Filling side ports with actual aluminum?
 

In thinking of some ideas for my PP build, I had something come accross my mind and wanted to see if anyone had any thoughts. I know typically Devcon is used, but I have access to a local forge and don't think it would be too difficult to melt, say, an old aluminum water pump and pour it into the side housings to fill the ports. I plan to dremel in some small wells in the port path to give the alum something to fill into (sort of like tabs) to help it stay put to be safe. Also will need to pre-heat the housing so the temps are at least similar before cooling.
That said, I am limited on my knowledge of metals and whatnot, so I didn't know if anyone had any concerns on the process or how well this would hold up. I can't imagine it being worse than some sort of epoxy but it could be more complicated than that. Thoughts?

melting point of aluminum is 1218F, that is far too much heat for the cast iron, you will ruin any core plugs in the casting and almost certainly warp the irons, all of which would require lapping, renitriting and then recoring. none of which is worth it when the dev con works just fine.

I planned on having them lapped afterwards anyhow.
I planned to pre-heat the irons to help minimize any warping. I know there will still be a temperature difference at the port channels during pouring, but wasn't sure if that is enough to warp it on a larger level where the faces are.
It may be something I have to grab a bad iron and test with to get the process down. It was just a thought. What fun!

up to you, but devcon works great, so why reinvent the wheel

The problem you may run into is that aluminum has a wildly different thermal expansion.

It's not typically much of a problem in an engine, within certain bounds and with proper engineering care. However it's another thing when you fill a hole with 1300F aluminum and then it cools down to room temperature.

The contraction is significant enough that casting cores have to be made oversize relative to the final part dimensions, and there is a whole subset of engineering dedicated to designing castings so that they don't have residual internal stresses from cooling at different rates.

It is usually okay to plug exhaust crossover ports in iron cylinder heads with aluminum, because those passages tend to be complex and the aluminum plug isn't going to go anywhere if it loosens up. I'd be leery of doing it with an intake port where if it loosens up it could fall out.

As mentioned before, Devcon or the 24 hour setting JB Weld are proven to work well for this purpose.

Also, remember when filling the side ports to leave a depression in the side port opening "face" so that the seals on the rotor can blow down their sealing combustion gas pressure and so carry carbon back out from behind the seals.

Risk & cost vs reward & benefit

if you’re open-minded in your assessment then the reason why people with unlimited budget and capability aren’t doing it should be obvious. The idea isn’t novel. People aren’t going there for good reasons.

Also any heat treatment done to any part of the metal will be ruined. I am still a newbie on the rotary, but as far as i know the side irons have a steel plate on it right? And if you heat this part to much it wil reset the heat treatment it got from facotry, resulting in a softer surface that will wear much faster. Also as mentioned, you cant simply cast in to a hole, since the metal will shrink after cooling. Tho, the shrinking of a 50mm opening would only be 0.76mm from liquid aluminium (660C) to room teperature.

That's still a lot of shrinkage. I'm with peejay on this. The aluminum will not like going through the nitriding process and the nitride will not like the heat from the molten aluminum. My two cents...

I had the same idea quite a few years ago wrt sealing up the Renesis side exhaust ports while using 13B peripheral exhaust port rotor housings instead. As much as I wanted to believe it might be possible the mechanical engineer in me finally accepted how ludicrous it’d be to attempt it that way. In that case epoxy isn’t a suitable alternative due to being exposed to the heat and so on. Ultimately you need plates designed with that in mind instead. Which in the case of an earlier 13B based engine if you really think it’s that important than just go buy a set of RB aluminum plates that have those blocked just for that purpose and get the weight savings too. The potential problems & issues with the proposed aluminum casting technique are much greater and with significant impact on durability, longevety, etc. than the OP is recognizing.

Ok, thank you for all of the replies. I did realize that if it were easily done then it would have been, but I wanted to ask anyhow since I have the forge available.
I wish the alum side housing weren't so expensive, as that's the obvious perfect match for the PP and seal, but it is what it is.
Have a great Thanksgiving!

Nice, but not the same thing as being discussed with the side plates

But you *could* mill the intake ports and then mill press in block off plates as Chips Motorsports does for the rotor housings.

Stock ports probably wouldn't need to be milled, but I am not sure of stock Mazda milling accuracy.

Wow, those clamps are like $200 a pop...

I have no idea why someone would go to all the trouble, but the "machinist's" idea would be-

Machine a plug for the intake port that is an interference fit into the port (machine the port if necessary for precision) and has a cylindrical (or oval if you want to it to guide the plug axially as well) stepped "finger" attached to the back side.

The "finger" is to guide the plug in straight, prevent sideways kick-out (since the side housing material that interfaces the plug is quite thin) and set the plug depth with the step).

Plunge cut the side housing inside the port "bowl" for the finger, face cut the "bowl" for the "finger" step, freeze the plug while heating the side housing to its limit, press the plug in and machine off from the "outside" of the engine the protruding "finger" if it interferes with flywheel or waterpump housing. For the intermediate plate you would have to have a pair of matching "finger" and a... erm... "ear" (yeah, ear...) machined into the backsides of the plugs.

Personally,
I would just clean the side housings well chemically and mechanically, rough up the ports if using a previously ported plate (which I wouldn't because I always port to drop the leading side seal which is not desireable for a P-port motor) and fill the port in with a fuel proof, metal filled epoxy with the same thermal expansion rate as the cast iron side housing.

ie, the way it has been done for the last 50 years of racing.

Even Mazda Factory Race engines for production classes would have used the epoxy technique (though with epoxy matching the Aluminum side housings thermal expansion rate as they used early production Cosmo Sport 10A Aluminum side housings).

You should see some of the old Pro Stock heads back when they HAD to use production line castings. Maybe ten-fifteen pounds of epoxy in the intake ports, no welding. Fill in the bottom half of the intake port, then port upwards until you almost hit air, then get some more epoxy up on top, then continue porting...

All we're doing is filling in a hole, nothing fancy. Don't need to get fancy if your problem isn't fancy.

- end thread -