Starting techniques for injected engines

I am running a new IO360M1B for the first time. This is without the cowl on. It starts almost immediately when cold (a couple of blades) and primed with the fuel pump. After running for a few minutes, I stopped the engine (to make an adjustment) and then went to restart. The engine did not like to restart. It takes three or four times of 10 sec cranking to get it started and then only when the mixture is pulled halfway back. Previously with an injected engine which is cowled, I flood it and then start with the mixture full out and the throttle full in. Using this is usually for eliminating vapor in the fuel lines. Without the cowl on and after running only 2-3 minutes, I had not expected vapor to be an issue.
I am interested in what procedures people use for a warm restart and whether this should even be an issue without the cowl on.

When warm, I start by setting throttle to approx 1000 rpm position and mixture at ICO. Engage starter and once spinning advance mixture until it starts.

Ditto

ColoRv said:

When warm, I start by setting throttle to approx 1000 rpm position and mixture at ICO. Engage starter and once spinning advance mixture until it starts.

Click to expand...

This is the same procedure I use w/one Slick and one Lightspeed. No need to flood mine.

Never been stranded yet... knock on wood.

Joe

It is my experience there are no procedures that work all the time because of the variable engine temperature situations. You've got to do what needs doing and cranking without a light off after 5 or 6 blades only runs the battery down.

The engine needs the right fuel/air mixture ratio to start and if it is not right, it won't start. The flooded engine trick works because it hits a sweet spot when clearing and starts.

I find the prime has to be constant, i.e., throttle and mixture set the same all the time. If not the amount of prime is not constant and the formula for getting fuel in the cylinders is different and the amount of time to prime is different.

I get into trouble when I don't prime enough on a cold start...happened yesterday and I nearly killed the battery before the flooded engine technique finally got it going. If the prime time is just right, it will start almost instantly.

With a hot start, no prime works best for me. It lit off on the first blade yesterday. That's always a good feeling. But next time something may not be just right...so you've got to hang loose and not run the battery down away from home. If it doesn't light off soon try something different like flooding it before its too late battery wise.

My .02

For what its worth here's my procedure for io-520's. I don't know if it applies here but this is the procedure that the 135 operator I worked for had in place for their aircraft. The seaplanes in our fleet operated in an oilfield environment (i.e. close quarter tie ups to sides of boats, drilling rigs, etc) where we HAD to have the engine crank after pushing off or the plane would get bent by drifting into the many obstructions around. The second hot start procedure took some time to master. Both procedures become second nature as you can feel/hear the engine which in time and practice, leads to a smooth engine start.




Cold Start: throttle and mixture full, fuel boost pump for approx 2-3 seconds or until a rise on fuel flow gauge, crank engine with hand on throttle, as the engine catches pull back the throttle.

Hot Start: there were two different scenarios here depending on how long the engine had been shut down and how "hot" the ambient temperature was (summer vs winter).

1) Immediate engine restart or less than 5 min: throttle full, mixture idle cutoff, crank engine as above

2) Alternate method for extreme hot temps: throttle full, mixture rich, try to crank, if engine does not catch
try boost pump for 1-2 seconds and then crank again. If engine catches be ready for it to kill and have
your hand on the boost pump ready to jockey it (on/off) until the engine catches and smoothes out.


**After second thought, our PA-23's IO540's were completely different with their "shower of sparks." So maybe this won't work for your situation.

Here is what I have settled on over the last 350 hrs with my M1B:

When cold, the book says prime with the boost pump for a few seconds and then go to ICO to start cranking. Then you do the quick play on the levers to keep it running. I don't like that.

A much easier cold start (less hand movement) way is to set full rich, turn on the boost pump, count to 3 ( or 5 if it is winter ) and start cranking with boost pump still on. Fires right up and zero hand movement needed. If something goes awry, you gotta' turn the boost pump of right away but that has yet to occur because it fires every time.

When hot I think David is correct, there is some variability however I have found that my installation needs a short shot of fuel to fill the injection system or it will not fire quickly. So... ICO, boost on, full rich for a count of 1 then back to ICO. A little throttle (like the 1000 rpm setting somebody said above) and start cranking. Mixture forward after it fires. This may not apply to your case of an uncowled engine since I am compensating for fuel boiled out of the injection lines.

Hot start

Sounds like I was over priming by leaving the host on more than a couple of seconds. I will try the procedures above. May I assume that ICO is mixture cut off (all the way back)?

jay.pearlman said:

May I assume that ICO is mixture cut off (all the way back)?

Click to expand...

You certainly may.

ICO------Idle Cut Off.

I always treat a warm engine as if it's flooded, because as soon as you shut down the injectors drip and that's exactly what you have. Therefore, every warm start is a "clearing proceedure": no boost, ICO, full throttle. In the 10 years that I've been flying behind Bendix FI, this works 99.9% of the time. When it doesn't start, the engine is now "clear" and you can do a modified cold start.

The key is to clear a warm engine... If it happens to start during this process, then all the better.

FWIW, I just took the lycoming service school, so I'll relate what the Guru Jim Doebler said (and there's no doubt, hes a guru; that said, engines do have a mind of their own) - Bendix style injection, and working properly is assumed....

Normal: Fuel pump on; mix to full rich; prime until fuel flow meter comes up; prime off, mix to idle cut off; hit the starter. When catches, mix to full rich.

Flooded: throttle to wide open, mix to idle cut off; motor starter until catches, throttle back mix to full rich.

Hot start (ie shortly after shut down): NO prime (system pressure is up and ready to go); mix full rich, crack throttle, hit starter.

If its long after shut down, the pressure should have bled down so youre at a normal starting condition - in theory.

AlsoiIn theory, the system can never vapor lock as the servo diaphram area has a small passage in it to bleed off backed up vapor pressure.

With all the other responses, hope this helps a little.

It's time

When the weather clears, I will try the approaches suggested. A friend and mechanic said that each engine has its own personality and, like other friends, you have to get to know it.

The variety in the procedures people are using pretty much confirms what I've found -- nobody's engine behaves the same way in these various conditions.

I have vertical induction and have been flying the 7A M1B since '12 and I'm STILL just figuring out how to baby the thing in a way she appreciates.

Many -- and I mean, MANY -- of the suggestions over the years, which run the gamut -- just don't seem to work. But they provides a starting point.

I finally started using one as a place to start and logged conditions and result from every start, and eventually I came up with the best method to use after fine tuning.

What was the temp? What were the CHTs for hot starts, where was mixture and throttle, did it start? How long did it take? What happened when I added mixture? When it started and then quit, what did I do?

Eventually, I came up with a fairly handy chart given all the variables for what I should do.

Next step

Can you provide your chart to us? Sounds like a great place to "start"