Help! Beaglebone Blue seems to have died

[Ted Carancho]

Help!  My BBBL just arrived Monday and was in the middle of porting my quadcopter code to it.  I was able to get 9DOF measurements from it and decided to take it to the library with me, so I put it back into the static bag it came with, placed it in the original box in between the ESD foam it came with and placed it into my backpack.  When I took it out, I found that it was unresponsive.  I thought maybe I somehow bricked it by not doing a shutdown -h before unplugging it, so I tried to follow online directions on booting from an external uSD card.

If I power the board from USB or an external 3S Lipo, all I see is the blue ON LED, the LIPO status LED stuck at 75, and the G and R LED's on (picture below).  If I press the SD button with the uSD already plugged in and apply power, I don't see any blinking lights that tell me it's booting from the uSD card.

I've tried the Debian 8.7 2017-03-19 IoT and Debian 8.7 2017-03-19 LXQT firmware, I tried a 8GB and a 32GB uSDHC card, I tried 2 different USB cables.  When plugged into a USB port, I can't find the network access point to be able to login using 192.168.7.2, and also I can't ping the device over it's wireless network after I had configured the wifi using connman.

I'm starting to think that the hardware was somehow damaged (but I felt I took appropriate ESD considerations, and transported it in its box).

This is my first beagle bone product ever (but have used many Arduino's, RPI and Intel Edison).  Is there some newbie mistake I'm making?  Thanks for your help in advance!

[William Hermans]

Hello,

I do not have a beaglebone blue personally, but if you mentioned it I'm not seeing if you switched the boot switch or not.

https://github.com/beagleboard/beaglebone-blue/blob/master/docs/BeagleBone_Blue_ShortSpec.pdf

Scroll down to the "balloon" diagram, and look at the switch that is nearly opposite of the 5v barrel jack.

--
For more options, visit http://beagleboard.org/discuss
---
You received this message because you are subscribed to the Google Groups "BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/650cec52-89c4-4503-89cb-6e83631afc8a%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ted Carancho]

Thanks for the response William,

I followed the directions online to put in a formatted uSD card with firmware, held the boot button down (it's a momentary push button, not an on/off switch) while plugging in a 3S Lipo (12V).  No booting from uSD happens . To double check I viewed the process on YouTube (for BB Black) to make sure I was doing what was expected.

What do you think?  I realize these boards are new (says Beaglebone Blue Rev A2 on back) but I hope there isn't some sort of defect with them.  I used a DMM to verify that the push buttons are shorted when depressed.

Anything obvious I'm missing?

To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard...@googlegroups.com.

[Chad Baker]

The Blue uses a 2S LiPo battery. ~7VDC vs ~12VDC. In the specs that William ref'ed look for section e, Power management, ii => 2 cell (2S) LiPo battery charger.
I haven't checked the schematic, but you might have caused a problem.
Chad

To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/b79e2918-e7b4-44cf-b350-9991b9483847%40googlegroups.com.

[Ted Carancho]

Per the data sheet William link, it states Power Source input of 9-18V for DC Jack... plus there is a 12V sticker on it too.

I had also checked the schematic and it uses the AP1509 voltage regulator which says to have a max Vin of 24V.  I was interested in seeing if it could support a 6S LiPo ultimately for a future multicopter project.

Any other thing to look at guys?

[Chad Baker]

Looking at the schematic for the Blue, "https://github.com/beagleboard/beaglebone-blue/blob/master/BeagleBone_Blue_sch.pdf", 1st page upper right hand corner, there is a "2S Lithium Balancer and Protection" circuit. This uses a BQ29209 chip. The "Electrical Characteristics" for this device, "http://www.ti.com/lit/ds/symlink/bq29209.pdf", give maximum values but the device is based on 2S not 3S LiPo's. Using a 3S or 6S might invalidate the warranty.

The barrel connector and LiPo battery both connect to a AP1509 5VREG, there are diodes protecting against reverse current flow. 12VDC feed from the barrel jack basically shuts off current flow from the ~7.4 VDC battery into the AP1509 while without the 12VDC feed, the LiPo battery will feed the AP1509. You can have one or the other or both sources, these do not affect the 2S balancer.
Chad

To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/a0c83976-d96a-4404-a686-c1acb7fe650b%40googlegroups.com.

