4" Electronics Bay Design for my Level 2 Cert Flight

[g.pitts]

I generally print most of my avbay bits at 15% hex infill, 3 top/bottom layers and 2 perimeter shells. They seem quite strong, but light.

FYI, I've printed one sled with 30 percent triangle infill, and it did turn out significantly lighter with no obvious reduction in strength. Thanks for the tip!

I also refined my PVA removal technique. Instead of baking the print at 130'F, I took it down to 90'F and alternate 90 minutes of ultrasonic with a couple of hours of soaking in the warm water. Dump the water, refill, and repeat until the print is clean. There's no FAST way to remove PVA. Slow and steady seems to be the best process to use.

 

[tollyman]

Hello! I am working on a semi similar project. Instead of using 3d printed sled, my PCB is the sled itself. What size traces did you have? and im guessing the large resistors are being used as current limiting resistors to prevent the lipo from supplying too much current? My design incorporates mounting holes for astrologers and the same style connectors to plug the altimeters in. PCBs should be in this week!

 

[g.pitts]

Hello! I am working on a semi similar project. Instead of using 3d printed sled, my PCB is the sled itself. What size traces did you have? and im guessing the large resistors are being used as current limiting resistors to prevent the lipo from supplying too much current? My design incorporates mounting holes for astrologers and the same style connectors to plug the altimeters in. PCBs should be in this week!

I took the same approach for my L1 certification - the PCB was the sled. Worked great, but you're spending PCB $ for area that you aren't using very effectively (lighter than my 3D printed sled, however). That's when I decided to go the route that I did for the L2 flight. I still have the KiCad schematics and board files for that somewhere.

The resistors are intended to limit the current sunk by the MOSFETs on the altimeters "just in case" (so really more a protection of the altimeter outputs). "Just in case" most likely would mean a shorted e-match, which I suspect is not the normal failure mode, but... I chose these resistors. They have a 3 Watt continuous rating, but have pulse withstanding specified since the theoretical max power is around 40 Watts (V^2 / R = 8^2 / 1.6) for the length of time I have the Quantum outputs set for, which is two seconds.

As for trace width, I chose 100 mils. I probably could have gone a bit wider, but when I threw the limit switches into the circuitry it got a little busy in that area so I stayed with 100 mils.

Hey, I found the layout for the "sled on a PCB"! Here was my initial implementation. I used tie-wraps to secure the two batteries - one battery was mounted on either side since I wanted complete redundancy. Both ends were designed to break away from the finished PCB, and be glued into the PCB milled slots (secured with epoxy). I had an Eggtimer switch on each altimeter, and used the Quantum altimeters. This was before the relatively recent TRA ruling that declared "thou shalt have a mechanical switch to break power to the charges". There was a fair amount of current required - going from memory, I believe the switch and the Quantum each drew about 100 mA.

 

[OverTheTop]

Remember you can always specify thicker copper if you don't want to make the tracks wider. 1oz copper is standard, 2oz usually doesn't cost much extra, and is half the resistance of 1oz. You can't do it on PCBs with fine-pitch tracks though (the design above would not be a problem). The limit on track spacing for this depends on the PCB manufacturer and they can tell you what they can work down to for 2oz.

 

[tollyman]

My PCB trace widths are 1.5mm, idk what that is in mills. I have an option to put a resistor in line with the battery but i think in most cases i will short that out with a soldiered connection. I am going to do tests with various batteries to see how much current the traces can handle. Where ever possible i mirrored the traces front and back to double the current carrying capability. I did think about going to 2oz trace thickness but it nearly tripled the price using EasyEDA (through jlcPCB). Here is my messy schmatic...I have it set up so i can have dual stratologgers or a stratologger and a raven.

 

[OverTheTop]

1.5mm, idk what that is in mills

60 mils

1mm = 40 mils

I am going to do tests with various batteries to see how much current the traces can handle.

There are various on-line calculators for working out maximum currents in traces. Personally I would not go over a 30degC temperature rise when looking up the currents.

 

[g.pitts]

There are various on-line calculators for working out maximum currents in traces. Personally I would not go over a 30degC temperature rise when looking up the currents.

That was my approach in sticking with 100 mil traces. I was nowhere close to 30'C rise, but I don't recall the exact number.