Should be fixed now. I had to do some DMA re-mapping, and so there is a new Arduino-proffieboard plugin (v3.4) and also some minor changes in ProffieOS.
3 Likes
Ill do dowmload the fixed OS , update plugin, and test tonight. Thanks Fredrik. I was wanting to center this beta around things we missed on 2.2. Everyone seems to be using subblades now, but I still like the many data pads!
1 Like
Fixed Fredrik! Matter resolved!!!
Time to read blocks: Average speed: 1997.42 kb/s, 22.65 simultaneous audio streams.
::
::
::
::
::
::
::
::
::
::..
x100us 1 2 3 4 5
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0
Programmed with āsmallest codeā in tools.
1 Like
Motion gestures all work, as does Battle Mode 1. So, I guess the SD card read speed has massively improved blst.wav and swng.wav retrieval speeds. You can do blasters so fast, it sounds like Boba Fettās Slave I blasting away. Accent swings can trigger more rapidly as well, its noticeable. That is a huge improvement! Still swinging it around. I added SA22Cās Prop file from 5.7. I can clone Fernandoās but, I donāt want to get too advanced here. Just trying to check all the functions on an installed saber, seeing if I get glitches. Iāve been throwing mutiple functions at it rapidly, and no freezes or anything so far.
Once I run the battery down, Iāll test charging through USB.
4 Likes
Still trying to find time to really work on things but I did build one of @profezzorn ās graflex helper boards today and tested the remote USB pads.
5 Likes
Ok, Iāve got things 60% wired up, Curious about the onboard LED color/blink, is there a list of what colours and blink patterns mean what?
Anyone willing to intentionally corrupt an SD card for testing? (Simple reformat of course should restore )
I should mention this is using MacOS 10.14.6 Mojave, an older system, but I donāt think that matters.
Using SDIO High Speed (didnāt test with default speed) I clearly get file corruption if the SD is not gracefully ejected from a mass storage mounted condition before uploading. While of course this is ātestingā the result of a user error and not something that should happen in the first place, itās an interesting failure to me as Iāve inadvertently done this many time before with no consequence (was I lucky all along? )
It might be this brand of card, it might be my OS, or both, or neither as maybe the v3 Proffieboard has something to do with it. Hence, the request for someone else to see their results.
The corruption Iāve seen people have usually shows the file names and folder all weird characters.
This is different, where the names are all intact, but the file type association ONLY FOR DIRECTORIES seem to have get lost, but not all of them.
Ive already wasted 2 SDs on my build, cjassis wasnt really designed for the sd sticking out on 2.2s and 3.7s. Me and my soldering Iron did some improptu remodeling after the 2nd sd card snapped.
Tom
2 Likes
I was going to make a comment about the whole SD holder coming off the board, but I wonātā¦or just did.
2 Likes
Well, I think Iāve got the dubious honour of killing the first V3 board. No idea how I did it, but something is shorted internally: button one is always triggered even with nothing soldered and the CPU gets hot when power is applied. Iāve removed everything just to be sure and looked at the pads and traces, no obvious sign of a short.
The CPU gets hot enough that itās uncomfortable to the touch after about 2 seconds.
With nothing connected to any pad, just plugged into USB:
Welcome to ProffieOS, type 'help' for more info.
EVENT: Power-HeldLong#1 millis=2137
EVENT: Power-HeldLong millis=2137
Amplifier off.
I have to say, this is pretty embarrassing.
1 Like
Maybe dumb question but are you sure you chose DOSFS: SDIO (High Speed) ?
Will that impact the CPU?
I may have used regular speed while I was playing around with the new plugin
Edit: In any event, this happened all of a sudden, everything was working fine and I was finishing the install of the OLED and bluetooth when all of a sudden the power button stopped working, so I started removing stuff to get to the button pad ( the OLED sits on top of the proffieboard in this build) and when I touched the cpu, it was hot.
Looking closer at the board, I see that I must have slipped with my iron and slightly damaged the corder of one of the reset/boot buttons but both still function. I must have put too much heat into the board somehow and broke something internal.
Aside from button 1 being bridged and the CPU getting hot, it does still work so I may still install the board again and test features until it dies. I can remap the power button to one of the free padsā¦
Are you sure itās the CPU thatās getting hot?
The charging chip gets pretty hot while charging a battery, so hot isnāt always an indication of a problem on a V3 board.
Also, could you try a very basic config and see if the power button is still stuck? There are about a thousand different ways that a problem in the code somewhere can cause buttons to not work properly. In fact, we had an instance of that earlier in this thread.
1 Like
Iām relatively certain itās the CPU, as the warming occurs whenever the board gets plugged in, even when nothing is connected to the board. The config Iāve got applied is straight from the configurator, with three pixel strings, an OLED and the Bluetooth.
Matt, have you updated to the latest plug-in? 3.4.0 now. A button/serial conflict was corrected in 3.3.0.
One of the worries about using BGA components is that the tiny gap between the chip and the board might be magnet for small metal bits and/or single strands of copper wire. The power button connects to the chip in the outmost row (in two places) so there is a chance that something like this could have happened.
In addition to checking those spots (and maybe blowing it out with some compressed air) there is also a via between Data1 and Free3 that connects to the BT1 pad, so that is another place to inspect for shorts. (top and bottom)
(The doodle-on-a-picture feature can be quite handy sometimesā¦)
3 Likes
Checked and sadly, no dice.
Do you want me to mail it back to you?
Please try a config file without serial/oled (or anything else out of the ordinary), just to make sure thatās not causing something unexpected. If itās not a short and itās not software, I doubt Iāll be able to figure out what went wrong with it.
Also, Iām heading to Sweden on the 23rd, so I probably wouldnāt be able to take a look at it until I get back.
Basically, there is probably no point in sending it back.
1 Like
Tried just a basic config with no change.