Nothing fabulous in today's update; I found myself troubleshooting a couple APMs that connect, but respond with the dreaded "Failed to Boot MPU6000 5 Times" error.
NOTICE:
6/3/2014 - Updated MPU6000, MS5611-01BA in above Diagram. Uploaded updated diagrams as promised.
In the course of this work, I found the APM_v252_Release schematic to be a bit counter-intuitive due to poor arrangement of the pinouts, particularly around the TXB0104 Level Translators. So I made updated schematic.pngs for the two TXB104s at the DataFlash and MPU/Pressure locations so that their pinouts reflect the actual layout of the IC being used; I've also made a complete schematic with those updates. I have not changed the circuit or attempted to resolve any design issues; I've only rearranged the appearance of the diagram.
When I used to do this for a living, it was a matter of pride to make sure your schematic reflected the pinout of the actual part being used; nowadays we just let the computer do all that. It doesn't care that the pinouts are out of sequence; it just tries to make the neatest diagram possible.
This makes it a lot easier for actual errors to slip past the vetting process; they aren't nearly so glaring when your brain is already trying to deal with a jumbled mess due to out-of-sequence pinouts.
So if you're troubleshooting failures in these areas, hopefully this will help... I may find myself attacking the other IC pinouts in those areas as well; if I do, I'll update here with those revised schematics.
Please feel free to double check my work against the original and let me know if you find any errors; a second set of eyeballs often catches something originally missed.
Peace!
mnem
*Revisionated*
APM_v252_MPU-Pressure%20%2B%20TXB0104.png
APM_v252_DataFlash%20%2B%20TXB0104.png
APM_v252_Updated_With_TXB0104_s.png
Updated Diagrams:
Comments
Ahhhh... I see. I looked quickly and they appeared similar; I should have looked closer. I've spoken with Stefan; he's a clever bloke. My Code-Fu is weak; I could just get what he was doing, but mostly I understood it from the description in his notes.
Paul
thank you. Stefan Gofferge's Arduino Mode Switch is another solution.
it connect between PPM out of CPU and PPM input of transmitter radio module
mode switch data replace chanel data directly in PPM sum.
in my solution arduino set analog voltage level just as it makes in your resistor divider. but levels for each mode can to be easy selected using calibration commands (connect D8 to GND for - or D9 to GND for +)
in additional to D7 pin may be connected piezo buzzer
Thank you; it actually took longer to shoot pix and write up the DIY than to make the first switch. Now that I know what I'm doing, could probably finish one in an hour or two. I plan to try for my 2nd revision here soon; a version with potentiometers.
Stefan Gofferge's Arduino Mode Switch: http://diydrones.com/profiles/blogs/arduino-based-mode-switch
The UART0 flaw is not an issue in most cases, because UART0 is usually used for the 3DR Radio link, which plugs in on the other side of the board and the UART0 pin header you're concerned with remains dormant. ;)
Paul
unfortunately I do not know who is Stefan Goffer.
I used a very simple code that I wrote for pair hours.
I really liked how you are able to do manual for beginners in pictures
APM 2.5.2 have yet another bug
uart0 +5v pin and 3v3 regulator input get power by line going under hmc5883 chip
any changes of current make compass error.
you can make test: connect +5v uart0 power to any powerful device (500ma-1000ma) and check internal compass heading changes.
Alexey -
I didn't realize it was you I was talking to; I looked very closely at the collaboration between you and Roman on the talking APM Mode Switch. Very nice work! Was your switch based on Stefan Gofferje's code or did you develop your own?
As you can see, I took a much more "Low-Tech" approach in developing my switch; I wanted it to be reproducible by ordinary folks, not just geeks like us! LOL
I see your two boards with the 470uf radial electrolytic caps, but not a 470uf on the green board. That would be HUGE! The reason a 100uf ceramic cap was chosen as the buffer/protection cap for the retrofit was they have a much lower ESR than ANY electrolytic cap can manage. Much better at flattening a momentary dip in voltage. I've found that 150-200uf total of ceramic caps on the 5V pin at UART0 and the 5V pin at INPUT 1 is enough to not only protect the TPS7133, but to keep the board from even hiccuping and losing USB connection in Mission Planner when a high-surge-current RX like the RX3S Gyro w/DSM2 is hot-plugged.
Great to speak to a fellow tinkerer from across the great ocean! I have козаки́ family tree back several generations, as well as the obvious Polish in my name.
Paul
fantastic blogs!
we have similar ideas
my mode switch
http://sites.google.com/site/hobbyarduino/project-updates/theresult...
http://forum.rcdesign.ru/blogs/99365/blog14414.html#comments
my repair 3v3 regulator manual http://forum.rcdesign.ru/blogs/99365/blog17682.html
my own clone of apm with SMD cap 470Uf (green board)
Alexy -
Thanks for viewing my blog; I hope you find the material here useful!
There are good reasons for having a buffer between the CPU and the dataflash. There is nothing wrong with the idea, it was the execution that was flawed.
I am VERY familiar with the issue surrounding the TPS79133 regulator; if you look at my other blogs you'll see that I've already addressed it in several ways.
Thanks for stopping by,
Paul
main problem of apm 2.5 / 2.5.2 is using bad 3.3 voltage regilator chip markes as pesi
in new design boards -redesign EN pin to +5V for using mic5333-33 ldo (YM5 case)
or mic5205-3.3
i recover 100+ apm 2.x controllers but not found burned TXB0104 chip or 45db161 data flash.
data flash chip designed for using with removable memory card and have protection on input/output pins.
Absolute maximum rating for input/output is 6.25V
So current PSB design no brought problem to users, but if you design own controller using APM schematic you can remove this level converter
If you find out what's wrong with those APM units that have that dreaded error, please post! One of mine has that and it wasn't the 3.3V regulator so I'm interested!
-
1
-
2
of 2 Next