It's an AIM recommendation, and not a regulatory requirement to intercept the glideslope at the "published glide slope intercept altitude" (absent a a mandatory altitude) but there are countless other recommendations within the AIM, especially w/r/t instrument approach procedures which are the instrument pilot's bread and butter of "best practices." In my opinion the best practice on an ILS is only intercept the glideslope at the published glideslope intercept altitude.
I dont view the paragraph outlined from the AIM as a "recommendation." Nowhere in the paragraph does it use the language "recommend." It is merely a statement of intended use of the glideslope's vertical guidance.
No. I'm not saying that. I am suggesting that in its entirety, it's pretty clear that if you're going to intercept the glideslope at the published glideslope intercept altitude, you're going to have to comply with the published minimum altitude on that segment of the approach. Again, it's just my opinion.
My interpretation of what the AIM is saying is that if you intercept the glideslope before the final approach segment, you still have to comply with the altitudes published for any fixes between your intercept point and the PFAF where the glideslope is intended for use in the descent. What the AIM guindance seems intended to address is a situation where following the glideslope outside the PFAF/FAS could cause you to get below the published minimum for that portion of the approach and fix. The FAA only clears the ILS glideslope out to the FAF so there could easily be a tower just past the FAS that extends above the glide slope and for which a higher fix altitude is required.
The AIM guidance does not to me suggest that you have to be at altitude when intercepting the glideslope, just that you cannot necessarily follow the glideslope down if you intercept prior to the PFAF because you may violate an altitude restriction.
Lets take the approach provided by
@midlifeflyer as an example.
If I maintain 4000' after crossing PECIT, I can expect glideslope intercept at approximately 12.5NM DME. That's 1NM beyond PECIT and 5NM before PURME. Great. Since there are no fixes between PECIT and PURME, I can fly the glideslope as the AIM indicates because the next altitude restriction of 2400ft is coincident with my PFAF and I therefore have no risk of descending below the minimum altitude for the segment between PECIT and PURME by following the glideslope.
But lets say I have a tower between PECIT and PURME. That tower is a tall tower at 1900' and sitting on top of a hill that is 200ft above the surrounding terrain of 700'. That puts the tower right at 2800 ft so I now need a fix at a higher altitude between PECIT and PURME, so the FAA creates a fix called TOWER with a minimum altitude of 3,000ft at DME 9.0, just 1.5NM before PURME. If I were to follow the glideslope from 4000' at my intercept point at 12.5NM, 3.5NM later I would be between 2850 and 2950' and below the minimum altitude of my new fix TOWER and dangerously close to the tower at 2800'.
This also works in reverse with overlying airspace. If I intercept at 4000 and 12.5NM but have a Class B shelf that extends down to 3000' at DME 10NM and therefore have a fix with a maximum altitude of 3000' ft at DME 10NM, following the glideslope down would have me bust the maximum altitude and the overlying Bravo by 250ft.
The FAA cleared the remaining 7.5NM of the final approach course inside the PFAF when they surveyed the approach so that a 3degree glide slope would work; minimum altitudes at fixes inside the FAF do not apply as long as you are on the glideslope but outside the PFAF, as I just illustrated its possible that fixes extend above or below the glideslope and this area has not been specifically surveyed for compliance with the glideslope.
That to me is what the AIM means when they say "Interception and tracking of the glide slope prior to the published glide slope interception altitude
does not necessarily ensure that minimum, maximum, and/or mandatory altitudes published for any preceding fixes will be complied with during the descent."
It is therefore on you as the pilot to ensure compliance with minimum/maximum/mandatory altitudes at fixes outside the PFAF if you choose to intercept and track the glideslope prior to the PFAF. Notably the AIM does not suggest that you cannot or should not intercept prior to the PFAF, only that the glideslope is not intended to be used for vertical guidance before the PFAF and therefore you may find yourself in conflict with fix altitudes between the point of intercept and the PFAF.
In all situations, I would say the definition provided by
@luvflyin applies:
"GLIDESLOPE INTERCEPT ALTITUDE− The published
minimum altitude to intercept the glideslope in the intermediate segment of an instrument approach"
Intercepting at a higher altitude above the minimum altitude further out in the intermediate segment is fine so long as you are aware of the AIM guidance which says you might not want to follow the glideslope once intercepted from further out because it might not comply with all altitude restrictions between the intercept point and the PFAF