X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Fri, 12 Aug 2011 09:05:33 -0400 Message-ID: X-Original-Return-Path: Received: from imr-ma02.mx.aol.com ([64.12.206.40] verified) by logan.com (CommuniGate Pro SMTP 5.4.1) with ESMTP id 5089464 for lml@lancaironline.net; Thu, 11 Aug 2011 17:49:48 -0400 Received-SPF: pass receiver=logan.com; client-ip=64.12.206.40; envelope-from=Sky2high@aol.com Received: from mtaomg-db06.r1000.mx.aol.com (mtaomg-db06.r1000.mx.aol.com [172.29.51.204]) by imr-ma02.mx.aol.com (8.14.1/8.14.1) with ESMTP id p7BLn1M9008050 for ; Thu, 11 Aug 2011 17:49:01 -0400 Received: from core-mta001b.r1000.mail.aol.com (core-mta001.r1000.mail.aol.com [172.29.234.129]) by mtaomg-db06.r1000.mx.aol.com (OMAG/Core Interface) with ESMTP id 47A93E00009B for ; Thu, 11 Aug 2011 17:49:01 -0400 (EDT) From: Sky2high@aol.com X-Original-Message-ID: <201cf.5735d5ee.3b75a84c@aol.com> X-Original-Date: Thu, 11 Aug 2011 17:49:01 -0400 (EDT) Subject: Re: [LML] Re: Re-doing my panel - carefully thinking through failures X-Original-To: lml@lancaironline.net MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="part1_201cf.5735d5ee.3b75a84c_boundary" X-Mailer: AOL 9.6 sub 5004 X-AOL-IP: 24.1.9.48 X-Originating-IP: [24.1.9.48] x-aol-global-disposition: G X-AOL-SCOLL-SCORE: 0:2:437848768:93952408 X-AOL-SCOLL-URL_COUNT: 0 x-aol-sid: 3039ac1d33cc4e444e4d296a --part1_201cf.5735d5ee.3b75a84c_boundary Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Content-Language: en Bill, =20 Choose wisely indeed! =20 We are all a product of our experiences regardless of any scientific bias.= =20 Those of us that are much older seemed to use much of that experience to= =20 attain elder status, albeit a bit wrinkly and gnarly. =20 =20 Vacuum systems fail slowly? Well the FAA approved Skymaster I flew had a= =20 "pressure differential" AI (If I remember correctly, the pump was only on = =20 one engine) but, more to the point, the pressure differential gauge was dow= n=20 by my left knee - not exactly in the normal instrument scan. Thus, in my= =20 wee Lancair, the vacuum gauge was located immediately above the spinning= =20 mass AI so they both are seen as one instrument. So there......... If the= =20 needle ain't in the green, it's gonna be a mean lean. =20 Now, about those wonderfully designed and throughly tested electrical =20 devices - Take the STEC 50 with the electric turn coordinator, please. If = the=20 pitch control board senses sufficient rpm of the spinning tilted gyro, it= =20 puts out +10 VDC on a pin so that both the Oops flag on the TC is retracte= d=20 AND the AP system can become ready. The board was messed with by that=20 tingly electric cloud I mentioned in an earlier email in such a way that c= aused=20 the 10 VDC to always be present at the pin. "How did I find that out?" A= =20 curious mind might enquire. Wellll, some on-ground panel testing had the= =20 TC breaker pulled and it not reset (Brain MK I goof) for a subsequent test= =20 flight. Without the TC gyro spinning the AP was available in the "ready"= =20 state and, when engaged, resulted in a gentle declining left turn from bad= =20 gyro data. Hmmmmmmm. Resetting the breaker soon brought stabilization = as=20 the gyro actually spun up..=20 =20 I can hear it now, "So what?" Well, in spite of the sophisticated testing,= =20 be prepared for odd failures not supported by all the data gathered from= =20 instruments relying on different input and power sources. Do your own=20 integration. Use your own creative imaging. Don't completely rely on nut= t'n.=20 =20 Grayhawk =20 PS Oh, it is really good to look out the window every now and then. Traffic= =20 alerters notwithstanding.=20 =20 =20 In a message dated 8/11/2011 3:33:41 P.M. Central Daylight Time, =20 gt_phantom@hotmail.com writes: A point that has not been brought up by the proponents of spinning mass is= =20 the fact that their degradation modes are often far more insidious /=20 undetectable than power failure or EFIS failure. Whether powered by vacuum or by electrons, spinning gyros can become=20 "insidiously inaccurate" for a number of reasons. Off flags meant to warn= of=20 such problems fail. Gyros may retain some spin, but gently wobble (or be= =20 simply wrong). Stopped gyros still display some attitude, even if wrong. Each of these modes of failure are common, and have resulted in deaths. I will not even attempt to argue which is "more likely" - EFIS or gyro=20 failure - simply pointing out that both have failure modes which render th= em=20 inaccurate (or worse, appear to be accurate). As someone else pointed out, these days often the most likely piece of=20 equipment to become unreliable in flight is the Mk I brain. Thanks to your tax contributions, I received more comprehensive flight=20 training than most of you on this forum - and it instilled in me a strong= =20 sense of responsibility to know everything about every system on my plane,= =20 including possible degraded modes and insidious failure modes. Along with= =20 other situations I've mentioned, I've also had the displeasure of flying a= =20 Phantom home in the soup over the North Atlantic with only my turn and ban= k and=20 a ground controller calling my turns. Of all the arguments for a spinning= =20 gyro, I think I'd buy one for an electric T&B over an electric AI, because= =20 it is only attempting to tell you relative rate of rotation around your=20 vertical axis and will continue to work in degraded modes. As for spinnin= g=20 gyro AIs, I've simply seen too many of them telling people to "go the wron= g=20 way" to consider them "better." Not "worse," just not "better." If you want to be prepared for any possible situation in life, there is=20 nothing wrong with having one of every piece of equipment ever dreamed up = on=20 your airplane. Odds will definitely be better that way, with the dual=20 costs imposed both to your wallet and to your obligation as a responsible = pilot=20 to completely know, and to be able to recognize and respond appropriately= =20 to, all of the failure modes of every piece of equipment. If, however, yo= u=20 are willing to accept the 1:1,000,000 risk associated with "only" having 