Decision Height

[RotorAndWing]

You are correct, I am not instrument rated. Although I still maintain that if a target tolerance is -0/+100, wisdom advises one to make their actual target +50, ±50. I'm enlightened to learn that -0/+100 applies to non-precision approach MDA, not precision approach decision height.

 

[Kritchlow]

There is a TERPs thing called the Glideslope Qualification Surface ("GQS") is the reason many airports have LP rather than LPV.

No doubt... But that still doesn't change what I said.

 

[aterpster]

No doubt... But that still doesn't change what I said.

Well, you did throw down the gauntlet.

 

[DJTorrente]

We all can't be as infallible as you, doctor.

 

[Palmpilot]

"Doctor"?

 

[Everskyward]

As to the 50' trees, it was a joke my CFI told (and likely many others) that takeoff and landing are calculated over a 50' obstacle because the FAA mandates all trees be 50' tall. I was also thinking of this incident at Saratoga Springs where the trees had encroached on the glideslope. Admittedly uncommon.

And Dr. House/Rotor/Wing, you play your chosen part to perfection. You are correct, I am not instrument rated. Although I still maintain that if a target tolerance is -0/+100, wisdom advises one to make their actual target +50, ±50. I'm enlightened to learn that -0/+100 applies to non-precision approach MDA, not precision approach decision height.

You, as an individual, can make your decision at DA +50', your choice. But the approach was designed so that the decision is made at DA. As far as the article you quoted, it is somewhat irrelevant since the pilots had the runway in sight before the DA. The hit the trees because they didn't see them as it was night and the trees had grown into the approach path. I guess I'm surprised no one had noticed this before, or if they had noticed they hadn't reported it.

 

[dmspilot]

If you weren't allowed to go below DA, period, then why would it need a different name than MDA...you would just call them all MDAs.

 

[Palmpilot]

In support of the above, we have 14 CFR 1.1:

Decision altitude (DA) is a specified altitude in an instrument approach procedure at which the pilot must decide whether to initiate an immediate missed approach if the pilot does not see the required visual reference, or to continue the approach. Decision altitude is expressed in feet above mean sea level.

Decision height (DH) is a specified height above the ground in an instrument approach procedure at which the pilot must decide whether to initiate an immediate missed approach if the pilot does not see the required visual reference, or to continue the approach. Decision height is expressed in feet above ground level.

Minimum descent altitude (MDA) is the lowest altitude specified in an instrument approach procedure, expressed in feet above mean sea level, to which descent is authorized on final approach or during circle-to-land maneuvering until the pilot sees the required visual references for the heliport or runway of intended landing.​

 

[Gucci Pilot]

I do believe a constant glide path is fairly new, and some of the verbiage may not reflect that. I believe many aircraft today are still not equipped for constant glide path.

Any plane can do a constant glidepath on a non-precision approach. It just requires a little more math from the pilot.

 

[RotorAndWing]

"Doctor"?

Dammit, now everyone knows I'm a proctologist!

 

[Kritchlow]

Any plane can do a constant glidepath on a non-precision approach. It just requires a little more math from the pilot.

That's true, although not as precise and likely not practiced all the time. It is only in recent years that the dive & drive has been frowned upon.

 

[joemurffy]

"Doctor"?

Dr. Simon Tam reference, I presume.

BTW, great topic. There was very little Internet discussion back in 2006 when I was working on my IR about approaches with DAs and the natural decent lower than them when initiating one's missed approach. My CFII did think it was a hard minimum, and I actually wrote to Rod Machado for his opinion. He wrote back, and was of course of the same mindset as almost everybody in this thread. (Not that anybody here needs his validation, mind you)

 

[aterpster]

Just a bit of TERPs to demonstrate that criteria account for height loss during the transition from the glideslope to the missed approach segment, beginning at DA. This is the ILS case. (LNAV/VNAV is different but nonetheless height loss is accounted for in that case, too):

 

[Palmpilot]

Dr. Simon Tam reference, I presume.

Never mind - I just figured it out.

 

[Witmo]

Just a bit of TERPs to demonstrate that criteria account for height loss during the transition from the glideslope to the missed approach segment, beginning at DA. This is the ILS case. (LNAV/VNAV is different but nonetheless height loss is accounted for in that case, too):

Don't know about anyone else, but the above, "explanation," gave me a headache

 

[Everskyward]

Don't know about anyone else, but the above, "explanation," gave me a headache

I think labeling is missing from the diagram... or something. But, for the purpose of this thread, it's probably the highlighted passages which make the point.

