My hangar neighbor Ron Hart came over this morning to loan me some 5 gallons gas cans as well as a 2 gallon pitcher marked in quarts to make the fuel flow measurements easier. You can see in the pictures below that his pitcher has "1 gallon" marks in sharpie that are above the measurements it was manufactured with. This is because he didn't trust the pre-printed measurements and measured a gallon of 100LL by weight (6 lbs per gallon) to make those marks before using it to calibrate the fuel system on his RV6. The weight of 10LL is nominally 6.02 lbs per gallon, but that's at standard temperature and pressure which today wasn't. For today I just interpolated between his marks and the pre-printed ones. As you can see below, the different is pretty negligible when you're trying to decide if you pumped 3.2 quarts or 3.3 quarts. I mean really we're talking a couple of ounces here.
I started by removing the fuel quick drains I had installed last week (wasn't thinking ahead) and rinsing out the fuel tanks. This was accomplished by pouring about 2 gallons of gas into the filler and catching it in a funnel lined with a shop cloth before going into another can at the drain location. There was really very little debris in the shop cloth afterward, just a couple of specks of pro seal.
Once that was buttoned back up, we removed the fuel line at the firewall and attached a 3/8" vinyl tube I had rigged with a length of 3/8" AN aluminum tubing with flare and B nut on the end to the firewall fitting. The other end got a funnel and by holding it above the engine, I was able to gravity feed backward thru the boost pump to prime it.
Once the pump was good and wet, we then put a gallon of gas in the left wing. I had Ron hold the hose about 18" above the engine, which gave me about 3' of head rise from the firewall fitting to simulate an extreme pitch up attitude.
In that configuration I then started a timer set for 1 minute and turned on the boost pump. We followed the same process for the left wing.
Unsurprisingly, both wings had an almost identical flow, approximately 3.25 quarts per minute, which equates to 195 quarts per hour.
There was no air being sucked at the end of the flow test on either wing, So starting with 1 gallon in the wing was also the test proof that unusable fuel in a 3 point attitude is something less than .75 quarts. Given that the fuel pickup locations are at the extreme aft inboard corner of the fuel tank, this will only get more favorable in a steep climb, i.e. the most extreme attitude.
I've talked to guys who have elevated the mains, put the airplane up on jacks, dug a hole in the ground to get the tail down in, all sorts of things to try to get the airplane in an extreme pitch up to measure this, but it seems to me that the chance of a calamity here far outweighs the benefit. In my case, I really won't know until after I fly what sort of pitch will result in a Vx climb. With this big engine and a 3 blade prop pulling hard, it will be a bunch. It's really not practical to try to achieve something like a 30 degree pitch up attitude on the ground, so the process we set up is the next best thing.
Based on all this, there's less than a quart unusable in each 21 gallon tank, but to be conservative I'm going to call it a gallon unusable in each tank. This also conveniently results in a nice easy to remember 40 gallons of total usable fuel.
Even though this isn't an FAR23 airplane, I'm falling back on the FAR23 standard for fuel flow, so here's where the math comes in.
Lycoming says that at max power, this engine should burn 120 lbs of fuel per hour. At 6 lbs per gallon, that equates to 20 gallons.
The FAA standard says that an engine should burn 0.55 lbs per hp per hour. This is a 210 hp engine, so 210 x 0.55 equals 115.5 lbs per hour, which, at 6 lbs per gallon equals 19.25 gallons.
We'll take the higher of the two and call it 20 gallons per hour as the max fuel burn.
FAR23 also says that for an engine with a boost pump, the minimum flow should be 125% of the maximum fuel burn.
So, with all that, the minimum acceptable fuel flow for this specific setup should be 20 x 125% = 25 gallons per hour minimum fuel flow.
Gong back to our flow test, I had 3.25 quarts per minute, x 60 minutes equals 195 quarts. 195 quarts divided by 4 quarts per gallon equals 48.75. gallons per hour against a minimum requirement of 25 gallons per hour.
Put simply, the minimum requirement is 25 gph, and I've got almost twice that. Success!