My little altimeter

I don't see any sign of the Excel file, but here is a picture of the altitude graph, based on acceleration and pressure. The acceleration-based altitude is affected by tilt of the rocket from weathercocking and tilting the launch rod a bit into the wind. You can see the pressure spike from the ejection charge. The rocket that did this flight is just an Estes Hi-Flyer launched with an A8-3 engine, so there wasn't a sealed avionics bay!

Turns out the Excel file is too big to attach, so here are some graphs of the accelerometer output. I also have measurements of the battery voltage, chip temperature, and sensor input voltages to make it possible to do a calibration that corrects for those variables.

Here are 38 Parrots and 1 Peregrine, fresh out of Advanced Circuits (4pcb.com). There are 2 more panels just like this one. Props to them for letting me pick them up at 7:30 on a Saturday. Tonight or tomorrow I'll find out if I did the updates correctly. Fingers crossed!

Looks great

I'm still amazed at how small these alts are - they are incredibly functional for their size.

mtwieg said:

Anyways, a question that occurred to me is if you plan on adding deployment, won't the Li-poly battery be insufficient? I'm not an expert on them, but I doubt their ESR is low enough to drive pyrotechnics directly. Also, I'm assuming the voltage on your battery is around 3.7V, which won't be enough to drive some igniters anyways. If you wanted to keep the battery, you would probably have to boost the voltage and use a large discharge cap to fire the pyros, which would still limit it to low current loads.

I think the li-poly is great for logging, but not so hot for deployment.

Click to expand...

I'm not really into electronics, but i just wanted to put my $0.02 worth in on this. I would imagine that most rockets that even need a dual deployment system are at least 29mm minimum diameter. Anything smaller and the motor ejection and a standard recovery should be enough. So why not just use high current relays on the altimeter and use an external power source to light igniters or pyrotechnic charges for staging/DD? This would alleviate the problem of needing more voltage onboard. Again, this isn't really my thing, so correct me if I'm wrong.

I would imagine that most rockets that even need a dual deployment system are at least 29mm minimum diameter. Anything smaller and the motor ejection and a standard recovery should be enough.

Click to expand...

I tend to agree, and that's why I'm making the dual-deploy version, called the Peregrine, wide enough so that the board can be as short as possible while still fitting inside a 29mm rocket. See the My little dual-deploy Altimeter for more information on that board. I hope to have a prototype of the Peregrine ready for testing in a few weeks.

So why not just use high current relays on the altimeter and use an external power source to light igniters or pyrotechnic charges for staging/DD?

Click to expand...

As far as I know, all of today's altimeters use Field Effect Transistors (FETs) for power switching the deployment charges because for the current rating we're talking about (~2-5 Amps) they're smaller, cheaper, and more reliable than mechanical relays. The factors that will make the Peregrine uniquely small and light are the tiny components, tight packaging, and integrated li-ion cells. Keeping the board this small opens up more options for packaging, even in a larger rocket. For example, I'm working out a minimum-diameter 38mm rocket with this altimeter and the main parachute both located in the nosecone.:driver:

Well, I flew one of the Parrot altimeters today, and it seems to be a wonderful unit. I can't post full data yet, as I don't have the full calibration for the baro sensor (it output in counts rather than feet, and Adrian is planning on giving me the calibration so I can convert sometime soon), but once I do have the full data, I'll be posting it (possibly in another thread). The altimeter worked great though, and stood up to my punishment (60 gees or so continuous during the first half second of flight, and over mach 1.6). I will say that it is amazingly small for what it does, and seems to be a really capable unit.

Several graphs of data from cjl's flight is posted here in a thread over in the high power forum. Attached one of them.

Since cjl was edging up close to a club record, I wanted to be doubly sure that I got the barometric calibration right. I was already planning to drive from Denver to Summit County on Sunday, so I threw my laptop and the Parrot in the back of the car, and recorded the real-time data as I went. I took the long way over Loveland Pass (11993 feet) and recorded the times that I passed various landmarks. Then I looked up the landmarks on USGS topo maps when I got home later and used that as the basis for a revised sensor calibration. It turned out that this direct calbration method estimated the apogee within 20 feet of my original calibration, so I was pretty happy about that.

With a better resistor selection the altitude resolution will improve by about 50% (it's about 15 feet at 17,000 feet above sea level now), while still keeping 100,000 ft altitudes within the measurable range.

If you happen to still be looking for someone to professionally manufacture these, I happen to work for a small prototype company and if you send your bill of materials and assembly drawing/gerber files to [email protected] I can give you a quote for making these. I can't say for sure, but I think we'll have a much better price than you've found elsewhere. Especially if I can acquire a few of the finished product at cost . And hey, if you end up needing a couple thousand I can make them dirt cheap...

