new model on my Tx, setting it up as a simple 4channel model, to eliminate any possible Tx trim or mixer issues, it still lurches forward.

 

[FinalGlideAus]

Ok this is getting annoying
Have you checked where the centre points of the sticks rest on the receiver screen like I said earlier? If the elevator green bar is not sitting on 1500 (or within a few points) it will be like having trim clicked into the tx.

 

[FinalGlideAus]

Ok this is getting annoying
Have you checked where the centre points of the sticks rest on the receiver screen like I said earlier? If the elevator green bar is not sitting on 1500 (or within a few points) it will be like having trim clicked into the tx.

 

[califrag]

Ok this is getting annoying
Have you checked where the centre points of the sticks rest on the receiver screen like I said earlier? If the elevator green bar is not sitting on 1500 (or within a few points) it will be like having trim clicked into the tx.

Seriously!!! On top of this whole mess I think I may have messed up one of my minimosd boards somehow during flashing. Oh well live and learn.

Here's what my receiver settings look like at rest. All values seem pretty OK

 

[califrag]

well this is beginning to get really frustrating now. I downloaded the latest baseflight code, compiled it in cygwin, flashed to the naze, set up everything again and it's still pitching forward. Gonna proceed with trying to get it running as a quad first then gonna worry about hex after.

 

[FinalGlideAus]

Yes..... frustrating..... Apologies that I can't track it down for you. It is always harder trying to solve issues through a computer screen...

 

[cranialrectosis]

I am running a Polakium hex with DYS 1806 motors on a Naze 32 and I have no such pitching forward issues. I am rebuilding my Twitchity Hex today (carbon fiber frame should arrive today ) and had no such issues with it running SunnySky 2204s and a Naze 32.

Baseflight is a chrome app. I don't think you can run anything but the latest version on your PC.

I am running firmware version 2.2 August 3 on my Polakium and 2.2 Feb 24 on my Twitchity hex (will update when I rebuild). Get the version by typing 'Version' into the CLI in Baseflight. If you are running 2.2 February or later, your issue is not with Baseflight or the firmware on the Naze.

Do you have the Hex set up as a Hex6 X in Baseflight? It will show on the first tab on the left side of your screen.

Having seen your video and seen my own Polakium do the same exact thing on my maiden in February, I believe you have ESC wires 1 and 4 or 2 and 3 swapped where connected to the Naze. It is REALLY easy to do. I have done it 4 or 5 times in my year of building. It happens to everyone.

Do you have a good photo of the copter showing the connections to the Naze?

Do you have a good photo of the copter showing all 6 rotors while they are stopped?

I may be wrong but it is a simpler fix than rebuilding your hex into a quad if I am right.

 

[califrag]

Yes..... frustrating..... Apologies that I can't track it down for you. It is always harder trying to solve issues through a computer screen...

I completely understand and I really appreciate all the help and suggestions you've given thus far.

I am running a Polakium hex with DYS 1806 motors on a Naze 32 and I have no such pitching forward issues. I am rebuilding my Twitchity Hex today (carbon fiber frame should arrive today ) and had no such issues with it running SunnySky 2204s and a Naze 32.

Baseflight is a chrome app. I don't think you can run anything but the latest version on your PC.

I am running firmware version 2.2 August 3 on my Polakium and 2.2 Feb 24 on my Twitchity hex (will update when I rebuild). Get the version by typing 'Version' into the CLI in Baseflight. If you are running 2.2 February or later, your issue is not with Baseflight or the firmware on the Naze.

Do you have the Hex set up as a Hex6 X in Baseflight? It will show on the first tab on the left side of your screen.

# version
Afro32 CLI version 2.2 Sep 3 2014 / 04:47:14
# mixer
Current mixer: HEX6X

Having seen your video and seen my own Polakium do the same exact thing on my maiden in February, I believe you have ESC wires 1 and 4 or 2 and 3 swapped where connected to the Naze. It is REALLY easy to do. I have done it 4 or 5 times in my year of building. It happens to everyone.

Do you have a good photo of the copter showing the connections to the Naze?

Do you have a good photo of the copter showing all 6 rotors while they are stopped?

I may be wrong but it is a simpler fix than rebuilding your hex into a quad if I am right.

Here are some pictures of the connections and rotors. If you need better close ups let me know.

The motors are labeled CW1-3 and CCW1-3 to note their spin direction and placement on the frame. Motors CW1 and CCW1 have the shortest leads to the ESC (front arms). CW2/CCW2 and CW3/CCW3 have the longer leads to ESC and are for the middle and rear arms.

You can see on the post-it how I've matched them up:

Motor 1: CCW3
Motor 2: CCW1

Motor 3: CW3
Motor 4: CW1

Motor 5: CW2
Motor 6: CCW2






I was able to get my minimosd connected to it, and I noticed that after calibrating the accelerometer level, the minimosd shows the false horizon line slightly below center. It thinks it is pitched back slightly, which I think explains why it is trying to pitch forward. I am gonna try re-calibrating the accelerometer a fourth time see if it makes any difference.

 

[cranialrectosis]

I stand corrected. Your ESCs are corrected right and your rotors are spinning the right direction.

You even use the same blue tape I use to keep them all straight. Sorry this didn't help out.

 

[califrag]

I stand corrected. Your ESCs are corrected right and your rotors are spinning the right direction.

You even use the same blue tape I use to keep them all straight. Sorry this didn't help out.

No reason to apologize, I am really grateful for all of the help and suggestions everyone has given so far. The double checking of everything still helps. It was just a big jumble of wires before and now it's a lot easier to spot any connection errors.

