Exhaust manifold cracking, can ceramics, other coatings, or AI help?

[SpinningDorito]

Let me start out by saying that this is my first post, so please, be gentle. I have searched the forum on my main points, I found the threads to be informative but I want more depth of knowledge.

The car is a mildly modified 1993 Mazda Rx-7 (engine code: 13b-rew), I have an Apex'i Power FC taking care of fuel maps and a Profec B Spec-II monitoring boost. The stock intake box is slightly modified stock intercooler (garbage), exhaust 3" from the turbos back with a high flow catalyst. All emissions gear is in place. Turbos are jury rigged non-sequential (parallel).

Which brings me to my question:

I'm going to be removing the TT assembly and finishing the non-sequential conversion- Entirely removing the flapper door that stops the secondary turbine from pre-spooling and boring out the wastegate. I considering going further and having the manifold extrude honed but the OEM cast piece has a history of cracking and extrude hone will only aggravate that.

I want to ceramic coat the manifold and turbine sections to lower radiant heat and potentially keep the heat in the mani. Is there anything that can stand rotary egts that could be sprayed on the interior of the manifold to prevent the manifold from heatsoaking and cracking?

I've been considering alcohol injection to lower my egts but a good kit is upwards of $1000, considerably higher than most enthusiast grade coatings.

Anyway- if you can decipher my post please share your thoughts.

Link to pictures of 20b-rew with separate pictures of it's very similar TT assembly:

http://rotan20b.tripod.com/id12.html

Link to explanation of cracks:

http://www.mazdatrix.com/faq/tbcrack.htm

 

[btrueblood]

Wow. That center splitter vane will get really, really hot, probably glowing. And it's trapped on the ends by relatively cooler walls and flange. Any way to eliminate the split?

 

[btrueblood]

And if not, then consider trying to build the part as two seperate, thin-wall tubes attached (welded?) to the flange. You'll want a suitable high-temp. alloy. The idea of two pieces is that each tube is then free to thermally expand/contract "radially" without restraint from the opposite wall, and the temperature difference from the splitter wall to the outer wall of the tube should be reduced (somewhat) by air circulating between them.

 

[Bribyk]

Could you not cut a slot across the splitter vane so that it's free to expand in from either side of the slot? This would eliminate your thermal stress on this part.

We have a vane like that in our shop right now where two turbo exhausts converge and this once-straight vane is now s-shaped and the metal is chunking off like tempered glass.

 

[Baldturbofreak]

Perhaps time for a Inconel header Und honkin single?
I have used Swaintech coatings on everything that sees fire in my top turbo builds for years(does wonders for BSFC). But the thing that prevents a plasma applied exhaust coating like whitelightnin from being applied to the inside of a header/manifold is that it must be sprayed at a 90 deg angle. If it's not it wont "burn in" to the surface and be free to dislodge. Ceramic thru a turbine spinning at 150K is no good no matter how small.

Ive never seen a divided turbo header or turbine housings for that matter that sees high egt's (turbo rotaries are 1800-2000F) that ISN't cracked in the divider. So long as the crack doesnt propagate to the outside, in my opinion dont worry about it.

 

[SpinningDorito]

I've been considering having a custom barrel style mani fabbed out of inconnel.

Picture two l shaped tubes that fit together against each other, with a flange on the end for some smaller GT20ish turbines with two downpipes into a collector. I'll try to put together a simple gif to show what i mean.

I don't know if boring a hole between the two is possible (already done?). I'll bring it up on the rx7 forum. I'm hesitant to bring this question up because there was some interest in a custom tube mani for the stock twins but no one ever came through with a prototype. Most responses end in "switch to a single". I'm trying to fool an uneducated search of the engine bay and maintain emissions legality.

 

[wgknestrick]

Why is the manifold divided anyway? I would make my own out of a 3" (or whatever the size would be) SS heavy wall tube bend. You can crush it to fit the square-ish sides of the port. I would think this would offer a performance increase as well, then just get it coated.

Just draw up your flange in a CAD software or calculate the inner circumference around the flange. You just need to match that to the nearest round ID circumference of a standard pipe size.

 

[SpinningDorito]

Better picture:

http://i17.photobucket.com/albums/b88/ArmenMAxx/DSC_0008.jpg

The reason i can't do a simple barrel mani is space/ accessories, and the turbine housings are more or less integrated in to the manifold.

I appologize for the poor rendering of my ideas but you get the idea.