Hey Adrian,
That altimeter looks outstanding! Definitely let us know when it is available for purchase. And if you would like someone to test it out in a rocket let me know

Adrian! I have been following this thread quietly as I do not have much to contribute in the sense of knowledge regarding design and electronics. I am more of a mechanical guy (you know - look at the parts, hands on type). Electronics for me are stuff that either works or that you put in the trash bin. I do not construct or repair electronics!

Anyway, I just wanted you to know that once I decide to get an altimeter, yours is very high on my list!

Thanks for the encouragement. I'll definitely update this thread here when they're available. In my testing I have been finding more battery discharge with the unit off than I had been hoping for. In my Peregrine prototype I experimented with a screw switch integrated on the board, and I have been happy with how the switch turned out, so I think I will do one last rev of the Parrot board to incorporate the switch for a positive battery disconnection.

Well, the rev 3 Parrot PCBs are done, and I took the boards down the street to my assembly vendor along with an expensive box of parts. Now one last double-check that everything is just the way I want it, before sending the assembly people the data they need and the go-ahead and...

I got some "encouragement" from one of my NCR friends ("I won't buy one unless it does apogee deployment" :no, so I started thinking about what it would take to just do a simple apogee deployment with the existing board. It already has a spare switched output, but the FET resistance is too high for ignitors or most ematches. But there's a nice dual FET part available that fits in the same location as the old one and has much lower Rdson. The trouble is that doesn't have the same pinout. So should I scrap my new rev 3 boards?

Today I had some fun lighting Estes ignitors with my little Li-poly cells, and trying to see if it's possible to add this feature without making the board bigger. The little 65 mAhr, 1 gram Lipoly cell I use for the Parrot can smoke one or two Estes ignitors in series, so long as the microcontroller doesn't have to stay running at the same time from the same cell. In order to connect igniter wire and a separate pyro battery return without making the board bigger, I worked out a way to do it by making the mounting holes at the front end do double duty as board mounting holes for 2/56 screws, and wire terminals. One is now connected to the ground plane for the pyro battery return, and the other will go to one end of the ignitor. (the + end of the battery and the other end of the ignitor have to get connected together off-board somewhere, perhaps with a separate arm switch) So hardware-wise, I think I can do it. Next I'll do some of the last code additions I have been planning to see if I can fit it all in if I add logic for Mach-inhibited apogee deployment.

Spinning another rev of the boards is painful given the cost and schedule, but now that I know that adding apogee deployment can be done, I can't help but go for it.

Adrian A said:

Also, anyone have a suggestion for a name? I've made a few flyers with the title "Featherweight Altimeter" but I'm open to suggestions. 50% off for someone comes up with a good one!

Click to expand...

How about The Hummin' Bird Altimeter? It's small like a humming bird...

Well, it's already named (the Parrot), but that is an interesting suggestion

Thanks, but I'm saving the hummingbird name for when I make a small altimeter. If I make the Hummingbird, it will be no-frills but be less than half the size of the Parrot.
The Parrot is small-ish, colorful, and talkative.
I have some designs for a 3-output altimeter that I'll call the Peregrine (fast dive before it fires the mains), but I don't know if I will have the patience to develop it to production.

The first flock of Parrots:



Now the challenge is to finish assembling, programming, and calibrating all these in time for the Northern Colorado Rocketry Club's Oktoberfest.

Nice! Did you do the SMT? How do you cut the big blog of alitmeters into featherweight ones?

Nice! Did you do the SMT? How do you cut the big blog of alitmeters into featherweight ones?

Click to expand...

I had the SMT parts professionally assembled by Advanced Assembly. The panel looks really clean and hopefully it won't have any of the solder bridges or opens that I had trouble avoiding when I was doing prototypes in my basement.

When I'm ready to cut the boards apart, I'll do it with a Dremel tool's cutting wheel, and then trim them up with a belt sander that's clamped to a workbench.

Adrian A said:

When I'm ready to cut the boards apart, I'll do it with a Dremel tool's cutting wheel, and then trim them up with a belt sander that's clamped to a workbench.

Click to expand...

Do you have a diamond wheel? If not, I highly recommend one for cutting fiberglass.

Do you have a diamond wheel? If not, I highly recommend one for cutting fiberglass.

Click to expand...

