Looking at a sample of typical POHs, excluding a start with an external power source, the ALT field is ON before start, and is switched off after engine shutdown. With an external power start, the ALT is brought on line after start.
In a twin engine you might check an individual engine at idle, to see if the other engine’s alternator picks up the electrical load, and vice versa.
I recently came across a pilot who had been told to switch the ALT field off after starting in a twin to check the load on the other ALT, and vice versa. I pointed out this was not in the POH and could cause an electrical spike when the ALT was brought back on line, which might damage the regulator, and I sort of recall an engineer telling me it could even damage the accessory drive. I am guessing this scenario is with the engine in the cruise, and not after start at relatively low RPM.
There is a great body of knowledge on the forum on electric systems, so hopefully someone can point to the most authoritative answer to this scenario.
The husky POH specifies alternator ON after start which is what I’ve always done and haven’t noticed any ill effects so far.
What is the alternative to cycling the alternators after start if you have two of them (doesn’t matter if twin or two on one engine) and need to check that actually both are working and not only one?
Funny to read because the only SOP i saw in all french clubs I have been a member of is ALT On after start and Off before shutdown. After On you check if the battery recharges with the volt or ammeter.
I have to admit I don’t know what happens when I push that switch 
Have discussed with an engineer and they confirm this shortens the life of the Alternator, and in extremis might shear the shaft. Hence, at least on this type, the ALT is ON before start, and OFF after closing down. I believe the average puddle jumper, ie Pistons generally, have your bog standard Ford or Chrysler belt driven design. No doubt there are some improved designs out there.
On a multi bringing power to idle on one engine will show the voltage regulator transferring the load to the other engine alternator.
On a PA46 or other complex SEP with two alternators, an ALT failure, either one or two, would generate a CAS message
The most important is to follow the POH procedures for alternator operation, I have read a few POH’s which somewhat “limit” the switching of the alternator.
That said, the important thing is to consider what load you’re applying to the alternator (how much stuff you have turned on), and at what engine RPM. If you have the engine running slowly, and you apply a big load to the alternator, the drive torque goes way up. For a belt driven alternator, slipping a belt is a risk, though somewhat protects from damaging more expensive parts. (Bear in mind that if you burn off an alternator drive belt on the Lycoming, you’ll have to remove the prop to change the belt = big job). For front mount alternator Continental 520’s (Cessna 310, for example), loading up the alternator at low RPM (and creating high torque) can slip a drive coupling inside the engine, which once slipped (and engine oil gets in) never really works well again, and will require expensive replacement).
A very good way to reduce alternator load on GA planes is to install LED lights all the way around. For a VFR GA plane, it’s the lights which are most likely the big consumers, so reduce that, and you’ve helped. On my C 150M and my flying boat, I will check the alternator operation by switching it, and have never had a problem.
Most multi engined planes I have flown have a switching means to indicate alternator operation, without turning them on and off….
I don’t buy the bit about shearing the alternator shaft, because an alternator which can’t handle whatever the torque is at the max field current (i.e. full bus voltage applied across the field winding) is junk. If you put say 100-200V DC across the field, that would break the shaft right off (or break something allright) and that principle is actually used in the Honeywell Y-plan heating system valve (don’t click on this unless you are an anorak) 
We have done it before e.g. here and here.
I can see no reason for having ALT off for engine start, except to guard against
There is a good reason for having the Avionics Master OFF during engine start – here. It is to protect the avionics from the starter motor spike
I can see no reason for having ALT OFF during engine shutdown, except to guard against
The problem with this “excess torque” argument is that it leads to a recommendation to have ALT OFF during normal engine idle – because at idle the alternator will (normally) be making not enough to hold up the bus even with max field current, so raising the revs up from there will create a “dangerous” condition w.r.t. torque. That’s why I don’t buy it. The alternator is simply not capable of creating that much torque.
Pilot_DAR wrote:
That said, the important thing is to consider what load you’re applying to the alternator (how much stuff you have turned on), and at what engine RPM. If you have the engine running slowly, and you apply a big load to the alternator, the drive torque goes way up.
I really don’t understand the torque thing. If I remember correctly, then
P=2pi*T*n (with P=Power, T-Torque, n=RPS=RPM/60)
A normal Alternator has a max 60A@24V so 1440W. Idle RPM is in the ballpark of 800/min=13/s.
Therefore Torque at full output power is
T=P/(2pi*n) = 1440/(2*3*13)=78Nm = 78.000Nmm
In engineering classes one learns that for a solid shaft one should keep
d > 1.72 * root_3(Mt/tt) with (d= diameter, Mt= torsion moment and tt=max shear change stress of the material).
tt for simple steel is about 150 (N/mm12)
So in our case the shaft should have a diameter of at least
1.72*root_3(78.000/150)=1,72*8=14mm
So even with the cheaper steel a 14mm Shaft in the Alternator should not brake due to the stress. If you use more stress resistant Steel (e.g. CrMO or CRNiMo steels) you can easily get tt over 300 so that even a 6mm shaft should do the job.
Idle RPM is in the ballpark of 800/min
About 3x more due to the belt ratio.
Thank you all for taking the time to answer, a lot of good interesting points.
Some MEPs have 85A and 100A alternators, still belt driven, so the torque may be more significant? Or it may be the strength of the accessory pad which may be an issue?
