Wind Vector on HSI w/o OAT Probe

[jdubner]

I asked this question of Ron U. at the Arlington fly-in last week and thought I'd post it here to make it "official".

In the EFIS D10A HSI screen, the wind vector functionality is inoperative when no OAT probe is installed. (I have not verified this with the new 3.0 firmware yet). In this case, could we have it simply use the manually-entered OAT (that's used for the TAS calculation)?

Thanks,
Joe
Long-EZ 821RP
Lewiston, ID

 

[dynonsupport]

We've discussed this multiple times internally and decided against it. Winds are very sensitive to any errors in any data, and we just don't feel like we should present data that we aren't really sure of. It's not that we forgot to use the manually entered OAT value for winds, we did actively decide not to display winds without a real OAT.

 

[jdubner]

That's a good point.

But I do have a real OAT gauge -- just not a Dynon. See http://mail2600.com/tmp/Panel%206.jpg It's visible in the lower right. I decided against the Dynon OAT probe because this gauge was already installed (and paid for .

I highly value the Dynon EFIS TAS calculation using my manually entered OAT and would like to have the wind vector too. Please reconsider.

Thanks,
Joe
Long-EZ 821RP
Lewiston, ID

 

[khorton]

While I understand the desire to only use input data that is known to be valid, we must keep in mind that the hazard created by an erroneous displayed wing value is very small.  And, there are many other pieces of data required to calculate wind, each of which almost certainly has some error already, yet they are used with apparently no concern.  E.g.:

1. Heading may have errors.  Many users on this forum have heading errors.
2. CAS may have errors, due to pitot and static system leaks, and static source position error.  I am aware of some builders who have found that they had static source position errors equivalent to an error of 10 kt CAS.
3. OAT may have errors due to a poorly chosen location for the probe.  The OAT indication will likely have errors due to ram temperature rise (this would be approximately 4 deg C at 180 KTAS, depending on the recovery factor of the OAT probe).

I also desire that a manually entered OAT be a possible input for wind calculations, as I also have a non-Dynon OAT indication.

 

[JetJockey]

Count me in also. I've already got two OAT gauges in my Yak and I'm not going to add another one. If manually entered OAT is good enough for the D-10A to calculate TAS and DALT, why not a wind calculation? I would get the same erroneous winds on my E6B if I used the wrong OAT. :'(

 

[PilotKris]

We've discussed this multiple times internally and decided against it. Winds are very sensitive to any errors in any data, and we just don't feel like we should present data that we aren't really sure of. It's not that we forgot to use the manually entered OAT value for winds, we did actively decide not to display winds without a real OAT.

One very relevant input that you’re not using is the inclinometer. You’ll get a VERY inaccurate indication if you’re slipping or skidding.

Is there any way you can adjust for that? Or, do we just assume that the pilot is keeping the ball centered?

 

[PilotKris]

That's a good point.  

But I do have a real OAT gauge -- just not a Dynon.  See http://mail2600.com/tmp/Panel%206.jpg  It's visible in the lower right.  I decided against the Dynon OAT probe because this gauge was already installed (and paid for ...

Thanks,
Joe
Long-EZ 821RP
Lewiston, ID

Come on guys... it's 37 bucks...

I get all kinds of crap from you guys for not wanting to redesign my entire instrument panel and structure to install the new HS34 and you won't spend $37 to get the wind vector arrow... Stop Whining...

I'm 100% with Dynon on this.

 

[jdubner]

I get all kinds of crap from you guys for not wanting to redesign my entire instrument panel and structure to install the new HS34 and you won't spend $37 to get the wind vector arrow... Stop Whining...

OK, so lets see if I got this right. Dynon will sell me their (normally $65 priced) OAT probe for $37 (normally priced at $65) and ship it free (instead of charging their customary $10). Meanwhile, I'll remove my existing probe, wiring, and gauge to make room for Dynon's probe. When it comes, I'll install the probe in the same hole (hopefully), route the wiring to my EDC-D10A connector, and cover up the hole left in my instrument panel. When I'm done I can read OAT in tiny letters on the Dynon LCD instead of the large 7-segment display I've had since before Dynon invented the D10.

C'mon -- get real.

And Dynon, c'mon -- do the right thing here. Kevin Horton stated it much more elegantly than I ever could in his message (above). There are multiple sources of potential error in the wind calculations and the accuracy of the output isn't specified anyway. What's the big deal?