3= =20 or 4 working pieces of equipment in the aircraft, then my recommendation= =20 is simply: Choose wisely. Fly Safe! Bill On 01/-10/-28163 02:59 PM, Gary Casey wrote: =20 =20 I think there might be a difference of understanding regarding failure =20 modes. I don=E2=80=99t think anyone is proposing that we abandon all electr= onic =20 devices and go back to vacuum-powered gyros. What is being proposed by Bren= t (and=20 me, I suppose) is that a vacuum-power spinning gyro is perhaps the most =20 reliable backup system available. How can the =E2=80=9Cleast reliable=E2=80= =9D system be the =20 =E2=80=9Cmost reliable=E2=80=9D backup? Two reasons: First, the spinning gy= ro is not =20 susceptible to catastrophic electrical failures (like a lightning strike) = and=20 will keep running as long as the engine is turning. Second, it is logical = to=20 limit one=E2=80=99s thinking to =E2=80=9Csingle failure modes=E2=80=9D. In = other words the ONLY=20 requirement of the gyro system is that it keep working from the time of th= e=20 electrical failure to the earliest possible landing. What is the likelihoo= d=20 that the gyro will fail in those 15 minutes? What is the likelihood that= =20 the electrical system would quit and then the engine stop turning? All thi= s=20 is dependent on the vacuum system being =E2=80=9Crequired equipment=E2=80= =9D for IMC=20 operation (if the vacuum system fails on the ground, it is a no-go. If it = fails=20 in the air under IMC, it is a =E2=80=9Cland immediately=E2=80=9D condition= ). The poorest choice for a primary system then becomes the best choice as a= =20 backup. Logical, I think. Gary Casey =20 =20 =20 From: _GT Phantom_ (mailto:gt_phantom@hotmail.com) =20 Sent: Tuesday, August 09, 2011 5:55 AM To: _lml@lancaironline.net_ (mailto:lml@lancaironline.net) =20 Subject: Re: Re: Re-doing my panel - carefully thinking through failures Brent, I suppose I should have been more specific - I assume that pilots of =20 experimental aircraft will exercise sufficient judgment not be flying into = =20 thunderstorms, and thus the likelihood of in-flight lightning strike is ne= arly=20 nil. Not absolutely nil, naturally, but approaching or below the likelihoo= d=20 of vacuum failure, which is fairly common. Of course, perhaps I should not= =20 make such an assumption given that a very famous pilot died just last year= =20 flying his Bonanza into thunderstorms. Absent thunderstorms, we will simply have to agree to disagree. Vacuum=20 pumps and vacuum-operated artificial horizons are notoriously fallible, an= d a=20 poor vacuum can give insidious symptoms causing catastrophic results. Whil= e=20 there have been some experimental EFIS units (notably one you mention,=20 also the original pioneer Blue Mountain) which had early individual failur= e=20 rates much higher than vacuum equipment, still the likelihood of two or mo= re=20 going down simultaneously is rather rare. In my personal experience my TSO= =20 Garmins failed just as often as my experimental equipment - anecdotal=20 evidence, to be sure, but 3 TSO failures in 600 hrs not counting vacuum pu= mp=20 failure and attitude indicator partial failure ("lazy" attitude, "sort of"= =20 working). None of this absolves individuals contemplating use of experimental=20 equipment from the burden of research to draw their own conclusions about= =20 reliability. Your statement that TSO is required for legal flight is simply untrue. If= =20 you wish to dispute this, please feel free to point out the section of the= =20 FARs that you believe says otherwise (it does not exist, but knock yoursel= f=20 out). I don't expect to convince you of that; it seems that there are some= =20 folks who have made up their minds and aren't interested in anyone else's= =20 opinion. That's fine, you are entitled to yours. I, like many, have=20 reviewed the pertinent FARs along with (among others) my mechanic who was = a chief=20 avionics safety inspector for a major airline. For the type of flying for= =20 which Experimental aircraft are authorized there is no such rule stating= =20 that each piece of equipment must be certified to pass TSO. The altimeter= =20 must, or pass the test for equivalent accuracy (performed during the annua= l=20 pitot-static check). Doesn't mean it's a bad thing, only unnecessary for l= egal=20 flight. Just as you admonish people who are not engineers (I too was a=20 software engineer by trade) to form unwarranted opinions about avionics, y= ou=20 too should not consider yourself an expert on FARs simply because you buil= d=20 avionics. Glass houses, etc. Blocked pitot or static tubes are no longer a killer for correct attitude= =20 indication on any of the three leading experimental EFIS units (GRT, Dynon= ,=20 MGL). May also not be on others, haven't kept up. Of course, you will not= =20 get accurate airspeed with either blocked and will not get accurate=20 barometric altitude with static blocked, but that would happen irrespectiv= e of the=20 type of avionics you use. However, you will still have accurate horizon,= =20 and GPS can provide altitude and ground speed which, combined with a pilot= 's=20 knowledge of their aircraft power settings, etc. should enable you to fly= =20 safely to landing.=20 At the end of the day, you are putting your own life on the line. If you= =20 feel more comfortable with spinning gyros, by all means load up. However, = if=20 you feel you have done your research and would rather replace that vacuum= =20 pump for a second alternator to prevent power-out and ditch the gyro for a= =20 small self-contained backup EFIS, then your odds of total failure will=20 ultimately be about the same - just different causes. Fly safe! Bill On 01/-10/-28163 02:59 PM, Brent Regan wrote: =20 Bill speculates: <> The primary assumption here is false. It is not "given" that "two EFIS=20 units with battery backup are more reliable than a single vacuum pump".=20 Analysis and data show the opposite is true. =20 -- For archives and unsub http://mail.lancaironline.net:81/lists/lml/List.html --part1_201cf.5735d5ee.3b75a84c_boundary Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Content-Language: en
Bill,
 
