Warm start more difficult than hot start? What gives?

[A64Pilot]

  On 4/2/2025 at 5:10 AM, 0TreeLemur said:

 According to the video above, the normal state for a Lycoming IO-360 before hot or warm start is flooded.    I did not realize that.   Don said that the flow divider acts like a coffee "percolator" and it drains fuel through the injectors after shutdown.   Based on that I believe that by switchiing on the boost pump I didn't prevent cavitation, I caused the flooded condition to persist longer than I needed to.

Note to self:  boost pump not used for hot or warm starts.   I was over thinking it.

 

Expand  

Maybe not, I hate to say it depends but it does.

This may sound dumb but I open the cowl flaps on final and if it’s Summer I leave the oil filler door open after shut down, both seem to help and if I had to go around having the cowl flaps open before hand is one less thing to remember to do.

1. Don’t do what I hear so many Bonanza and Cirrus driver do for some reason, and that’s start immediately at a high RPM, sure use full throttle but pull it as soon as she starts. I don’t know why those guys do that, the IO-520 in my C-210 was easy to start when hot or cold. Must be different in them.

2. Boost if she is running rough right after starting seems to help, I believe as has been stated it helps prevent vapor lock / pump cavitation, so once running if she starts surging turn the boost on and see, why not give it a try and see. What happens of course is due to there being vapor and not liquid fuel in the system she goes lean almost immediately after starting eventhough the mixture is at full rich and eventually if it keeps running that vapor will clear and she will run smooth, but your right after firing is not good for the muffler. Excessively lean mixtures may cause after firing. Excessively rich may cause back firing.

I haven’t watched any video, but add cold starts to your statement about the normal state for starting is flooded, priming when cold slightly floods the engine unless you get lucky and hit that sweet spot, most don’t. It’s just like the old cars with carburetors, pumping the gas pedal caused the accelerator pump to spray, slightly flooding the engine.

 

Edited April 2 by A64Pilot

[PT20J]

I believe @EricJ is correct. Many of us have noticed after installing an engine monitor that the fuel pressure has a lot of wild variations in a M20J (maybe other models also). The Garmin EIS seems more sensitive to this than the JPI which may have better filtering. The analog OEM gauges don't seem to be able to react fast enough to display this variation. There are posts on vansairforce about this phenomena with the IO-360 in RVs. So, I think it's going to happen in any airplane with a Lycoming IO-360 and a tight cowl. I prevailed upon @kortopates to peruse Savvy data and he said it is common. 

The fuel lines in the engine compartment are all fire sleeved and thus well insulated. However, the mechanical fuel pump is not insulated and picks up conducted heat from the engine and radiated heat from the engine compartment. This can cause vapor bubbles to form in the pump. You can easily see this if you direct a heat gun to the fuel pump with the engine off and the mixture in ICO so that there is no flow. As the pump heats up, the fuel pressure will rise. Directing the heat gun at other fuel components has no effect.

During cruise power operation, there is enough fuel flow to minimize vapor issues, and the RSA servo is designed to be insensitive to inlet fuel pressure variations and to separate fuel from vapor, so the vapor induced pressure fluctuations don't seem to affect engine operation. In my airplane, if I run the engine at around 1000 rpm on the ground for an extended time on a hot day the engine compartment will get hot enough to cause vapor and make the engine run rough. If I increase power to about 1500 rpm for a few seconds, it seems to clear it out.

[Hank]

Even the much lower fuel pressure in my C varies significantly in flight. Sometimes the needle will point up near or even above 2 psi, sometimes it will drop as low as 0.5 psi, right at the bottom of the green arc.

Not sure I want a second-by-second graph of what it's doing, the swings of the needle are enough!

[takair]

It’s interesting how many variations there are to start techniques.  I will say mine has evolved over the years and the equipment installed.  In its latest and most reliable iteration for my configuration (SureFly mag), I have been using the same technique hot, warm, and cold (gasp).  Throttle about 1/4” open, pump on, mixture rich of slow count to 5, mixture closed, starter until it catches, mixture rich assuming solid catch, RPm to 1000.  Occasionally on a warm start it will have to try twice.  Warm start always seems to be the odd man out.  I suspect this consistent starting as attributed to the SureFly and starting rich.  Your mileage may and likely will vary.  Some choice language may or may not help too….

[takair]

…..also…have you ever noticed….when there is an audience….you yell “CLEAR” very loud so that everybody knows you read the AIM.  However, yelling “clear” too loud can scare the engine into having performance anxiety and leads to that awkward silence at the end of a few turns of the prop.  You will notice that with each subsequent attempt, you say “clear” more quietly until the engine finally  determines that all conditions for it to run are met.

[ArtVandelay]

  On 4/5/2025 at 12:37 PM, Hank said:

Even the much lower fuel pressure in my C varies significantly in flight. Sometimes the needle will point up near or even above 2 psi, sometimes it will drop as low as 0.5 psi, right at the bottom of the green arc.
Not sure I want a second-by-second graph of what it's doing, the swings of the needle are enough!

What is interesting in my case, the swings are reduced as the fuel levels decrease, you can see below where I switched tanks.