Thanks, Mike and Warren, for the constructive feedback.

Mike, help is on the way in the form of a PC interface program that Jim Yehle has been kind enough to provide, which will simplify the downloading and interpretation. It's about 1 software rev away from me advertising that Hyperterminal is no longer necessary, and Excel is optional. It's a few revs away from also eliminating the need for Excel entirely.

While an altimeter-derived peak altitude readout would undoubtedly simplify an altimeter's use in contests, the validity of the peak is really questionable. The very flight that started this thread would have a reported altitude of about 9050 according to a peak-readout altimeter, rather than 8241. Why should that be o.k.? Look at the graph again on the first page. The post-ejection dynamics induced a pressure in the av-bay that was lower than ambient. And this isn't the first flight I've seen this behavior. If the altimeter doing the peak recording is the same one that performed the ejection, it can be programmed to ignore any altitudes after it starts flowing current to the deployment charge. But that's definitely not the case for an Alt15k (which is accepted for records but doesn't do deployments), and I don't know if that's how the other altimeters are programmed. Also, when two altimeters are flown together, one may fire a charge while the other is still recording the peak. A human reading a graph can correctly interpret the true peak altitude better than an altimeter will ever be able to. Not to mention the issue of ambient temperature compensation, which would also require an off-altimeter implementation if the community ever decides it wants to improve its record standards to be more accurate than about 10%.

As an afterthought, I proposed in my R&D entry at NARAM 50, that for the NAR to accept altimeters for record attempts, to get rid of the need for theodilites, that a pressure chamber be present at a meet, and every altimeter compared to a 'gold standard' altimeter for that day. Any percent of variance to the standard would be added or subtracted from the altitude reading to establish the actual meet results. Competitors would have their altimeter serial numberd, calibrated with its offset and approved before competing at that meet, or season.

That would certainly work to eliminate questions about altimeter calibration. A test rig that would be adequate for this would cost about $250 for a reference pressure transducer. My setup cost about $500 additional for a pump, chamber and vacuum relief valve so that I can hold the pressure steady. There are probably cheaper ways to do it if a steady reference pressure isn't necessary. But before long we'll be seeing relatively cheap and small altimeters with factory-calibrated digital sensors that will eliminate the calibration issue altogether. If the spec accuracy is to be believed, these pressure transducers are as accurate as a reference gauge over a wide range of sensor temperatures. That's what I had in my prototype last January that I never got around to developnig. But for an accurate altitude, you still need to eliminate deployment spikes and compensate for ambient atmospheric temperature.

Adrian, In your model rocket configurations, the altimeter is in the same envelope of the airframe so it sees the deployment charge and is directly affected by it. Would the reading be more accurate if it was isolated from the charge by being in a payload, so that it was only reading outside pressure?

Although that adds a little weight and length to the record attempt, it might prevent the anamonoly from occuring. I don't know much about this, so if I am off base, forgive me for my ignorance. If this is a true assertion, then that would help 'sell' it to the sanctioning bodies.

Adrian, In your model rocket configurations, the altimeter is in the same envelope of the airframe so it sees the deployment charge and is directly affected by it. Would the reading be more accurate if it was isolated from the charge by being in a payload, so that it was only reading outside pressure?

Although that adds a little weight and length to the record attempt, it might prevent the anamonoly from occuring. I don't know much about this, so if I am off base, forgive me for my ignorance. If this is a true assertion, then that would help 'sell' it to the sanctioning bodies.

I'm pretty sure his configurations use a payload bay. At least Barac-it did.

Actually this was something I had interest in as well. Were I able to put the altimeters in with the laundry, I'd be able to save considerable space. Over in "lost and found" Adrian had commented that he didn't see the altimeter in the photo of the found rocket, but wrote "But i think I threaded them onto the shock cord, so maybe they're deep inside next to the motor? (fingers crossed)"

I've always had a sealed ebay, as other altimeter manufacturers continually talk about damage to the unt, corrosiveness of bp, pressure spikes, etc.

Bottom line, if you don't have to seal this area off, it would greatly modify current building techniques. I'm curious on Adrian's thoughts here, as MW, Transolve, PF, etc. caution against this.

Give an experienced flyer your rocket, the same motor and your electronics and do it again. Or do it yourself with high supervision. Prove that you can do it, or come close. Science can tell you anything, except when the human factor of skepticism comes along. So prove what you need to, to take out the human factor.

I always put a sealed bulkhead between the altimeter and the deployment charge. In the case of the booster converted to a single stage that went to 8241 feet, there were actually 3 separate pistons so that it could do booster duty as a single structural piece. The inside insides of the rocket went, from the bottom up:

-G37 motor with the transmitter antenna alongside the motor
-the av-bay with the transmitter and the Parrot side-by-side, and tied to the top of the motor,
-a bulkhead with a section of a coupler as a seal
-the apogee charge with a section of shock cord
-another piston to protect the streamer from the ejection charge
-a streamer
-another piston to protect the streamer from the booster separation charge,
-another streamer
-the nosecone, with 80 grams of tungsten-epoxy.

Here's a picture of the whole 2-stage configuration before I finished painting it:

The insides of the booster are kinda complicated because I wanted the booster to be in one external piece for stiffness, since I knew it was a long, skinny rocket. So when I flew it in a 2-stage configuration, the booster fired the separation charge at the top of the rocket through an electrical connector, and then ejected the front 2 pistons and the streamer at apogee using a charge that was on the other side of the bulkhead from the electronics. For the single-stage attempt, I replaced the sustainer with a heavy nosecone and I didn't want a stage separation charge, so I put in the second streamer between the top piston and the nosecone. It didn't appear to help much; the two streamers and the shock cord came in as a tangled mess at about 50 fts/second. No damage though, other than some paint chips on the fins.

