steam v TV

Let'sgoflying! said:
an article
http://www.avweb.com/news/features/Steam-Gauges-Are-Safer-225682-1.html
Click to expand...

I knew it! In fact I never for one minute, despite all the balleyhoo in favor of it and the recitation of improving accident statistics, doubted it. I absolutely KNEW IT in my bones! So, why the hell don't the braniacs who design this stuff and sell this stuff and fiddle with this stuff know it too? The first step in learning how to use it should be how to turn it off.

dtuuri
 
Oh, you can bet the designers do know the limitations. Good ones at least.

Marketers, on the other hand, aren't going to tell you anything that might prevent a sale.

There is a whole cottage industry in "human factors" as it relates to aviation, though it's largely limited to airliners and various ATC facilities (at least, it is here).
 
MAKG1 said:
Oh, you can bet the designers do know the limitations. Good ones at least.
Click to expand...
I remember when RNAV first came out. Pilots wanted to input the nearest VOR's radial and distance to the outer marker. "Why would you want to do that?" they asked us. Radial and distance information wasn't available anywhere, even for airports in the early days. Unless they work in the cockpit they can't know what pilots find useful. The FAA went to the trouble of charting RNAV routes for us to use, like everybody would want to fly where the FAA wants instead of where their passengers want. Now, to make room for information, stuff you really don't need, they turn the round dials sidewise, like looking at the edge of a coin, and paint numbers on the edge. Instead of instant recognition, you have to stop and interpret the number, "Uh, is that number higher or lower than the one I should be targeting, uh, what was the target again?"

MAKG1 said:
There is a whole cottage industry in "human factors" as it relates to aviation, though it's largely limited to airliners and various ATC facilities (at least, it is here).
Click to expand...
Yep, they've failed too. Fatal accidents are twice as high in TAAs than steam. FITS was another BS program. Bet the "solution", though, is more of the same.

dtuuri
 
Last edited:
When you've used the stream gauges,for years,you feel even though you have glass,that the steam gauges be available for backup. The new glass seems to work best around an autopilot.
 
That article is poorly written and wouldn't pass muster in my wife's 5th grade class. What studies is he referring to and where is he getting his numbers? Since I was curious, I tracked down a NTSB study and found this interesting:

"Over the entire
period from 2002 through 2008, aircraft in the glass cockpit cohort showed a disproportionately
lower rate of total accidents per registered aircraft but a disproportionately higher rate of fatal
accidents per registered aircraft than those in the conventional cohort. "

Much like pilots with an IR rating; less accidents but the accidents that occur are more often fatal.

I think training is still an issue as well. When I did my G1000 checkout, I knew more than the CFI who was checking me out. That experience made me recognize that my training is my responsibility. Cirrus obviously recognized this and I wonder if the study was done today if the improved accident rate with those aircraft would skew the results at all.

Here's a link to the NTSB study:

http://www.ntsb.gov/safety/safety-studies/Documents/SS1001.pdf
 
I learned on steam gauges and finished my checkride PPL in a plane with an Aspen glass panel. Prefer the steam gauges except for night time flying the clarity of glass is a nice feature. Glass cockpit equipped planes cost 2x as much to buy so out of my budget.
 
uncreative said:
That article is poorly written and wouldn't pass muster in my wife's 5th grade class. What studies is he referring to and where is he getting his numbers? Since I was curious, I tracked down a NTSB study and found this interesting:

"Over the entire
period from 2002 through 2008, aircraft in the glass cockpit cohort showed a disproportionately
lower rate of total accidents per registered aircraft but a disproportionately higher rate of fatal
accidents per registered aircraft than those in the conventional cohort. "

Much like pilots with an IR rating; less accidents but the accidents that occur are more often fatal.

I think training is still an issue as well. When I did my G1000 checkout, I knew more than the CFI who was checking me out. That experience made me recognize that my training is my responsibility. Cirrus obviously recognized this and I wonder if the study was done today if the improved accident rate with those aircraft would skew the results at all.

Here's a link to the NTSB study:

http://www.ntsb.gov/safety/safety-studies/Documents/SS1001.pdf
Click to expand...