Thanks for the tip. I may give one of those a try. The Dremel abrasive wheels are cheap and they're nice and thin for greater margin for error, but they sure don't last long. My cutting area is pretty much littered with the crunchy bones of broken abrasive wheels already, just from cutting out prototypes.

yes, definantly diamond >>>dremel abrisives. I don't know if the is a harbor frieght near you (although they do have a webstore), but the have the perfect saw for work like this--a little 4" tabletop circular saw, that comes with a diamond blade. It is about 200% faster than dremel

Also, I really like that pic of the sheet of altimeters. Really shows off the super dense routing in there :surprised:


Also, have you at least powered one of them up yet to make sure the sweatshop guys weren't soldering any of the chips in upside down, wrong spot, etc?

yes, definantly diamond >>>dremel abrisives. I don't know if the is a harbor frieght near you (although they do have a webstore), but the have the perfect saw for work like this--a little 4" tabletop circular saw, that comes with a diamond blade. It is about 200% faster than dremel

Click to expand...

Interesting. Actually, that sounds kind of like a small, homeowner-grade tile saw that I already have that came from Home Depot a few years ago. That one is intended for wet sawing ceramic tiles, but I wonder if I could use it dry on these boards. The batteries will be attached by the time I cut them up, so wet isn't an option.

Also, I really like that pic of the sheet of altimeters. Really shows off the super dense routing in there

Click to expand...

Thanks. I thought that I would be at this point last June, with rev 1 of the board. Now I'm up to rev 5.:trytofly: But I have squeezed in a few more features than the original, along with the bug fixes.

Also, have you at least powered one of them up yet to make sure the sweatshop guys weren't soldering any of the chips in upside down, wrong spot, etc?

Click to expand...

Yes, I have one running now that I'm using to work out the code on a calibration program and the final flight program. And no sweatshops were involved; the parts are all machine-placed here in the Denver area, and for a lot less money than my time and patience would be worth. This is the short version of the story.

I agree on the saw blade, specifically a wet/dry turbo cutting diamond saw blade is a good choice for cutting the circuit boards. The only real issue is whether you have enough space between boards for the blade width.

The cheapest is a 4" pack of three diamond blades from Harbor Freight. You can buy them individually for $6.49 each or in a 3 pack (with two others not suited for populated circuit boards) for $7.99.



https://www.harborfreight.com/cpi/ctaf/Displayitem.taf?itemnumber=41740

https://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=46152

If you don't have a tile saw HF also has 4" ones (don't use the wet blade) for $59.99

https://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=3733

Perhaps a better one is the 7" unit for the same price of $59.99 but again you need a different blade. https://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=91511

You could use larger 4 1/2" and 7" turbo blades with this one. The larger blades are likely to last longer but cost $14.95 and $29.95 respectively

https://www.harborfreight.com/cpi/ctaf/Displayitem.taf?itemnumber=41741

https://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=41742

If these blades are too wide, you need to like at slittling blades. The 4" and 4 1/2" diamnond blades are probably 0.080" wide and the 7" blades are probably 0.090" wide. Mcmaster-Carr sells metal and carbide slitting saws that are much thinner.



https://www.mcmaster.com/pdf/113/2464.pdf

Bob

Not that it will help you now, but in the future I suggest specifying perforations on your pallet to make the separation easier. Then you can just pop em apart and use a small file to take off the burs. Playing with a dremel near PCB's scares me.

Thanks, Bob, for the great info. The tile saw I have at home is just like the one you linked to. I think I'll pick up one of those wet/dry blades at a Harborfreight on my way home and try it out. I have 0.125" (or was it 0.1"?) between the boards, but with the guide on the tile saw I shouldn't need much room for error.

Not that it will help you now, but in the future I suggest specifying perforations on your pallet to make the separation easier. Then you can just pop em apart and use a small file to take off the burs. Playing with a dremel near PCB's scares me.

Click to expand...

The company I wanted to go with for the boards (Advanced PCB, 4pcb.com) specifically prohibits that sort of thing with the special I was using, otherwise I would do that or perhaps a V-score. The dremel method isn't too bad though, but now that I will have Li-poly cells attached to each board it will be a little more exciting.

Great work! I am taking a peek at your home page every now and then.

Comes Christmas I just might have to... (looking at my wallet and hear it moan... :surprised: ) ...or I might have to put that on hold until spring...

Great work! I am taking a peek at your home page every now and then.

Click to expand...

Thanks for the encouragement. I should update my web page, too.

Well, I picked up the turbo diamond cutting wheel, along with a set of arbor adapters, but when I got home I found out that the arbor on my saw is 7/8, whereas the 4" wheels were 1/2-3/4 depending on the adapter. So I decided to go for it with my original tile saw wet cutting wheel anyway, and it worked fine so far on some test cuts, and is definitely easier to be more precise than with the handheld Dremel. Thanks for the idea!