--
Joe
Long-EZ 821RP
Lewiston, ID

 

[PilotKris]

OK, so lets see if I got this right.  Dynon will sell me their (normally $65 priced) OAT probe for $37 (normally priced at $65)  and ship it free (instead of charging their customary $10).

C'mon -- get real.  

There are multiple sources of potential error in the wind calculations and the accuracy of the output isn't specified anyway.  What's the big deal?

--
Joe
Long-EZ 821RP
Lewiston, ID

I stand corrected. $37 is the price for the D180/D120 OAT probe. You have the D10A so it would be $65 (and there would have been no extra shipping if you ordered the probe at the same time as the D10A).

$10 for shipping is your penalty for not thinking ahead.

I guess for SOME people, $75 is a MUCH bigger problem than $37...

I thought the wind vector indication was valuable enough to install a $2,000 (certified) Air Data Computer in my Aztec to get it on my GNS530/430. I was more than happy to pay $37 to have it in my Sky Arrow (one of the best SAFETY features of the D180 in my opinion).

And yes, there are multiple places for potential errors to be introduced into the winds aloft calculation. Why would you want to intentionally add more?

 

[khorton]

And yes, there are multiple places for potential errors to be introduced into the winds aloft calculation. Why would you want to intentionally add more?

Actually, if you could enter the OAT manually, you could correct for the ram temperature rise.  Currently, there is no way to do that, as near as I can tell, as the system appears to use the OAT sensor value with no correction.  So a manually entered OAT could be more accurate than having it reading a directly connected one, unless Dynon chooses with a future software update to add correction for ram temperature rise.  If they go down this road, hopefully they will allow for a user entered probe recovery factor, with a suitable default value to use unless the user overrides it.

 

[PilotKris]

Well Kevin,

I'd never have enough time to enter the OAT value manually as the only possible way I'd ever get a high enough airspeed to worry about "ram temperature rise" is in a full power, pure vertical dive... after the wings have broken off!

Besides, without extensive wind tunnel testing, how the heck are you going to know what the "ram temperature rise" is going to be for a given probe location?

I'm no expert on mach-number aerodynamics but I agree with Dynon that I'd be far more likely to introduce errors by manually entering (and "correcting") OAT values.

Speaking of introduced data errors, do you compensate for different pencil widths when using your wizzwheel (E6b)?
Wait, I'm sorry, you probably do the Density alt., TAS, wind triangle computations in your head.

Let's see... the cosine of the angle of the dangle + the angle magnetic variation (or is it - the magnetic variation).... My head hurts already...  :-/

What I would do if I were as smart as you, I'd figure out what the "ram temperature rise" would be for the airspeed range that I most frequently flew and use that static correction in the setup that Dynon currently uses.

There are ALOT of other things I rather the Dynon people be working on than "dynamic ram temperature rise / probe recovery factor OAT correction".

But I do have to hand it to you Kevin. You, by far, have the most elaborate excuse for not spending $65 that I've ever heard.

 

[khorton]

I'd never have enough time to enter the OAT value manually as the only possible way I'd ever get a high enough airspeed to worry about "ram temperature rise" is in a full power, pure vertical dive... after the wings have broken off!

Many of these systems are being installed in RVs, which could have ram temperature rise of around 3 deg C at typical cruise speeds.  I agree that this size error would only have a small effect on the calculated TAS (and hence the calculated wind).  I simply raised it as one example of the many possible errors that are not being accounted for, as a counterpoint to Dynon's apparent concern about the possible errors associated with pilot-entered OAT.

Besides, without extensive wind tunnel testing, how the heck are you going to know what the "ram temperature rise" is going to be for a given probe location?

Actually, all you need to do is fly at max speed at constant altitude, wait until the indicated OAT stabilizes, then record indicated OAT, IAS and altitude.  Slow to minimum possible speed at the same altitude, let the indicated OAT stabilize, and record the data again.  Do this on a few different flights, so you have a bunch of data.  Take the data from each flight and plot indicated OAT vs (TAS squared)/7592 (this assumes OAT is in deg C, and TAS in kt).  The slope of the line from each flight is the probe's recovery factor.  Average the calculated recovery factors from several flights.

I'm no expert on mach-number aerodynamics but I agree with Dynon that I'd be far more likely to introduce errors by manually entering (and "correcting") OAT values.

The pilot would need to make a 30 deg C error in his entry to have the same effect as the already existing errors that some aircraft have on their static source.  Yet Dynon is apparently worried about the effect of OAT error, but not about providing a way to correct the much larger, and much more important possible static source error.  Go figure.