:confused: The tone of your post seems to take exception to the article, yet here's a few quotes from the study you reference which seem to validate the article:
"However, general aviation activity has also been decreasing. Normalizing the number of accidents by annual exposure data—in this case, the FAA’s annual general aviation flight hour estimates—results in a rate that more accurately represents safety risk. ... In contrast to annual accident totals, general aviation accident rates and fatal accident rates per 100,000 flight hours have remained relatively steady over the last decade."

"As shown in figure 9, the percentage of accidents resulting in fatality was about twice as high for the glass cockpit cohort as for the conventional cohort."​
So, no, safety has not been improved and, yes, steam is twice as safe.

dtuuri
 
Last edited:
This is exactly what I don't like about glass panels. I can glance at my ASI needle and tell it's right. More importantly, I can glance at the Altimeter needle to see if it's Straight Up/Down and the IVSI to see if it's horizontal. With glass, I have to look much longer to recognize the number and se if it's the right one or not.

Speaking of programming idiosyncrasies of non-pilots, why do so many of the VSI readings disappear? When IMC, I use that more than the altimeter to hold level. Guess they think I don't need it? Then at some non-user-selectable rate (often 50-100 fpm), it pops up to tell me that I'm having a problem. But 10fpm for several minutes will also create a problem, and the stupid thing won't tell me it's happening unless I happen to stare at and recognize that little number in the altimeter tape.

Another thing I hate about the tapes: there's a little window with your actual speed/altitude inside, but if changing the numbers are often half out of the window so you can't read them unless you continue to stare at it to see what the next readable number is, then a little longer to see which way it's moving (up,or down).

Nah, I'll keep my steam gages.
 
On the glass, it helps to use the bugs for everything. But that "best practice" is far from universal. It allows you to tell at a glance if you're where you need to be.

The main problem with glass is that there are many ways to do everything, which leads to interface complexity, and there are quite a number of non-obvious mode transitions. You have to know to check that the next mode is armed, and a few aren't shown (e.g., G1000 CDI transitions on ILS and VOR approaches).
 
Hank S said:
This is exactly what I don't like about glass panels. I can glance at my ASI needle and tell it's right. More importantly, I can glance at the Altimeter needle to see if it's Straight Up/Down and the IVSI to see if it's horizontal. With glass, I have to look much longer to recognize the number and se if it's the right one or not. ...
Click to expand...
Yup. There's a reason why most people wear digital watches but demand analog displays.

I'm sure the that "the tapes" came into being when display sizes were very limited and it was judged to be most important to have a large attitude indicator. The tapes got whatever room was left without anyone giving much thought to the human factors considerations.

Another factor is screen clutter. The programmers are determined to show you every single trick in their book as a default. When I climb into a G1000 airplane my first actions are to shut stuff off. No micro-map on the PFD, no wind vectors unless surface winds are an issue, etc.
 
I'll take a G1000 any day of the week and twice on Sunday hand flying in IMC. The large AI, the HSI and the RMI all in one compact area to scan are all plusses to me. I notice that the median age in that study was 47. I wonder if those of us who grew up playing flight sim on PCs that use EADIs with speed and altitude tapes have less of an adjustment?

My personal take away is that it is a training issue, not an issue with the equipment itself.
 
I started my IR training on steam gauges in an older model 172 and finished up in my current airplane, which is probably best classified as a TAA from the generation before glass - the primary flight display is a Sandel 3308, driven by a CNX-80 GPS. I've definitely had that experience of "suddenly realizing that my mental model is wrong" - once since getting my rating. I was flying an ILS and found the Sandel reverse sensing... it turned out the OBS had failed to auto-slew to the LOC course as I had expected. I've posted about this before, but the reason was that I had punched in my destination using direct-to instead of FPL, and the CNX-80 inserts approaches at a different point in the flight plan than usual in that case - AFTER the destination airport in fact. That was something I didn't know, because I had never received training from an expert on the avionics in my plane, and this little detail wasn't documented anywhere at the time (it's since been added to Keith Thomassen's book on the GPS). I caught the problem and manually turned the OBS to the front course, but it was a scary moment in my flying experience and could definitely have ended badly, with a pilot deviation or worse.

So I agree with both conclusions of the NTSB study - that TAAs are not necessarily safer, and that it's critical to safety that pilots know the ins and outs of their avionics thoroughly.
 