[Dennis Lee Bieber]

On Thu, 30 Mar 2017 21:22:44 -0700 (PDT), Ted Carancho
<tedca...@gmail.com> declaimed the
following:

>on/off switch) while plugging in a 3S Lipo (12V). No booting from uSD

Only specs I've found in a quick search are
https://beagleboard.org/blue but a key factor on that page is that the
/charger input/ is 9-18V, which implies (allowing for need to push a higher
voltage than the battery to charge it) the battery is less than 9V.
Specifies "2-cell LiPo" -- your "3S" 12V probably fried some circuits.
--
Wulfraed Dennis Lee Bieber AF6VN
wlf...@ix.netcom.com HTTP://wlfraed.home.netcom.com/

[William Hermans]

On Thu, Mar 30, 2017 at 9:22 PM, Ted Carancho <tedca...@gmail.com> wrote:

Thanks for the response William,

I followed the directions online to put in a formatted uSD card with firmware, held the boot button down (it's a momentary push button, not an on/off switch) while plugging in a 3S Lipo (12V).  No booting from uSD happens . To double check I viewed the process on YouTube (for BB Black) to make sure I was doing what was expected.

The only documentation I've seen seems to indicate that the Beaglebone blue has a 2 cell 6-16v charger circuit, which tells me that battery input voltage can also be in that range.Or more correctly 12v nominal battery input. However, with that said, the documentation I've seen to date is not what I'd consider to be complete. Which is to say, I don't have one of these boards, and I just have not looked all that hard.

What do you think?  I realize these boards are new (says Beaglebone Blue Rev A2 on back) but I hope there isn't some sort of defect with them.  I used a DMM to verify that the push buttons are shorted when depressed.

Anything obvious I'm missing?

I do not think so. About the only thing I can think of is that you could hook up a 3v3 logic USB to UART module / cable to see if uboot gives you any additional information as to what's happening. I'd probably contact the reseller you purchased this from however, and look into getting an RMA. Because if everything is as you say( no left out details ), it sounds like a potentially defective board. Which I've yet to see a defective beaglebone black, or green to date, and I've handled quite a few greens( 50 +).

I'm considering buying one myself just so I can help others troubleshoot the board once in a while. But that may / may not happen.

[William Hermans]

Additionally, I presume you have a Robotics application which you're wanting to experiment with. With that said, maybe it could be beneficial to get one of the other variants initially. Then if you're like me, opt out of the wireless variants, and if you need wireless get a well supported dongle with an external antenna cable connection.

By well supported, I mean a dongle that uses a chipset that is well supported in Linux( Debian ), and is not too short. The length could potentially play a factor in interference with the HDMI, *if* you're using the HDMI output at all If not it is my understanding that you can simply disable HDMI to bypass that problem.

I did want to say that I also own an rpi 3.0, and I know there is a lot more documentation all over the web concerning that platform. Some of the software, and utilities seem to be pretty good too. With that said, in the context of robotics, and general embedded Linux. I'd have to say that the beaglebone SBC's are light years ahead, in terms of flexibility, and connectivity. Where the beaglebone platform is also a platform that requires more knowledge from the user or engineer using it. Basically, aside from what you'll find googling, and what information people will share here. There is no hand holding on this platform. Which means you need to check everything you're working with.

For instance, if you had any IO connected to this board through any of the GPIO, or other periphery pins while you powered the board up, or down. You could have damaged the processor. So all the pins need to be isolated in that case. Ethernet, USB, and the serial debug port are exceptions to this rule. But are not what I'd exactly consider "periphery" in the context of embedded systems.

[William Hermans]

On Fri, Mar 31, 2017 at 1:47 PM, William Hermans <yyr...@gmail.com> wrote:

For instance, if you had any IO connected to this board through any of the GPIO, or other periphery pins while you powered the board up, or down. You could have damaged the processor. So all the pins need to be isolated in that case. Ethernet, USB, and the serial debug port are exceptions to this rule. But are not what I'd exactly consider "periphery" in the context of embedded systems.

The IO pins are also only 3v3 tolerant, and can only source / sink a minimal amount of current. I want to say as far as sourcing or sinking, ~5mA is your limit. I do know that some pins can handle more, and others still less. So always check the datasheet for the AM355x processor. Or in this case, perhaps the Ovtavo SoM has some differences ? Not sure on that. But in either case, you need to find all that out, before connecting to, and using any of the input / output on the board.

