I now have a little experience using Red Devil Fireplace & Stove Silicate Cement, though I haven't seen it "under fire". It is rated at 1000 degF. This might work well for what you're talking about; however, I might question its strength when used sparingly: I have taken a fully cured pencil of it in hand and broken it easily by bending, so the final tensile strength can't be all that great.
I don't think the problem with using something like this for bonded construction will have anything to do with the temperature of the cement per se, nor with the mechanical stresses from chamber pressure [which on average is pretty low in a pulsejet], but rather, the difference in thermal expansion between the cement and the metal it's bonded to. I'm sure the local stresses caused by this could turn out to be pretty high when things start to warm up. So, I think for this project I'll stick with welding, which for me is walking on known steady ground.
The welding is difficult, though, with such thin material -- I'm sure I would never recommend this project for beginners. That's one thing that would change radically if a good bonding agent could be found, just as you've suggested, Bruno -- the project would be more like model building than blacksmithing, sort of like I was getting at with Maggie Muggs. Since then, I've found that the J-B Weld is not really adequate to the task, as Al has put forth above. The first photo below shows the flare I made on Elektra I out of J-B Weld -- beautifully formed, I think [I used a carefully undersized rubber flat washer forced on as an outer mold], but it was unstable when conducted heat got to it, becoming rubbery and detaching from the tube edge [I should have reinforced it somewhat down around the tube wall]. The fuel pipe was added later, by filing a groove and bonding it in with more resin.
The second photo shows forming of the fireplace cement flare. It handles like damp clay or putty, so you can just mold it around the tube as shown, building the fuel pipe assembly right in as you go. In the third photo, after curing for 24 hrs, you just smooth up the inside surface with a half-round file. The finished surface is more like fine sandpaper than a polished surface, as you might expect, but certainly should be good enough to guide lots of air down the chute. I expect this to hold up a lot better than the earlier epoxy attempt.
- The intake flare molded from J-B Weld epoxy, finished except for adding the fuel pipe assembly. Photo Copyright 2004 Larry Cottrill
- ElI_JBWeld_flare_finished_crop1.jpg (32.78 KiB) Viewed 6040 times
- Hand forming the flare out of silicate fireplace cement. The fuel pipe assembly is simply worked into the process as you go. Photo Copyright 2004 Larry Cottrill
- ElI_making_cement_flare_crop1.jpg (61.17 KiB) Viewed 6041 times
- Finishing the inside surface of the fireplace cement flare with a small half-round file. Photo Copyright 2004 Larry Cottrill
- ElI_finishing_cement_flare_crop1.jpg (52.46 KiB) Viewed 6041 times