Naze32 hardware discussion thread

timecop · Post by **timecop** » Thu Oct 09, 2014 8:13 am

Hey how about you take javaflight discussion in the relevant thread, so people don't get confused of all the renamed variables? Thanks.

nepalien · Post by **nepalien** » Thu Oct 09, 2014 6:32 pm

I noticed in Baseflight (0.54) the battery voltage monitor values keep changing to 0.3 for min and 0.4 for max. I also noticed it happens every time I hit the save button in the Reciever Tab. Bug?? Has anybody else have this issue?

I am a total newbie so it my just be me.

BaseFlight ver. 0.54
Naze32 Acro with
# version
Afro32 CLI version 2.3 Oct 1 2014 / 07:45:46

jamieFL · Post by **jamieFL** » Thu Oct 09, 2014 9:31 pm

timecop,

I plugged a lipo into the board and got dreaded magic smoke and a message on my computer that a USB device was attempting to draw too much voltage. Earlier, USB only worked fine. The computer is fine. See photo of the damaged chip. This is not my board because I cannot take such clear closeup photos. Is it the voltage regulator and how do I purchase another to fix this?

jamieFL

timecop · Post by **timecop** » Fri Oct 10, 2014 5:49 am

Do you see that crater on the STM32 mcu? It's gone.
Not worth fixing either - unless you want a mess of melted connectors.

Chrisj800 · Post by **Chrisj800** » Fri Oct 10, 2014 7:39 am

Ive got a rev 5 pink board nearly a year old now and been through many quad crashes. Its now playing up. When I plug into my PC it no longer gets the moving orientation graphic when I move the quad around. Ive got 2 newer Naze32's that do still move the graphic properly so its not the PC! And now this old one no longer seems to start the motors in sync. Always one beeps later and I cant take off without twisting fwd and crashing. Thinking its a bad ESC or motor I replaced those but same issue. If I plug into a different Motor port then the delayed beep changes to a different motor which makes me think its the Naze32 board playing up. Any ideas anyone?

timecop · Post by **timecop** » Fri Oct 10, 2014 7:49 am

Have you bothered upgrading firmware?
Also, model won't move until green led is done blinking at power up.

Chrisj800 · Post by **Chrisj800** » Fri Oct 10, 2014 10:35 am

Thanks tc, I had done a firmware update back in Feb but not since so just did one and thats fixed the display. Ive still got an issue but think its now an esc or motor or solder joint. Love your board, still got 2 working fine!

nepalien · Post by **nepalien** » Fri Oct 10, 2014 7:17 pm

timecop, my battrey voltage monitor keeps randomly changing to 0.3 and 0.4. I set it to 3.4 and 4.2 and saving it and random it changes to 0.3 and 0.4. I noticed these values change when i hit the save button on the reciever tab but i am not sure what makes it change when it is not connected to PC.

BaseFlight ver. 0.54
Naze32 Acro with
# version
Afro32 CLI version 2.3 Oct 1 2014 / 07:45:46

timecop · Post by **timecop** » Sat Oct 11, 2014 7:05 am

Maybe configurator bug? Not sure, I'm not actually sure what the problem/question is, try modifying the values through CLI and make sure you're running latest version configurator too (chrome->menu->extensions->update all)

CVMichael · Post by **CVMichael** » Sat Oct 11, 2014 10:49 pm

Hi timecop,

I had only 2 flights with the Naze32 so far with my new wood H-Quad that I built about 2 weeks ago, and I am still amazed how stable and easy it is to setup and fly it, so thank you for that

I have a small problem though, and I don't know what settings to change to make it right. When I fly forward and want to stop (or go backwards), the quad always rotates to the right/clockwise (on the yaw axis). Also the same thing when I drop in height, for example, if I go about 50 meters, then I bring the throttle almost all the way down, the quad drops, and then I push the throttle all the way high to stop the fall, the quad always rotates to the right. Sometimes it rotates almost 90 degrees...
Do you know how to fix that ?

