gallois wrote:
One of main problems is keeping the speed above Vmca, which is why many pilots delay Vr until the aircraft is above Vmca.
Indeed and many training places will fudge Vr and delay it to near Vyse, way above Vmca, then the problem of control & climb one engine after rotation is pretty much solved like a walk in the garden but you need extra runway length vs AFM/POH, there is no free lunch !
Peter wrote:
Sure, but all certified piston twins are supposed to “climb” on one engine
This isn’t true – Vy(se) can be a negative value at gross weight in a certified twin. Even if Vy(se) isn’t negative at ISA and sea level, often it’s a very low value that can only be achieved after you’ve feathered the prop and got rid of other drag (e.g. retracted the flaps and landing gear – and you can’t retract the gear if you’re still on the runway, even if you’re nearly at flying speed).
It may be true of heavy twins – but not of the typical light MEP.
This is what the FAA’s “Airplane Flying Handbook” has to say about it:
“With regard to climb performance, the multiengine airplane, particularly in the takeoff or landing configuration, may be considered to be a single-engine airplane with its powerplant divided into two units. There is nothing in 14 CFR part 23 that requires a multiengine airplane to maintain altitude while in the takeoff or landing configuration with OEI. In fact, many twins are not required to do this in any configuration even at sea level […] [aircraft] 6,000 pounds or less maximum weight and VSO 61 knots or less: the single-engine rate of climb at 5,000 feet MSL must simply be determined. The rate of climb could be a negative number. There is no requirement for a single-engine positive rate of climb at 5,000 feet or any other altitude. For light-twins type certificated February 4, 1991, or thereafter, the single-engine climb gradient (positive or negative) is simply determined.”
In the Apache that I did my multi in, if one stopped at any stage during the takeoff, even at rotation, you rejected the takeoff. Even if you were going off the end of the runway and into the woods; it’s much better to run off the end of the runway and crash into obstacles, while braking and slowing and keeping the aircraft under control, than perhaps ten feet off the ground and at flying speed. The crash energy that must be dissipated increases at the square of the speed you hit it (not to mention additional crash forces from falling if you’re off the ground) and you can often wipe off quite a lot of speed with the wheel brakes.
From high places (I once went out west in the Apache) we briefed that we’d close both throttles and land ahead if we were less than 500 feet as the Apache with 3 of us in it and all our gear and full fuel would not likely maintain altitude on one engine once departing from a 5000’ elevation airfield.
In reality it’s rare for a climb gradient in a twin to be negative when clean unless above single engine service ceiling, in which case it will descend slowly to that altitude.
The old CPL required Net Take off Flight Path calculations for Performance B MEP, seems to have gone by the wayside with the passage of time. While most MEPs can manage to accelerate to Vyse and carry out a feather drill if the EFATO occurs after take off and gear in transit, the climb gradient will be negligible/negative during the drill, and marginal afterwards (typically 1 to 2 %). The take off brief saying the aircraft will remain in the circuit demonstrates the NTOFP has not been calculated, as it will take several miles before the aircraft can achieve a safety altitude and turn towards the circuit, well outside the ATZ upwind.
The PA-30 AFM kindly has a full page warning that simulated asymmetric training is the most dangerous form of flight training that can be carried out.
In UK, most places do CPL now on complex SEP (mighty Arrows or new TB20s), was it on MEP back in the old days?
In France, the norm is to do MEP+CPL combined (most likely DA42), there is only one ATO that does CPL on singles, they do CPL+IR+MEP on Wassmer CE 43 and PA34, legacy steam gauges with old VFR nav and IFR on ADF, maybe a good selling point if one want to join Islanders fleets 
At the ATOs I know you can do SEIR + CPL + transition to MEIR. Most on DA40/42 plus DA40/42 simulator combinations. It all depends what your starting point is.
@RobertL18C yes simulated assymetric training was possibly the most “dangerous” part of MEP training but that is the reason the FAA and EASA dropped assymetric below Vmca simulations as part of in flight training.
I seem to be a lone voice in promoting the cause of twins but as I really like them I’ll try.
Rejecting a take off in a twin is really not much different than that in a single except there are more options open to you so you perhaps plan and brief more than you would with a single.
RobertL18C talks about briefing for an engine failure after take off when the engine fails mid gear retraction. In both singles and twins I think we can all agree that engine failures these days are rare events, to happen on take off even rarer, and to happen during the approx 10second or so gear retraction process and rare as hens teeth.
(I know some aircraft have longer gear retraction times and gear out times. But whether we are flying an RG single or twin we know how long that should take for our aircraft don’t we?)
In a single the options you have is to land straight ahead give or take 30° and whether or not you have time to put the gear back out or whether you land gear up. This situation would be regardless of whether you are at or below MTOW. This decision remains the same whether the pilot is in the habit of retracting the gear as soon as a positive climb is achieved or waits until there is no runway in front.
In a twin your first job is control the aircraft by stamping on the pedal of the working engine to control the yaw towards the dead engine and lowering the nose, once you have done that if you have briefed yourself this way you can do exactly the same options as you would in a single, although you would also need to reduce power on the working engine. You could on the other hand push all levers forward to attain blue line speed. The gear will by now be up you will have no flaps, all that remains now is to fly straight ahead or climb slightly if you have enough power available. All twins even old ones should do this even at MTOW. So unless the ground ahead is rising there is no problem but if it is rising you will again look for somewhere to put down, again as you with the same options you have with an RG single. However, you also know that if you can identify and verify the dead engine and feather the prop during this phase it is possible you will have probably reduced the drag enough to climb, if only slowly. And how often is this all going to happen at MTOW? So it might climb even better. There is no requirement to climb to circuit height to return and land back on the runway but you do have the option, obstacles allowing, to call an emergency, make a low level circuit and land back on the active runway.
The briefings before take off should not be recited as some sort of mantra, the options should be thought out and decisions made at each airport and with each aircraft before lining up. But then we all do that, for every flight, don’t we?
gallois wrote:
At the ATOs I know you can do SEIR + CPL + transition to MEIR. Most on DA40/42 plus DA40/42 simulator combinations. It all depends what your starting point is
Do they do “CPL only” on DA40? I would be really interested
Cirrus pilots frequently abort takeoffs as the door doesn’t close properly and you’ll notice only once you get airspeed. Its something with the door seal rubber.
I now have the habit to ask my wife to check the LH door before she gets in.
Try Iroise Aero Formation at Brest LFBR or Astonfly at LFPN. Both are very helpful and I get the impression that you have native French but both also teach in English if you want.
The problem might be that IAF I know is very busy and has been searching for instructors, so I assume Astonfly to be the same.
My favourite is IAF and they do a great deal on their simulators.
gallois wrote:
Try Iroise Aero Formation at Brest LFBR or Astonfly at LFPN
Yes they both do CPL courses but it’s in DA42s not in complex SEP? the singles they have are non-complex (P2002, DA40 & C172 & SR22), so not suited for EASA CPL training, I flew in both places renting their SEPs and I highly recommand them, although Brest is way more scenic, relaxed, friendly and good value for money
