Flying below the glide slope as a tactic

[narchee]

I was talking with an airline pilot a few days ago. He flies international routes to Europe for a major airline. Something he mentioned was very interesting. He said if the ceilings are low and you want the best chance at seeing the runway environment, that you should fly the approach a quarter dot below the glide path because that brings you down to the numbers and you will see the ALS sooner. What are your opinions on that?

 

[frfly172]

The dots are there for a reason,when you cheat on an approach,you might get bitten.

 

[bflynn]

I get what he's saying - it's easier to pick out the rabbit in front of you rather than behind you. But that assumes you can't fly the glide slope down to the runway and you need to look outside and see the ALS. If you're that far off then you're asking for trouble.

Flying below the dots won't really do anything extra except to add risk. You won't see anything until you break out / close to break out. To get more visibility, you'd actually have to fly lower than the MDA, which I hope a professional pilot is not suggesting.

 

[Unit74]

I think what he is trying to do is get the MDA sooner and hold DA for longer in the hopes of acquiring the visual for a better chance of landing. As long as he is not below MDA between the FAF and MAP I'd say, although not industry standard, is legal and safe.

 

[Kritchlow]

I'm calling BS. First, if it's that low, it's likely a coupled approach. I know of no way to to set up an AP to do that.
Second, for 121 and 135 operations it is against the FARs, and would be an automatic bust on a check ride.
Third, I do NOT see where it would help you.
Fourth, it's stupid.

 

[Fearless Tower]

I'm calling BS. First, if it's that low, it's likely a coupled approach. I know of no way to to set up an AP to do that.
Second, for 121 and 135 operations it is against the FARs, and would be an automatic bust on a check ride.
Third, I do NOT see where it would help you.
Fourth, it's stupid.

Sounds like an old timer story from the days of 4 engine propliners and early jets.

 

[aterpster]

I was talking with an airline pilot a few days ago. He flies international routes to Europe for a major airline. Something he mentioned was very interesting. He said if the ceilings are low and you want the best chance at seeing the runway environment, that you should fly the approach a quarter dot below the glide path because that brings you down to the numbers and you will see the ALS sooner. What are your opinions on that?

Really, really bad advice!

It's a good way to get the landing gear of a large airplane taking out some approach lights.

 

[aterpster]

Sounds like an old timer story from the days of 4 engine propliners and early jets.

I flew F/O in those early jets. I never saw an old timer do that.

 

[azure]

Sounds risky to me. So you see the rabbit and descend further, still below the GS. Now the tolerances are even tighter. Are you 100% confident you can keep from going a half dot or more below? There could be close-in obstacles, what is the TERPS-allowed obstacle clearance? How accurate is your altimeter?

It also depends on whether that quarter dot is really just a quarter. If it is, I don't think you gain any advantage anyway. More than that and I think the issue I mentioned above applies. But I could be wrong.

 

[Kritchlow]

Sounds risky to me. So you see the rabbit and descend further, still below the GS. Now the tolerances are even tighter. Are you 100% confident you can keep from going a half dot or more below? There could be close-in obstacles, what is the TERPS-allowed obstacle clearance? How accurate is your altimeter?

It also depends on whether that quarter dot is really just a quarter. If it is, I don't think you gain any advantage anyway. More than that and I think the issue I mentioned above applies. But I could be wrong.

Not only do I not see an advantage at 1/4 dot, I don't see an advantage at full scale deflection as far as seeing the runway. The only advantage I can understand is if you wanted more pavement to land on.
There's just too much wrong with this scenario.

 

[Jaybird180]

What's the PTS tolerance on GS?

 

[Kritchlow]

What's the PTS tolerance on GS?

I'm betting for *intentional* flying below it would be zero.
That aside, the PTS is not really relkevant here. This is not a certificate check ride. These are part 121 operations.

 

[RussR]

What's the PTS tolerance on GS?

For ATP, 1/4 scale deflection is the tolerance.

But I agree with Kritchlow, this isn't a certification ride so PTS is not relevant (though perhaps a useful guide).

 

[azure]

Not only do I not see an advantage at 1/4 dot, I don't see an advantage at full scale deflection as far as seeing the runway. The only advantage I can understand is if you wanted more pavement to land on.
There's just too much wrong with this scenario.

I think the reasoning is the same as doing "dive and drive" on a non-precision approach, get low so you have more time to see and identify the runway environment. But I agree it's stupid to do that on an ILS or LPV unless you stay above the straight-in MDA for the co-located non-precision approach.