I did some flips today, and I loved it. But the flip is... big... (don't know how to describe it), so I have to be quite high (over 30 meters) to make sure that I don't hit the ground by the time the flip is complete. (Yes, I learned that from crashing it LOL)
I watched this video, and I set the same settings in my Naze32:
https://www.youtube.com/watch?v=BNtCKpBNrBY
Basically: Yaw rate at 0.35, and RC Rate at 1.25
I have seen other videos where other people flip it "on the spot', without loosing much height, and much faster flip. How do you do that?

timecop · Post by **timecop** » Sun Oct 12, 2014 3:58 am

You need higher pitch/roll rates to flip on the spot. Also higher RC_Rate if you want, but pitch/roll rate is what will determine faster spin.

Not sure about yawing, load latest firmware? there was a yaw-related bug that was fixed months ago

CVMichael · Post by **CVMichael** » Sun Oct 12, 2014 2:34 pm

Thanks, I will increase the pitch/roll rate.

I think I have the latest firmware, when I open baseflight it says "Running firmware is Oct 1, 2014"

cester · Post by **cester** » Sun Oct 12, 2014 6:43 pm

Hi, two of my Naze32 boards have a weird issue both with the 4th motor output, which doesn't send any signal to the esc while the other all other motors output work flawless. I tried everything, even resetting the board, but nothing helps. because it is the 4th pin I can't even use servo_flip to use other pins. Is it possible to use another motor output instead? like 1,2,3 and 5?
thank you!
EDIT: I would add that the first presented this problem immediately, the second after a month of use

jaxxzer · Post by **jaxxzer** » Sun Oct 12, 2014 11:10 pm

Hi.

I have a naze32 flashed with the latest firmware

# version
Afro32 CLI version 2.3 Oct 1 2014 / 07:45:46

If I power the board through usb, or a bec, and then power up my escs, all is well and everything works. HOWEVER.

If I power the board and escs simultaneously, I only get the startup tone from the escs, then no beeps for cell count or ready. Nothing, they stay silent.

The escs aren't receiving any pwm signal from the board until sometime after the board is turned on. If they don't receive pwm on startup, they wont accept input later.

I have successfully built and flown a acro earlier in the year, and I also have a new acro naze flashed with the same firmware that does not exhibit this behavior.
I am familiar with the cli set variables, mincommand, calibrating escs and all that, the same exact setup works fine when plugged into the acro with the same default settings. The only thing different between the two boards as far as I can tell is the extra sensors (mag and baro). I never plugged anything in backwards or any of that.

What is going on here?

Steve_in_NJ · Post by **Steve_in_NJ** » Mon Oct 13, 2014 3:35 am

I am setting up my first NAZE32. I have the board installed on my quad, and powered up via the receiver. I have it plugged into the configurator. Everything seems to be going fine but I have discovered what appears to be a discrepancy in the way the app identifies the motors. Here is the diagram I am presented with on the "setup" screen:

When I test the motors using the "motor testing" screen, with the sliders here...

...I find the motors spin up in a more standard layout. In this picture the front is at the top.
Here's what corresponds to 1,2,3,4 on the motor testing screen...

Why the difference? I assume that 1 is CCW, 2 is CW, 3 is CCW, 4 is CW - is that right?

Thanks much,
Steve

timecop · Post by **timecop** » Mon Oct 13, 2014 4:09 am

yea, pretty sure motor test just outputs in normal order. While mixer is different.
If it wasn't for years of retarded backwards compatibility, i'd say we should change mixer to use normal CW or CCW motor indexing without the dumb multiwii order. But too late now.

jaxxzer · Post by **jaxxzer** » Mon Oct 13, 2014 4:23 am

timecop, do you have any thoughts on my issue a couple posts back about no pwm output on powerup? I have searched for the answer, but turned up nothing on these overloaded threads.

timecop · Post by **timecop** » Mon Oct 13, 2014 4:45 am

jaxxzer wrote:Hi.

I have a naze32 flashed with the latest firmware

# version
Afro32 CLI version 2.3 Oct 1 2014 / 07:45:46

If I power the board through usb, or a bec, and then power up my escs, all is well and everything works. HOWEVER.

If I power the board and escs simultaneously, I only get the startup tone from the escs, then no beeps for cell count or ready. Nothing, they stay silent.

The escs aren't receiving any pwm signal from the board until sometime after the board is turned on. If they don't receive pwm on startup, they wont accept input later.