Choose wisely indeed!
 
We are all a product of our experiences regardless of any scientific= =20 bias.  Those of us that are much older seemed to use much of that=20 experience to attain elder status, albeit a bit wrinkly and gnarly. &n= bsp;=20
 
Vacuum systems fail slowly?  Well the FAA approved Skymaster I fl= ew=20 had a "pressure differential" AI (If I remember correctly, the pump was onl= y on=20 one engine) but, more to the point, the pressure differential gauge was dow= n by=20 my left knee - not exactly in the normal instrument scan.  Thus, in my= wee=20 Lancair, the vacuum gauge was located immediately above the spinning mass A= I so=20 they both are seen as one instrument.  So there......... If the needle= =20 ain't in the  green, it's gonna be a mean lean.
 
Now, about those wonderfully designed and throughly tested electrical= =20 devices - Take the STEC 50 with the electric turn coordinator, please. = ; If=20 the pitch control board senses sufficient rpm of the spinning tilted g= yro,=20 it puts out +10 VDC on a pin so that both the Oops flag on the TC= is=20 retracted AND the AP system can become ready.  The board was messed wi= th by=20 that tingly electric cloud I mentioned in an earlier email in such a way th= at=20 caused the 10 VDC to always be present at the pin.  "How did I fi= nd=20 that out?"  A curious mind might enquire.  Wellll, some on-ground= =20 panel testing had the TC breaker pulled and it not reset (Brain M= K I=20 goof) for a subsequent test flight.  Without the TC=20 gyro spinning the AP was available in the "ready" state and,= =20 when engaged, resulted in a gentle declining left turn from bad gyro= =20 data.  Hmmmmmmm.   Resetting the breaker soon brought=20 stabilization as the gyro actually spun up.. 
 