But I do have to hand it to you Kevin. You, by far, have the most elaborate excuse for not spending $65 that I've ever heard.

I'd purchase the OAT probe if I wanted the wind indication.  But it isn't important enough to me to justify running more wires from the wing to the instrument panel.

I am much more interested in eventually having a way to correct for static source position error, before the static pressure is used to calculate CAS and pressure altitude.  But, I recognize that not one in a hundred customers is interested in this feature, so it won't happen.

 

[PilotKris]

I did ask, didn't I?...

You're an Engineer, aren't you Kevin?...

Most customers not only don't care, they don't have the skills (mental and piloting) to do a proper calibration to quantify the static source position error. I suppose that there even if Dynon included a way to adjust for static errors, most users would introduce more errors than they would eliminate.

I'd re-evaluate the value of the wind vector. How much would you want that piece of information if your engine quits at 1500 AGL? A 15kt headwind vs. a 15kt tailwind on the emergency deadstick landing can mean the difference between an interesting story and an obituary...

 

[khorton]

I'd re-evaluate the value of the wind vector. How much would you want that piece of information if your engine quits at 1500 AGL? A 15kt headwind vs. a 15kt tailwind on the emergency deadstick landing can mean the difference between an interesting story and an obituary...

I'd be a bit careful about relying on the wind vector for this purpose.  While there is often a rough correlation between winds at 1500 ft and the winds at the surface, there can also be cases where this is not so.  It would be best to rely on traditional means to select a forced landing direction (i.e. knowledge of the surface winds based on visual clues, or based on info that you gleaned from your weather briefing).  Let's face it - the vast majority of light aircraft hours are flown on local flights, and every pilot should know which direction the local winds are coming from by the time he is airborne.

 

[PilotKris]

I'd be a bit careful about relying on the wind vector for this purpose.... It would be best to rely on traditional means to select a forced landing direction (i.e. knowledge of the surface winds based on visual clues, or based on info that you gleaned from your weather briefing).

Ah Kevin, you are an interesting contradiction...

You complain the modern equipment isn't as accurate as it could be yet will disregard it all in an emergency.

I often fly over the Baja dessert were weather reports are nonexistent and, unless there is someone kicking up dust, "traditional" methods of determining wind direction/speed are next to useless.

Having that extra piece of information instantly available can only increase my situational awareness hence safety. Since the wind vector is constantly calculated based on current observations, it's also faster/easier to identify wind sear and helps with mountain flying as well.

Is technology "better" or a replacement for good pilotage? Of course not! But it certainly adds to situational awareness if available (and used).

 

[dynonsupport]

How is this for a compromise:

The wind vector will only show when you also have the TAS/DA/OAT info item displayed (when you don't have a real OAT).

The issue we have is that the winds can be displayed with a totally wrong OAT when it's manually set since the OAT isn't on the screen anywhere. You could have set the OAT three months ago at 18K feet. By doing it so that you can only see the winds when the OAT is also on the screen, this is a lot less of an issue, since it's much harder to ignore a really wrong OAT when it's right there in front of you. This is the same philosophy we had when we allowed you to manually set the OAT to get TAS and DA.

Does this sound reasonable for those of you that want winds without an OAT?

 

[JetJockey]

Or how about just changing the color of the wind vector or putting an "E" (for estimated) in front of the winds, i.e. E270/14?

 

[khorton]

How is this for a compromise:

The wind vector will only show when you also have the TAS/DA/OAT info item displayed (when you don't have a real OAT).

This seems like a reasonable approach to me.  It provides the requested functionality, while making it obvious which OAT is being used.

Another approach would be for the system to note the altitude and time when the OAT is entered.  If the altitude subsequently changes by more than XXXX ft (1000 ft?), or YY minutes (60 minutes?) have elapsed, then the OAT value is considered stale, and the winds information is removed.  This would prevent seriously stale OAT from being used, but would help reduce the screen clutter a bit, by allowing the TAS/DA/OAT info item to be removed.

 

[jdubner]

I pretty much agree with Kevin Horton's thoughts in message #17 and especially like his alternate approach.

But I would simplify things a tiny bit by eliminating the time constraint: i.e. if OAT has been set while within a thousand feet of the current pressure altitude anytime since power was cycled, display the wind info. If not, display something like (SET OAT) in place of the wind info.

I realize this complicates the firmware task with all its attendant costs. If you simply wanted to display wind info irregardless (or based upon a softkey option) that would suit me too.

--
Joe
Long-EZ 821RP
Lewiston, ID