Go Down

Topic: Electromagnetic interference? (Read 4578 times) previous topic - next topic

exceleronsports

Hello!  Big Arduino fan here with a mind boggling problem that I'm hoping someone more experienced than myself can solve.  Admittedly, I'm not great with electronics, but I am a software engineer by trade so I have just enough technical expertise to be dangerous.  My issue is related to some strange behaviour we've seen with our weather balloon tracker (http://overlookhorizon.com) that seems to inexplicable fail on all our recent flights.  My latest theory is that we have some electromagnetic interference that is occurring with our setup perhaps?  Here's what's happening:

  • The tracker worked fine prior to the introduction of the cameras and camera battery packs.
  • The tracker works perfectly when connected inside and/or outside of our payload box and with USB power input
  • The tracker works when connected outside the payload box with battery power
  • The tracker works when inside the payload box with used batteries, nearing end-of-life
  • The tracker fails, loses GPS signal, and occasionally "freezes" the Arduino when inside the payload box with brand new batteries
  • The tracker works when inside the payload box, with brand new batteries, IF the camera batteries are turned OFF (see diagram, below)
  • The moment the camera batteries are turned ON, the tracker will lose GPS signal, even if already  acquired.  Also, the moment you turn OFF the camera battery packs, GPS signal is reacquired almost immediately.
  • Unproven, but in our recent flight, we had a backup external GPS unit that was completely unconnected to the Arduino.  This also could not receive GPS coordinates after landing (only expected to work at ground level).  Once the camera battery pack died, though, it acquired a good GPS signal.


I tried to include lots of detail, but can certainly provide more if it helps.  Just looking for some suggestions on what might be occurring as it has been frustrating to try to pinpoint and gets expensive to replicate since it only seems to occur with fresh batteries.  Thanks for your responses in advance!

Here's a diagram of our "stack" (sorry for the basic drawing) that goes into our payload box.  The payload box has internal dimensions of 6"x5"x6"


123Splat

Let's try a schematic of what you think you have hooked up, shall we?

exceleronsports

#2
Aug 22, 2016, 07:12 pm Last Edit: Aug 22, 2016, 07:18 pm by exceleronsports
I can certainly do that... do you have a recommendation on software to use to draw the schematic?  Everything I've looked at doesn't have an Arduino Mega component to start with...

exceleronsports

I just found fritzing.org ... let me try to draw something up that's a little better than my terrible patchwork drawing.  :-)

123Splat

Do NOT use fritzing. nobody likes it.
Use something like Eagle.  and just use a terminal strip (pin header) to represent the mega pinouts (label them).  It doesn't have to be perfect just legible.

If nothing else photo a hand drawn schematic.

allanhurst

#5
Aug 22, 2016, 10:38 pm Last Edit: Aug 22, 2016, 10:38 pm by allanhurst
Sounds like the camera takes enough power to drop the battery voltage enough to upset the other electronics.

 So :

-  it might be the start up current is high - add a big reservoir capacitor.

- as you get  higher, the temperature drops - does the battery ( and the rest of the gear?) handle that OK?

regards

Allan


exceleronsports

Thanks for the thoughts...

Temperature isn't an issue as the temperature inside the payload never drops below 50° F (10° C).  The payload is well insulated.

In terms of the voltage drop... the batteries that power the camera (4 AA's per camera) aren't being used to power the Arduino.  The Arduino has its own separate battery pack (8 AA's) so turning on/off the cameras shouldn't drop the voltage of the Arduino's power supply.

allanhurst

OK well idea 1/  dead!

GPS runs at 1574 MHz with  VERY low signal levels - they're coming from satellites 25,000 miles away..

It's quite possible that harmonics from the camera's clock are upsetting this.

So :   put ferrite beads on all your power, earth and signal leads.
    :   add earthed shielding around the camera.

regards

Allan

exceleronsports

Perfect!  This is a direction I haven't tried... I'll try some ferrite beads.

In terms of "earthed shielding", I'm not quite sure what that is.  Have any sort of example of this?  Apologies for my lack of electronics knowledge...

I'm working on a formal schematic of my design now, although it'll take me a bit.  Hopefully, I can post a schematic tomorrow that might give some better insight.

Wawa

  • The tracker works when inside the payload box with used batteries, nearing end-of-life
  • The tracker fails, loses GPS signal, and occasionally "freezes" the Arduino when inside the payload box with brand new batteries

Without looking at anything else than "a 12volt source on the DC socket", I would say the regulator of the Mega is overheating. Feel it.
I probably would have choosen for a single 12volt battery pack with several 5volt micro-power buck converters.
Leo..

TomGeorge

Hi,

Can you please post a copy of your circuit, in CAD or a picture of a hand drawn circuit in jpg, png?
Especially your power supply components.

Definitely not a fritzy....

Have you used bypass capacitors in the arduino and peripherals circuitry.

Thanks..  Tom... :)
Everything runs on smoke, let the smoke out, it stops running....

exceleronsports

Here's my first attempt at an actual schematic.  Hopefully everything is actually accurate.  Never created one before so hopefully it's not too atrocious.   :)


dlloyd

Could be the reduced current that's available @5V with the increased voltage at Vin.  With 8 batteries fully charged, could get up to 1.75 x 8 = 14V (too high). Why not use just 6 batteries for 9V to power the Arduino? You'll have the same current capacity (mAh) and could actually gain some running time due to less power loss.

TomGeorge

#13
Aug 23, 2016, 04:17 am Last Edit: Aug 23, 2016, 04:18 am by TomGeorge
Hi,
You have no bypass capacitors around the 78 series regulators, please look at the data sheet and see the suggested capacitor values.
Also caps on  the supply lines to your peripherals.
The physical layout of your wiring is very important.

Thanks.. Tom.. :)
Everything runs on smoke, let the smoke out, it stops running....

exceleronsports

Thanks!  I could certainly do the 9V option.  I'll certainly look at changing that, but I'm not sure that addresses my main issue, though.  The Arduino itself operates normally with the 8 battery setup now.  The issue arises when the camera batteries (the two 4-battery banks) are switched on to power the cameras.  The Arduino continues to work (kind of).  The main issue is that the GPS will instantly lose signal acquisition as soon as the camera battery packs are turned on.  Turn the camera battery packs off - the signal is reacquired almost immediately.

Go Up