2019 Miata engine investigations.

The fact that there are people who are this much more knowledgeable than I am and have the time and energy to investigate things like this on this board is no surprise. What's always made me smile is that most of them will use their knowledge and time to analyze things like this and then come to the board and share what they found back to us.

Thanks, Keith, Alpha, Knurled and all the others who regularly do this.

 

Talking to the Mazda powertrain engineers (and reading between the lines), it seems like they work on one of the production engines at a time. Once the ND came out, it was all about the 2.5T in the CX-9. Then once that was done they apparently turned their attention back to the 2.0.

The skunk works are the weirdo engines. Mazda's put more weird gas engines into production than just about anyone in the past 40 years, especially considering their size. Miller cycle supercharging, rotaries with sequential turbos and soon HCCI.

In reply to Keith Tanner :

Maybe because of their size.  Smaller companies have to fight harder for market share, and they don't need much of it.

Wasn't what nearly killed Subaru their attempt to be Toyota?  That was a dumb move, Toyota is already the best at being Toyota.  They changed marketing tactics in the late 90s and picked up nicely.

Part 2! This is a lot less exciting, as the exhaust side is a lot simpler.

This is new (right) vs old (left). Same cat, same flanges, same fundamental design, but the primaries and secondaries are both 2mm larger in diameter.

The ports themselves are 2mm larger by our measurements. Looks like they merge a little sooner than on the ND1. We already know that there's a large exhaust valve down there along with reshaped ports to avoid flow separation. You can see the diameter change by looking at the soot around the port - the gaskets are the same.

ND1.

ND2. Sorry, I didn't get quite the same angle on the engine. The ND1 was on a stand, this one's still in the car.

A nice detail update. The heatshield is exactly the same shape, but the grommets have changed from being able to slide around (the lower PEJ5 shield) to being springloaded for easier installation (PEDA). If you want a really nerdy upgrade for your ND1, here you go :)

A little less nerdy - the ND2 has gained an "oil cooler". It's a heat exchanger that allows the oil and water to trade heat energy. It's probably more useful assisting oil warm-up in cold climates than trying to keep a hard-working track car cool, but there you go. It's a similar unit to those that first appeared on Miatas in 1994. Oddly, the engine appears to have been designed for it from the start. This is the thermostat housing for the ND1. See that freeze plug? Pop that out and press in a nipple, and that's where the coolant comes from. There's another fitting on the front of the engine that's molded plastic. It's plugged on the ND1, but if you just drill out the plug then you can install the new heat exchanger. The cooler itself is a sandwich unit under the oil filter.

We already know that this car responds very well to exhaust modifications - this is a bolt-on test of our midpipe and active muffler vs stock.

An important take-away from this: our BBR turbo system will fit the ND2. The exhaust ports are a bit small on the manifold, but that can be opened up without too much trouble. Woo!

Keith, thanks so much for sharing all of this information along with the beautiful photos. It's really interesting, even for someone who doesn't own a Miata!

Keith I know there is the CARB thing hanging around these days, with the gains you are seeing with midpipe/muffler, do you think there is anything you can gain with a CARB-legal exhaust header?

...maybe. It wouldn't be a cheap date, as you'd either have to retain the stock cat (best choice, welding is needed) or come up with a new one that can meet the specs. Technically, it's not legal on the federal level to change out a functioning cat but I can't see that being a big problem if it had been through the EO process.

So you really only get to change what's ahead of the cat, and you have to make sure it gets up to temp quickly. That doesn't give a lot of room to mess around. So...maybe. Unless that bit ahead of the cat has a turbocharger in it, in which case it will definitely make more power

This car may respond really well to a primary cat delete...I mean, a header. We won't be testing that. Thought about it when we were putting it back together this afternoon, as we have a couple sitting on the shelf. But we have to justify what we do from a business standpoint, and since we can't sell either of the options we have there's no point in putting the money into testing them.

Makes perfect sense, my curious side just always wonders.

Last question, can't remember if I've asked you before regarding the '19s.

Do they have the same overhead in the fuel system, for a tuner that is willing to take the risk, for an E85 tune?