 

[Gucci Pilot]

I think what he is trying to do is get the MDA sooner and hold DA for longer in the hopes of acquiring the visual for a better chance of landing. As long as he is not below MDA between the FAF and MAP I'd say, although not industry standard, is legal and safe.

He's talking about a precision approach though. On an ILS it is expected that you dip below the DA when reaching it to execute a go-around because you are keeping a constant, stable descent. You aren't diving and driving on an ILS and especially in a large heavy turbine. If you put his glideslope profile next to an on glideslope profile, when he reaches the DA, he would be lower to the ground, and further from the runway than the on glideslope individual. Things I am thinking about with this is the margin of safety from obstacles is decreased, and my go-around obstacle clearance is decreased as well. I could see a owner/operator with get-home-it is doing this. Not a ATP professional pilot that gets paid by flight time.

 

[Kritchlow]

I think the reasoning is the same as doing "dive and drive" on a non-precision approach, get low so you have more time to see and identify the runway environment. But I agree it's stupid to do that on an ILS or LPV unless you stay above the straight-in MDA for the co-located non-precision approach.

Along the lines of what Gucci said, there is no benefit to "dive and drive" on an ILS.

 

[jesse]

He's talking about a precision approach though. On an ILS it is expected that you dip below the DA when reaching it to execute a go-around because you are keeping a constant, stable descent. You aren't diving and driving on an ILS and especially in a large heavy turbine. If you put his glideslope profile next to an on glideslope profile, when he reaches the DA, he would be lower to the ground, and further from the runway than the on glideslope individual. Things I am thinking about with this is the margin of safety from obstacles is decreased, and my go-around obstacle clearance is decreased as well. I could see a owner/operator with get-home-it is doing this. Not a ATP professional pilot that gets paid by flight time.

Right, makes no sense, just puts you further from the runway you're trying to find when you hit DA.

 

[Jaybird180]

For ATP, 1/4 scale deflection is the tolerance.

But I agree with Kritchlow, this isn't a certification ride so PTS is not relevant (though perhaps a useful guide).

So in the grand scheme of things, the OP was talking about a 1/4 dot. The FAA will bust you at 1/4 scale. Sure, you give up margin but does it have merit?

He's talking about a precision approach though. On an ILS it is expected that you dip below the DA when reaching it to execute a go-around because you are keeping a constant, stable descent. You aren't diving and driving on an ILS and especially in a large heavy turbine. If you put his glideslope profile next to an on glideslope profile, when he reaches the DA, he would be lower to the ground, and further from the runway than the on glideslope individual. Things I am thinking about with this is the margin of safety from obstacles is decreased, and my go-around obstacle clearance is decreased as well. I could see a owner/operator with get-home-it is doing this. Not a ATP professional pilot that gets paid by flight time.

This takes away the merit question, IMO.

 

[azure]

Along the lines of what Gucci said, there is no benefit to "dive and drive" on an ILS.

I agree, and said pretty much the same thing in my first reply. If you're doing dive and drive, you're really flying the LOC approach, and giving up all the advantages of an ILS. If you're below MDA *and* the GS, you're playing with fire - even in a piston single.

 

[narchee]

Sounds like an old timer story from the days of 4 engine propliners and early jets.

No he is currently flying for a major US carrier and has been for 25 years. Ex military. If the ATP standard is a 1/4 scale deflection then maybe in his mind this is legal because it is within the limits and he was adamant that it is in your favor to have the rabbit in front of you when you straining to see it rather than under you.

I spent an evening drinking and asking him questions. It was quite enlightening to me to say the least. But that will be for another thread. Maybe. Don't want to start a war.

 

[aterpster]

For ATP, 1/4 scale deflection is the tolerance.

But I agree with Kritchlow, this isn't a certification ride so PTS is not relevant (though perhaps a useful guide).

Usually a Part 121 certificate holder's standards are higher than the ATP PTS. Plus, most, if not all Ops Specs require use of the flight director or auto-flight when the weather is below something on the order of 400 and 3/4. I suppose you could fly a flight director incorrectly to get below the G/S but not an auto-coupler.

 

[aterpster]

No he is currently flying for a major US carrier and has been for 25 years. Ex military. If the ATP standard is a 1/4 scale deflection then maybe in his mind this is legal because it is within the limits and he was adamant that it is in your favor to have the rabbit in front of you when you straining to see it rather than under you.

I spent an evening drinking and asking him questions. It was quite enlightening to me to say the least. But that will be for another thread. Maybe. Don't want to start a war.

