Altimeter Accuracy - RC Groups
Thread Tools
Mar 11, 2012, 07:18 PM
Aurora Builder

Altimeter Accuracy

People, myself included, tend to put a high level of faith in the altimeter readings we get with our model airplanes. This is the altimeter I use, and feel it is a good representation of the existing altimeters:

Today I was out launching and flying with some friends. I was flying the Aurora wing on an Akcent-2 fuselage. Due to some structural issues with this particular wing, testing yesterday indicated a max height of 147'. I suspect 160' might be possible with the aileron linkage changes I made yesterday, but that's about it as the wing is simply too twisty (not a problem with a carbon d-box )I have previously recorded 184' with the same altimeter in the same fuselage with the Akcent-2 wing. I launch Validol an easy 10' higher (190-200') than that, and am still 20' shy of some top launchers (~220'?). All this adds up to me thinking the altimeter is working great and accurate to within a few feet.

Then today I record a 227' launch with the Aurora wing. Side-by-side with some Tabooish's that I doubt were going over 175' (probably closer to 150-160'). Yes, the Aurora design in d-box form on a great fuselage with great tails should be more than capable of that altitude. Right now though, I just can't trust that reading. And no, I did not climb in altitude at all during the flight, it was a very challenging day and getting over 1:30 was hard. Perhaps the altimeter just didn't reset? My thinking however is with a tight canopy seal, the flow on launch will generate suction out the tail boom, hence lowering the pressure inside the fuselage and giving a false reading to the altimeter. Thoughts?

Sign up now
to remove ads between posts
Mar 11, 2012, 07:59 PM
Registered User
I don't trust and altimeter readings from anyone that does not show data. The truth of the matter is altimeters are always the least reliable instrument in a cockpit. This is because the altimeters are based off of atmospheric pressure readings which unless properly calibrated and then corrected based on local atmospheric conditions, can lead to large deviations in altitude.

The how high series of altimeters are probably the worse of them. Due to the dynamic nature of our launches, a large negative spike in altitude is seen in the pressure data. This well documented with various theories as to why this happens. I have seen this negative pressure spike to be on the magnitude of 10 to 30 ft! The how high algorithm basically takes the lowest and highest pressure reading, subtracts and converts the values to the unit of measure desired. Due to the huge negative pressure reading, folks are getting an extra 10 to 30 ft of altitude.

It is my opinion that for any altimeter to be trusted, the data has to be available to be looked at so that a proper null value can be taken. Rams series altimeters allow this as well as other post processing so that a semi accurate altitude can be determined. That along with a good test procedure, leaving the airplane on the ground for a few seconds before launch, allows for the best realistic launch data.

For myself, while I am roughly 10 ft below the top launchers in the US, using the procedure I listed above, I record between 160 to 180 depending on conditions and airplane configurations. Personally I think most altitude claims are over exaggerated because I will throw oflver people who claim to throw 185+ft.

Mar 11, 2012, 08:08 PM
Will fly for food
davidjensen's Avatar
I have the same unit and so far its been consistant for me. My first readings were in the 120's and since then I have worked it to the 140's to low 150's. It has shown me that at altitudes of over 4000' I can get an additional 25' from my launches. The best I've thrown was 169' in a blustry 5+mph wind and I'm certain there was some updraft in it. I've wondered about the pressure inside the fuse and how it may affect the sensor. Apon launch there must be a low pressure area at the back of the boom sucking out the air from within, however at the apex of the launch that should not be a factor.
Mar 11, 2012, 08:31 PM
Jim C
ShadowFalken's Avatar
I have had atlimeter data show that my spin up went below ground level then to altitude. A quick pushover at the top to level flight still shows as a spike and drop of 30 feet. Would you trust that? This was data collected using the method of allowing the model to sit on the ground for a period between launches. Pick the plane up, you can see the increase to 3 feet of altitude before the spin starts. Then the spike NEGATIVE during the spin to below ground level.

If you select that negative spike as your zero point, I throw 220 feet too. I stopped using it for anything other than personal setup comparison. Make changes and look at the top spike. I ignore the reading and look at the trends.
Mar 11, 2012, 09:20 PM
Aurora Builder
Guess I should get a second unit that allows for post processing. At the end of the day, if I can compare models, that's what matters but right now I don't trust the data enough to do that.
Mar 11, 2012, 09:21 PM
Registered User
I don't believe any of these altimeters are temperature compensated. That means that unless you are comparing readings taken at the same air temperature, you will get different results.
Mar 11, 2012, 09:26 PM
Registered User
I had use of a How-Hi for a while last year. I went from suspicious, to trusting, and then gave it up when it became wildly erratic. I felt I could use it in the course of a day to gage things like launch technique, but as an actual source of data, I just couldn't bring myself to rely on it.