I have successfully built and flown a acro earlier in the year, and I also have a new acro naze flashed with the same firmware that does not exhibit this behavior.
I am familiar with the cli set variables, mincommand, calibrating escs and all that, the same exact setup works fine when plugged into the acro with the same default settings. The only thing different between the two boards as far as I can tell is the extra sensors (mag and baro). I never plugged anything in backwards or any of that.

What is going on here?

No idea, really.
Flash w / "Full Chip Erase".
If that doesn't fix it, no clue, because this stuff is in "can't happen" territory.

jaxxzer · Post by **jaxxzer** » Mon Oct 13, 2014 4:59 am

yeah i did the full chip erase, I have no clue. It's really weird it works perfect when the fc is powered up, then everything else. I registered here just to post, in all of doing this stuff I've been able to figure everything out, and this makes no sense at all to me. I've gone over it and over it and can't figure out what I'm overlooking. BTW I appreciate your approach to feedback on these threads lol I've seen a lot but never taken part.

jaxxzer · Post by **jaxxzer** » Mon Oct 13, 2014 6:41 am

If it makes a difference (which I don't think it should, since this doesn't happen with the acro version), the escs are cheap cheap hobbywing 30amp copies from china with linear BECs and original firmware, there doesn't seem to be any delay in 5V out, though.

cester · Post by **cester** » Mon Oct 13, 2014 8:08 am

Hey Timecop could you help me with this?

cester wrote:Hi, two of my Naze32 boards have a weird issue both with the 4th motor output, which doesn't send any signal to the esc while the other all other motors output work flawless. I tried everything, even resetting the board, but nothing helps. because it is the 4th pin I can't even use servo_flip to use other pins. Is it possible to use another motor output instead? like 1,2,3 and 5?
thank you!
EDIT: I would add that the first presented this problem immediately, the second after a month of use

Steve_in_NJ · Post by **Steve_in_NJ** » Mon Oct 13, 2014 3:45 pm

timecop wrote:yea, pretty sure motor test just outputs in normal order. While mixer is different.
If it wasn't for years of retarded backwards compatibility, i'd say we should change mixer to use normal CW or CCW motor indexing without the dumb multiwii order. But too late now.

Well the good part is that at least both the "multiwii order" and the "normal order" both have the motors spinning in the same directions!

jamieFL · Post by **jamieFL** » Mon Oct 13, 2014 4:00 pm

timecop wrote:Do you see that crater on the STM32 mcu? It's gone.
Not worth fixing either - unless you want a mess of melted connectors.

timecop,

The board pictured is not mine. I cannot take good closeup photos so I found one on the net. The fryed component is the one circled. What is it and where can I get one?

jaxxzer · Post by **jaxxzer** » Mon Oct 13, 2014 5:45 pm

It sounds to me like you have your escs plugged into the wrong pins. If you dont think this is the mixup an easy fix would be to plug the front left motor which should be 4 to the pin that outputs when you test motor 4, and the same with the rest. And when you are done you have discovered the right pins.

kilby · Post by **kilby** » Mon Oct 13, 2014 7:39 pm

jaxxzer wrote:It sounds to me like you have your escs plugged into the wrong pins. If you dont think this is the mixup an easy fix would be to plug the front left motor which should be 4 to the pin that outputs when you test motor 4, and the same with the rest. And when you are done you have discovered the right pins.

I would agree but I have also seen people's esc numbering not matching the motor numbering.

Fairly obvious though that output 1 doesn't correspond with motor 1

Steve_in_NJ · Post by **Steve_in_NJ** » Mon Oct 13, 2014 7:50 pm

jaxxzer wrote:It sounds to me like you have your escs plugged into the wrong pins. If you dont think this is the mixup an easy fix would be to plug the front left motor which should be 4 to the pin that outputs when you test motor 4, and the same with the rest. And when you are done you have discovered the right pins.

Thanks for that. I tried your technique, but when I test the config by holding on to the quad while pushing it front / back / left / right with the TX, it still does not respond correctly.

SO THEN, below is a picture of the NAZE32 with the pins labelled the same as on the back of the board. Will someone please tell me which motor of an X quad goes to which pin? And please tell me if you are using the "regular" method of identifying motors, or the "multiwii" method, also as shown below?

kilby · Post by **kilby** » Mon Oct 13, 2014 8:30 pm

Viewing from the rear of the quad.