[Ted Carancho]

Oh, I think I understand the confusion.  I didn't connect a 3S LiPo to the 2S LiPo connector (3 pin JST I think).  I connected the 3S LiPo to the DC barrel jack that has a 12V sticker on it (I think you can see it in the BB Blue pictures online).

[Ted Carancho]

Thanks so much for all your help William!

I went into purchasing the BB Blue with eyes wide open, knowing it's brand new.  When the board was working I was very impressed with the cloud9 IDE and the capability to even use a terminal directly from a web browser.  I only worked with the onboard IMU and barometer and didn't connect any other external devices yet.

I contacted the vendor I bought this from (Newark/Element 14) and they provided me a full refund within 2 days!  Great customer support!  I'll wait awhile to see if others have success with their batch of the board and give the Beaglebone Blue a try again in the future. (I also entertained buying one of the more well used versions, but I really like the integrated IMU so I'll wait)

Thanks for all the help guys!

[William Hermans]

On Fri, Mar 31, 2017 at 8:25 PM, Ted Carancho <tedca...@gmail.com> wrote:

Thanks so much for all your help William!

I went into purchasing the BB Blue with eyes wide open, knowing it's brand new.  When the board was working I was very impressed with the cloud9 IDE and the capability to even use a terminal directly from a web browser.  I only worked with the onboard IMU and barometer and didn't connect any other external devices yet.

 

 It seems a lot of people like the cloud9 IDE, but I do not really like it much personally. I've always been a minimalist type when I can be, and I just use ssh to compile my projects natively using gcc from the command line. One of the issues I have with cloud9, is that it's not always responsive enough for me. Another is that it is not documented well enough on this platform for my own wants. e.g. I have my own development system in place that works very well for me, with no real motivation to switch to something that seems to want to fight against the user when that user want's to customize it's features.

I contacted the vendor I bought this from (Newark/Element 14) and they provided me a full refund within 2 days!  Great customer support!  I'll wait awhile to see if others have success with their batch of the board and give the Beaglebone Blue a try again in the future. (I also entertained buying one of the more well used versions, but I really like the integrated IMU so I'll wait)

Well that's good. I'd expect no less from Element14 though.

[cdu...@pharmadva.com]

I'm seeing a problem that may be related.  We have now fried 3 BeagleBone Blues Rev A2.  We started using ours with a 15V 2.6A supply without connecting a LiPo battery.  After only a couple of power cycles, the battery charging IC gets very hot and pops with a puff of smoke and a burned trace on the board.  I've not seen any similar/related posts.  Are the posted docs incomplete or incorrect?  Is 12V the max?  Is it required to have a LiPo battery attached at all times when powered?

[a156...@owlpic.com]

.. same here.

BB Blue was in the process of booting with usb-power. Attached to the board was:

- an Adafruit GPS module (GPS connector)
- six NAZA ESC, data and GND pinned to the 8x3header (1..6), flight power battery to the ESC unconnected.
- one Spektrum satellite receiver to E4 connector
there was no battery attached to the bb blue LIPO connector.

When I plugged in a 12V from an AC/DC Adapter to the barrel connector the LIPO charger IC smoked away.
I powered up this collection several times before, even flew the copter a few seconds.
Anyway in the schematic I cannot find a plausible cause for a high current at that situation.
What a disappointment.

Jo

[Chuck DuBois]

I've had to move on to another project for the short term, but since changing our process, we have not had an issue with the LIPO charging IC.  Discontinuing use of the USB power connector seems to be key, and this setup has worked for multiple days without incident.  We now cable to USB for initial configuration without standalone power, then use Wi-Fi for connectivity and a 12-15v power adapter exclusively.

[chris Bruner]

Mine was working fine, until I let the battery that was running it die down. Then it wouldn't restart (same lights as in parent photo). Also seemed to be running hot. I was very excited about this board, but I need more reliability then this.

On Thursday, March 30, 2017 at 10:53:30 AM UTC-4, Ted Carancho wrote:

[Clark Sann]

William

If I power my IO through the 3.3V bus on the BB Blue, which turns on as the BB Blue turns on, will that be sufficient to protect the GPIO?

Clark

[Jason Kridner]

On Thu, Apr 13, 2017 at 8:43 AM <cdu...@pharmadva.com> wrote:

I'm seeing a problem that may be related.  We have now fried 3 BeagleBone Blues Rev A2.  