I can hear it now, "So what?" Well, in spite of the=20 sophisticated testing, be prepared for odd failures not supported by all th= e=20 data gathered from instruments relying on different input and power=20 sources.  Do your own integration.  Use your own creative=20 imaging.  Don't completely rely on nutt'n. 
 
Grayhawk
 
PS Oh, it is really good to look out the window every now and then. Tr= affic=20 alerters notwithstanding. 
 
In a message dated 8/11/2011 3:33:41 P.M. Central Daylight Time,=20 gt_phantom@hotmail.com writes:
= A point=20 that has not been brought up by the proponents of spinning mass is the fa= ct=20 that their degradation modes are often far more insidious / undetectable = than=20 power failure or EFIS failure.

Whether powered by vacuum or by=20 electrons, spinning gyros can become "insidiously inaccurate" for a numbe= r of=20 reasons.  Off flags meant to warn of such problems fail.  Gyros= may=20 retain some spin, but gently wobble (or be simply wrong).  St= opped=20 gyros still display some attitude, even if wrong.

Each of t= hese=20 modes of failure are common, and have resulted in deaths.

I will n= ot=20 even attempt to argue which is "more likely" - EFIS or gyro failure - sim= ply=20 pointing out that both have failure modes which render them inaccurate (o= r=20 worse, appear to be accurate).

As someone else pointed out,= =20 these days often the most likely piece of equipment to become unreliable = in=20 flight is the Mk I brain.

Thanks to your tax contributions, I rece= ived=20 more comprehensive flight training than most of you on this forum - and i= t=20 instilled in me a strong sense of responsibility to know everything about= =20 every system on my plane, including possible degraded modes and insidious= =20 failure modes.  Along with other situations I've mentioned, I've als= o had=20 the displeasure of flying a Phantom home in the soup over the North Atlan= tic=20 with only my turn and bank and a ground controller calling my turns. = ; Of=20 all the arguments for a spinning gyro, I think I'd buy one for an electri= c=20 T&B over an electric AI, because it is only attempting to tell= you=20 relative rate of rotation around your vertical axis and will continue to = work=20 in degraded modes.  As for spinning gyro AIs, I've simply seen too m= any=20 of them telling people to "go the wrong way" to consider them "better."&n= bsp;=20 Not "worse," just not "better."

If you want to be prepared for any= =20 possible situation in life, there is nothing wrong with having one of eve= ry=20 piece of equipment ever dreamed up on your airplane.  Odds will=20 definitely be better that way, with the dual costs imposed both to your w= allet=20 and to your obligation as a responsible pilot to completely know, and to = be=20 able to recognize and respond appropriately to, all of the failure modes = of=20 every piece of equipment.  If, however, you are willing to accept th= e=20 1:1,000,000 risk associated with "only" having 3 or 4 working pieces of= =20 equipment in the aircraft, then my recommendation is simply: Choose=20 wisely.

Fly Safe!

Bill



On 01/-10/-28163 02:5= 9 PM,=20 Gary Casey wrote:=20
I think there might be a difference of understanding regarding fai= lure=20 modes. I don=E2=80=99t think anyone is proposing that we abandon all el= ectronic=20 devices and go back to vacuum-powered gyros. What is being proposed by = Brent=20 (and me, I suppose) is that a vacuum-power spinning gyro is perhaps the= most=20 reliable backup system available. How can the =E2=80=9Cleast reliable= =E2=80=9D system be the=20 =E2=80=9Cmost reliable=E2=80=9D backup? Two reasons: First, the spinnin= g gyro is not=20 susceptible to catastrophic electrical failures (like a lightning strik= e)=20 and will keep running as long as the engine is turning. Second, it is= =20 logical to limit one=E2=80=99s thinking to =E2=80=9Csingle failure mode= s=E2=80=9D. In other words=20 the ONLY requirement of the gyro system is that it keep working from th= e=20 time of the electrical failure to the earliest possible landing. What i= s the=20 likelihood that the gyro will fail in those 15 minutes? What is the=20 likelihood that the electrical system would quit and then the engine st= op=20 turning? All this is dependent on the vacuum system being =E2=80=9Crequ= ired=20 equipment=E2=80=9D for IMC operation (if the vacuum system fails on the= ground, it=20 is a no-go. If it fails in the air under IMC, it is a =E2=80=9Cland imm= ediately=E2=80=9D=20 condition).
The poorest choice for a primary system then becomes the best choi= ce as=20 a backup. Logical, I think.
Gary Casey
From: GT Phantom
Sent: Tuesday, August 09, 2011 5:55 AM
Subject: Re: Re: Re-doing my panel - carefully thinking thr= ough=20 failures
Brent,