Back to the altimeter discussion, though: I brought the heavy nosecone with me to the launch just in case I wanted to fly the booster high as a single stage, but I was focused on getting the 2-stage configuration to work right. So I didn't anticipate the post-deployment dynamics that caused a lower pressure reading after apogee. At apogee, the apogee charge sent the laundry and 2 pistons out the front. When the shock cord went tight, it pulled on the top of the avionics sled that has its built-in piston bulkhead. That made the av-bay move away from the motor a little bit (everything was friction-fit), which increased the volume the altimeter was sitting in, reducing the pressure in there until the vent hole could equalize the av-bay area again.

This isn't the first time I've seen an altitude spike caused by the ejection, however. Pressure transducers also respond to high accelerations like you get from an ejection charge, and also sometimes you get the vent hole facing different ways into the wind, etc, that make any readings immediately after the ejection charge totally unreliable. Those really need to be ignored one way or another.

If you look at the graph on the first page, you can also see a pressure spike before apogee. That happened when the motor burned through the delay. I had removed the motor's ejection charge, but I forgot to fill the hole, so some hot gas got into the av-bay area when the delay burned through. You can also see this in the temperature plot that the Parrot recorded during the flight. After seeing this, I realized now that this was what caused the weird plot in the G25 flight I did this summer. Apparently when the delay charge burned through, it got past the epoxy potting I used on the end of the motor on that flight, and may have contributed to the glued-on shock cord coming off of the motor, which caused me to lose the motor and my nice 29mm all-carbon fin can somewhere out on the prairie (Good thing for you SS competitors!)

So in answer to John's question, the two times I have let hot BP or AP residue get into an av-bay, I re-tested the altimeters and they had less than 0.5% error compared to my reference, so it doesn't seem to do any immediate damage to Parrots. I can't vouch that the same would be true for other sensors, however.

Give an experienced flyer your rocket, the same motor and your electronics and do it again. Or do it yourself with high supervision. Prove that you can do it, or come close. Science can tell you anything, except when the human factor of skepticism comes along. So prove what you need to, to take out the human factor.

If I had a transmitter I trusted, I'd put that in there and give it to someone to fly at the November launch. I think I'm going to start working with the Adept transmitters next, despite their wide bandwidth, frequency drift, and larger size. I lost 5 beeline transmitters in the last year, and I think only 1 was my fault. You know what they say: "Fool me once, shame on...you, fool me 5 times..... you can't get fooled again. " I'll send this beeline back to see if Greg can find out what's going on and if I'm screwing something up.

Maybe I'll get a new transmitter and some revised internal parts and re-fly this at the Atlas site in December if the weather holds. With plenty of supervision and a fully-plugged G37 this time. I wonder if the vacuum pump for my little calibration chamber would run off of my inverter (probably not). Then I could do an on-site calibration verification with witnesses. And compare to somebody else's altimeter while I'm at it. Or do it at somebody's house with a wall outlet on the way back if they're interested.

this is a perfect example for my to point out what point I was trying to make:

Why does it need to be re-calibrated? Once you have programed the altimeter and factory calibrated it why would you need to do anything other then maybe verify readings testing it against a standard? Are you doing this before every flight? This is were I think you being a manufacture gives you the unfair advantage. Does every use if he parrot need to go through this every flight? if so it is definitely not user friendly and if not calibrated correctly or if it was not done then you will get different data then if you had, this gives you a margin for error on the operators side or an opportunity for someone to cheat. If you are doing R&D then that is OK, you are still developing your firmware but I do not think you should be submitting it for altitude records.

I am hoping this will help you see the point that I have been trying to make. And again I am not saying you have, are or will but you do need to factor in both the "human" and the "stupid" factor to keep users from submitting data from an improperly calibrated altimeter.

James,

Of course the Parrot doesn't need to be recalibrated. I already said upthread that I rechecked the accuracy after the flight, just to be sure, and the accuracy was damn near perfect. The only reason I proposed bringing my chamber with me for the next attempt is for folks who are skeptical and might feel better about all this to see that the altimeter is accurate with their own eyes. Nobody has to do that. I'm trying to be as transparent and accomodating as I can, offering to go to more trouble than would be asked of anyone else, and amazingly, you're taking that as evidence that I have an unfair advantage?

Maybe this was just a misunderstanding and I should have said "accuracy check" in my last post instead of "calibration verification.". It's just doing a check of the factory cal, just like you suggested.

if you are able to change calibration on site, soothing/manipulating data before sending in your data for an altitude record, it is. NO one is doing any of that before they submit their data, only you. Everyone else turns on their altimeter, fly the rocket, bring it back to LCO/RSO has altitude verified and then the form is filled out and sent in when they get home.

I am not trying to beat you up over this or call you a cheat, just want you to understand what you saying and are doing look questionable to some.

if you are able to change calibration on site, soothing/manipulating data before sending in your data for an altitude record, it is. NO one is doing any of that before they submit their data, only you.

I'll repeat what I said earlier, in case you missed it:

Parrot altimeters report pressure when you download the data. There is a formula that everyone uses to convert pressure to altitude, and that's what the spreadsheet does when you plug in the Parrot's data. There's no adjustment, no tweaking, no scaling of the pressure data. I do use the data "straight off", and that's what I reported.

Please PM me if you want to discuss this further.

Adrian,

My thought is why don't you submit the Parrot to Tripoli as THE official recording altimeter for record flights? Let them test them against calibrated standards and mandate that IT is the "official" altimeter for TRA altitude records?

Get in touch with the records committee and send them a couple to test. The combination of baro sensing and 3 axis accelerometers is better than any of the other standards.

Warren