I'm a younger guy and have flown and instructed in both, frankly it's all the same to me, but I did initially learn on steam and I think the more basic you build your foundation on the better off you will be.

There is something to be said for having hands and gauges, it's a lot easier at a glance compared to digits

Also as someone who flys a steam plane IFR VFR I love my 3308, it's the best of both worlds IMO
 
Last edited:
I am a younger guy as well, mid twenties. I did all of my Training and IR with steam and paper. I know a couple of people that fly with g1000 and a bunch that use foreflight as a primary display (vfr). All of those planes drive me all the way up a wall. The amount of time they spend gathering information that is just flat unnecessary is mind boggling. Now I use a gtn-750 and it holds a lot of information but how much of it do I really need.

I think training for glass panels and glass in general needs to focus on only taking what you need and being able to get it quickly. That's great you can pull up every frequency and automatically load it but the amount of time it takes most people to fiddle too it someone could look at their notepad and tune it.
Because of the mass addition of information, I find that it is more of a distraction than it is help. But used in the right ways, the glass could be underwhelming and save a life.


Sent from my iPhone using Tapatalk
 
Most commonly heard sentence in a G1000 cockpit: "What's it doing now?"

Next most common/experienced pilots: "It's doing it again!"
 
write-stuff said:
A cry out into the darkness for better instruction.
Click to expand...
Well, I don't know. Even with the best and most thorough instruction, memories decay. People not flying the G1000 very frequently will begin to forget things. Also, in an airplane flown by more than one pilot, settings get changed.

I got into a G430 Archer one time and didn't catch that some previous pilot had changed the parameter layout on the Nav1 page. TRK was not in its usual place and I fly with TRK more than with the DG. It had been moved somewhere else on the screen. Why? I don't know.

I figured that out during a VFR climbout but had there been a low ceiling and hence the extra workload going into IMC I'm sure I would have been confused for a longer time. I doubt that I would have killed myself but I would not have been a happy camper.
 
Steam or glass works for me. I fail both regularly on my students. If my G1000 or Persepctive student is looking too much on the inside, the PFD goes off. If my steam student is fixating too much on the inside, my flight bag covers the instruments. A mix of glass and steam is good. I took my IR in a 172 with 2 G430W and even that was pretty easy. A good instructor will let you use both and know when to fail one or the other. For my long IFR X country we did 2 legs simulating a GPS failure so it was all VOR tracking and seeing if I could track a Victor airway without the GPS. Nothing wrong with glass as long as you know how to fly/navigate without it.
 
At 80 hours, I found steam easier to grok than glass. At 200 hours, I don't really notice the difference any more. Bug the heading, bug the altitude, FTFA and fuggedaboutit.
 
Been puffing on steam since 1975 and still here ...

I would like to have synvis and keep my steam however.
 
I am a guy who taught L/F Range to my students in a Link Trainer and the airplane.

Most of my career was steam gauge except for the B767 which was glass hybrid.

I no longer fly but I am still involved in instrument procedures and play around with Garmin's various trainers.

I am a huge fan of the RNAV system we have today. Having said that, I sense it is a system more for the benefit of the airlines than GA. at least light airplane GA.

All the knob twisting of a G-1000, for instance, is way too much heads down for a single pilot.

OTOH, the value of a G-1000 with a compatible autopilot in IMC is an invaluable improvement in the light airplane IFR platform. I almost died on an IMC night at KSBA because of poor avionics and a moment of inattention.

I vote for the TAA cockpit, but only for the TA pilot. :)
 
aterpster said:
I vote for the TAA cockpit, but only for the TA pilot. :)
Click to expand...
You are correct. The avionics can either be your best friend if you are proficient or your worst enemy if you aren't proficient. When I do Cirrus transition training the first thing I tell my student is that they will absolutely have no problem flying the airplane but the avionics will be the steepest learning curve. We usually do about 3 ground lessons with the GPU plugged in and twisting knows and pushing buttons.
 
What do you do if your airspeed is wrong? I know what I do, I look at my GPS (ground) speed. I get laughed at because "that doesnt include the wind". It is one heck of a lot better than nothing at all. I usually know if I have a tailwind or headwind, because I look at it all the time. I can also keep my wings level by just looking at the IFRGPS right/left line. There is a lot there that can be used in an emergency, but you have to train for it. Are there any books or articles advocating these backup techniques?
 