Esc output 1 to closest right land side motor

Esc output 2 to furthest right hand motor

Esc output 3 to nearest left hand motor

Esc output 4 to furthest away left hand motor.

It is simplicity itself provided you have your esc's labeled properly and your esc's are connected to the correct motors.

Your image shows all you need to know.

Don't worry about the direction of rotation (yet)

I also assume that you have the arrow on the naze pointing towards the front of the quad (yes I did that once)

Steve_in_NJ · Post by **Steve_in_NJ** » Mon Oct 13, 2014 9:45 pm

kilby wrote:Viewing from the rear of the quad.

Esc output 1 to closest right land side motor

Esc output 2 to furthest right hand motor

Esc output 3 to nearest left hand motor

Esc output 4 to furthest away left hand motor.

It is simplicity itself provided you have your esc's labeled properly and your esc's are connected to the correct motors.

Your image shows all you need to know.

Don't worry about the direction of rotation (yet)

I also assume that you have the arrow on the naze pointing towards the front of the quad (yes I did that once)

Kilby,

OK, thanks.
Yes the arrow is pointing forward.

I already set up the motors to rotate in the correct directions

I connected the motors as you described.

OK, I found the problem. I was screwed up in two different ways - I needed direction to get over the "multiwii vs normal" motor assignments, and just as bad, I discovered that I had not changed my radio from the default R-E-T-A setup on channels 1-4 to the correct A-E-T-R. UGH. That created some interesting effects!

I just ran outside to give this thing another test. Much better results than last time! Thanks again for your help. I have more, new questions, but they can wait.

Here's the happier setup:

Steve

Steve_in_NJ · Post by **Steve_in_NJ** » Wed Oct 15, 2014 2:20 am

So I just had a couple of flight tests of my freshly-built "FPV250v3" frame with a NAZE32 at it's center. It's 572gm AUW. I am using 5030 2-blade gemfan props, Sunnysky X2204S 2300kv motors, 12amp SimonK ESCs, and a 1600mah 3S lipo. I am finding that it is really quite sluggish to respond to more or less throttle - that is, if I give it a bit more, it takes a few seconds before it will start to rise, and then start to shoot up. Give it a little less throttle and it will continue to rise for a moment, then drop - more than one would expect. All other controls seem just fine -very responsive. Any suggestions on how to tweak out this odd throttle behavior? All settings are factory default at the moment.

I have had one person suggest the fault is the 5030 props - I am told "they are weak for the weight". So then, is there anything to do with the NAZE32 here, or is my problem a "hardware" one? thanks...

timecop · Post by **timecop** » Wed Oct 15, 2014 4:23 am

Did you enable althold by default?

jamieFL · Post by **jamieFL** » Wed Oct 15, 2014 11:07 pm

timecop,

You replied to this once before, but didn't answer my question...The photo is not my board but one I found on the net.

I plugged a lipo into the board and got dreaded magic smoke and a message on my computer that a USB device was attempting to draw too much voltage. Earlier, USB only worked fine. The computer is fine.

See photo of the damaged chip. Is it the voltage regulator and how/where can I purchase another to fix this?

jamieFL

: What is circled?

cester · Post by **cester** » Wed Oct 15, 2014 11:09 pm

Timecop, do you have any ideas about my issue I posted some days ago? I'm waiting for an answer before throw away these two boards and buy new one (Can't afford it right now)
Thank you.

kilby · Post by **kilby** » Wed Oct 15, 2014 11:30 pm

Steve_in_NJ wrote:
OK, I found the problem. I was screwed up in two different ways - I needed direction to get over the "multiwii vs normal" motor assignments, and just as bad, I discovered that I had not changed my radio from the default R-E-T-A setup on channels 1-4 to the correct A-E-T-R. UGH. That created some interesting effects!

Glad you got that fixed.

Remember if you wipe the naze and restore your settings the CLI values are not included, you will have to remap your channels, spin motors on arming and all such other goodies

kilby · Post by **kilby** » Wed Oct 15, 2014 11:30 pm

Oh Rats, Double post

kilby · Post by **kilby** » Wed Oct 15, 2014 11:44 pm

JamieFL,

Probably a regulator (can't be arsed to look at my own naze, but it's got 3 lens and looks like it's in a likely position for a regulator) you have blown the crap out of.