I can't imagine any U.S. air carrier condoning that practice. Also, Part 91 rules pertaining to Class B, C, and D airports require remaining on, or above the G/S to Decision Altitude, regardless of the weather. That applies to all jets, not just airline jets.

 

[Henning]

I'm having trouble picturing any value in this technique, especially when one considers the actual distance value of a quarter dot of deflection at the distance DA is from the glide slope antenna. Doesn't seem to me like it's going to make much difference. If you're going to cheat an ILS it makes more sense to have the needles centered and keep on coming.

 

[narchee]

If you're going to cheat an ILS it makes more sense to have the needles centered and keep on coming.

Well I think his point is though that if you keep the needles centered and keep on coming, you will come down to a point beyond the numbers and some distance down the runway. Whereas if you use this technique and keep on coming as you say the approach end with the more intense lighting will be what you are heading towards.

 

[Kritchlow]

I'm having trouble picturing any value in this technique, especially when one considers the actual distance value of a quarter dot of deflection at the distance DA is from the glide slope antenna. Doesn't seem to me like it's going to make much difference. If you're going to cheat an ILS it makes more sense to have the needles centered and keep on coming.

Correct. Absolutely ZERO advantage. Following the GS down will land you at 1,000 down the runway (assuming no float).
If you wanted to cheat, following the GS would indeed be the best method.

 

[jordane93]

I'm calling BS. First, if it's that low, it's likely a coupled approach. I know of no way to to set up an AP to do that.
Second, for 121 and 135 operations it is against the FARs, and would be an automatic bust on a check ride.
Third, I do NOT see where it would help you.
Fourth, it's stupid.

 

[Henning]

Well I think his point is though that if you keep the needles centered and keep on coming, you will come down to a point beyond the numbers and some distance down the runway. Whereas if you use this technique and keep on coming as you say the approach end with the more intense lighting will be what you are heading towards.

But it doesn't really, the arc difference of a quarter dot at that distance is very small.

 

[Kritchlow]

But it doesn't really, the arc difference of a quarter dot at that distance is very small.

Even at full scale all you would be doing is looking laterally at the lights instead of vertically. The distance would really be the same, and often the vis is worse laterally than vertically.

On a side note, I have yet to see an approach where I did NOT see the SFL (rabbit) if they were operational. Even in the sim.

 

[narchee]

Even at full scale all you would be doing is looking laterally at the lights instead of vertically.

I've never tried (and never will) but somehow he was arguing that in some cases the visibility can be so poor that by the time you would be able to see it, you're over it and your view is occluded by the plane being in the way. So you may have the same lateral distance versus vertical but you don't have the plane in the way now.

By the way I do not condone this in the slightest. I just found it very interesting to hear an ATP rated pilot with tons of hours flying to international destinations on a major carrier mention this as a possible technique. I don't know if he was being serious or just telling a good drinking story.

 

[Kritchlow]

I've never tried (and never will) but somehow he was arguing that in some cases the visibility can be so poor that by the time you would be able to see it, you're over it and your view is occluded by the plane being in the way. So you may have the same lateral distance versus vertical but you don't have the plane in the way now.

By the way I do not condone this in the slightest. I just found it very interesting to hear an ATP rated pilot with tons of hours flying to international destinations on a major carrier mention this as a possible technique. I don't know if he was being serious or just telling a good drinking story.

I have done approaches to true minimums, and the geometry of the lighting system is set so you can see it. If the airplane wS in the way it would make no sense.
I do realize you don't condone this..

 

[Henning]

I've never tried (and never will) but somehow he was arguing that in some cases the visibility can be so poor that by the time you would be able to see it, you're over it and your view is occluded by the plane being in the way. So you may have the same lateral distance versus vertical but you don't have the plane in the way now.

By the way I do not condone this in the slightest. I just found it very interesting to hear an ATP rated pilot with tons of hours flying to international destinations on a major carrier mention this as a possible technique. I don't know if he was being serious or just telling a good drinking story.

The thing is, this isn't accurate. At DH the lights are still in front of you unless maybe you are on a CAT II approach. Fly the glide slope in VFR and take a look, at 200' the approach lights are still ahead of you. The rabbit runs all the way to the threshold anyway.

 

[peerlesscowboy]

All it'd do is put you farther from the runway at DH.

 

[unsafervguy]

My flight manual says at or above. So 1/4 dot below would be unsat on any checking event.

Bob

 

[petrolero]

Right, makes no sense, just puts you further from the runway you're trying to find when you hit DA.

This.

DA is DA. If you're low, you'll reach DA farther from, not closer to, the runway. It would theoretically make slightly more sense to fly a quarter-dot high if the idea is to reach DA closer to the runway. The difference would be small. I don't know about others, but I can't maintain quarter-dot accuracy in crap wx anyway.

