VNE vs altitude: glider specs vs rules of thumb

[Bret Hess]

Since some gliders (like mine) don't have tables of VNE vs altitude, I made a plot of relative IAS-VNE vs altitude from seven random gliders that do have tables in their manuals. I got most of it from BGA datasheets.

See the plots at: https://docs.google.com/document/d/1t6kgxATcXP7nIhvwkP3hySKrUU5P_jrUFx8OA41uH0A/edit?usp=sharing

Some interesting points
1. Most of the seven gliders follow the same curve, but two (Ventus A,B and Discus BT) have constant IAS-VNE to much higher altitudes before they decrease. Is this because they were simply tested to a higher altitude? or were they designed differently?

2. The common rule of thumb I've heard is that IAS-VNE drops by “2 percent per kft after 10 kft�. I plotted this with the data in the first plat, and it's a pretty bad description.

In kft, the correct description is “Constant IAS-VNE until 6.6kft; then drops 1.4% per kft�.

Or an easier-to-calculate rule of thumb is, “Constant IAS-VNE until 7kft; then drops 1.5% per kft�.

This is good to 40,000 ft.

3. In a discussion on RAS in 2002 (https://groups.google.com/forum/#!topic/rec.aviation.soaring/iRgr7pc44xg), Ian Strachan wrote about "The normal German (LBA) protocol used for the glider Vne schedule" which I'll rephrase as “Constant IAS-VNE until 2km; then IAS-VNE drops as TAS-VNE is held constant.� Indeed this is the curve that most gliders follow as shown in the 2nd plot. In km the description is “Constant IAS-VNE until 2 km; then drops 4.4%/km�. It's good through 12 km.

Does anyone know the rationale for the details in the protocol? Is 2000 ft just a convenient flutter testing altitude, so most gliders are tested there and no higher?

[Andy Blackburn[_3_]]

On Monday, June 8, 2020 at 7:17:13 AM UTC-7, Bret Hess wrote:
>
> Does anyone know the rationale for the details in the protocol? Is 2000 ft just a convenient flutter testing altitude, so most gliders are tested there and no higher?

My understanding on this topic is that for gliders in particular flutter is a significant concern at higher speeds. Flutter is an aeroelastic effect. It is partly a function of IAS (the relationship between lift, Cl and alpha) and partly a TAS effect (the time it takes a parcel of air to traverse the chord of the wing - which affects the frequency response of the aerodynamic parts). It's also a function of the structural rigidity of the wing, hinge moments of control surfaces, Cl and Cm vs alpha, depending on the type of flutter. Which is to say it's complex to calculate and somewhat dangerous to test.

TAS vs IAS does increase at about 2% per 1,000 ft so the totally conservative thing to do would be to tie Vne to TAS, but this leaves you with less room between stall and Vne at higher altitudes, so they sort of split the difference, while remaining conservative (the method is likely more sophisticated that splitting the difference literally). I suspect the reason they hold Vne constant up to a certain altitude is that it's easier for pilots to have a single number for the most common operating altitudes and leave the table lookup to less common operating scenarios. My best guess is that holding IAS-Vne constant with altitude means you have more flutter margin at lower altitudes and that margin falls to whatever limit they calculate/test for at the top of the constant speed band.

The other thing about high speeds at lower altitudes is that the relationship between control inputs and G-loading goes up with IAS, which is why you have a maneuvering speed (and similar logic for the yellow arc) which is calculated off of the aircraft's V-N diagram.

Andy Blackburn
9B

[Bret Hess]

On Monday, June 8, 2020 at 1:27:57 PM UTC-6, Andy Blackburn wrote:
>
> TAS vs IAS does increase at about 2% per 1,000 ft

Although looking at the data of the protocol and the glider specs, it's really closer to 1.4% per 1000 ft. Certainly 2% is easier to use and 1% is too little. The "10Kft, 2% rule" kind of splits the error. V is too high at 10K and too low at 40Kft. I have to say while it's not accurate enough for the table I will make for my glider, it's a good rule of thumb for on-the-fly estimates.

[Matt Herron Jr.]

On Monday, June 8, 2020 at 2:41:56 PM UTC-7, Bret Hess wrote:
> On Monday, June 8, 2020 at 1:27:57 PM UTC-6, Andy Blackburn wrote:
> >
> > TAS vs IAS does increase at about 2% per 1,000 ft
>
> Although looking at the data of the protocol and the glider specs, it's really closer to 1.4% per 1000 ft. Certainly 2% is easier to use and 1% is too little. The "10Kft, 2% rule" kind of splits the error. V is too high at 10K and too low at 40Kft. I have to say while it's not accurate enough for the table I will make for my glider, it's a good rule of thumb for on-the-fly estimates.

The difference between 1.4% and 2.0% is a dangerous place to play...

[Bret Hess]

The glider specs use 1.4%, so it's neither play nor dangerous.

[cdeerinck]

Funny, I was just writing some code to do this earlier today. I lifted this from somewhere on the web, and it is very accurate. You need to consider OAT to get an accurate TAS.

Quote:
I've used a formula that takes care of the PA / DA difference and introduces a temperature correction.
Was given to me a long time ago by an old retired navigator and it's surprisingly accurate.
Here it is :Â*TAS = IAS + 1 % per 600 ft +/- 1 % per 5°C diff with ISA
For instance : IAS = 97 kt ; OAT = 75 °F or 22° C ; FL 75.
1/- pressure correction : 7500 / 600 = 12.5 %
2/- Temp correctionÂ*
At 7500 ft, ISA = 0°C --> 22/5 = 4.4
Total correction 12.5 + 4.4 ~ 17 % or ~16 kt
Therefore, TAS = 97 + 16 = 113 kt

If OAT = -75 °F or - 60°C the temp corr is -60 / 5 = -12 %, both corrections annull each other and TAS = 97 kt

Just about everybody is right, you can't have a TAS unless you specify a DA, hence a temperature.
Unquote

Given:
IAS in Knots
MSL in Feet
OAT in Celcius

ISA = 15 + MSL/500
TAS = IAS + (IAS/60000) + (ISA/500)

I think Matt's point about 1.4% and 2% is very valid, and he meant to be conservative, and don't bank on that gap being safe. Given that it changes with OAT (albeit not by much) don't push VNE when up high.