I'd like to try one of the RAM units. I've seen others use one, and it seems to correlate with what we see on the outside.

Yours, Greg
Mar 11, 2012, 10:44 PM
Registered User
Interesting discussion. We have noticed that these altimeters are affected by the days temperature, the altitude of the field, the altitude they were originally calibrated at, etc, etc. I think Greg is absolutely right -they are pretty useless for absolute readings (ie how high was that launch) but very reliable for comparitive readings taken with the same altimeter -( ie how much higher was the launch with 1mm reflex than 1mm camber.)

I fly a lot of free flight and have the Altimax altimeter. I was always rather disappointed with the results I got in comparison to the claims of others. For instance, my Blaster 3 ST with Swindells tails looks really high on my best launches but reports only 42m. So I took my FIA glider out, attached both my Altimax and a friends and towed overhead to a known 51m altitude on a measured towline. (For comparison, 10 flights were made, all bunt launched to about 75m including a couple of minutes glide in thermally weather to compare the readings). My Altimax consistently reported 43 44m for the known 51m (14% under), my friends 47m. Although this was woefully inaccurate, the plots tracked each flight with an accurately matching discrepancy throughout. The altimeters were mounted on the fuselage side to avoid the pressure spike that shows on top of the wing. We also noticed that they were slightly affected by the orientation whether the sensor was facing into or away from the airstream which gave about 1.5 - 2m variation in the reports.

My suggestion is get one that over reports the height you will feel a lot better about your flying
Mar 12, 2012, 08:26 AM
Registered User
alan7037's Avatar

You need salt.

[QUOTE=murne;21007510]Interesting discussion. We have noticed that these altimeters are affected by the days temperature, the altitude of the field, the altitude they were originally calibrated at, etc, etc. I think Greg is absolutely right -they are pretty useless for absolute readings (ie how high was that launch) but very reliable for comparitive readings taken with the same altimeter -( ie how much higher was the launch with 1mm reflex than 1mm camber.)

After many hours working with FOUR altimeters with dlgs and electric sailplanes my advice is that we need to be very cautious about our altitude claims. I have done a number of build threads on RC Groups over the past few years, usually including some launch data, and each time it is clear that the data must be taken with a pinch of salt.

I have two How High altimeters, a RAM 2 and Eagle Tree's altimeter. The most ridiculous readings came from the Eagle Tree's device - it would suggest that my launch was a mere 75 feet when I was clearly over 140 feet according to the How Highs.

But the How High devices that I have are not giving accurate readings either. Take two dlgs out, alternate the planes with their respective HHs and there is a large difference between the readings - often by as much as thirty feet.

To test the How Highs I put both of them in the same plane, launched it and to my horror, found that the readings differed, but differed inconsistently. One flight would give a gap of twenty feet, while another would produce a larger gap between the readings. Go figure.

I have not compared the RAM2 against any of the other altimeters that I have. But given that they are erratic I am not sure that anything of any value would emerge from a comparison.

Jim is right - try to look for a trend. But if the data appears to be inconsistent this becomes a frustrating exercise. (Now if the inconsistency followed a pattern...)


Mar 12, 2012, 09:25 AM
flying is not a dream
lebenj's Avatar
you must know, that barometric pressure change during the day, and sometimes, more quickly than we think...
also, the pressure is not so stable.

the how high, take a reference pressure when you power it, so after several launches, the reference is maybe wrong ...
after 1 hour, you can have more than 50ft or error!!!

with a RAM3 or openlatimeter, when you read the graph at home, if you take care of the reference, you can trust of your launch height, but there tolerencies that i think can give you a good 1m of error.

ram3 VS openaltimeter grafic at home:
you can see the terific pressure variation during the flying cession (....

Openaltimeter grafic at home:
in 2 hours more than 50feet of error

on the field, the Openaltimeter, can give you the height by beeps. it's accurate, because, when it detect the launch, it takes a 0 reference, with few samples befores the launch... BUT, if you take a look on a graph, the reference have like an oscilation on the field, so depending of the waves.... you can have easily 1m more or less fot the same throw!!!

