APM v2.5.2 "Bad Gyro Health" message

[Skywalker]

I got the dreaded "Bad Gyro Health" message with my Hobby Works HK v2.5. My first mistake was buying the HobbyKing board. These things are junk. Terrible solder joints, bad lamination....junk. I should have spent the extra couple of bucks and got the board from 3DR. I haven't even flown the craft yet, and the second time I plugged it in, I got the message on the Mission Planner.
After some research, It turns out that the latest firmware can smoke one of the voltage regulators, or the regulator was defective to begin with and the firmware is now "noticing" it. This seems to be happening mostly on clone boards. When the regulator fails, you will have a full 5 volts running your sensors instead of the regulated 3.3 volts.
According to the 3DR website, your only options are to send the board back or replace the regulator. Since this is a HobbyKing product, I don't think it is worth trying to deal with their customer service...
The good news is these tiny regulators are only about $1.50 from Digi-Key
http://www.digikey.com/product-detail/en/TPS79133DBVR/296-11866-1-ND/381441
I was able to fairly easily get the old one off with a small Radio Shack soldering iron. I should be getting the new regulator in the mail in a day or two, so I will follow up if I can successfully get it back on the board. View attachment 15450View attachment 15451
I'm curious to see if the voltage regulator on the HobbyKing board was the correct Texas Instruments regulator, or some kind of China copy.

 

[Bartman]

thanks for posting that information. sometimes those little components can be a real MF to remove and replace, that one looks like it went without much of a fight.

i have to wonder though when these things happen did the component fail or did some other factor or combination of factors cause the component to fail. it'll be interesting to see if the component repair provides a good long term fix or if the new one will fail also.

good luck with it.
bart

 

[Skywalker]

I agree. According to 3DR, They claim that the ones that have been replaced have not failed twice. I looks like they really cranked up the SPI clock speed with the new firmware. If you look at the regulator on the 2.5 board compared to the 2.5.2 board, the newer board has a much bigger regulator. I wonder what the size of the regulator is on that new system they just released? The pads on the board are actually big enough that I could probably hard wire something beefier if I had to. I probably would not even try to attempt replacing the one on the 2.5 board. way too small.
http://diydrones.com/forum/topics/a...oblem?id=705844:Topic:1457156&page=2#comments

 

[Skywalker]

Now I'm frustrated. I got 4 new 3.3v regulators from Digi-Key today. Soldered one on and checked the voltage. 4.4v?, Tried a second one, 4.4v? Tried a 3rd one 4.4v??. I then checked the voltage without the regulator installed 4.8v. I tried running Mission planner with the 4.4v and the "Bad Gyro Health" message still comes on after a few minutes. Any ideas?View attachment 15492View attachment 15493View attachment 15494View attachment 15495

My soldering may not be the best, but its seems like it should work. I guess I can try to test my last voltage regulator with 5v and see if it steps it down to 3.3v before I solder it on the board.View attachment 15496

 

[Bartman]

i wouldn't bother trying another one. call digi-key tomorrow and talk to a tech assistance rep. they're very helpful and may be able to explain what's going on. it may not be the voltage regulator that is screwing things up, it may be a problem elsewhere. i don't have the chops or the patience to try to fix circuit boards so I generally throw them away before I can get the urge to make an attempt. i tried to fix a few Mikrokopter speed controllers but there's that fear that the thing that popped wasn't the problem so you're just on borrowed time until it pops again.

call digi-key and see what they have to say. you may have the wrong part #, even an incorrect prefix or suffix can screw things up with electronics parts as there are so many variations for a given size shape component.

 

[Skywalker]

Good point. I checked the data sheet for the regulator and there are 3 versions for fixed voltage. 1.8v, 3.3v, and 4.7v. Maybe They just sent me the wrong one.

 

[Skywalker]

I called Digi-Key, excellent customer service. I am going to check my remaining regulator off the board to see what the voltage is. Im hoping they just sent the wrong one. The markings on the case of the regulator are slightly different that the original one.

 

[Skywalker]

I'm throwing in the towel on this one. I tested my remaining regulator off the board, and it measured 1.99v(should be 3.3v) so that makes no sense. I checked the voltage on the pads of the board without the regulator installed, and I think this burned something else up. Now it wont work at all. Oh well. I should have just left it alone when I had 4.4 volts and used it on a cheap quad build or something. Looks like time to order a new one, and not from Hobby King!

 

[R_Lefebvre]

After some research, It turns out that the latest firmware can smoke one of the voltage regulators, or the regulator was defective to begin with and the firmware is now "noticing" it.

It's extremely unlikely that the firmware is what is actually causing the regulator failure. Nobody has come up with any kind of plausible mechanism for to happen. It is basically a consensus that the regulators fail for some unknown reason, and it's just that the new firmware (running the SPI bus much faster) is just noticing the problem. IIRC, this was back-checked, by slowing the bus down again, and then the board works OK even though the voltage is still too high. It appears the gyro chip is mostly tolerant of 5V, it just becomes unable to communicate at the higher speeds.

EVERYBODY is looking into the cause of the failures, 3DR, the cloners, even crazy forum people doing all kinds of wacky things, and nobody can figure it out.

 

[caotri]

I got the same problem and replace with the same regu IC. It died again. I did it 3 times witht he same result. Finally I replace with biger regu IC 117- 3.3 volt which is 500mah max.
View attachment 16694

 

[R_Lefebvre]

That's a clone right? The clones seem to have a much greater rate of this happening. Nobody knows why.

 

[caotri]

The board in picture are clone from hobbyking.
I got another from 3DR also has the same problem. I just try to fix this clone first and next will be 3DR board since it take to much trouble and cost for ship back for warranty.

 

[R_Lefebvre]

FYI, there has been a recent development in the research to this problem:

http://www.diydrones.com/forum/topi...blem?id=705844:Topic:1457156&page=61#comments

For the first time, Philip has worked out how to cause the regulator to fail on purpose, and the method is something that many people may have done. Luckily, there are two easy solutions:

1) Never Hot Plug components to the APM
2) Install a 100uF capacitor on Vcc.

 

[caotri]

FYI, there has been a recent development in the research to this problem:

http://www.diydrones.com/forum/topi...blem?id=705844:Topic:1457156&page=61#comments

For the first time, Philip has worked out how to cause the regulator to fail on purpose, and the method is something that many people may have done. Luckily, there are two easy solutions:

1) Never Hot Plug components to the APM
2) Install a 100uF capacitor on Vcc.

I raplace the regu with LM117-3.3 volt 500 mA max and it works fine I fot 4 flight this morning and alot of setup hours , it still working well.

 

[Bartman]

FYI, there has been a recent development in the research to this problem:

http://www.diydrones.com/forum/topics/ac3-1-rc5-spi-speed-problem?id=705844%3ATopic%3A1457156&page=61#comments

For the first time, Philip has worked out how to cause the regulator to fail on purpose, and the method is something that many people may have done. Luckily, there are two easy solutions:

1) Never Hot Plug components to the APM
2) Install a 100uF capacitor on Vcc.

good info, thanks for posting that.