Quantcast

Need help with Eggtimer Proton

The Rocketry Forum

Help Support The Rocketry Forum:

DeepOvertone

Well-Known Member
TRF Supporter
Joined
Oct 29, 2017
Messages
253
Reaction score
48
Hello All,

I wanted to know if any users/assemblers have ever experienced the issue I'm dealing with on a Proton altimeter. The altimeter worked correctly after assembly(I've assembled many before this one) Then it just stopped working. It beeps upon applying power, the wifi SSID shows up in wifi connection managers and I'm able to connect to it and obtain an IP from the wifi module. I can even ping the server IP(192.168.4.1) and get a valid response. BUT, I'm not able to actually browse to the internal web server. It just hangs and hangs. And most recently I've had the proton reboot itself upon attempting to get into the web server. I've hooked it up to a terminal and booted it up and the little info that gets displayed looks fine. I even updated the firmware(it flashed just fine) but that didn't help anything. The only other info I can offer is that when the proton reboots itself while hooked up to the terminal, it spits out a bunch of other data but I'm not sure what to make of it.

Thanks for all your help you guys.
 

mbeels

Yes balsa
TRF Supporter
Joined
Feb 9, 2019
Messages
2,543
Reaction score
1,329
I have not had that experience. Did you check with Cris (Eggtimer)?
 

Voyager1

Well-Known Member
TRF Supporter
Joined
Oct 23, 2015
Messages
611
Reaction score
203
Location
ACT, Australia
I've built a couple of Protons and haven't had these problems. As suggested, contact Cris. You will get a little bit of garbage data out the serial port when rebooting.

In the meantime, I assume that you've gone through the fault-finding check list in the assembly manual. In particular, this section:

Can’t bring up Proton web page
• Bad WiFi connection (check your WiFi manager)
• Incorrect URL (use http://192.168.4.1 )
• Weak battery (use a freshly charged one)
• Bad solder joint on the ESP8266-12 module
• Problem with the two 4.7K resistors next to the baro sensor
• CAT25M01 EEPROM mounted incorrectly
• Browser setting issue (try using the "private" mode on your browser)


Your issue reminds me of a similar problem I had when developing an application with another WiFi processor. It would reboot when a sensor communication process wasn't able to complete causing the processor to timeout. Perhaps the coms with the baro sensor is faulty, or the EEPROM? At any rate, check your solder connections thoroughly, bad joints and solder bridges.

Good luck! I hope you get it resolved.
 

Attachments

Last edited:

DeepOvertone

Well-Known Member
TRF Supporter
Joined
Oct 29, 2017
Messages
253
Reaction score
48
Cris did get back in touch. He said its failing the check on the accelerometer. I checked the solder joints that he told me to and even re-worked them just in case. It still does the same thing. Haven't heard back from him on where to go from here. The weird thing is that this altimeter DID work when I first assembled it. Then the second or third time I powered it up, it just stopped working properly. Nothing was changed.. Its almost as if the accelerometer just stopped working. Oh well. I may just see about sending it in for him to look at if he does that sort of thing.
 

Voyager1

Well-Known Member
TRF Supporter
Joined
Oct 23, 2015
Messages
611
Reaction score
203
Location
ACT, Australia
Ah, yes, I forgot about the accelerometer. It's one of the peripheral serial ICs that are connected to the processor. It's not unusual for a kit build to work initially, then stop. It could be a bad joint with an intermittent connection, a damaged PCB track, solder bridge under the IC, etc. It's also possible that the IC was heat damaged when installing, but was able to function briefly on the first power up before failing. You might need to replace the accelerometer IC.
 

DeepOvertone

Well-Known Member
TRF Supporter
Joined
Oct 29, 2017
Messages
253
Reaction score
48
Here are a couple images that I sent Cris. I'm not sure if it'll help you guys at all but maybe you can see something I can't. Thanks again for all the help.
62836427449__8D5EEFA6-5A3B-49F0-AA30-E29E6089C153.jpeg
IMG_4729-1.jpeg
 

cerving

Owner, Eggtimer Rocketry
TRF Sponsor
TRF Supporter
Joined
Feb 3, 2012
Messages
4,093
Reaction score
1,339
A few of the WiFi processor pads look pretty "flat"... like the leads didn't get through the board. Check them for continuity against the pads on the top of the processor.

The three pads pointed out above would affect firing/continuity but they wouldn't keep it from booting.
 

DeepOvertone

Well-Known Member
TRF Supporter
Joined
Oct 29, 2017
Messages
253
Reaction score
48
Checked all the circled pads in a 60x scope. Its just the picture.... They are all very shiny and nicely filleted in real life. Checked continuity on every wifi module pad front to back... all solid. Honestly, I've gone over the board many times with that scope and don't see any cold joints or jumped tracks. Cris, I know you're super busy right now, but could I send this guy in to you and see if you can figure it out once you slow down? I hate to give up on this rather expensive piece of kit. Thanks
 

DeepOvertone

Well-Known Member
TRF Supporter
Joined
Oct 29, 2017
Messages
253
Reaction score
48
Sure, I'll look at it. Hopefully by the time it gets here I'll have the rest of the sale orders shipped out... there's only a few dozen left, waiting for parts so I can kit up the missing item(s).
Send to the snail mail address on the website?
 

OverTheTop

Well-Known Member
TRF Supporter
Joined
Jul 10, 2007
Messages
4,659
Reaction score
1,928
Location
Melbourne Australia
Just looking at some of the joints there appears to be no heel fillets on the leads in some locations. See picture below.
IMG_4729-1.jpeg


The toe appears to have a nice fillet, but I can't see any solder between the lead and the pad (along the side of the lead), and the heel (left end of where the pad bends up away from the pad) has no fillet at all. In regards to checking for solder joint quality according to IPC-610 there should be a heel fillet always. There is no requirement for a toe fillet at all in checking quality of a solder joint. Make sure you hold the iron on the legs for long enough to encourage the solder to flow under the pin towards the heel.
Untitled.png

Heel fillets are important to relieve the mechanical stress caused by thermal expansion/contraction of the SOIC parts, and also vibration. The fillet provides a reasonable radius that reduces the effect of stress raisers. Stress concentration is inversely proportional to the radius of curvature of the fillet (smaller fillet = higher mechanical stress).

If you are thinking about how to make the solder flow remember that solder follows heat, flux and other solder. Work out which of these you can use to get the solder where you need it ;). Make sure both the pad and the pin are at temperature so the solder can form a properly bonded joint.
 
Last edited:

DeepOvertone

Well-Known Member
TRF Supporter
Joined
Oct 29, 2017
Messages
253
Reaction score
48
Update: I ended up sending the proton back to Cris for troubleshooting and it took him a bit but he finally found the issue!! Turned out to be a weird rare production error on the PCB itself. If I understood his description and photo correctly, it was a bad VIA that wasn't properly plated through to the other layer which caused an open circuit to the accelerometer.

I wanted to give Cris some props for taking the time to troubleshoot this thing and really dig in to figure out why it was acting so odd. His products are top notch and customer service is outstanding. He took care of the issue for me and I'm proud to purchase and own his products.

Thank you to everyone who commented and offered their help on this problem.
 
Top