Glass is nice but most rental aircraft and planes that I could afford to buy are steam gauges.
 
Blueangel said:
Glass is nice but most rental aircraft and planes that I could afford to buy are steam gauges.
Click to expand...

Alas, as is the case with much of the U.S. the well-to-do get new TAAs. Thus, the slow death of light airplane GA.
 
coloradobluesky said:
What do you do if your airspeed is wrong? I know what I do, I look at my GPS (ground) speed. I get laughed at because "that doesnt include the wind". It is one heck of a lot better than nothing at all. I usually know if I have a tailwind or headwind, because I look at it all the time. I can also keep my wings level by just looking at the IFRGPS right/left line. There is a lot there that can be used in an emergency, but you have to train for it. Are there any books or articles advocating these backup techniques?
Click to expand...

You fly the numbers. That will get you much closer than ground speed under some fairly common scenarios.
 
uncreative said:
I'll take a G1000 any day of the week and twice on Sunday hand flying in IMC. The large AI, the HSI and the RMI all in one compact area to scan are all plusses to me. I notice that the median age in that study was 47. I wonder if those of us who grew up playing flight sim on PCs that use EADIs with speed and altitude tapes have less of an adjustment?

My personal take away is that it is a training issue, not an issue with the equipment itself.
Click to expand...

I think it's a "training issue" because the G-1000 has a poor interface. It's non-intuitive, clumsy, and "rate" values are presented particularly poorly.

Having flown steam and glass IMC, I think it's a wash, assuming the steam has an autopilot. . .
 
Opinions are like *******s. This article is about 98% sphincter and 2% fact.

Three of the four factual sentences (the third is the one that states when FITS began) that I could find in this entire piece were:

Pilots of TAA kill themselves more often than steam gauge aviators—almost twice the rate, according to the NTSB. Technology advances address many of the leading causes of GA fatalities: loss of control, controlled flight into terrain, fuel problems, midair collisions and weather.
Click to expand...
...pilots flying TAA have higher ratings and more experience. A majority are instrument rated.
Click to expand...
From that he derives a crap load of opinions, causes, and solutions.

But in order to conclude that "steam is twice as safe" you have to first assume that everything else is the same. This is a typical error people make when drawing conclusions from statistics.

In particular, TAA are on average much more powerful and faster than the average steam gauge plane is. I know there are glass LSAs but there are also steam piper cubs.

To tease out the real safety difference between TAA and non-TAA perhaps they could compare glass vs non-glass of the same aircraft type and with similar pilot populations?

So that is one possible cause, unrelated to avionics, that could explain some of the difference in fatality rates. He didn't address it at all. He launched into an avionics rant.

In addition to the cross-sectional problem there is also a time series problem with the thesis that "steam is twice as safe."

First, TAA training is relatively new and has had to evolve. Steam gauges went through that evolution half a century ago. Maybe compare TAA accident rates vs time to steam accident rates vs time. What about when "new" steam gizmos were introduced such as HSIs? Was there any difference in those vs conventional steam gauges?

Second, why focus solely on fatal accidents?

Third, accident probable causes? How many of these accidents even had avionics as a contributing factor. Engine failure on takeoff is not a TAA problem for example even if, in some aircraft, it's a chute-decision problem. Does the author discuss this? Nope. Rant on.

There very well may be a higher accident rate with TAA but this article fails to support the thesis. Focusing solely on fatals tells us little. Ignoring other possible variables that could explain some of the statistical differences doesn't teach us much.

So if you found yourself nodding your head while reading that piece it's probably because he was confirming your pre-existing biases.
 
Last edited:
Well, you're right that the article says nothing about how TAAs will evolve. And it's true that some older GPSs suck toads compared to modern ones (as anyone who has gotten lost in Apollo menus can attest).

But it does answer one important question: Right now, does TAA help prevent you from dying? Even though TAA sometimes includes airframe parachutes, the answer is no. And fatals are the right thing to look at. Fatals from all airborne causes, as that's the claim TAA makes. Such as a "nearest" button if the engine dies in IMC. It would indeed be very interesting to break down by specific cause, as that may shed some light on what the problem is. For instance, does information overload contribute? Misunderstood mode changes? Poor proficiency? How?

