Yikes! We’re getting into deep stuff here that I don’t understand.
All I did was add the prof’s suggested delay line (below) to my normal config file - I didn’t dive in to any of the other files.
#define CONFIG_STARTUP_DELAY 500
Yikes! We’re getting into deep stuff here that I don’t understand.
All I did was add the prof’s suggested delay line (below) to my normal config file - I didn’t dive in to any of the other files.
#define CONFIG_STARTUP_DELAY 500
Right. So what basically happens if you look is…
it checks if you’ve defined CONFIG_STARTUP_DELAY or not.
If you have, it checks if it’s greater than 1000.
If greater, it uses that custom value, otherwise it uses 1000.
The point is, this all used to happen later in the file (after the prop file is set).
Now, with the code block moved up to where it is currently, it might be a clue to your problem.
@profezzorn might have more insight as to how/if it could be related.
No, the default is whatever the arduino-proffieboard plugin specifies. (Which I think is 100 ms.) If your arduino-proffieboard plugin is too old to specify a startup delay, then it defaults to 1000.
Plugin is 3.6.0 which I think is the latest one (correct me if I’m wrong).
But as mentioned, I added the startup delay define to my normal config at 500 last time and it didn’t seem to make a difference.
Thinking about it, the fact that I’ve had this problem with two of the last four boards I’ve installed makes me think this problem must be quite common.
Brian, I know you said you’ve known of some cases of this, but I wonder how many other cases have cropped up? Are people simply living with it given that even on affected boards, four out of five (or more) boots are successful?
For the record, in my case the virgin boards were supplied by KR in their Thrawn Hunter kits to completely separate buyers who then sent the kits to me to install. In the first case I wrote the config, in the second the customer sent me his own preferred config, both of which were quite different apart from the wiring setup.
No idea if these things are relevant or not, but thought I better share just in case.
So do people just ignore it? Because I don’t hear a lot of complaints about it.
I have to say that I have not come across this yet with any of my set ups thus far but I will let you know if I experience it.
It is rather strange that its intermittent
I’ve actually had this same issue on a Warsabers Revanted I installed for myself. Interestingly, it’s also a KR Proffieboard V2.
It started happening some time after I updated to OS6, but I don’t think the update itself caused it. I immediately checked my SD card’s file names, reformatted the SD card, and then tried a different SD card entirely, but the issue still persisted. After noticing my SD card would slide out more easily compared to my other Proffie sabers, I concluded it was the board itself.
This was a saber I initially wanted to install with Proffieboard V3, but I got impatient and used a spare V2 instead. This issue, combined with some dissatisfaction in my own install work, led me to decide my best option was to wait on V3 and reinstall.
However, after stumbling across this thread, I decided I’d try some of the proposed solutions. I added #define CONFIG_STARTUP_DELAY 500 to my config, and then tried downgrading to OS5.9. The issue still persisted.
In my mind, this further concludes that it’s my board, but I thought I’d share to see if my case can help illuminate anything. I’d be more than happy to provide further information, or try out any other proposed solutions.
If no-one else is having this issue, I can’t see what I could be doing differently to everyone else to create the problem. I’ve installed over 70 Proffies in various hilts, and the only time I’ve ever had an SD card error was on a particularly tight install where a wire was pressed up against the SD card tray, causing a bad contact. But that was consistent in that it didn’t work at all, and once the wire routing was sorted the problem was fixed. But now I’ve had it on two installs both with completely different configs.
Sorry, I posted just at the same time that Porgchamp did. So it seems I’m not alone. But crucially, Porgchamp had the problem with 5.9 as well, but I’ve yet to experience the issue with 5.9 - only ever with 6.x.
Curiouser and curiouser…
I’m just trying to figure out how common this problem is, or if there are any common factors, like maybe a particular sd card, batch of boards, or some particular set of configuration options.
The SD cards in my case were the standard red Sandisk ones that Saber Armoury ship with all their Proffies. Not sure about the boards. Do they have batch numbers or any other identfying markings anywhere?
The only unusual common factor in my case was that with both the boards exhibiting the problem, I was using three LED pads for the blade instead of two. (With a white blade I thought the extra fet couldn’t hurt).
Config-wise, they were completely different: one had modified SA-22C buttons, no edit mode, and the other had Fernando’s prop file and all his associated bells and whistles.
Is it worth my asking if anyone else has had this problem on the Facebook Proffie pages? I think quite a few people there don’t know about this forum and I know you don’t visit there much these days.
I have one board from a set of five from electronics123, that fails to boot correctly about 2/3 of the time. Every so often, it boots saying “no SD card found”. This is in a saber I built for my daughter. She power cycles it until it boots correctly. I have not had time to take it apart, to figure out what is wrong, but it is not the SD, config or wiring. I have swapped the SD and reflashed the board. Something wrong with the board, since my son has nearly identical setup that works correctly every time.
I suspect some power problem that keeps the microcontroller from starting correctly.
Yes, but searching back for any posts is virtually impossible there.
Hal314, which ProffieOS are you using? Have you tried loading 5.9 on? Does it make a difference?
I am using 5.9 on the board. I have not tried 6. This one board did not behave correctly, even when I tried to loaded it the first time. It failed to download twice. The strange things is that it was only able to download with an SD card inserted. I did not realize there was something wrong with it until later. I was under the impression I did something wrong.
It is soldered to another board which has a battery charger and a large P MOS to control power to the blade, so it is not easy to take part. To make matters worst, I got the battery holder to close to the board on the 3D printed holder, so I was unable to connect the USB to an external connector.
Further to Brian’s post on Facebook, sorry, – I’ve dropped the ball a bit on all this. (Had a lot going on lately – hopefully should calm down soon).
The only update I have is that I haven’t yet found an instance of this happening with OS 6.5.
I’ve only done one build myself with 6.5 before 6.6 came out, but talking to other people it seems 6.6 was the earliest that the phenomenon was seen. Of course it could be total coincidence.
Will try and get some more tests done on some loose boards soon.
Using a PNY 95mb/s card and verified to within 4-5mb of that speed. I get the SD card not found but only after doing anything involving hooking up and removing a USB cable (updating config, using webusb, copying sound fonts). But one off to on killswitch and it works fine. Tried it on my other sabers (with Kingston, Sandisk from Best Buy, etc), same thing. Moving back to 5.9 fixes this and it is always only after a config upload, webusb, anything involving hooking to a PC. I normally don’t mind since just switching the kill switch off and on or removing and putting the battery back in will solve the issue.
@hal314 @Sabersense @Bruinrogue where did you get your board from and when? Per profezzorn’s question the few posts I’ve seen on this kinda point to a bad batch of boards as they all seem to have popped up together at the same time but don’t show up for users who’ve had their boards longer.
Electronics123 on July 27th, 2020: I ordered 5 boards, and only one has trouble booting.