You may know this - apologies if so - but power-cycling an iPhone doesn’t actually quit an app, as the app state is retained and re-loaded at start-up. If BlinkUp fails, try manually quitting the app then restarting. This will then allow you to launch Electric Imp fresh, and may yield better results - or at least some further clues as to what’s breaking here.
Have you tried 8.0.2? And I see 8.1 is out in beta, so this may change things shortly.
Have emailed you some instructions on how to collect the opto waveform so we can look at this some more. Really seems like an iOS8 regression from your description above.
@smittytone - no need to apologize. At this point in time, I’ll take just about any ideas short of signing my soul away. But yes, I learned that as I’ve used my iPhone–a bit of a surprise to me. I didn’t detail that quite completely above, but believe I had been careful to always quit all applications before power cycling the iPhone. (updated) By the way, I don’t believe I understood the persistence of the apps across a reboot, so some of my earlier testing may not have been “valid”.
@Hugo, I’ll take a looksie at what you sent and will be happy to do this. It may be on my own time, as my employer is getting weary hearing BlinkUp throughout the day and wondering how much more time I’m going to invest in this platform.
Strange…my I rebooted my iPhone and noticed it had connected to the 5 GHz WiFi in the building. I had logged out of Electric Imp. When I logged back in, I had to add the 2.4 GHz network I’m trying to BlinkUp to. BlinkUp worked once. I mention this only due to the fact that I’m assuming it’s “normal” for the iPhone to be on the same WiFi network as the imp and am simply passing this along in case it has some relevance to the issue. I’ve now had several successful BlinkUps. Quite honestly, I’m afraid to do Clear Wireless Configuration, as it seems it’s what starts the problem.
It seems if I log out of the Electric Imp app. Quit it. Restart it. Log back in. Things seem to work OK. So…it would appear not to be a hardware issue and it’s looking more and more to me like a BlinkUp app issue that may somehow be related to iPhone being on one (5 GHz?) WiFi network and trying to BlinkUp on another. I can also assure you that I’m ALWAYS trying to blink up onto a 2.4 GHz network. This is also likely an uncommon setup which could explain why others aren’t seeing it.
At this point @Hugo, I don’t know if trying to do the capture is of help or not. Thoughts? I have a couple of imp001 with April cards (soldered on my boards) and a couple of P3V3 boards. I may need some guidance as to the optimum way of connecting these to get what you’re looking for.
Testing the P3V3’s is going to require cutting traces and making the BlinkUp circuit inoperable. I have a C3V0 that is set up for this, I use it to test various phototransistors and bias resistor combinations.
@Hugo, If you think that collecting this data will be helpful, I’ll ship my testing rig down to @ctmorrisonhvacsp so he can get some data from his phone. It is set up to work with the imp-o-scope code, and I can send you baseline data from my iPhone 5s.
@ctmorrisonhvacsp Are you OK with doing testing that way? We will have to coordinate testing together, since the BlinkUp circuit no longer functions.
@MakeDeck – sounds like a super idea. Let’s see if @Hugo believes it’s worth the effort and I’ll gladly do whatever from my end. I’ve invested way too much time to let this go unresolved. Thanks!
@ctmorrisonhvacsp: the phone doesn’t need to be on the same network as the device being configured - the phone can even have wifi disabled and blinkup is ok (it needs network access, but via cellular is fine). But… strange that it works ok for you after quit/re-open.
Collecting raw blinkup data from a failed blinkup would be good to determine if somehow wrong data is being sent due to a bug that only exhibits in your config. We can send you a phototransistor board for this.
We should also rebuild the imp app with the latest Xcode libs and get you on the beta list to check it out. Will email.