I recommend starting a new thread for a different issue. Otherwise, I find it difficult to see if issues are getting useful responses. I understand you saw this one as possibly related, but you say you've seen smoke and I don't want to miss issues like that. The outer thread still looks like it could be file system corruption to me.

 

We started using ours with a 15V 2.6A supply without connecting a LiPo battery.  After only a couple of power cycles, the battery charging IC gets very hot and pops with a puff of smoke and a burned trace on the board.  I've not seen any similar/related posts.

I believe this operation is meant to be fine. The charging circuit should provide power to the system in this case.

 

 Are the posted docs incomplete or incorrect?  Is 12V the max?

18V is the max.

 

 Is it required to have a LiPo battery attached at all times when powered?

We generally expect you to have the LiPo battery, but I believe the case where no battery is present should be OK.

Did the board come with the wire on CHGIC or is that damage as well? What kind of load did you have on the system when it was damaged?

 

On Thursday, March 30, 2017 at 10:53:30 AM UTC-4, Ted Carancho wrote:

Help!  My BBBL just arrived Monday and was in the middle of porting my quadcopter code to it.  I was able to get 9DOF measurements from it and decided to take it to the library with me, so I put it back into the static bag it came with, placed it in the original box in between the ESD foam it came with and placed it into my backpack.  When I took it out, I found that it was unresponsive.  I thought maybe I somehow bricked it by not doing a shutdown -h before unplugging it, so I tried to follow online directions on booting from an external uSD card.

If I power the board from USB or an external 3S Lipo, all I see is the blue ON LED, the LIPO status LED stuck at 75, and the G and R LED's on (picture below).  If I press the SD button with the uSD already plugged in and apply power, I don't see any blinking lights that tell me it's booting from the uSD card.

I've tried the Debian 8.7 2017-03-19 IoT and Debian 8.7 2017-03-19 LXQT firmware, I tried a 8GB and a 32GB uSDHC card, I tried 2 different USB cables.  When plugged into a USB port, I can't find the network access point to be able to login using 192.168.7.2, and also I can't ping the device over it's wireless network after I had configured the wifi using connman.

I'm starting to think that the hardware was somehow damaged (but I felt I took appropriate ESD considerations, and transported it in its box).

This is my first beagle bone product ever (but have used many Arduino's, RPI and Intel Edison).  Is there some newbie mistake I'm making?  Thanks for your help in advance!

--
For more options, visit http://beagleboard.org/discuss
---
You received this message because you are subscribed to the Google Groups "BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/6f1a5c15-eebd-41ae-9256-2360afa2435f%40googlegroups.com.

[William Hermans]

On Fri, May 19, 2017 at 8:42 AM, Clark Sann <acs...@icloud.com> wrote:

William

If I power my IO through the 3.3V bus on the BB Blue, which turns on as the BB Blue turns on, will that be sufficient to protect the GPIO?

As Jason mentioned. Please make a separate new post / thread so as not to confuse the issue. With that said, I'm not sure we're talking about the same thing. In your new post, please be more verbose.

[chris Bruner]

Well this was fun. Just tried plugging the 12V in again, to see if anything changed, and the magic smoke came out of the chip near the 12V power plug. I assume that's the voltage regulator. And I had such high hopes for this board.

[cheng....@gmail.com]

I've currently burned 3 different CHGIC chips on boards. I've been running the board powered off a power supply set to 18V. I have other boards booted at the same time, and have booted up all of these boards on the same power before. There are different ports connected, but none of them are being used. I believe the trend I noticed right before they break is that they power on, but are unresponsive. I power cycle them and they continue to be unresponsive. Then I plug them in a 3rd time, and the CHGIC chip pops and smokes. 

Is there something I can do? I can't have boards continuing to break. And from the last message, it sounded like his board burned from even just 15V. 

[jesusandall...@gmail.com]

My BBB no longer works from the barrel jack; I plugged it into a 12V switching power supply and the smoke escaped (see picture). It still runs from a 2S LIPO.

[cojocar...@gmail.com]

Can you check how much current it draws from the lipo ? Had a feeling the motor drivers , randomly , short themselves to ground causing a large amount of current to flow from the barrel via VDC into the CHGIC and from there to VBATT ( which is connected directly to the two motor drivers ).

This seems to be more than the CHGIC can handle without blowing up.