Reducing false peaks in altimeter readings

Discussion about deployment systems including altimeters, timers, air speed flaps, servo systems, and chemical reactions.
SaskAlex
WRA2 Member
WRA2 Member
Posts: 238
Joined: Tue Apr 07, 2009 2:36 pm

Reducing false peaks in altimeter readings

Post by SaskAlex »

Well I launched my perfectflite alt15K for the first time last night. The altitude versus time plot is below. As you can see, I have a ton of false spikes. I thought I'd probe for tips on reducing these before I go experimenting.

Now, there are several elements of my altimeter mounting that are far from ideal. First of all, I have one rather big hole in the payload section to allow the removal of a tommy timer gear jam at launch. This hole is considerably larger than the recommended size for this size of payload bay. It's also considerably less than the recommended 3x rocket diameter distance back from the transition of the nose cone. Also, my nose cone is on a string and I could hear it (in the onboard video) banging against the side of the rocket as it descended. I can't see this having much effect, though, because it hits the rocket below the payload section. Finally, at 26 km/h, it was rather windy last night.

Does anybody have any experience in what it takes to get a real smooth altitude plot? Do you think making three more equally spaced holes the size of my existing one would help? Or would that just make it worse since my hole is already too big?

For this rocket I might just not worry about it. I can just look at the raw data and take a conservative value for the height. But it might be worth designing my next payload section with the altimeter in mind.

Thanks for your thoughts,
Alex
Attachments
altimeter plot.JPG
altimeter plot.JPG (40.05 KiB) Viewed 104 times
User avatar
Daan.[D&P]Rockets*
Current WRA2 Flight Duration World Record Holder
Current WRA2 Flight Duration World Record Holder
Posts: 138
Joined: Fri Jan 11, 2008 4:19 pm

Re: Reducing false peaks in altimeter readings

Post by Daan.[D&P]Rockets* »

Hi alex,

Identical problems over here to! We are using the same altimeter and got the same weird spikes.
The last launch we did it gave us a height of 1252 feet. Have not been able to download the plot yet..

On my post in Photos and Videos.. George from aircommand give some options for reducing or causes of false readings.
SaskAlex
WRA2 Member
WRA2 Member
Posts: 238
Joined: Tue Apr 07, 2009 2:36 pm

Re: Reducing false peaks in altimeter readings

Post by SaskAlex »

Hi Daan,

I had a look at George's comments, and I'm thinking the sun might be part of the problem. It was almost sunset, so as the rocket spun around the pressure sensor would be exposed to direct sunlight. If the sunlight has to be more or less 90° to the sensor's surface in order to have an affect, this would explain why my peaks are so narrow but the overall curve is still quite clear. If it is sunlight causing the problem, it's the easiest one to fix. I'll try a paper cover over the pressure sensor next launch.

I can't say much before we see your plot, but I'm not sure if we are having the same problem. My peaks were so narrow they mostly got filtered out for the audible report of altitude. On the plot you can see I had one spike way up around 1500 feet, but the audible report was only 1021. That is actually quite close to what I believe to be the true apogee of around 1005 feet.
User avatar
WRA2
Site Admin
Site Admin
Posts: 1089
Joined: Sat Dec 30, 2006 11:30 pm

Re: Reducing false peaks in altimeter readings

Post by WRA2 »

SaskAlex wrote:Well I launched my perfectflite alt15K for the first time last night. The altitude versus time plot is below. As you can see, I have a ton of false spikes. I thought I'd probe for tips on reducing these before I go experimenting.

Now, there are several elements of my altimeter mounting that are far from ideal. First of all, I have one rather big hole in the payload section to allow the removal of a tommy timer gear jam at launch. This hole is considerably larger than the recommended size for this size of payload bay. It's also considerably less than the recommended 3x rocket diameter distance back from the transition of the nose cone. Also, my nose cone is on a string and I could hear it (in the onboard video) banging against the side of the rocket as it descended. I can't see this having much effect, though, because it hits the rocket below the payload section. Finally, at 26 km/h, it was rather windy last night.

Does anybody have any experience in what it takes to get a real smooth altitude plot? Do you think making three more equally spaced holes the size of my existing one would help? Or would that just make it worse since my hole is already too big?

For this rocket I might just not worry about it. I can just look at the raw data and take a conservative value for the height. But it might be worth designing my next payload section with the altimeter in mind.

Thanks for your thoughts,
Alex
I think that you are correct that the hole is too large.
Remember that the wind speed usually increases as you go higher in altitude. (there are no buildings or trees to slow it down) You may be able to use the video and coincide each time the rocket faces toward and away from the wind and whenever there is a gust. (especially if you know the wind direction as it relates to landmarks seen in the video. If the wind blows into the hole you get high pressure, away should make low pressure (this will depend on the diameter of the rocket) A larger diameter rocket may effectively block the wind from the hole when facing away. The real answer may be when the wind blows across the hole (which happens twice during every rotation) which increases the probability that a gust will blow across the hole.

