Cooling Duct Glassed In

As I had said before, when you think you're done with the lower cowl, that's when the work on it really begins.
In my case it was even a bit more challenging as I wanted to do the oil cooler modification at the same time - thinking it would be easier to do it at this stage in the build.

The modified cooling duct turned out to be usable after the second coat of filler had hardened. I gave it another light sanding job and declared it good enough for production.

Then I installed the brace that will support the center aft edge of the lower cowl. Just in case that might affect the following fitting session.

The cowl was installed with the cooling duct clamped to the upper left edge of the cowl.

The fit was pretty good and I decided to leave as it was.

The only issue I had was with moving the oil cooler forward. It turned out that there are two small extrusions on the bottom of the Rotax that are not being used for anything but they are interfering with the yet to be installed oil cooler frame.
To provide sufficient clearance I used some wooden buffers.

That even provided a slightly better fit with the front air inlet.

So I went ahead and drilled the upper flange of the air duct and clecoed it.

After adjusting the interface frame that ensures an air tight seal to the radiator, I drilled the first hole into the lower flange. Finding that flange is easy if you get a light inside the cowl.

The cowl came off and I took a closer look at how my modification would change the fit and the glassing procedure. You can see that the gap between the bottom of the oil cooler flange and the cowl needs some significant filling to build it up.

I marked the 5/16" of required clearance between the aft end of the duct and the radiator face by eyeballing the necessary trimming amount. This is not very delicate as it will get glassed into the interface frame and you get quite some leeway that way.

It looked like I had that trimmed enough for a first attempt to get the interface frame in there.

Off came the cowl again (I stopped counting at some point) and I turned it over to drill the rest of the bottom flange of the cooling duct.

I put the spacers on the front face of the radiator as per plans. I later learned that putting an additional spacer on the bottom part of the frame does help.

After a few repetitions of cowl-on/cowl-off the required clearance was achieved. Please note that my interface was offset to the left of the aircraft when referenced to the radiator face. That would have created a bad seal and I needed quite some time and trimming to put it more to the inside of the cowl - while maintaining the 1/4" clearance.

You can see the gap that this adjustment created on the inside fit of the interface to the cooling duct in the following picture. This would need to get filled with flox.

Drilling one hole in the side flange of the interface was quite a tricky task as hardly any area was accessible. This cleco was important though to maintain the delicate positioning for final drilling and a good air seal.

I took it all off again, finished drilling the interface frame and marked the outline of the cooling duct flange to the lower cowl.

The cowl went back on to check the final clearance between interface and radiator face.

What looked simple in this pictures is the result of multiple on-off cycles of the cowl, a lot of additional trimming and re-drilling and clecoing holes in the interface frame. So, don't let this end result here fool you. You might need to do something to make it fit well enough.
This was the final step before I glassed the cooling duct in. Due to my modifications, I slightly deviated from the plans in that I was trying to attach the duct flanges to the cowl and the interface frame to the duct but not filling the gap between interface and cowl at this point. I simply had not enough time to get all that done before the pot life of the resin came to an end.
I did glass over the front upper part of the duct and the cowl though. The need for this was because the installation of the duct created a weird misalignment of the lower cowl that removed the clearance to the spinner backplate in the right side that I had diligently created during the initial cowl fitting. I pushed back the cowl in that area while the resin was curing and the glass over the duct in that area helped keeping it in that correct position.
Sorry, no photos of that stage as time was way too short to mess with the camera.

After a day of resin curing, DO NOT FORGET TO INSTALL THE COWL AFTER GLASSING THE DUCT IN, I took the cowl off and cleaned up the leftover work from the day before.
This consisted of glassing the area between the interface frame and the cowl and adding the glass strip on the inside of the cooling duct, between the cowl and the interface frame.
Also, I got a chance to fill the base of the oil cooler flange with flox from inside the cowl and from the air inlet side.

Now, the cowl is installed on the plane again and the resin is curing. The last step will be to install the oil cooler and see if all the measures for ensuring sufficient clearing have succeeded.