I=20 suppose I should have been more specific - I assume that pilots of=20 experimental aircraft will exercise sufficient judgment not be flying i= nto=20 thunderstorms, and thus the likelihood of in-flight lightning strike is= =20 nearly nil. Not absolutely nil, naturally, but approaching or below the= =20 likelihood of vacuum failure, which is fairly common. Of course, perhap= s I=20 should not make such an assumption given that a very famous pilot died = just=20 last year flying his Bonanza into thunderstorms.

Absent=20 thunderstorms, we will simply have to agree to disagree. Vacuum pumps a= nd=20 vacuum-operated artificial horizons are notoriously fallible, and a poo= r=20 vacuum can give insidious symptoms causing catastrophic results. While = there=20 have been some experimental EFIS units (notably one you mention,= also=20 the original pioneer Blue Mountain) which had early individual failure = rates=20 much higher than vacuum equipment, still the likelihood of two or more = going=20 down simultaneously is rather rare. In my personal experience my TSO Ga= rmins=20 failed just as often as my experimental equipment - anecdotal evidence,= to=20 be sure, but 3 TSO failures in 600 hrs not counting vacuum pump failure= and=20 attitude indicator partial failure ("lazy" attitude, "sort of"=20 working).

None of this absolves individuals contemplating use of= =20 experimental equipment from the burden of research to draw their own=20 conclusions about reliability.

Your statement that TSO is requir= ed=20 for legal flight is simply untrue. If you wish to dispute this, please = feel=20 free to point out the section of the FARs that you believe says otherwi= se=20 (it does not exist, but knock yourself out). I don't expect to convince= you=20 of that; it seems that there are some folks who have made up their mind= s and=20 aren't interested in anyone else's opinion. That's fine, you are entitl= ed to=20 yours. I, like many, have reviewed the pertinent FARs along with (among= =20 others) my mechanic who was a chief avionics safety inspector for a maj= or=20 airline. For the type of flying for which Experimental aircraft are=20 authorized there is no such rule stating that each piece of equipment m= ust=20 be certified to pass TSO. The altimeter must, or pass the test for=20 equivalent accuracy (performed during the annual pitot-static check).= =20 Doesn't mean it's a bad thing, only unnecessary for legal flight. Just = as=20 you admonish people who are not engineers (I too was a software enginee= r by=20 trade) to form unwarranted opinions about avionics, you too should not= =20 consider yourself an expert on FARs simply because you build avionics. = Glass=20 houses, etc.

Blocked pitot or static tubes are no longer a kille= r for=20 correct attitude indication on any of the three leading experimental EF= IS=20 units (GRT, Dynon, MGL). May also not be on others, haven't kept up. Of= =20 course, you will not get accurate airspeed with either blocked and will= not=20 get accurate barometric altitude with static blocked, but that would ha= ppen=20 irrespective of the type of avionics you use. However, you will still h= ave=20 accurate horizon, and GPS can provide altitude and ground speed which,= =20 combined with a pilot's knowledge of their aircraft power settings, etc= .=20 should enable you to fly safely to landing.

At the end of the d= ay,=20 you are putting your own life on the line. If you feel more comfortable= with=20 spinning gyros, by all means load up. However, if you feel you have don= e=20 your research and would rather replace that vacuum pump for a second=20 alternator to prevent power-out and ditch the gyro for a small=20 self-contained backup EFIS, then your odds of total failure will ultima= tely=20 be about the same - just different causes.


Fly=20 safe!

Bill


On 01/-10/-28163 02:59 PM, Brent Regan wro= te:=20
B= ill=20 speculates:
<<Given that two EFIS units with battery backup = are=20 more reliable than a single vacuum pump, your argument that people=20 must have "TSO'd" equipment is logically ridiculous - especial= ly if=20 they also have as part of their panel an independent 2-axis=20 autopilot.>>

The primary assumption here is false. It is= not=20 "given" that "two EFIS units with battery backup are more reliable th= an a=20 single vacuum pump". Analysis and data show the opposite is true.=20 


--

For archives and unsub http://mail.lancaironline.net:81/lists/lml/List.html=
--part1_201cf.5735d5ee.3b75a84c_boundary--