Your exhaust plus an E85 tune could possibly be a legitimate 200whp NA ND2 Miata.

I really had to think about the header question when I was standing in front of the car. It would have been so easy to chuck one in, but the dyno runs and the subsequent removal just couldn't be justified.  If we weren't so strapped for time, I would have done it.

We don't know the answer to the fuel question yet. We're looking at the injectors to see if we can get any info on how they changed. All but the first ND1s had a fuel pump that can support 500+ hp, so that's unlikely to be a limitation.

As soon as there's a legitimate 200 whp, everyone will want 225 :)

The factory setup actually looks really, really good.  Nice collector, nice tubes.  Not like the stuff you see from other companies where they shove a whole bunch of tubes together, or put a lot of pinches in them for ease of machine assembly.

You know what?  Buy new manifolds from Mazda, bead blast them, have them ceramic coated on the outside, and sell them as upgrades...

Keith Tanner said:

.Technically, it's not legal on the federal level to change out a functioning cat but I can't see that being a big problem if it had been through the EO process.

This is a bit of law I've never quite understood.  Why should they care if I replace a working cat with another working one of proper spec? 

In reply to rslifkin :

Not all cats are equal.  OE cats tend to be of much better quality than aftermarket.  There are some vehicles that simply will not pass emissions with aftermarket catalysts.  They're technically catalysts but they are not good enough for the application.

Cheapy aftermarket cats aren't $300 and OE cats aren't $2000 just for the hell of it, as it turns out.

"Of proper spec" is the kicker, of course,  Which is where 90% of aftermarket cats now have the disclaimer 'This is not legal for sale or use on California spec cars', because California now has a law that says you can't sell crap and claim it is good enough.  The aftermarket mostly just bow out so they can continue to sell parts that are not OE-quality.

In reply to Keith Tanner :

There is SO much effort put into those headers to get the best trade off between flow and catalyst light off, it's not even funny.  That includes the shape of the catalyst inlet area- so angles make sure the whole front face is being used evenly (which keeps it working best AND reduces hot spots that will prematurely age it).  

Unless you guys have a really, really good CFD program to make that happen, I'd think that changing headers will be a LOT harder than changing the parts downstream of the catalyst.

In reply to rslifkin :

In addition to what Knurled pointed out- the availability of good catalyst in the post OEM market is thin to none.  Not that it helps that very few of the people who are selling aftermarket catalyst have any clue what is in the products they sell.

Which is another way to say the "proper spec" may just be the original one on the car, and that's about it.

If you want to know the difference between OE and aftermarket cats, just try selling them to recyclers. They'll pay through the nose for the OE parts and often won't even take the aftermarket ones.

Alfa, I love your input on this stuff. With our emissions testing over the years and through conversations with you and other OE powertrain engineers, I've gained a pretty good appreciation of just what the OEs have to balance and how hard it is. Also just how far ahead the OEs are right now, I swear there have been a couple of generational leaps in tools that haven't really made it to the aftermarket.Watching an ND in cold start mode or at cruise really tells you how complex the control is, as it juggles valve timing, ignition timing and throttle plate control to keep the cat at temp without affecting the driving experience. 

In order to get the BBR turbo through the EO testing procedure, we had to wed the OE cat into the downpipe. This means we have to do an exchange program with customers and weld THEIR cat into the downpipe. We can't do a core exchange program because 1) we can't guarantee the health of the core and 2) we can't sell a used cat. We can build them a new downpipe with a new OE cat which is technically not legal but unlikely to attract attention because it'll possibly even improve emissions over the used one they pull out.

Also, to give you an idea of the level of testing done by the OEs - there are very few labs where aftermarket companies like us can do the testing for an EO. Really, we all use the SEMA lab. They have one emissions testing cell, I believe. Mazda USA has something like 30, and that's a small automaker that does most of their powertrain R&D in Japan.

Here's our downpipe in the welding jig for those who are interested.

I wonder...  From a "cost is no object" perspective, what tricks like ceramic coating the header to retain heat could be used to help with fast cat light off?  

