U8 spycam - g-force effects?

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

car3107

Active Member
Joined
Feb 4, 2018
Messages
29
Reaction score
0
I've used these cheap little usb "spycams"on a few launches now, and for the most part they perform as you'd expect. On a coupla flights, though, they've gone weird when subjected to, in one case, launch g-forces or, in another case, being whipped around following chute ejection. In both cases video stopped being recorded but audio carried on.

I'm just a little intrigued on how shock/g-forces could affect such a low-mass item with no moving parts. It's not been a case of the battery disconnecting or the sd card ejecting. Any ideas of other ways g-forces could be affecting these cameras? Thanks...

Sent from my XT1039 using Rocketry Forum mobile app
 
The camera I have in my rocket is powered by a Raspberry Pi and on several occassions the recording has ceased before it should have - it was suggested to me that acceleration can interfere with crystals, which I'm sure it has, and so cause problems. Maybe your spycam has crystals :confused: .
 
So crystals as in crystal oscillators for time reference? If they don't like being subjected to high g's I guess that would indeed cause problems. Interesting that I can't get my cameras to fail when subjected to the deceleration g's of giving the camera a whack or two, though... Looks like a subject for further study. Thanks for the idea.
 
Oh, that's interesting. I know as those crystals age they kind of stiffen up and start oscillating a little faster. I wonder if under g forces they compress a bit and start running just a bit faster. That would give you all kinds of oddball effects, increasing the clock timing while it was in the middle of something...
 
Here is the other days's vid of the U8 usb spycam in the nosecone with a 45-degree mirror to give horizontal view. Camera fails at the moment of chute ejection charge. Previous fails have resulted in video recording stopping while audio continued ok. This is the first time that we get the rolling "unsynced raster scan" effect in the video and the square-wave audio garbage...

Oddball effects indeed.

https://youtu.be/uFhVnHVmsiE
 
I agree - but one thng I can't get my head around on the other day's video is that, despite the video doing the rolling-raster thing and the audio going all square-wavey, the timestamp continues to progress and display, and the three big faint "water spots" (caused, I have assumed, by my rig's mirror and/or windscreen being less than clean) remain visible. If the clock etc has crashed, I don't see how those could continue to be visible...
 
I could conceive of a failure mode where the oscillator continues but at either a relatively wafting or different frequency, or since the crystal is damaged, by just oscillating around some other frequency.
 
I like the form factor of U8 cameras, but have had very poor luck using them. I even had a few that I thought turned on the pad, but only got video and sound halfway through the flight (like the camera decided to turn on 5 seconds into the launch). To be honest, I'd rather chalk this up to a mechanical failure of the connections or switches in the camera, than some sort of gforce affecting the oscillator. I suppose it could happen, but given the cheap construction of them, usually mechanical failures are the first thing that has a problem on cheap electronics.
 
usually mechanical failures are the first thing that has a problem on cheap electronics.

I agree 100% but it's such an odd failure it's interesting to spitball something that might cause that. And re: OverTheTop - exactly, it would just cause the frequency to drift around. When you design electronics there's all kinds of havoc that can cause, especially when you're doing things on the cheap and you make a lot of assumptions about this thing will be done before this other thing happens without any margin for error.

There's nothing you can do about either problem other than just buying a different brand and hoping it has more margin of error in its construction to handle this kind of stuff.
 
Back
Top