Mar 12, 2012, 12:28 PM
Registered User
Does anyone know how the How-Hi devices actually work? I doubt seriously they are barometric devices, but I certainly don't know.

One of the big deals in full-scale test flying is "calibrating" the altimeter installation on a prototype aircraft. The sources of error are many - and subtle. The barometric altimeter is the least reliable instrument on any airplane's instrument panel, followed closely by the airspeed indicator. Both rely on a source of static pressure, and that's the issue. The compass is not usually accurate, either, although it is usually consistent in it misinformation and you can deal with it.

But, the How-hi (and other) devices are pure, electronic solid-state - aren't they? How do they work? Surely they aren't integrating acceleration.

Yours, Greg
Mar 12, 2012, 12:47 PM
Father of Fr3aK, DLG Pilot
tom43004's Avatar
I have tested the lolo and the how-hi against each other. They're both reasonably accurate, and reasonably inaccurate. It depends how you test them. It's pretty easy to make them both look bad statistically (larger enough variation to cause doubt) and it is also pretty easy to make them both look somewhat accurate.

Don't look at one or two or three launches to compare. Look at ten or twenty and throw out the highest 10%, the lowest 10% and average the remaining 80%. You'll find that they're pretty good at telling you whether you are improving your setup or going in the wrong direction.

If you look at small samples, testing data will look more random.

I trust both the LOLO and the How Hi for basic testing. Do I trust them to know whether I threw 193 or 195 feet? Nope. Do I trust them to know whether a different camber setting is better or worse than another? Sure.

Trust the trends and relative findings. Ignore the numbers if you don't trust them.
Mar 12, 2012, 12:47 PM
Duane, LSF IV
Wazmo's Avatar
Originally Posted by glidermang
Does anyone know how the How-Hi devices actually work? I doubt seriously they are barometric devices, but I certainly don't know.
Pretty sure the How High uses a barometric pressure sensor, probably MEMS type. I know the OpenAltimeter uses a Bosch MEMS barometric pressure and temperature sensor.
Mar 12, 2012, 01:05 PM
Soaring Circuits
rcbrust's Avatar
Guys, the major points have been touched upon already but I'll group them together here. The two largest contributors to altimeter inaccuracy are air temperature and barometric drift. Neither of these have anything to do with how well the altimeter senses pressure.

Let's look at air temperature. A barometric altimeter has to do two things. It has to accurately sense pressure and then it has to convert those pressure readings into altitude data. The algorithm that's used to convert pressure to altitude is temperature dependent. If you go out on a cold day and throw X feet high, without temperature compensation, the altimeter will read higher than if you go out on a hot day and throw that same X feet. The program that lets you view your data has to be able to compensate for this.

As shown above, the barometric pressure can very greatly throughout the day, even over an hour or so if a front is coming through. This can really throw off your zero altitude reference. As described previously, the best method to use is to let the model sit on the ground for a few seconds, then fly, then after landing let it sit for a few seconds again. That way, the viewing software can let you pin those two reference points to zero.

It can be a little bit of a pain, but if you follow this procedure and use good recording hardware and the proper viewing software, you can get altitude readings accurate within a couple percent or so.

Mar 12, 2012, 01:15 PM
Jim C
ShadowFalken's Avatar
All true, Randy. The soaring circuits software allows for all those variables and corrections. I know playing with it, it can sense a person walking up a flight of stairs. Impressive.

My issue is what would be suggested for prevention of the VERY visable spikes caused by the nature of the DLG launch. I fully trust the data in normal flight conditions.

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Question Thick lay up + vacuum bagging + leading edge accuracy Berrie Composites Fabrication 16 Jan 24, 2012 04:14 PM
Discussion A simple way of measuring Kv of unloaded motor with acceptable accuracy modisc Electric Motor Design and Construction 7 Jan 19, 2012 03:20 PM
Question ESC with GOOD cutoff accuracy ? cyclops2 Batteries and Chargers 4 Dec 03, 2011 04:02 PM
Discussion Accuracy of the ICE Logger? eJets Castle Creations 19 Oct 18, 2011 08:34 AM
Discussion Altimeter V3 accuracy golfietis Eagle Tree Systems 1 Jul 30, 2009 05:45 PM