I know there's at least one Miata turbo kit from another vendor on the market that had to resort to that to get through testing. Not that the ones you buy have ceramic coating...

In reply to rslifkin :

Retaining heat isn't actually a big deal for light off.  By the time you need the catalysts hot, the ceramic coating benefit won't be effective- even with a very thin wall manifold, there's enough heat that just has to heat the metal up before any external effects matter.  It all happens in 20 seconds.

We've looked into many cost is no object ideas.  

Flow and flow control is far more important than any heat retention actions.

Keith Tanner said:

I know there's at least one Miata turbo kit from another vendor on the market that had to resort to that to get through testing. Not that the ones you buy have ceramic coating...

Turbo flow is a little interesting.  What I will suggest is for you to look at the turbo design on a BMW that is SULEV certified.  Do your best to copy that design- it's pretty awesome.

So 16whp from a cat-back, or technically less than a cat-back? That's impressive. Mazda must be using some extra magic sauce to get their driveline loss below 10% too, because that's like super rare dawg. 

Keith Tanner said:

Also, to give you an idea of the level of testing done by the OEs - there are very few labs where aftermarket companies like us can do the testing for an EO. Really, we all use the SEMA lab. They have one emissions testing cell, I believe. Mazda USA has something like 30, and that's a small automaker that does most of their powertrain R&D in Japan.

To expand on this a bit, I work in an OE testing facility. We have around 90 engine dyno's. Some focused on emissions, altitude testing, Hot/cold testing, tilt stand testing, etc. We test 24/7, and my employer has so much demand for testing that we can't meet the need so they pay to do some additional testing in outside facilities. We spend in the neighborhood of $700k annually on fuel alone (and when you're buying fuel by the tanker full, you get a better deal). This doesn't count on-road testing, testing at other company facilities, or outside testing facilities. 

Thanks Kieth for taking the time to let us geek out with you. Really enjoying it!

 Yeah, my comment was on emissions specifically. I've got some friends at GM powertrain who drop by when they come through town doing high altitude testing, and seeing both the quantity of instrumentation and the level of specialization of each of those guys is fascinating. This is after all the computer aided work is done and the rubber is literally hitting the road, of course.

Vigo said:

So 16whp from a cat-back, or technically less than a cat-back? That's impressive. Mazda must be using some extra magic sauce to get their driveline loss below 10% too, because that's like super rare dawg. 

Right from the start, the NDs have dynoed high. It's an efficient drivetrain, and we're not even testing in the 1:1 gear. Why not? Because 6th is the 1:1, and that means wheel speeds of over 180 mph on the dyno plus the fact that the engine starts shutting down early as it gets into implausible operating ranges. I think that Mazda may be underreporting the output a bit as well.

What impressed us about that catback test was not just the peak gain, but the gain through the range. It's all killer, no filler. We went back and tested the ND1 with the same parts (turned out we'd never done that specific test for reasons) and it did well but not that well.

Edit: ND1 with the same exhaust mods.

In reply to STM317 :

We also have a dyno facility with somewhere around 100 engine dynos.  But most of the emissions testing really happens in a lab with somewhere around 20 chassis rolls, running all the time.  Some have environmental development, too- temperature and pressure.

That does not count the off site chassis dyno facility, which I don't know how many sites it has.  Nor the ones that everyone uses dotted in a few places around the country.

For each car, there are thousands of hours of engine dyno time, and thousands of miles of chassis dyno testing done.  I just looked up one truck, and it has over 500 emissions test run on it.

BTW, there are some more emissions site for hire places than SEMA.  Although, most of the ones I know about are here in SE MI (Roush, Prodrive, AVL, etc)

Right, we did look at one or two other labs but Michigan is twice as far as CA I seem to recall there were some other problems - they couldn't do specific tests as well IIRC. Or maybe that was one of the other labs a bit closer. Regardless, SEMA worked out to be the best option for us and they're really good to work with.

Roush knows a lot about building emissions legal stuff, we've had some good talks with their braniacs.

Now I'm going to go look at BMW manifold design...