 

[aterpster]

Don't know about anyone else, but the above, "explanation," gave me a headache

All I intended was to show that the design criteria provides for a presumed brief continued descent as the descent on the glideslope is discontinued at DA. I'm not trying to teach TERPs school.

 

[Sluggo63]

We do something similar at work, for non-precision approaches.

If we're shooting a non-precision approach, we fly it using constant descent procedures. We then have to calculate a "DDA" (Derived Decision Altitude). We add 50 feet to the MDA and use that as our new "DA". That ensures that if we go missed on a non-precision approach using VNAV, our inertia will not carry us below the actual MDA.

The one exception we have (in the US only) is if we are shooting a localizer approach to a runway that is served by an ILS (say the G/S portion was OTS), then we are allowed to use the MDA like a real DA, without adding the 50'. The rationale being that the approach has been TERPS'd for the ILS, so going below the MDA on the VNAV glidepath shouldn't have any obstacle clearance issues.

 

[Turningfinal]

The one exception we have (in the US only) is if we are shooting a localizer approach to a runway that is served by an ILS (say the G/S portion was OTS), then we are allowed to use the MDA like a real DA, without adding the 50'. The rationale being that the approach has been TERPS'd for the ILS, so going below the MDA on the VNAV glidepath shouldn't have any obstacle clearance issues.

What if the G/S was OTS, or the approach NOTAM'd LOC only, precisely because of an obstacle penetrating the ILS approach surfaces?!? While you're likely safe a few feet below the MDA, you're still violating FARs based on what I've read above.

Here around O'Hare, there's quite a few occasions where we've had to limit approaches to LOC or visual only due to construction cranes on and around the airfield.

 

[flyingron]

The one exception we have (in the US only) is if we are shooting a localizer approach to a runway that is served by an ILS (say the G/S portion was OTS), then we are allowed to use the MDA like a real DA, without adding the 50'. The rationale being that the approach has been TERPS'd for the ILS, so going below the MDA on the VNAV glidepath shouldn't have any obstacle clearance issues.

Hold the phone! Do you believe this is legal or safe?

The VNAV GS (starting at the FAF minimum altitude and crossing the MAP and the MDA) isn't necessarily the same as the ILS GS (in fact it's often steeper). You blow through the MDA you're not guaranteed not to hit things.

 

[unsafervguy]

We do something similar at work, for non-precision approaches.

If we're shooting a non-precision approach, we fly it using constant descent procedures. We then have to calculate a "DDA" (Derived Decision Altitude). We add 50 feet to the MDA and use that as our new "DA". That ensures that if we go missed on a non-precision approach using VNAV, our inertia will not carry us below the actual MDA.

The one exception we have (in the US only) is if we are shooting a localizer approach to a runway that is served by an ILS (say the G/S portion was OTS), then we are allowed to use the MDA like a real DA, without adding the 50'. The rationale being that the approach has been TERPS'd for the ILS, so going below the MDA on the VNAV glidepath shouldn't have any obstacle clearance issues.

and your POI signed off on that?

bob

 

[Sluggo63]

What if the G/S was OTS, or the approach NOTAM'd LOC only, precisely because of an obstacle penetrating the ILS approach surfaces?!?

Then there would be a NOTAM saying that using MDA in lieu of DA(H) is not allowed. Or we'd have a FCIF telling us that we would have to use DDA procedures for that particular approach.

While you're likely safe a few feet below the MDA, you're still violating FARs based on what I've read above.

See below...

Hold the phone! Do you believe this is legal or safe?

I believe it is both legal and safe

and your POI signed off on that?

bob

Yes. See FAA OpSpec C073. It spells it all out.
OpSpec C073--Vertical Navigation (VNAV) Instrument Approach Procedure (IAP) Using Minimum Descent Altitude (MDA) as a Decision Altitude DA)/Decision Height (DH)

 

[unsafervguy]

Then there would be a NOTAM saying that using MDA in lieu of DA(H) is not allowed. Or we'd have a FCIF telling us that we would have to use DDA procedures for that particular approach.
See below...

