Forgot to mention, using the 3 x height/1000 to work out the descent gradient from up high also requires us to allow for slowing down in the 757 (a pretty slippery beast!). What works well is 1nm for every 10kts you want to lose. So;
cruise at 37000' with a target of 210kts when intercepting the localiser. 37x3=111nm + 10nm to slow down = 121nm. That works fine as a ballpark check for the VNAV profile.
Once you start going down it's all about energy management, just keep the calculations going to see where you are, it doesn't matter if you are high or low on the profile, just use the energy in the a/c to your advantage; if you are low on the profile and there is the chance of a shortcut at some point don't try regaining the profile using thrust, you'll probably end up on profile when you get the shortcut or even above it and you've just burnt all that extra fuel! If you are not going to get a shortcut then you are better trying to regain the profile earlier rather than later (and lower) where you will burn more fuel
If you are high on profile once cleared you can increase speed in the descent to dive down below that required and once level allow the speed to wash back off to get yourself back on the correct profile just as your speed comes back to the required level. This happened the other night going into Manchester. Initially arriving via Mirsi for a landing on 24L we requested 06L as it was very quiet. At a very late stage our request was granted so all of a sudden we had just lost 25 track miles. My initial reaction was to bin VNAV, go FLCH, wind the speed up (no ATC speed restriction) and pull the speedbrake out, select HDGSEL and point us towards an 8nm final. Quickly rebrief for the VOR/DME to 06L and set up the navaids (although we had the field in sight they still like you to follow the approach aid). Using the MCT DME 'plus a few' as my rough guide to distance to go before the LNAV arrival track was reprogrammed for 06L I could once again see how high I was on the profile. It's not long before you have dived through it and can start thinking about the decelleration, at which point the automatics were ditched and I got the chance to practice a hand flown VOR approach, recalculating all the time using the rules mentioned previously with the added bonus of the PNF calling the descent profile height checks until we had the PAPIS in sight. Works a treat. The idea being to keep the thrust levers shut from TOD to around 4nm when you spool them up in case of the GA. That way we burn only about 0.3T of fuel, whereas if you have to use thrust at any stage that figure goes up dramatically, especially with a level portion low down. Can often be nearer 1.0T
PP