I think the issue may be with my accelerometer calibration simply due to what I saw in minimosd. I'll update if I make further progress, eliminate other possibilities, or after I test it as a quad.

 

[FinalGlideAus]

I completely understand and I really appreciate all the help and suggestions you've given thus far.



# version
Afro32 CLI version 2.2 Sep 3 2014 / 04:47:14
# mixer
Current mixer: HEX6X



Here are some pictures of the connections and rotors. If you need better close ups let me know.

The motors are labeled CW1-3 and CCW1-3 to note their spin direction and placement on the frame. Motors CW1 and CCW1 have the shortest leads to the ESC (front arms). CW2/CCW2 and CW3/CCW3 have the longer leads to ESC and are for the middle and rear arms.

You can see on the post-it how I've matched them up:

Motor 1: CCW3
Motor 2: CCW1

Motor 3: CW3
Motor 4: CW1

Motor 5: CW2
Motor 6: CCW2

View attachment 30518
View attachment 30519
View attachment 30520
View attachment 30521

I was able to get my minimosd connected to it, and I noticed that after calibrating the accelerometer level, the minimosd shows the false horizon line slightly below center. It thinks it is pitched back slightly, which I think explains why it is trying to pitch forward. I am gonna try re-calibrating the accelerometer a fourth time see if it makes any difference.

The interesting thing is though, if you fly in "rate" mode the accelerometers are not being used and that's why I asked if you had tried that flight mode. This is why I believe it is a trim issue as in rate mode the only thing I can think of that would hold a constant forward pitching force would be forward elevator stick or forward trim.
I'm guessing you're letting the hex be still after plugging in so it can calibrate the gyros? It beeps three times after it is done. Of course there is the possibility it's a bad board?????

 

[califrag]

The interesting thing is though, if you fly in "rate" mode the accelerometers are not being used and that's why I asked if you had tried that flight mode. This is why I believe it is a trim issue as in rate mode the only thing I can think of that would hold a constant forward pitching force would be forward elevator stick or forward trim.
I'm guessing you're letting the hex be still after plugging in so it can calibrate the gyros? It beeps three times after it is done. Of course there is the possibility it's a bad board?????

Ohhh hmm you're right I didn't even consider that the acc. were not being used... Well crap there goes that idea.

I don't have my buzzer connected to the Naze, but after plugging it in I would wait until all the ESC's beeped their tune before moving it. I'm pretty sure it was finished calibrating each time I tested but I'll plug the buzzer and listen for the beeps from the naze just to make sure.

The same board was in my SK450 quadcopter flying perfectly stable before I moved it into the hex so it's definitely not a bad board.

 

[ssteve]

Your video looks exactly like what I was experiencing when I set my hex up. It ended up being the esc's. That's after calibrating them a couple times through the transmitter and a pigtail I made. Try calibrating them through baseflight with min commands. Before that check and see where each motor is coming on and write it down. That one super long winded video is good for walking you through the process and there's a time stamp located in the info box so you can skip right to where the pertinent info is.

 

[califrag]

Your video looks exactly like what I was experiencing when I set my hex up. It ended up being the esc's. That's after calibrating them a couple times through the transmitter and a pigtail I made. Try calibrating them through baseflight with min commands. Before that check and see where each motor is coming on and write it down. That one super long winded video is good for walking you through the process and there's a time stamp located in the info box so you can skip right to where the pertinent info is.

Okay cool that's one thing I haven't double checked since I started this.

When I first connected everything I used a splitter to calibrate them:



After I had calibrated them directly using (two) of those and a receiver connected directly, I plugged the hex into the computer and did a calibration through baseflight using the following method:

Disconnect main battery and remove props
In Baseflight Configurator, go to Motor Testing tab, enable checkbox, set Master level to Full
Apply main battery to hex, wait for ESC initial beep then bring Master level to idle (0)
ESC's emit confirmation long beep, then go into normal 3beeps for cell count and ♪123 tune

I'm using EMax 12A SimonK Series which I think are actually using BLHeli ?

Also I noticed the screws for the prop adapters on the DYS BE1806 seem to be a little long, do you think they are too close to the coils and causing an issue?

 

[FinalGlideAus]

Ohhh hmm you're right I didn't even consider that the acc. were not being used... Well crap there goes that idea.

I don't have my buzzer connected to the Naze, but after plugging it in I would wait until all the ESC's beeped their tune before moving it. I'm pretty sure it was finished calibrating each time I tested but I'll plug the buzzer and listen for the beeps from the naze just to make sure.

The same board was in my SK450 quadcopter flying perfectly stable before I moved it into the hex so it's definitely not a bad board.

It should only take about 2 seconds after the hex has stopped moving after plugging in. I think the teen light also flashed 3 times so you can watch that as well. It's probably not it as it needs to calibrate before it allows you to fly but I thought I'd put it out there.

 

[jhitesma]

Those emax ESC's are a bit of a mystery...there's a debate in the blheli thread on RCGroups right now over just what firmware they're actually running. They're SiLabs chips and simonk doesn't run on SiLabs..but apparently emax claims they're NOT blheli but rather "based on simonk" which seems unlikely that they'd custom port simonk to SiLabs for a budget priced ESC. Then again China...it's not like they're hurting for programmers willing to do crazy jobs on the cheap.

Those screws do look long. I used the short ones on my prop adapters. I noticed when I used the long ones in the bottom I was able to feel the motor turn more stiffly just turning it by hand as they got closer and closer to the coils. If you can I'd try swapping to the shorter screws. Of course then you may have the longer ones on the bottom and have the same problem in reverse...but you could always add some washers or spacers (cut up some old hotel room keys or something) to take up the extra length on the bottom.