tips for a backup altimeter

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Hyperteks "sing" at about 60-120 hz. A low sampling frequency is OK if a passive low pass filter was used. Without it Nyquist will hurt you.
 
What frequency did the AltAcc use? What frequency did they think was needed to handle the Hypertek motor?

Based on Nyquist they should have had a sampling rate at least twice the highest frequency of the motor's buzz. John's figures for the Hypertek buzz seem reasonable (trying to relate them in my memory to the familiar 60 Hz buzz.)


Steve Shannon
 
Good point about the safety factor.

I cannot envision a zero risk factor, both as a flier or spectator, in this hobby, and it is something that I am very aware of.

That being said. As you put it, "with a little outlay (time and money)" I personally want to take what I feel are reasonable precautions, in an effort to mitigate what could be a regrettable situation.

I often fly a single altimeter, but agree two isn't a bad idea. We're all out there on a huge dart board, and rockets are tiny. We haven't been lucky, just playing within the extremely low odds of someone getting smacked. But with every failed flight that comes in, we move a little bit closer to an accident. So reducing the number of failures is a good thing.

As to someone who said BP doesn't fail- poorly packed charges fail, and sometimes fail to sheer pins/tubes bind/etc. Two full charges (or a plus sized backup) isn't a terrible idea. I don't claim to do it all the time. but if we're talking "best practice" I'd put down two charges for that.
 
Here's my redundant setup:
Main Altimeter:
-apogee charge, correct amount of BP
-main charge at X altitude, correct amount of BP
Backup Altimeter:
-Apogee + 1 second, BP+some (usually .5g extra)
-main charge at X-200ft, BP+some

My methodology:
Apogee:
Primary: BP amount as calculated to shear the actual number + plus one additional pins. This accounts for material specification variables.
Back-up: Primary amount + 100% extra. Deployment timing is Apogee + 1 or 2 seconds (as limited by altimeter settings)

Main:
Primary: BP amount as calculated to shear the actual number + plus one additional pins.
Back-up: Primary amount + 50% extra. Deployment timing is primary less 200'

I ground test the primary amounts and have not had a failure using the actual + 1 shear pin calculations.

Considering the significant risk associated with a failed apogee deployment, I don't skimp on the back-up BP amount.
 
About ten years ago or so, I had a rocket I launched quite a bit with a single dual deploy altimeter and used the motor delay to back-up the drogue. I used the (ultra reliable) flash bulbs for the deployment charges. Worked great for many flights. So I thought I was getting pretty good flyin' rockets and didn't use the motor as back-up on a J-350 flight.

Well up to apogee this flight was nominal But no drogue! As I watched it coming in ballistic I wondered what I could have done wrong. I'm usually extremely careful in prep 'cause I used to race cars and motorcycles. As I watched, right on cue at 250 feet the main came out and the rocked disintegrated in front of my eyes even tearing off one of the Rocketman chute lines. My wife and I spent an hour picking up parts so we could do a post-mortem.

Found out the drogue flash bulb didn't go off and had an open circuit. I still have the bulb.

All high power rockets I've built since then have two entirely separate altimeter systems in them. Haven't had a problem since or even needed the back-up. But when (not if) I do need it, it's there.

Joe
 
We haven't been lucky, just playing within the extremely low odds of someone getting smacked. But with every failed flight that comes in, we move a little bit closer to an accident.

I digress...
Are you familiar with the Gamblers Fallacy?
Basically... Think of it as flipping a coin. Just because the coin comes up heads 1000 times in a row, it doesn't mean that the next coin toss is any more likely to be tails... It's still 50/50.
So, in this case, it's extremely low odds of someone getting smacked, period. And the odds are going to continue being low no matter the outcome of any previous launch. It's erroneous logic to say that we're "moving a little bit closer to an accident." Not only is it logical fallacy, it's also unnecessary doom & gloom.
Now... None of this is to say that proper, logical redundancy it's unnecessary. Every safety measures that can be taken should be taken. With that, I wholeheartedly agree.
... And I return you to your regularly scheduled TRF thread...
 
Oh I hate to argue statistics...

Assuming we all do something to make every flight we do a little safer the mean of the probability distribution will tend to move more to the "safe" end of the spectrum. So if we all think and act a little more safely, the stats will show it.

Now, any distribution with enough samples approaches Gaussian distribution which can be quite easily worked with. The probabilities of the "bad" tail of the graph can be quite easily related to the statistics by mean and standard deviation. So by flying safer we drag the graph in the direction of safe, reducing the area of the "bad" tail and the possibly suboptimal outcomes (eg death) that it predicts.

So that is good and means we will never have any fatalities, doesn't it? No. There is a finite probability (it is not an infinitesimally small area) that death (for example) could occur.

Now increase the number of launches over time. That same probability of failure applies to a larger population, so eventually it just might happen. The longer you go, the more likely it becomes.

Same applies to coins. 50/50 per throw with no knowledge of the historical events. The final distribution will always approach the 50/50 with enough throws. Yes, you could possibly throw 1001 heads, but it is very unlikely. You will always end up with the results approaching 50% for each side (minus a small % for landing on an edge :wink:) )
 
Yeah... But...
(When philosophic logic meets statistical analysis/prediction)
Statistical probability of any one event occurring is still completely independent of all other such events and is only dependent on the circumstances/variables of that one single event.
However... A friend's put it like this: imagine a golf ball and a blade of grass. The probability that the golf ball will hit that one blade of grass is nearly infinite, but there is a nearly 100% chance that the golf ball will hit A blade of grass.
So, manage the variables on each individual rocket flight (like a golfer perfecting his swing, or a rocketeer using strong redundant deploy, motor backup, proper motor selection, safe distances, etc) and the probability for a negative outcome on that flight drop precipitously, completely independent of all other rocket launches everywhere over time. Therefore,I stand by my statement that it is a logical fallacy to claim that "we are moving closer" to a fatal rocket flight simply because we haven't had one yet (under TRA and NAR safety codes). By that logic, once it happens, we could all breathe easy because it happened and now, statically speaking, it won't happen again for another 70 years. Obviously a false statement. Just as obviously false as the original statement.
Obviously, it COULD happen at any moment on any launch, the probability is never zero on a rocket launch. Hence, safety first on every flight. And the best levels of safety depend on good systems and checks.
To bring it back to the OP... A good redundant electronic deployment setup would have two completely redundant systems at least using tested charges with plenty of wiggle room to account for as many eventualities as we can imagine, within reasonable limits.
 
Last edited:
I almost always have two means of apogee deployment. The simplest use motor deployment as your backup. Just plan a delay that is about 2 seconds longer than optimum. This is always possible or desirable.

If you need to go with redundant electronics, use separate batteries, switches, e-matches, and charges. Plan the apogee backup 1 or 2 seconds after the primary and main backup about 200 feet lower. Make the backup charges 25% to 50% larger than then ground-tested primary charges. And finally, use your most reliable (or familiar) device as your backup. Thinking hear is the primary is optional, but the backup HAS to work.

...Fred
 
And finally, use your most reliable (or familiar) device as your backup. Thinking hear is the primary is optional, but the backup HAS to work.

...Fred

Then use 2 of those... :)
 
Back
Top