Dilbert66 wrote:Quite interesting though that you are having the same issue that I was having with the ms5611 baro with the 10dof module I had (mpu6050,5883 and 5611). Try doing what I said with the -O option and also try one other thing if that doesnt work. Change the i2c init speed from 400khz to 100khz in the drv_i2c.c file and see if that works. It did in my case as well. Of course you don't want to work that way. Just prove what the issue is. If it does, then it proves that the compiler is doing something funky with the i2c timing somewhere. I don't know why this is the case. The reason it was failing for me, is the fact that it was reading incorrect values from the baro and failing the prom checksum.
This BMP180 drive me crazy ... anyhow ... good learning, but still no solution.
I did finally downloaded the proper latest baseflight firmware from
https://github.com/multiwii/baseflightAt least the latest version does work on rotate the MAG ... the previous version from Nov 2013 was not doing the job correctly .. so my 10DOF DIY .. at least now is 9DOF functional.
Still both original and as my modified version with 100k i2c speed ... don't allow to read anything from the Baro .. and I'm amazed I don't see any i2c error ... even if the BARO is recognized.
For double check again ... I used my Arduino Nano .. loaded the BMP180 test firmware .. and I get decent baro reading from my 10DOF using 3.3v power .. exactly as I power the DOF from STM32
Anyhow .. after a quick check between Arduino working code, and Baseflight ...
I see that the 085 register used in baseflight are not same used by working Arduino ... also the structure is looking different.
So ... @timecop ... either you 100 customer of AfroMini32 with BMP180 have not still check the baro ... or I'm still doing some dummy stuff on my side ... being me ... I start to guess that I maybe right ... and none of your 100 customer did care or notice the baro issue.
PS: I'm a bit off topic here ... so I will try to get some info from the baseflight thread.
PPS: Thanks a lot for the trick related to STM32 with 64k and 128k ... amazing ... so all STM32 chips are same, but some are market 64k and other 128k ... very strange policy ... I would guess that STM32 market 128 got a more intensive test at QC, while 64k version have a limited QC ... who knows ...
PPPS: From the test that I have been able to do over BMP180 ... it is looking a way more precise chip then BMP085