Think about it - the rpm will not drop in the air.. but yes, it can go past optimum rpm. This can act as a form of governor, I believe this is how piped CL stunt models are generally setup.

Yes, you need to open up the needle to anticipate the flow conditions when the pipe is on boost - rpm will jump, air and fuel flow requirements increase, and a more open setting is required to achieve the right air:fuel mix at higher rpm. You might think that it comes out in the wash - more rpm, higher air flow rate through the venturi = higher fuel "suck" = same A:F mix. But it is not a linear relationship. And the pipe is cramming mixture back into the cylinder as well, messing things up. If you remember the axiom, more rpm needs more open needle, you're in the right direction.

Here is how I deal with the process, now anyways:

Make a decision ahead of time. What flight prop are you going to use? This is much the same as deciding what rpm you are going to run it in the air, but choosing the prop based on some knowledge of the engine and the aircraft's needs is easier. For example if you are bopping around on a muffled .46 and it spins a 10-7, and the performance is great but you want to bump horsepower to get a bit more.. well then, stick with the 10-7 and tune the pipe system to work with it.

To tune the pipe in boost condition on the bench, assuming the pipe/header system is starting on the long side (i.e. the "metal safe" side) the easiest method I know of is to unload the engine by reducing the diameter of the prop. You can reduce pitch instead, but you are usually stuck with integer changes, and you have less control over the magnitude of the change as a result.

For engines in a somewhat "regular" state of pipe tune, trimming the prop to 90% of it's diameter should get you close. i.e. I would trim the example 10-7 to 9.1". For higher strung, peakier setups 85%. You'll find a mirror of this recommendation in ILS's dissertation on breaking in high performance engines as well. Same reason - getting to flight rpm on the bench.

Bench run the engine on the pipe system with the flight prop trimmed to 90% diameter. Warm it up, floor it, and tune momentarily for peak rpm. Record the tach reading.

Trim 1/8". Redo the above. Did the rpm climb? Good, you're going the right direction. No, or did it drop? Then you were short in your initial estimate of length. Better drop the flight prop size or get another header. Oops.

Repeat until the rpm changes are approaching nothing. If you hit no change stop. If your last change looks like almost no change, stop. And, with it running full bore on the bench prop on the pipe, set the needle on the rich side of peak as well, not right on it except for brief tach readings of peak. A clean 2 cycle, but back off from peak to the richer side of 2c.

Now without moving the needle, if you fit your flight prop the engine should now struggle a bit to jump onto the pipe, and should appear somewhat rich. A quick line pinch should see it jump up in rpm but drop off again. Don't move the needle. You are looking at a reasonable facsmile of the ground setting for flight with that prop and pipe setup.

Then as the aircraft gains airspeed and the prop unloads, this allows the engine to get on pipe and the rpm will climb up to, hopefully, pretty much where you were running it on the bench. And if the needle was set to the correct ground-rich setting, the mixture should be decent in the air.

Setting the needle lean on the ground is common for those not totally accustomed to pipe tuning - it may appear comfortably rich running at zero airspeed (I don't like the word static) but as the rpm climbs, as mentioned, the A/F ratio will shift lean as fuel draw fails to keep up with the intake flow. Or more axiomatically: it needs more fuel, open the needle.

The amount of raggedness on the ground compared to in flight depends on the state of tune of the engine and system. The extreme example is of course, F2A speed.. 19-20k or something on the ground, then mad whipping to help the engine unload and stage onto the pipe. very, very critical, peaky setup - very narrow operating range. Sport stuff is far more forgiving in relation.

MJD