We are talking quarter dot and not quarter scale, right?

 

[Everskyward]

I've never tried (and never will) but somehow he was arguing that in some cases the visibility can be so poor that by the time you would be able to see it, you're over it and your view is occluded by the plane being in the way. So you may have the same lateral distance versus vertical but you don't have the plane in the way now.

By the way I do not condone this in the slightest. I just found it very interesting to hear an ATP rated pilot with tons of hours flying to international destinations on a major carrier mention this as a possible technique. I don't know if he was being serious or just telling a good drinking story.

I'm guessing this. Not only would he get to DA further away from the runway as others have mentioned, he would also be flying it raw data without the flight director since the flight director follows the GS even if the autopilot is not coupled.

 

[MauleSkinner]

I'm having trouble picturing any value in this technique, especially when one considers the actual distance value of a quarter dot of deflection at the distance DA is from the glide slope antenna. Doesn't seem to me like it's going to make much difference. If you're going to cheat an ILS it makes more sense to have the needles centered and keep on coming.

Let's see...the AIM says he glideslope is 1.4 degrees wide, top to bottom. 1/4 scale would be 1/8 of that, so let's say .2 degrees, rounded up.

If the antenna is 1200 feet from the approach end of the runway, and DH is reached at another 2400, that's 3600 feet.

The rule of 60 says that for every 60 feet, one degree is 1 foot. 3600 is 60x60, so one degree of deflection at that distance would be 60 feet. But we determined that 1/4 scale is .2 degrees, so .2x60=12 feet below glideslope centerline. Probably closer to 10 feet if you eliminate our rounding errors.

If he's talking 1/4 dot instead of 1/4 scale, that's only 5 feet below centerline. Either way, it doesn't seem useful to me.

 

[Henning]

Let's see...the AIM says he glideslope is 1.4 degrees wide, top to bottom. 1/4 scale would be 1/8 of that, so let's say .2 degrees, rounded up.

If the antenna is 1200 feet from the approach end of the runway, and DH is reached at another 2400, that's 3600 feet.

The rule of 60 says that for every 60 feet, one degree is 1 foot. 3600 is 60x60, so one degree of deflection at that distance would be 60 feet. But we determined that 1/4 scale is .2 degrees, so .2x60=12 feet below glideslope centerline. Probably closer to 10 feet if you eliminate our rounding errors.

If he's talking 1/4 dot instead of 1/4 scale, that's only 5 feet below centerline. Either way, it doesn't seem useful to me.

Thanks for doing the math, he was talking a quarter dot IIRC, and the topic being discussed is hitting DH further out so the rabbit is in front of you. Still makes no sense to me as a useful technique to get in at minimums.

 

[Kritchlow]

Let's see...the AIM says he glideslope is 1.4 degrees wide, top to bottom. 1/4 scale would be 1/8 of that, so let's say .2 degrees, rounded up.

If the antenna is 1200 feet from the approach end of the runway, and DH is reached at another 2400, that's 3600 feet.

The rule of 60 says that for every 60 feet, one degree is 1 foot. 3600 is 60x60, so one degree of deflection at that distance would be 60 feet. But we determined that 1/4 scale is .2 degrees, so .2x60=12 feet below glideslope centerline. Probably closer to 10 feet if you eliminate our rounding errors.

If he's talking 1/4 dot instead of 1/4 scale, that's only 5 feet below centerline. Either way, it doesn't seem useful to me.

Regardless, and I appreciate the nath, it doesn't matter.
At some point both pilots are looking for lights at 200 feet above TDZE. It's the same distance.

 

[Henning]

I think any way you look at it, this one really doesn't make sense as a technique to use. If you are SOL and need to get in, trim for 1.2 Vso and fly the needles to the ground.

 

[azure]

Even at full scale all you would be doing is looking laterally at the lights instead of vertically. The distance would really be the same, and often the vis is worse laterally than vertically.

This is a really good point - you are much likelier to see the rabbit below you, looking to the left of the nose or out the side window, than you are to see it in front of you just because the optical depth of the soup is so much less vertically.

And also, Henning is right, do the math and you'll see that at a quarter dot low you're only 3 feet or so low and maybe 60 feet further back, assuming a 3 degree glideslope.

After listening to all the arguments here I'm convinced that it's not just stupid because of the risk, but offers no operational advantage either, even if you went a dot or two low.

Edit: whoops, I see MauleSkinner already posted about the geometry of the glide slope.