The answer may be different in 5 years. That's a different question. Except I see interfaces getting worse, not better, due to feature bloat. Like adding custom hold plotting to GTNs. All you really need is OBS mode.
 
Last edited:
petrolero said:
So if you found yourself nodding your head while reading that piece it's probably because he was confirming your pre-existing biases.
Click to expand...

Except that the underlying study was quite unbiased:
"Potential confounds related to aircraft age, equipment, and usage were controlled for to the extent possible in the present study by identifying groups of similar aircraft of similar age, with and without glass cockpits, and then gathering the information necessary to further identify any differences in use or user population."​
You can read it here: http://www.ntsb.gov/safety/safety-studies/Documents/SS1001.pdf

Maybe pictures would explain it better:
attachment.php


dtuuri
 

Attachments

  • Crazy.JPG
    Crazy.JPG
    73.2 KB · Views: 121
Last edited:
So would we need an STC to pull out the Garmin Perspective system out of our 22s and put in steam?
 
I think if you have a standard six pack and could put one Aspen Synvis with HSI in the center that would be the best of both worlds and not overkill.

You retain your steam ASI, VSI, Altimeter, and Turn coordinator.

Coarse, you need a WAAS reciever to drive it.
 
dtuuri said:
Except that the underlying study was quite unbiased:
"Potential confounds related to aircraft age, equipment, and usage were controlled for to the extent possible in the present study by identifying groups of similar aircraft of similar age, with and without glass cockpits, and then gathering the information necessary to further identify any differences in use or user population."​
You can read it here: http://www.ntsb.gov/safety/safety-studies/Documents/SS1001.pdf

Maybe pictures would explain it better:
attachment.php


dtuuri
Click to expand...

The problem is a single pilot who is not well trained combined with using this stuff without an auto-pilot.

AHRS is such an improvement over the crap gyros, and RNAV with all the goodies offers a major improvement in accuracy and situational awareness.

Beyond that, I don't have the answer. The airlines have two pilots, one of which is heads down far too often modifying inputs to a jazzy FMS alpha-numeric keypad.
 
I think the correct statement is that solid state gyros and extra redundancy should lead to a reduction in fatalities, yet it doesn't. Why?

I have my own suspicions rooted in excessive complexity and information overload, but it would be nice to have a real answer. Those are solveable. But not by hyping.
 
dtuuri said:
Did you see it? Watch her eyes--she never even once looked outside the cockpit. Heads down the whole time. :(



dtuuri
Click to expand...


I'm trying to find a web link for one of the recordings presented as a NATCA Award a number of years ago of a Cirrus that receives a clearance to descend in IMC to 2700, and immediately keys up with puffing/breathing noises - when those quit, the controller quickly gives a low altitude alert and a request to climb.

After he confirms the aircraft has stabilized at a safe altitude, he asks if everything is alright.

The response is chilling: "No, everything's not alright, I took it off the autopilot and it just went nuts..."

The controller realizes the pilot is totally rattled and says, "I'm going to climb you and see if I can get you on top, climb and maintain 9er-thousand..." Pilot starts the climb and acknowledged and then the controller moves mountains to get five airliners out of the way for the climb.

Call sign in the recording is "Cirrus 8DF", and pilot is a female voice. Accent of he controller indicates Boston or similar area back east with corresponding heavy traffic to move for a Cirrus to pop up to 9000, if anyone knows where NATCA has hidden their audio archives on the web...

I suspect the aircraft was building ice and the "it went nuts" was when she punched the AP off way out of trim, but it's hard to say.

The takeaway listening to it is that the pilot was fat dumb and happy on AP in a very well equipped cockpit right up until she hit the disconnect button and about got killed for inattentiveness to whatever was going on with the aircraft.

Controller's idea to climb her was likely a lifesaver. Pilot didn't ask for it. (I think if the "thing goes nuts" and you go below MEA recovering it, it's probably a really good idea to try to get the hell out of IMC after that... if at all possible... Especially if you don't know why it "went nuts".)

I'll see if I can find a link or upload it somewhere that can handle a PoA onslaught later.
 
By the way if someone knows where the archives got hidden, it's the same year as these other recordings...

166d586e4e8221dd46d90e6fff8b97ad.jpg
 
You must log in or register to reply here.
Back
Top