Mazda Miata ITB Project

Regarding the question of underhood air temperatures, the general rule of thumb is 30-40 deg C above ambient depending on the measurement location and the relative position of the hotter engine components. As long as the vehicle is moving and you have airflow that increase over ambient is reduced. Worst case underhood temps are reached with hot ambient temp, and during a hot soak after full load operation (engine off, sitting with hood closed). During such hot soaks under hood temps increase significantly with excursion temps as high as 100-140 deg C.

In addition to the plastic's rated deflection temperature, keep in mind that heat cycling of the plastic (even though the rated temperature is never exceeded) may make the plastic brittle overtime.

What material do they make those phenolic spacer things out of?

If i was building a plenum from scratch I'd either use all aluminum or a two piece design with aluminum base and fiberglass cover.

If you run into clearance issues keep in mind you can get shorter velocity stacks. That said, i think we can all agree that those Alpenhorn looking BMW stacks are cool AF.

peter wrote: I'll be replacing the stud in the head with a bolt/screw that has a narrower head. The surface for that head will be a step below the throttle-body-to-flange mating surface, so the TB can overlap the bolt by a bit. So far, it's actually looking pretty decent.

What I might recommend, is instead of a bolt, install short studs to locate the TB's and use a longer bolt to sandwich over the TB flanges and adapter directly to the head.

It will require a load spreading device and a spacer on the adapter to work properly, but will save you the headache of having to remove the TB's individually to chase leaks between the adapter and the head.

Really, there are a lot of ways to approach it from that perspective.

Thanks for the input folks! Very useful.

I haven't had much time in the garage this week. But I did steal a few minutes this evening to putter around.

Tomorrow night I plan to go use the laser cutter for a couple of things. The big plan is to cut/etch a couple dozen drink coasters for my employer. They're pretty simple to do, once the artwork and settings are right. I take our logo, make a few tweaks and hit "print". Then just sit around for 90 minutes or so and make sure nothing catches on fire.

I did a few coasters for my own personal use. 1/8" MDF is cheap and cuts/etches beautifully.

I'll also use some scrap to cut a mock-up for the throttle wheel. This will give me an endpoint for the throttle cable, which will in turn help me design the bracket that will hold the cable in place. I've measured pretty darn close to 45mm of throttle cable travel from fully-closed to fully-open. Since that linear motion needs to turn into 90 degrees of rotational motion, I made the circumference of the wheel 180mm. Or thereabouts. These five pieces will get assembled with either glue and bamboo skewers or 4-40 screws. Sweet.

The small amount of time I spent in the garage tonight was used to get the layout of the two bolts I plan to use for my throttle cable bracket. They're the top bolt hole in this picture, plus the hole for the bolt in the center of this pic. I took this to capture how Mazda braced the engine hoist bracket against the head, in case I want to steal their idea. These photos are courtesy of the '94 engine I have on a stand next to the car. It's a pretty useful stand-in for the '01 motor that's in the car, but I do have to be wary of the differences. I think I'm pretty good for these two bolts.

Pencil rubbings aren't just for grade-schoolers in the local historic cemetery.

I hope to knock something together that I can cut tomorrow night on the laser. I know that real men make these sorts of templates out of cardboard with an Xacto knife and scissors, but... I feel much more comfortable with things that meet at a solid 90 degrees and are cut square and to exact dimensions. That's a personal limitation that I am slowly working on relaxing

Crackers wrote: What I might recommend, is instead of a bolt, install short studs to locate the TB's and use a longer bolt to sandwich over the TB flanges and adapter directly to the head. It will require a load spreading device and a spacer on the adapter to work properly, but will save you the headache of having to remove the TB's individually to chase leaks between the adapter and the head.

Last thoughts for the night. I have been really hoping I could build something where I bolt the throttles to the adapter on the bench, then carry the whole shebang over to the car and bolt it on there. I've had some similar thoughts, but hadn't gotten to the idea of a load-spreading device like you describe. Definitely something I'm mulling now.