I believe it is both legal and safe

Yes. See FAA OpSpec C073. It spells it all out.
OpSpec C073--Vertical Navigation (VNAV) Instrument Approach Procedure (IAP) Using Minimum Descent Altitude (MDA) as a Decision Altitude DA)/Decision Height (DH)

i didn't see that your certified for vnav mins. we use constant decent procedures but are not certified to VNAV min, so there is no way our POI would even think about it.

bob

 

[Sluggo63]

i didn't see that your certified for vnav mins. we use constant decent procedures but are not certified to VNAV min, so there is no way our POI would even think about it.

bob

Yeah. Sorry, I left that out of the original post. We're full up VNAV. No RNP, though. I'm sure the company had to jump thorough the hoops to get using this OpSpec, but it was invisible to us. Only our MD-11's aren't able to do this because of the way their VNAV (PROF) works.

 

[flyingron]

Then there would be a NOTAM saying that using MDA in lieu of DA(H) is not allowed. Or we'd have a FCIF telling us that we would have to use DDA procedures for that particular approach.
See below...

I believe it is both legal and safe

Yes. See FAA OpSpec C073. It spells it all out.
OpSpec C073--Vertical Navigation (VNAV) Instrument Approach Procedure (IAP) Using Minimum Descent Altitude (MDA) as a Decision Altitude DA)/Decision Height (DH)

It doesn't say you can always use the MDA as as DH on ILS's just because you feel it's safe (which is what you said). In fact, it specifically says you need to verify each approach fo the exact issue I mentioned (that you decided to remove from your quote from me) where the glide slopes are different plus a bunch of other requirements.

 

[dmspilot]

For those using a continuous descent for nonprecision approaches, if the MAP is at the airport and there is no VDP, how are you calculating the descent rate/angle?

Is anyone using the continuous descent technique in simple single engine equipment?

 

[Gucci Pilot]

For those using a continuous descent for nonprecision approaches, if the MAP is at the airport and there is no VDP, how are you calculating the descent rate/angle?

Is anyone using the continuous descent technique in simple single engine equipment?

Calculate my own VDP.
VDP(nm from runway) = Height above touchdown / gradient. Most are 3 degree gradients so I use 300 unless it is a different angle.
Decent gradient(ft/nm) = altitude to lose / distance

Multiply your descent gradient by your miles per minute to get a Vertical Speed.

I do it in the small stuff too.

 

[unsafervguy]

For those using a continuous descent for nonprecision approaches, if the MAP is at the airport and there is no VDP, how are you calculating the descent rate/angle?

Is anyone using the continuous descent technique in simple single engine equipment?

The fms does it.

Bob

 

[dmspilot]

Calculate my own VDP.

Makes sense, thanks.

 

[denverpilot]

For the OP let's add a couple of thoughts to add to this for him to ponder.

Keep in mind the words have specific meaning to a pro. And we private schlubs may not be thinking this way.

"A decision to CONTINUE the APPROACH"... Emphasis added.

Here's why...

"The approach light system, except that the pilot may not descend below 100 feet above the touchdown zone elevation using the approach lights as a reference unless the red terminating bars or the red side row bars are also distinctly visible and identifiable."

You can descend below DA and still have to go missed with only the approach lights visible but you have to have the red terminator bars or the side row bars visible to go below 100' AGL above the touchdown zone elevation.

Note, it does not say the runway itself is visible at DA. Nor that you can legally land out of it. Just that you can continue down below DA to 100' AGL above the number printed next to TDZE on your plate.

Is this WAY below most folk's personal minimums if they can possibly avoid it?

Absolutely.

But... I got a rare chance to do it, in actual IMC, with Jesse sitting right next to me, and a solid plan to go elsewhere if it didn't work.

Ground fog at KLNK during one of my training flights, and severe clear and VFR only a few hundred feet above us, for a quarter or more, of the State of Nebraska. Twinkling stars, lights of Omaha clearly visible when on top of it.

The decision at DA was to CONTINUE the APPROACH. Not to land.

Now why the emphasis on APPROACH? That's what killed the FedEx crew. They switched to "VMC brain mode" because they saw the runway and drove toward it.

The APPROACH still had minimums that needed to be heeded and understood to remain clear of obstacles all the way to the runway. They got bit.

So there were actually two approaches for us.

One was a solid missed approach.

We assessed and Jesse asked my opinion on a second attempt. (Checking my decision making skills, and comfort level. This is the real deal, dummy... Make the call!)

The controller asked at about the same time.

I said "standby" and said out loud... "All the accident reports in this stuff seem to always have the story that the pilot tried the approach numerous times and then either screwed it up or ran out of gas. We have the fuel to fly a long time, we have an 'out' nearby, so I'll take ONE more shot at this and then we're headed to Omaha."