Even if you had the ability remove and solder the replacement one in place, there's every chance something else is fucked too.

Unless you set a very low $ value on your time, buy a new one they're affordable

strepto · Post by **strepto** » Thu Oct 16, 2014 1:07 am

kilby wrote:JamieFL,

Probably a regulator (can't be arsed to look at my own naze, but it's got 3 lens and looks like it's in a likely position for a regulator) you have blown the crap out of.

Even if you had the ability remove and solder the replacement one in place, there's every chance something else is fucked too.

Unless you set a very low $ value on your time, buy a new one they're affordable

Actually I'm pretty sure it's a diode array thingie to stop power from USB and other sources crossfeeding. The rev 4 schematic I have saved somewhere (can't seem to find it online now) lists that as "D4 BAT54CTA".

Of course, I could be wrong and talking out my arse here.

If it's fried you may well have fried the regulator(s) too though (the 5 pin chips). If you remove it, and bridge two of the three connection, you could possibly use the board through one power source only. But then again I could also be talking out my arse and you could blow everything up. Caveat emptor!

timecop · Post by **timecop** » Thu Oct 16, 2014 2:03 am

jamieFL wrote:timecop,

You replied to this once before, but didn't answer my question...The photo is not my board but one I found on the net.

Naze32-Rev5---magic-smoke.jpg

BAT54C diode.

timecop · Post by **timecop** » Thu Oct 16, 2014 2:08 am

cester wrote:Hey Timecop could you help me with this?
cester wrote:Hi, two of my Naze32 boards have a weird issue both with the 4th motor output, which doesn't send any signal to the esc while the other all other motors output work flawless. I tried everything, even resetting the board, but nothing helps. because it is the 4th pin I can't even use servo_flip to use other pins. Is it possible to use another motor output instead? like 1,2,3 and 5?
thank you!
EDIT: I would add that the first presented this problem immediately, the second after a month of use

Don't know, usually if I don't answer something I have even less idea about it than you do.
Recompile firmware to skip that pin, or write a custom mixer to skip it, etc.
Don't know.

KC_703 · Post by **KC_703** » Thu Oct 16, 2014 5:47 am

cester wrote:Timecop, do you have any ideas about my issue I posted some days ago? I'm waiting for an answer before throw away these two boards and buy new one (Can't afford it right now)
Thank you.

Someone posted the same issue with M4 on a different forum. The solution was to use a custom mix, copying motor 4 to motor 5... as suggested by timecop. It seemed to work for the guy.

Here's the basic commands necessary:
mixer custom
cmix load quadx
cmix 5 1 1 -1 -1

Then plug motor 4 into M5. Not sure what the NG NG NG means... obviously different from the normal OK OK OK.

Hope that helps.

timecop · Post by **timecop** » Thu Oct 16, 2014 7:12 am

NG just means mixer isn't "sane" because you're missing 100% thrust on each axis due to missing motor. Anyway, it still allows flying so.
edit: once you copied the non-working motor pin, to regain mixer sanity, I think you can just replace the dead motor with 0/0/0 for RPY power.

wobbletop · Post by **wobbletop** » Thu Oct 16, 2014 6:37 pm

rev5 red board, latest firmware

Previously I notice the mag sensor was acting erratic. The model image in the configurator would jump around and the raw sensor data for MAG would have drastic swings with the quad stationary. Now, there is no change in the MAG sensor data unless I bring a magnet close to the sensor (like the buzzer). So the MAG sensor is working, but just with large fields.

Anything else I can do with this board or is it toast?

I notice there is a self test run when the MAG sensor is initialized in the code, but I don't know enough programming to see if the values from the self test are checked in the firmware. Is it?

KoekieMonster · Post by **KoekieMonster** » Thu Oct 16, 2014 8:11 pm

Hi TC,

My board (naze32 acro) does not update, connect to base flight configurator or send data to my minimosd anymore. It does arm(green LED goes on when yaw right) and aux channels also work (red LED goes on when in angle or horizon, beeper beeps when switched on). don't know if it'll fly since quad isn't finished yet.
I've tried updating both normal and with boot loader pads shorted, only blue light comes on then but configuration says: no response from boot loader: programming FAILED.
Is there any way to flash the board again?

