Eggtimer Quantum Didn't Survive First Test?

[mccordmw]

A while ago, I built an Eggtimer Quantum, and successfully ran a deployment test.

[video=youtube;eB8vqssPK-w]https://www.youtube.com/watch?v=eB8vqssPK-w[/video]

That was a few months ago. After the test, I put it away without handling it it any way except to disconnect the battery. Today, I charged the battery and reconnected it. I get 3 beeps, a pause, then a long beep. Then, the LED on the WIFI chip blinks once. However, I don't see the Eggtimer's WIFI address to connect to. I've checked with my phone and my laptop.

I thought maybe an iffy solder broke loose after the shock of a deployment test, so I put it under a loupe and checked eveyr solder point. Everything looks sound after several checks. Not sure what else could be wrong except maybe the shock broke something in the WIFI chip itself? Has anyone else had the WIFI chip stop broadcasting after the shock of an ejection?

 

[michigander]

I've had to change wifi connection on phone before , like switching from house wifi to egg wifi

 

[mccordmw]

Thanks. I tried that with both my phone and my laptop. I can't detect the network using either.

 

[smapdiage9]

Try reflowing the solder joints on the wifi chip?

 

[mccordmw]

Going to try Cris' suggestion and first re-flash the chip tonight. If that doesn't fix it, I'll try reflowing the solder. *fingers crossed*

I have a second one on order as a backup, but if this one can't be fixed, it will be my primary, and I'll dig out my old StratologgerCF as the backup. I wanted to avoid putting in switches for arming, though. Fortunately, I still have my screw switch from my Binder Design kit, so it's an easy option.

 

[woferry]

I had one fail after my second ground test, but it failed in a different way than you describe, it's totally dead (no beeping, WiFi module gets incredibly hot).

My first ground test was of the drogue, so the Quantum moved away from the fixed lower airframe. It continued to work fine after this test. But I then went on to testing the main, where the Quantum didn't move at all only the nose popped off, and it was an undersized charge so the nose just basically fell out of the rocket, no shock cord stretch, etc. After this the Quantum was dead, never beeped again. Cris replaced the kit but it certainly wasn't reassuring about the reliability of the WiFi module. I took shots of the board with an IR camera and the module and VR were the hot-spots on the board. I had the shield of the WiFi module carefully removed by a professional circuit technician I know at work, and looking again with the IR camera after the shield was removed it's the ESP IC itself that's getting hot, there aren't any shorts on the board or the like.

There shouldn't even have been any mechanical shock on that second test since the bay didn't move and the charge was undersized (I have video of both firings that I could post). So I don't know if it was something electrical that killed it. This particular board was rigged for independent batteries, so there shouldn't even have been a voltage dip on the Quantum's VR when the charge fired. So I'm stumped as to what could have caused it to fail under such gentile conditions, seems like much less than it would be subjected to in a typical flight. I've re-built the replacement kit though I haven't flown it yet, I'll probably only use it with backup until I have some more confidence that it will survive.

 

[cerving]

I've historically had about a 2% failure rate on the ESP8266 modules, i.e. 2 out of every 100 modules that I get. That's a bit high, but it's a module rather than a monolithic part so the rate will be higher. The latest batch that I've gotten seems to be better; they're packaged in factory-standard carrier tapes instead of being put into little antistatic baggies by the reseller. I open up the package to flash them, so I have some degree of confidence that if the module is dorked then I was the one who probably did it. So far, I haven't had any issues with the ones that come packaged in a tape.

 

[mccordmw]

Yes. I was a bit worried about shock sensitivity of the ESP8266-12 WiFi controller, so I checked on Google for any info. Appears to be a very widely used Wifi controller, but I couldn't find any common complaints about durability, so I'm not too concerned about its suitability for use in rockets. Maybe you just got the rare bad one?

 

[mccordmw]

After troubleshooting with Cris, we guessed it was a faulty Wifi chip, but there was no way to be sure. So I got a replacement to assemble. Thanks for Cris for all his help troubleshooting. Great support.

I also had already ordered a backup Eggtimer Quantum and that came in the mail along with a replacement for my primary. I assembled them both at the same time. I forgot what a pain it was to solder the optoisolators. This time, it was a breeze. I used extra flux on the pads even though the solder already has flux in it. In this case, copious amounts of flux really helped the solder wick under the legs and form really nice fillets on top and bottom of the connections.

I'd highly recommend getting a no-clean flux pen like Kester 951. It makes the solder wick beautifully onto the pads and legs. Although it's no-clean, I do like to give the end product a shot of flux remover to make it all pretty. I had a can of FLUX-OFF handy for that.

End result is I have another 2 Eggtimer Quantums tested and working fine.

If you assemble an Eggfinder, you might also want to use extra flux and look up how to do drag soldering. That should really help with assembly of that unit.

Oh and do yourself a favor and get a $2 spool of 2.5mm solder wick. It's a lifesaver for removing blobs of solder when you get too much on the board.