Jesse smiled and said, "Ok."

(Meanwhile a big hint that had been going on for a while that I would have picked up on now, but was way too behind things THEN, was the controller rattling off changing RVR numbers for both our knowledge and the inbound commercial IFR operators that night. They know those guys can't even start the approach without a high enough RVR, so he was announcing them as it changed to no one in particular. Note: If you hear a controller doing this, you're probably in for a very low approach... Heh.)

Second approach there were two callouts. Right before DA Jesse reminded me of the above rule. (Just about froze my brain solid right then, too. Haha.)

At DA, we saw the rabbit, and the approach lights (as they're passing below us basically) but not the red stop bar or side bars. The runway was still obscured in the fog. (By the way, this was at night.)

I announced "Continue" with the full intention that a go-around was VERY imminent and with a fast scan including the ILS staying centered, and the ability to see the altimeter winding down to that magic 100' AGL number above the TZDE.

While taking quick glances outside, caught the rest of the runway environment appearing out of the fog.

Second callout, "Landing". As in, "we just met the criteria to go below that 100' AGL number".

At this point I could now transition to being visual outside 100%. If the fog bank had changed the visibility again, I was still hair triggered to perform an instant missed approach.

So... Yes. Below DA can happen in real world weather, and you have to know the terms have a meaning to the pros.

"Continue" and "Approach" aren't "fly to the runway". And they're not a commitment to land.

Now. In the training environment I've NEVER heard of a DPE attempting to test this on a Private ride.

You're coming down the approach with a view limiting device and they say "you have only the rabbit and the approach lights, you do not see anything else" and then before 100' above the TDZE they say "you see the red stop bar", and then "you see the runway". They could test it, but it's going to be fruitless because at the "approach lights" notification you'd have to lift the device and add the window to your scan, and you'd see the runway anyway.

So... All that said. DA+80 is dumb. Someday you're going to need to know you can "continue" even below DA to 100' above TDZE. It's still just an approach at this point, and you can still need to go missed.

 

[Sluggo63]

It doesn't say you can always use the MDA as as DH on ILS's just because you feel it's safe (which is what you said).

Ok... I'm sensing that you like to argue about this stuff (or maybe you just don't like me), I don't know. I'm not doing it because I feel it's safe. This isn't a Sluggo63 personal technique I use at my work. I'm not the airline guy flying 1/4 dot low to "get in" or squawking 7600 to fly whatever oceanic routing I want. This is an established procedure codified in OpSpec C073, our company's FOM and our PHB. We're not trying to "get one over" on the FAA. This has been all approved at many different levels, and I'm sure has been looked at intensively. I'm sorry if I'm going to trust my company's flight operations engineers and the FAA's TERPS guys over all the experts on the POA website.

In fact, it specifically says you need to verify each approach fo the exact issue I mentioned (that you decided to remove from your quote from me) where the glide slopes are different plus a bunch of other requirements.

Yes, it does. Here is the quote "I decided to remove" from your previous post:

The VNAV GS (starting at the FAF minimum altitude and crossing the MAP and the MDA) isn't necessarily the same as the ILS GS (in fact it's often steeper). You blow through the MDA you're not guaranteed not to hit things.

Right. I didn't want to get into the nuts and bolts of how we do this whole thing. We have about a half-dozen pages in our PHB describing how we can fly various non-precision approaches. I didn't think all that in-the-weeds stuff was relative (or interesting, for that matter) to the audience here. But, yes... in order to fly a ILS G/S OTS approach using MDA as DA(H), we have to select the approach from the FMS database, and confirm that the inbound course matches the inbound course on the plate, and that the GP that gets pulled from the database matches the GP on the ILS. There are a couple other caveats dealing with step-downs, extreme hot and cold temperatures, etc., but in general, domestically, we can fly a ILS, GP OTS as a localizer NPA using the MDA as a DA(H) using our VNAV system. I'd be interested to know, if you can find me an approach where you think this procedure would be dangerous, which approach it is, and I'll look on our Jepp plates and see if we have a note prohibiting this OpSpec for that approach.

I hope this clarifies it for you and others.

Truce?

P.S. Actually, if ATERPSTER is still reading this thread, I'd like to hear his thoughts on this OpSpec and what protections we may or may not have using an MDA as DA(H).