I took this picture as straight-on as I could. I fed a short M8 bolt in from behind the MDF prototype, with the throttle body bolted on (using incorrect hardware - the bolt/nut dimensions are not right), to illustrate the interference at the frontmost stud. On paper, the edges of the holes for the Mazda stud and ze German stud/bolt are 3.15mm apart. If I were to do what I think you're saying, I'd still have to trim off a lot of the metal around the throttle body bolt-hole, in order for the stud to pass through/by it. Will continue to think...

Here's a similar picture of the interference at one of the inboard studs. Simple to fix with a rotary tool and the proper bit. I can probably get away with cutting off just the little hump there, leaving the bolting surface there untouched. (ah damn, uploaded the wrong pic, this should have a bolt in the MDF, as above. Too tired to fix)

Another late-night post. Got some time to finish up drawings and do some cutting.

From the front of the car, the throttle wheel rotates counter-clockwise to open the throttles. I think I'm going to end up with the throttle cable very nearly vertical at its engine-bay end. If I can manage it, I'll try to do some junkyard shopping tomorrow morning for a new cable.

Here's the end of the throttle rod, with the MDF wheel attached, and the MDF plate bolted to the back of the head. The throttle shaft has been set one cylinder back, so that all four throttle arms line up with original holes and there's a spare cylinder's worth of shaft at the front and at the back. This setup clears the TPS sensor nicely and miraculously doesn't try to occupy the space something else owns.

I'm mulling over how to attach the wheel to the shaft. Any and all input is appreciated. The end of the shaft is threaded for a tiny screw, and there's a snap-ring groove approximately 15mm forward of that. On the original car, these straddled the boss on the rearmost cylinder, positively locating the shaft and limiting fore-aft travel. I don't think the tiny screw is up to the task of forcing a friction-fit throttle wheel against the snap-ring. There's a hole approximately another 15mm or so forward of the snap-ring groove - that's where the throttle arm for the #6 cylinder used to attach. I could have a really long snout on the throttle wheel I guess...

It's been a busy month since I last posted - it's hard to believe it's been that long.

This project is still my priority, but progress has been sporadic.

I took a couple of "Safety and Basic Use" classes over the Fourth of July weekend. Re-learned how to use TechShop's lathe and how to use their water-jet. I made a top that doesn't spin and a bottle-opener that doesn't open bottles.

I also finished my second workbench - the first was a scratch build, this one I got two lab bench cabinets from a used office furniture liquidator and topped it with a slab of IKEA butcher block countertop left over from the first bench.

I finally got the very short-legged 90-degree cast aluminum elbows I ordered off eBay. I'd done some measuring and thought I'd be able to use them to connect the throttle bodies to a plenum and the bellmouths (which one of my friends kindly refers to as "she-wee"s). I built a plenum mockup and quickly realized that this was not going to work. At least my skills with Fusion 360 are improving. I highly recommend the package if you want to do CAD and CAM - I haven't used its other features yet. (You can get a free hobbyist license).

I decided to make a template of the area I have available to me between the throttle bodies and the body.
While doing that, I realized that inverting the throttle bodies has a number of advantages - it would put the throttle rod and balancing tube on the bottom, where I can't put the plenum anyhow, leaving me more open space above. Through a happy coincidence, it also results in the throttle bodies being darn close to level, which reduces the change in direction required to enter the throttles.

Inverting the throttle bodies causes some challenges with the mating flange. The frontmost throttle body bolt hole interferes significantly with the frontmost intake stud. I came up with a two-layer design. The first layer bolts to the head using the stock studs on the bottom and countersunk pan-head hex-drive bolts at the top. The throttle bodies bolt to the second layer, which is attached to the first via a set of M8 studs. I cut a partial mock-up of the rig last night on the shop's laser cutter. This has the final faces to the head/throttle bodies, plus the correct total thickness. My mockup throttle body shows this as a go. I'll double-check for interference with the throttle shaft and balance tube, then test my CAM path by making these in MDF.

Update?

It looks like you've already gone a ways down the throttle cable conversion but this is an off the shelf way I've seen folks have success with for swapping this motor and converting to various other ecus:

https://www.achillesmotorsports.com/BMW-S54-Drive-By-Wire-Throttle-Cable-Conversion-p/11-am-tck-s54.htm

Subscribed for more updates!

In reply to peter :

Here is a video I put together on ITB tuning.  See what you guys think.

https://www.youtube.com/watch?v=_GDZxUVTplk