[Bret Hess]

Chuck you're misunderstanding what I'm doing. I haven't neglected temperature. The German protocol has included the entire model atmosphere in their calculation of IAS does as altitude changes when TAS is held constant after 2000m. It includes T and P changes and hence air density changes. All I'm reporting is that the model gives linear data that has a 1.4 percent decline per kft in IAS, quite accurate up to 40,000 ft (it changes above that).. This line gives the data in the charts that the glider manufacturers use, so I'm not messing with the safety margin, I'm simply describing it. I added a third chart to https://docs.google.com/document/d/1t6kgxATcXP7nIhvwkP3hySKrUU5P_jrUFx8OA41uH0A/edit?usp=sharing to show that the 1.4% decline per kft is the right fit to the what the glider manufacturers use.

If your method is accurate, I'm guessing that if you fit a line to your results you'll get a 1.4 percent decline per kft in IAS if you hold TAS constant.

[Mike N.]

This is interesting for me as I am programming alerts into my S80 and ClearNav varios. In my Ventus Ct

For the purposes of reporting and alerting on air speeds, both the ClearNav and S80 have OAT probes near the pitot.

I am wondering of the TAS calculations in the vario(s) calculate TAS on both temperature and altitude? I assume as they are high end varios they do.

Also I am wondering if based on ballast, I can set the VNE alert in the vario(s) to shift appropriately based on ballast entered?

Yes I know, RTFM, but the questions seems relevant to the discussion at hand so if anyone can chip in on how STF varios calculate TAS it would be interesting.

Mike

[cdeerinck]

Mike - VNE should not change based on Wing Loading, so ballast should not affect it.
I'm pretty sure both of the units you measured calculate TAS.

My LX9070 actually has an Overspeed warning, and it is calibrated to go off based on TAS.
Don't ask me how I found this out.

[Mike N.]

So the S80 does have an alert for VNE.I assume at calculated TAS.

So I'm looking to put in the VNE alert with a safety factor of 85 to 90% of POH specified VNE

[Dan Marotta]

Flying a glider at or near Vne is pretty risky given that we fly most
often in turbulent air.Â* Besides, your rate of descent increases more
rapidly the faster you go.Â* Better to do better final glide planning.

My 2 cents...

On 6/10/2020 6:51 PM, Mike N. wrote:
> So the S80 does have an alert for VNE.I assume at calculated TAS.
>
> So I'm looking to put in the VNE alert with a safety factor of 85 to 90% of POH specified VNE

--
Dan, 5J

[Mike N.]

Agreed completely.
My primary reason for adding the VNE alert is when flying through areas of heavy sink.
Flying faster in sink is of course SOP in most cases.
So obviously as I fly faster through sink I will be seriously watching my airspeed, but having a secondary alert to my instrument scan while flying fast through sink, or even final glide, is an additional safety measure to help mitigate human errors.

[Bret Hess]

If you're interested in knowing more about the science behind the protocol, a very good paper is "Critical Flutter Speed of Sailplanes Calculated
for High Altitude: Examples of computation" by Wojciech Celcstyn, 1993, a featured article in Technical Soaring: https://journals.sfu.ca/ts/index.php/ts/article/view/216/200.

Fig 1 top half shows shaded flutter region for one oscillation mode and the dashed line how one would enter it flying fast enough IAS at 2000m (EAS and IAS are the same for glider speeds). If you flew at higher altitudes you would enter it at a lower speed, given by the parabolic looking curve going upward to the left, which gives the true altitude change of IAS-VNE flutter. The real change is NOT given by constant TAS, but the TAS line is shown for comparison. It's simply a convenient line that has enough slope so it will keep you out of danger in most cases.

The other figures show other modes. If you read the conclusion, it's clear that the standard (German and everyone else I guess) VNE protocol is not based on knowing where flutter is, either by testing or calculation. It's based on testing to a certain altitude and finding no flutter. Then IAS-VNE is chosen to drop off as TAS is held constant to stay away from unknown flutter at higher altitudes.

[Bret Hess]

I don't fly above VNE and don't plan to. But I hear that the most aggressive racing pilots do, often as their average speed! This makes some sense now because VNE is simply based on the testing altitude chosen, not knowledge of flutter (see my post just above this).

So you aggressive racing pilots, you are basically test pilots. Do you share the velocity and height space you've explored without flutter? Or do you just see how fast others with your same model fly? We could define new faster IAS-VNE plots for your model based on your experience.

[Steve Koerner]

On Tuesday, June 16, 2020 at 3:23:59 PM UTC-7, Bret Hess wrote:
> I don't fly above VNE and don't plan to. But I hear that the most aggressive racing pilots do, often as their average speed! This makes some sense now because VNE is simply based on the testing altitude chosen, not knowledge of flutter (see my post just above this).
>
> So you aggressive racing pilots, you are basically test pilots. Do you share the velocity and height space you've explored without flutter? Or do you just see how fast others with your same model fly? We could define new faster IAS-VNE plots for your model based on your experience.

Not so. Racing pilots have no competitive motive to fly above Vne. Decades ago there was a potential motive to do so when optical start gates were in use. That all ended when GPS came on the scene.