Interesting. Was the tool down on the turtledeck? Or up higher, actually within the circumference of the prop? I could see that the airflow falls off very quickly outside the propeller disc- in fact, there should be tip vortices and the air just outside the propeller disc should be moving forward, at least in blade pulses. A good part of the exhaust cowl flap is within the propeller circumference, so I was hoping it would pull good airflow past it. At least based on my cartoon-level understanding of propeller theory. I will draw a sketch of PLAN B and post it. That is excellent you're in the air again! How do the extra horses feel?
I'm still a little sad and disappointed that the cowl flap idea didn't work. I was under the impression that a cowl flap providing suction was even more effective per unit area for developing air flow than ram air into an intake scoop. 1) Maybe the cowl flap size is overkill? The airflow separates? I should have had an adjustable flap rather than the fixed mockup. 2) Propeller inflow isn't as high-velocity as I pictured it being? That close to the blades, it has to be a pretty good percentage of the outflow speed, right? 3) A leaf blower simulating the slipstream isn't accurate? Maybe the flow over the cowl flap needs to be several feet "deep" to really generate the aerodynamic effects, not just a narrow splash of air from the leaf blower?
With some effort, I was able to place a Kawasaki EX500 radiator in the turtledeck area, low enough to clear the folding wings/flaperons. I mocked up the ducting and plenum with plywood, cardboard and hot glue, solidly enough for a static run. The radiator core is slightly small for the calculations, but I'd hoped to overcome any cooling system shortcomings with 1) a massive adjustable barn-door cowl flap that would create a low-pressure area in the slipstream and propeller inflow, and would draw huge amounts of cooling air through the radiator. I was so hopeful about this design that I began to believe that just the propeller inflow over the cowl flap would provide enough cooling for 3000RPM downwind displacement taxi. The cowl flap is shown mocked up in the full-open position. In cruise, I pictured that it would be nearly fully closed. 2) Electric cooling fans in the form of a couple cheap 70mm EDF motors adding air from the cabin into the radiator plenum. I didn't want to depend on these for anything except ground idle and maybe taxiing downwind. Definitely not to be depended on for flight operations. Well, I instrumented everything pretty well and did some static ground runs. #2 was a good success. The cooling fans can hold the coolant temp within limits on a 90 degree day, facing downind, up to about 3000RPM. #1 was a miserable failure. Even with nephews with leaf blowers simulating fairly generous airflow into the intake duct, and simulating slipstream over the cowl flap, the engine was not adequately cooled at any RPM. If it had shown any promise at all, I would have made the mockup more solid and proceeded with step-taxi tests, but I think it was bad enough where I'm back to the drawing board.
I believe the amphibian has some frame differences, too. I believe the easiest way to tell the difference is the width across the lower rear of the cockpit opening. My Catalina is 41.25", and I believe amphibians would be less. Perhaps someone can verify this idea, it's just something I thought I heard at some point.
Bought it used a while ago, used it a bit on my Catalina, gonna stick with the Warp drive on the new 670. https://seattle.craigslist.org/see/avo/d/ivoprop-ul-ifa-in-flight/6604342037.html Price OBO especially for avidfoxflyer folks
I'd be cautious of directly admitting air as a means to lean out the engine. It might be possible to dial it in for one moment, but changes in engine RPM or throttle position could change the vacuum at the air-injection point and dramatically change the mixture. There's been a lot of success with carb bowl vacuum systems. I'd stick with that. If you want simple, I wired up a very cheap 12VDC vacuum pump to a 100 Ohm rheostat, and plumbed it to a T in the normal vent tubing. The vent tubing continued on to its normal atmospheric opening, which I adjusted in size until the pump running full-out pulled enough vacuum on the vent line to increase EGT by about 80 degF. It increased my ceiling from about 9,000' to 13,999' and made the engine much happier. I didn't like pulling gas vapor through a crappy electric pump, but the pump head seemed pretty well isolated from the sparky parts. Also, if the vacuum pump was on, and the downstream vent tube got completely or partially plugged, it would excessively lean out the engine and could be catastrophic. It seems unlikely, but it's worth thinking about all the failure cases. Here's a pic of me holding 1000degF EGTs at (just under) 14,000'.
On my Catalina w/ 582, i've run a Warp Drive 70" w/ 3.00:1 vs a IVO 2 blade ultralight 72" at 2.62:1 and the IVO 3 blade 72" at 3.00:1 (against their recommendations, I only used this last configuration for a few quick test flights). The two IVO setups were in-flight adjustable, and even given that advantage, I never matched the static thrust, takeoff and climb performance of the fixed pitch Warp Drive. The 2 blade IVO may have had a very slight cruise advantage, but it was difficult to tell. It was definitely smoother and sounded better.
That version of the event makes a lot more sense than what I'd read somewhere: that it was the wing-mounted sponsons that got scraped up in crosswind landings. Sorry that you knew him, Jack, I hope my mention of it doesn't seem irreverent in this context. Yes, I'm also pretty paranoid about water in the bilge too. Here's some of my handful of tricks for worrying about it: 1) These battery-powered water-detector alarms are light and reliable, and I never had false alarms. https://www.amazon.com/Zircon-Alert-Electronic-Detector-Sensor/dp/B01J39MJ9I/ref=sr_1_1_sspa?ie=UTF8&qid=1509547198&sr=8-1-spons&keywords=moisture+detector&psc=1 I kept one in the tail, zip tied to the frame just inside the landing gear spring bolt inspection cover, and I also kept one just behind the step of the hull, where it would detect any water sloshing aft into the tail. Later, after I dealt with a pretty bad gash in the aft fabric from taxiing into ice, I realized how steep the angle is from the tail to the bilge is. Even in plow taxi, water drains very effectively from the tail area into the bilge, and I'm confident any leak in the aft area will be apparent as water in the bilge. 2) I have a little self-adhesive mounted mirror (non-fisheye worked best) that lets me glance backwards and down into the bilge without moving my head. I give the bilge a glance once I'm on step. 3) In my plane, there's a small void space between the sponson and the hull. I drilled holes from the interior of the hull into that void space, and through that into the foam-filled sponson itself, so any appreciable water in the sponsons will drain into the bilge, and the leak will be evident.
PS- plywood is for prototyping and static/taxi tests. (And for the CONSIDERABLE trial and error in getting all this stuff to clear the wing-folding geometry. I've thought I was done like 5 times and then hit yet another interference point) The plenum will be all fiberglass once it's done.