What you could do to solve this is to put the altimeter into its own container with the correct size hole and mount it inside the payload bay that has the large hole. This will smooth out the spikes without having to redesign the payload bay.
Lisa Walker,
:WRA2: Forum Administrator. :WRA2:
:WRA2:The Water Rocket Achievement World Record Association :WRA2:
User avatar
Daan.[D&P]Rockets*
Current WRA2 Flight Duration World Record Holder
Current WRA2 Flight Duration World Record Holder
Posts: 138
Joined: Fri Jan 11, 2008 4:19 pm

Re: Reducing false peaks in altimeter readings

Post by Daan.[D&P]Rockets* »

Here is our plot from our last launch:

As you can see there are a lot of High peaks. 1 goes to 3500 feet..
We have some pretty big holes in the altimeter bay. Maybe that is the problem. And the sun is maybe an extra bugger ;)
Let me hear what you get out of the paper in front of the sensor method!
Attachments
7,5 liter raket ongeveer 630 feet.jpg
7,5 liter raket ongeveer 630 feet.jpg (54.3 KiB) Viewed 84 times
SaskAlex
WRA2 Member
WRA2 Member
Posts: 238
Joined: Tue Apr 07, 2009 2:36 pm

Re: Reducing false peaks in altimeter readings

Post by SaskAlex »

Daan, you got a really smooth curve on your class C record flight, didn't you? Can you think of the big differences between that payload and this one? Did it have smaller holes? I'm still hoping the sunlight cover will help, but I think I'll have to do something about the vent holes if I want to make big improvements.
User avatar
U.S. Water Rockets1
WRA2 Member
WRA2 Member
Posts: 1778
Joined: Sat Feb 03, 2007 3:24 pm

Re: Reducing false peaks in altimeter readings

Post by U.S. Water Rockets1 »

Alex,

Your hole size and placement is not optimal for recording your flights but should not create the kinds of spikes you are seeing. Those spikes look like they may be caused by mechanical shock to the rocket body. If your rocket is bouncing around on the parachute, it can bend and distort in shape and that can cause pressure changes inside the rocket... usually in spikes that jump low (compression) to high (expanding back out). Thar characteristic is in your graph at some points.

As far as your hole size and placement goes, if you have difficulty getting the hole in according to to directions, you can try putting the altimeter inside a container with one small hole within the rocket. The container acts as a baffle to the turbulent air caused by improper hole size and placement.
Team U.S. Water Rockets
Visit USWaterRockets.com
Visit our Blog
Tune in to our YouTube Channel
Visit our Facebook page
Visit our Twitter Page
Opportunity is missed by most people because it is dressed in overalls and looks like work. --Thomas Edison
SaskAlex
WRA2 Member
WRA2 Member
Posts: 238
Joined: Tue Apr 07, 2009 2:36 pm

Re: Reducing false peaks in altimeter readings

Post by SaskAlex »

I did another launch today and got a really smooth altitude curve. I didn't change a single thing except for making a light cover for the pressure sensor. It was also windy again. I really can't believe that the sunlight was the entire source of my problems, but I just can't see what else has changed.

Here's the plot. The only significant errors are right at launch and chute deployment. It looks like the chute deployed just a tad early, but it doesn't look like it resulted in the loss of much altitude.
Attachments
flight2.JPG
flight2.JPG (36.13 KiB) Viewed 64 times
flight2(2).JPG
flight2(2).JPG (34.79 KiB) Viewed 64 times
User avatar
Daan.[D&P]Rockets*
Current WRA2 Flight Duration World Record Holder
Current WRA2 Flight Duration World Record Holder
Posts: 138
Joined: Fri Jan 11, 2008 4:19 pm

Re: Reducing false peaks in altimeter readings

Post by Daan.[D&P]Rockets* »

Aah realy nice!

We had a nice curve on our Class C record. But on that day there was no sunlight.. It was cloudy so it coulnd disturbe the altimeter. Can you post some photo's of the cover you made?

Gr Daan
SaskAlex
WRA2 Member
WRA2 Member
Posts: 238
Joined: Tue Apr 07, 2009 2:36 pm

Re: Reducing false peaks in altimeter readings

Post by SaskAlex »

Here's the light cover, Daan. It's made out of black construction paper and two-sided taped in place. Just make sure there is some space between it and the pressure sensor.
Attachments
DSCI0051.JPG
DSCI0051.JPG (56.13 KiB) Viewed 51 times
DSCI0059.JPG
DSCI0059.JPG (67.53 KiB) Viewed 51 times
User avatar
Daan.[D&P]Rockets*
Current WRA2 Flight Duration World Record Holder
Current WRA2 Flight Duration World Record Holder
Posts: 138
Joined: Fri Jan 11, 2008 4:19 pm

Re: Reducing false peaks in altimeter readings

Post by Daan.[D&P]Rockets* »

Aha,

Realy nice! We are going to test this method also very soon! Hope that it works as good as your launch!
User avatar
U.S. Water Rockets1
WRA2 Member
WRA2 Member
Posts: 1778
Joined: Sat Feb 03, 2007 3:24 pm

Re: Reducing false peaks in altimeter readings

Post by U.S. Water Rockets1 »

We were going to ask you a question before we saw your recent flights without the errors. We had been thinking of another source for errors that could be trouble, and we thought we'd share with you our experiences in case it may come in handy to you or someone else. You should always test any altimeter that you have crashed in a rocket before. That usually results in unreliable readings. We crashed one way back in 2004 and we were getting really inconsistent readings that were way off from the simulator predictions. We chalked it up to the simulator being wrong. After a while we started getting false launch detections just sitting on the pad. We replaced the altimeter and had the crashed one repaired/recalibrated.

The pressure sensors are extremely sensitive and it's not surprising to see them get hurt by g-shock and the kind of over-pressurization that results from a rocket crumpling on impact. If you don't believe us, you can look at the altimeter graph Air.Command published from his recent flight and see the senor calibration get whacked when it hits the ground. After that the altimeter calibration goes to +75 feet just resting on the ground.

Who knows what happens when you crash these altimeters. We have ones we use for experiments and a pair of "gold" ones (flown only a few times and never crashed) that we use for record flights.
Team U.S. Water Rockets
Visit USWaterRockets.com
Visit our Blog
Tune in to our YouTube Channel
Visit our Facebook page
Visit our Twitter Page
Opportunity is missed by most people because it is dressed in overalls and looks like work. --Thomas Edison