TimPainter · Post by **TimPainter** » Thu Oct 16, 2014 10:12 pm

Hi
I have been using a naze32 with mag compass flying in headfree mode for photography this was would drift to a confusing offset after slightly aggressive maneuvers
so i have flashed the October firmware (base flight gui) and did a carful calibration : and now headfree is completely unusable mode, nose out result in 90 deg shift and yawing confuses me completely

i have flown AMP boards in simple mode (same as head free) so i know what to expect.
base flight reports reasonable magnetic headings with no jumps or dead spots and fresky taranis telemetry heading reporting is also correct.
I assume attitude + mag + headfree should be enabled?

can anyone help me with this one??
Tim Painter

CVMichael · Post by **CVMichael** » Fri Oct 17, 2014 3:08 am

TimPainter wrote:..... and now headfree is completely unusable mode, nose out result in 90 deg shift and yawing confuses me completely ....
Tim Painter

I have the same problem (or similar?)

I am using the October version also, and my quad keeps turning clockwise on the yaw axis when I move around, and I constantly have to correct it!
Yesterday I decided to flash the firmware, and I did a "Full Chip Erase" (just in case). After the update I put the same settings as I had before, with one modification, I changed the pitch/roll rate from 1.25 (or 1.35? don't remember exactly what I had before) to 1.5 because I wanted to do quicker flips

. I went to fly today, and now it is MUCH worse!! it rotates on the yaw axis much faster, I changed trim on my turnigy 9x transmitter to all the way to the left, and when the quad is not moving it stays in the same direction, but when I move around it keeps turning clockwise on the yaw axis.

By the way, I am using the AfroFlight Naze32 Acro, so it does not have compass, I was actually considering getting the full version of Naze32 to fix this problem, but it seems the yaw rotation problem happens on the full one also!

Torque66 · Post by **Torque66** » Fri Oct 17, 2014 4:15 am

Uhm... is there a way to have the board beep during arm/disarm with motor_stop enabled....?

TimPainter · Post by **TimPainter** » Fri Oct 17, 2014 5:17 pm

Hi CVMichael
Juat a thorght may be your escs are not in balance ( not calibrated the same) this would probably cause yaw in high power situations?

CVMichael · Post by **CVMichael** » Fri Oct 17, 2014 5:28 pm

TimPainter wrote:Hi CVMichael
Juat a thorght may be your escs are not in balance ( not calibrated the same) this would probably cause yaw in high power situations?

Actually they are calibrated: http://www.netraam.eu/nazewiki/pmwiki.p ... alibration

First i did option #2, and the motors did not calibrate properly because they did not start at the same time, and then I did option #1 using the CLI, and it got calibrated properly

Skinwalk3r · Post by **Skinwalk3r** » Fri Oct 17, 2014 11:11 pm

I am having similar problem with drift in flight. In a fast forward or backward flight in a straight line I can watch the quadrotor turn quite abruptly 90 degrees to the left or to the right this is an acro mode I have the latest October firmware. I just finished setting up this quadrotor it is a 250 size and the settings are all stock except for RC rate

caspar21 · Post by **caspar21** » Sat Oct 18, 2014 3:29 am

Hrmm. Need some input here.
Naze32 has been doing an amazing job!
I just shorted the 3rd channel signal wire from esc to +12v.
I saw the magic smoke. now all i get is one blue light off battery and no lights when on usb.
the nice crater in the stm tells me it is dead. My question is what it might have taken with it.
i have access to a flow solder gun and the tools to replace the stm32 but i need to know what else might need replacing too.

(sounds like a job for super TimeCop!)

as always, all help is appreciated!

timecop · Post by **timecop** » Sat Oct 18, 2014 5:08 am

wobbletop wrote:rev5 red board, latest firmware

Previously I notice the mag sensor was acting erratic. The model image in the configurator would jump around and the raw sensor data for MAG would have drastic swings with the quad stationary. Now, there is no change in the MAG sensor data unless I bring a magnet close to the sensor (like the buzzer). So the MAG sensor is working, but just with large fields.

Anything else I can do with this board or is it toast?

I notice there is a self test run when the MAG sensor is initialized in the code, but I don't know enough programming to see if the values from the self test are checked in the firmware. Is it?

mag lines jumping around in sensor graphs = its toast.
remove it + use external or just don't enable anything mag related (i.e headfree etc)