In 1964, with inline Gypsy Major, I was taught to stop the engine with mags, and open throttle as I did so.
When I returned to flying in 1986, with horizontally opposed Lycoming I was taught to stop the engine using mixture.
Returning to OP. If starting fails due to low battery, is solenoid likely to remain in “On” position? This happened to us last year with an O200.
Maoraigh wrote:
In 1964, with inline Gypsy Major, I was taught to stop the engine with mags, and open throttle as I did so.
Anything with a Stromberg carb (e.g. earlier small Continentals) is stopped the same way. Opening the throttle kills the vacuum in the carb, stopping fuel flow. With the Stromberg if you’re aggressive enough in opening the throttle at idle you can often stop the engine this way without touching the mags.
If you’ve got an idle cutoff I see no reason not to use it, especially with an injected or accelerator pump equipped carb engine that presumably doesn’t stop supplying fuel when the throttle is rapidly opened.
Relays can “stick” but for a starter relay to be stuck in the contacts-closed position, it had to be in that position previously (obviously 
) so what was the engine doing then?
The only scenario I can see where this is possible is where the relay had been energised but – for some other reason e.g. wire disconnection, or starter motor removed – with the engine not being rotated. And the relay failed at that moment.
Relays rarely stick in the closed position but they frequently stick in the open position, which for a starter relay means nothing happens when you try to start. You might hear a click…
Probably a big reason is corrosion, with these components never being changed, and not being hermetically sealed. More here with pictures of the typical Stancor relays used in GA.
Maoraigh wrote:
If starting fails due to low battery, is solenoid likely to remain in “On” position?
I guess it can, and mine definitely did. We free’d it up with use of a tapometer I’ve also seen an EC-120 that had a starting issue resolved with the tap of a hammer too so it’s a universal method.
I did not waste any time in ordering a new starter contactor from Adams this morning.
That sounds like the contacts were welded. A tap with a hammer can free them.
They must have become welded just at the moment when the battery finally and totally died, and the continued heavy current would have totally flattened the battery after that.
You can buy Stancor relays from e.g. mouser.com, for far less than the aviation sources.
The two links above require a login to Aliexpress (chinese ebay, sort of) but this is the one, I think
and the data suggests a 2V drop which at a few hundred A is quite a lot, especially when you consider that for optimal power transfer the voltage at the load is 50% of the source open-circuit voltage, so in a 12V system the starter motor may be getting only 6V, and you are stealing 1/3 of that.
But for sure an SSR (solid state relay) would work. You would need some serious protection from the inductive spikes, which are massive (easily kilovolts) as these would destroy an SSR.
I would build one with MOSFETs, not thyristors which is probably what these use. You can get MOSFETs below 0.001 ohm which is 0.3V drop at 300A. Probably the best way would be to assemble a lot of small devices on a PCB, in parallel. There are 0.001 ohm devices which cannot carry anywhere near 300A due to other limitations, so this would be a good way.
The downside is that a spike could blow them all up and cause them to fail in a conducting mode. I wonder how all this was solved in the solid state circuit breaker products used in the homebuilt market – example.
I have built power supplies with MOSFETs such as this. This is what a 300A device would look like
Years ago I looked into replacing the TB ground power relay with a MOSFET design. It is a weak point as mentioned above. It is nontrivial but doable. There are some interesting old patents from the automotive sphere, designed to overcome the problem that a MOSFET conducts in the opposite direction even when turned off in the main direction (due to a parasitic diode).
@vic relays are still relevant when starting the Yaks. The starting spark mechanism has a an oscillating contact (on-off-on-off…) to provide sparks at top dead centre for starting. I had a problem when this contact stuck in the ‘on’ position which caused the ignition switch/circuit breaker to immediately trip every time you pushed ‘start’ (the lack of power then causes the air solenoid to close…no rotation!). We eventually had to hand prop it to bring it back to base. Funny enough the contact which evidently had temporarily stuck rattled free on the flight home. The mechanics subsequently polished up the contacts and called it good to go.
Unfortunately, I found this out the hard way and was stranded overnight on Alderney (an island). If you fly a Yak, it might be worth finding this device (right hand footwell on an 18T) so that you can rattle it (or tap it with a hammer) or disassemble it and fix it if you are ever stuck.
Also good to know is if your starting air solenoid has a manual override button on the solenoid. Ours did not, but I gather some do. If you have this, I think you can turn the mags ‘on’ and then manually trigger the air to rotate the engine. It might be harder to start (because of the advance in the mags) but at least you don’t have to hand prop it. I haven’t actually tried this, so take the advice with a grain of salt.
In relation to the original post, I was always concerned with the Yak. You spend a lot of time spinning the prop to ensure that you don’t have hydraulic lock. I never had an issue, but a member in the group once felt a blade kick. I don’t know if it actually started or not, but they jumped clear uninjured… no defect was found during the subsequent inspection, although of course the mags could have accidentally been on.
I’d have to imagine at these sorts of currents the choice of transistor would be an IGBT rather than a MOSFET. An IGBT would likely be a lot more resilient to spikes, too.
But relays are generally very reliable and last for years, and have extremely low on resistence. It doesn’t really seem worth swapping it for solid state (which would just bring new failure modes, and is not necessarily any more reliable).
IGBTs have a much bigger voltage drop than MOSFETs.
I think the power relay problems are related to corrosion, possibly combined with lack of use. I have just pulled out an unused Ametek ground power relay (300A N/O contacts and 100A N/C contacts) out of a box and it was stuck. It took maybe a dozen energisations before the armature would fully move. I had it maybe 10 years inside the house and I don’t know where it was before that. A lot of “new” aircraft parts are really old…
