Sunday, June 12, 2022

Project update: The probe

For good or for ill, the heart of our system is our airdata source, i.e., the probe. This is roughly what our latest-model probe looks like:

Notice that the probe parts are made of HP MJF Nylon, and the probe is equipped with an explicit static probe made of silver-soldered brass.

This probe has been shown to provide very consistent airspeeds in a wind tunnel, under varying orientations. Basically, any air data quality problems are now gone. Check out this video:

I have used JawsTec for my 3D printing and have had great results until recently, when parts started to show up insufficiently sintered and "shrunken", and with voids:

I have switched to using Shapeways even though it is a bit more expensive. I plan to re-design a bit so that the remainder of my parts are made of regular Prusa PETG at 100% infill.

I have experienced cracking in the plastics, which I traced to my use of Loctite thread locker. For my next builds, I will either use plain CA glue or no locker at all.

In my current builds, I have relied on a combination of soft and hard tubing to get the air pressures around. The pressure sensor barbs are a little small to fit tightly with soft 1/8" OD tubing, so I have to use hard tubing for these, but the remainder of my fittings would be super tight. All this adaptation and mess is kind of crazy. My plan is to re-engineer with soft 3/32" OD tubing and use staked-in 1/16" brass tubing instead of the black plastic hose fittings. Stay tuned for how this goes.

Otherwise, the current design of the probe is pretty much as you have seen on the pages of this blog so far.

The probes have a nasty habit of shutting down inflight. Once they reboot, they try to autozero, which they do with significant dynamic pressure, which means the readings are off. For that, we need two things.

First, we need a firmware fix that writes the autozero settings to NVRAM and, if significant pressure is detected at bootup (say, more than 10 knots) then use these settings and do not re-do the autozero.

Next, we need to root-cause the problem. I suspect the JST battery connector is loosening due to vibration, but I cannot repro this, no matter how hard I try. Below are videos of the "probe torture" experiments with vibration, in some cases combined with heat and cold. No repro.

The "fix" I plan is to simply de-solder the connector and solder battery leads directly to the PCB. This should eliminate problems if they are due to that component. If not, then they'll show up again and we can chase them down again.

Note the fact that we still have the brass static tube. Is there a way we could eliminate that? Previous testing in Wichita State revealed that our hoped-for configuration was not great. But we have another trick up our sleeve. Check out this probe nose design:

The idea is that there are two holes in the sides that feed into a large plenum, which should average out the pressure between the two, and the pressure in the plenum would be calibrated to provide static pressure via the normal correlations.

Doing this would save us a bunch of work and time when we make the next batch of (we hope!) 10 probes. We are gearing up to do a wind tunnel test of this work as soon as we can. Wish us luck!

No comments :

Post a Comment