 

[Sluggo63]

For those using a continuous descent for nonprecision approaches, if the MAP is at the airport and there is no VDP, how are you calculating the descent rate/angle?

Is anyone using the continuous descent technique in simple single engine equipment?

Like Gucci said, I calculate my own. Just one word of caution, though. One of the reasons there may not be a VDP published is that there is an obstacle that penetrates what would be a normal glidepath.

I calculate a VDP and try to fly CDA in GA also.

 

[Sluggo63]

Now why the emphasis on APPROACH? That's what killed the FedEx crew. They switched to "VMC brain mode" because they saw the runway and drove toward it.

Good post. Great stuff in there. I think you might be referencing the UPS crash in Birmingham. That accident report is a good read for anyone who is instrument rated and it deals with lots of human factors/automation issues.

 

[denverpilot]

Good post. Great stuff in there. I think you might be referencing the UPS crash in Birmingham. That accident report is a good read for anyone who is instrument rated and it deals with lots of human factors/automation issues.

It's is. Definitely.

 

[kayoh190]

If we're shooting a non-precision approach, we fly it using constant descent procedures. We then have to calculate a "DDA" (Derived Decision Altitude). We add 50 feet to the MDA and use that as our new "DA". That ensures that if we go missed on a non-precision approach using VNAV, our inertia will not carry us below the actual MDA.

We do the same thing, however the procedure changed a few months back, and we no longer need to add the 50 feet if we're flying to a runway served by a PAPI or VASI. I just did this the other day on the VOR approach to 4 in LGA. I guess the MDA isn't as sacred as I was always led to believe.

 

[kayoh190]

P.S. Actually, if ATERPSTER is still reading this thread, I'd like to hear his thoughts on this OpSpec and what protections we may or may not have using an MDA as DA(H).

I'd be curious as well. It sounds like my company recently took it a step further than yours, and back at recurrent we were scratching our heads about it. I can't think of the last time I've taken a non-precision approach to mins, but the idea that I can treat the MDA as a DA just because there's a PAPI or VASI doesn't pass the smell test for a lot of us.

EDIT: I just read the OpSpec you posed, and it's right there. You only need to meet *one* of those requirements, and item (3) covers the VASI and PAPI. Interesting! Learned something new.

 

[MauleSkinner]

We do the same thing, however the procedure changed a few months back, and we no longer need to add the 50 feet if we're flying to a runway served by a PAPI or VASI. I just did this the other day on the VOR approach to 4 in LGA. I guess the MDA isn't as sacred as I was always led to believe.

We're you led to believe that you couldnt leave MDA safely when you were on the PAPI/VASI? The OpSpec is simply allowing you to do what you've always done--descend below MDA only when at or above the PAPI/VASI. The difference is that as a Commercial Operator, you have a training program in place to determine when and how to determine this point without visual reference, and then only in IMC as you initiate the missed approach.

I'd be curious as well. It sounds like my company recently took it a step further than yours, and back at recurrent we were scratching our heads about it. I can't think of the last time I've taken a non-precision approach to mins, but the idea that I can treat the MDA as a DA just because there's a PAPI or VASI doesn't pass the smell test for a lot of us.

"Just because there's a PAPI or VASI" means that the approach path has been surveyed clear of obstructions (unlike the Birmingham GPS 36 approach that trips terrain warnings on the VNAV glide path below MDA).

As noted above, your company has a training program that gives you the tools to determine when this approach path is clear, as well as how to determine that you are in protected airspace, so that an MDA can be treated as a DA.

 

[kayoh190]

We're you led to believe that you couldnt leave MDA safely when you were on the PAPI/VASI? The OpSpec is simply allowing you to do what you've always done--descend below MDA only when at or above the PAPI/VASI.

That's just it - I won't ever really know. If I'm at the MDA and execute a miss because I don't see anything, my momentum *will* carry me below it, and I'll never have seen the PAPI/VASI.

From a practical standpoint, my automation should keep me from going below the proper glidepath - which is why we validate and brief that the coded descent path in the FMS is correct. Obviously we train for this stuff, and have all the tools to execute these approaches safely. But like Sluggo63, I'm just curious about the TERPS protections behind it all.

 

[Kritchlow]

We also add 50' to MDA non precisions... This is common practice I believe.

 

[Palmpilot]

We also add 50' to MDA non precisions... This is common practice I believe.

I think that's a wise practice, since flying right at a minimum altitude leaves zero room for error. In the physical world, there is no such thing as zero error.