No no no! I reckon that is going way overboard, or more correctly, back-to-front. I haven't watched the vid (45 mins!), but my initial reaction is that there is a better way, with less downside.

If you are going to duct it, I think it should be a slot, area about - puts wet finger in air - one third the area of the face of the cooler. The air coming in slows, and passes through the cooler taking out more heat per unit of air that if it is being forced through even faster that the airspeed of the car - which is what that design would do. It may work, probably will, but is terribly inefficient, stealing air that can be doing other work - cooling brakes, feeding the cooling stack, loading the splitter for DF, and probably adding to overall drag to boot..

There is more than one way of arranging it, but see the arrangement here for the general idea, narrow opening ducted to much larger radiator surface.

 

Johnny, I won't forget all you guys when I'm racing IMSA or F1 next year

Wait, you're telling me there's more complex physics and science at play here!? This is why I love this site. Thanks for taking the time, Gee! Makes perfect sense. I'd always thought of it the opposite way, just figured faster air = more air passing through the exchangers = more heat bled off. I wonder if any other factors would be at play if we were to make a shroud for the oil cooler the way you just described. Mainly because I believe the same effect you described is already going to be occurring with the airdam opening and area behind it, right?

I'm still gonna run some back to back tests with it still and report back to him. Although upcoming projects might render a second version unnecessary..

 

I think this all comes back to the 1:3 golden ratio (or was it 1:4?) for mouth opening:heat exchanger ratio. The same as your bumper cut to feed the radiator. That said, these are garage built miatas with DIY aero so not everything is going to be optimized, or can be optimized even. With where you mounted it I think that's gonna work well for you, the fast air will get it through the oil cooler and the air will still be cool enough to cool the rad. Not optimal but probably better than what you had - I guess your data will tell us if that's the case!

 

I've not heard of a 'golden ratio', but the drawings I have seen point to that.

I was going to comment on the obstruction to the radiator, but when I looked more closely at the photo, the ears may in fact smooth (duct?!) the entry to the front of the rad!!

Yeah, we don't have the car volume, knowledge, or resources to to match F1 practice, so we just do the best we can. In this respect, I'd be looking at CLT and oil temps, and any indications of front DF and drag impacts - the former pair straight data logging, the latter more subjective

 

I'm fairly sure Emilio quoted the ratio as 1:3, but I bet there's variables at play that result in the "golden" ratio not being set to one specific number.

I guess we'll see what the final result is once the car's on track again! Like I said, I've got plenty of headroom with coolant temps to trade off for lower oil temps. Both are perfectly in check, really. Last weekend I peaked at 198*F CLT and 225* oil temps in 85*F ambient temps. Although it's worth saying that was on a super high speed, wide open track. The weekend of 10/17, I'll be at my buddy's track event at the tight and technical "Streets of Willow" on Friday, then the NASA event is going to be at the new "Circuit" at Buttonwillow Sat/Sun. I'll try to do back to back sessions with and without the shroud at both tracks. Will probably make it up to the mountains this weekend as well and do some "testing". My temps honestly get higher during spirited mountain driving than on track so that might be the best test of all.



Got the car back in its "street" configuration earlier this week. Swapped my street pads, wheels and tires back on. My RT615's were dead and I currently have two sets of mildly-used RC1's so I mounted those up and will run those on the street for the foreseeable future. I also spliced a new connector in and reinstalled the factory TB onto the Fab9 manifold. I had to order a 2.75" to 3" coupler over the weekend to mate the TB to my 3" MAF tube. Got a blue one and it looks horrendously rice-y. Oh well.



I took some datalogs on the way to work yesterday and the MAF readings with the OEM TB are dang near identical to those with the 68mm Bosch. MAP readings might be 1KPA lower on average, it's hard to tell. Overall, I don't think I'm missing anything with this TB. A high compression or forced induction 2.4/2.5 might be the minimum needed to see any tangible gains.

 

That looks like something that might be fun to swap out on a dyno some day, back to back comparisons.. for science!

 

The research I've done on this supports a bumper opening of about 1/3rd the frontal area of the radiator core. We have successfully run less on well optimized systems in some of our race cars. Those optimizations might include fully sealed custom ducting, removing any brackets, frame reinforcements or other stuff that might interrupt airflow, E85 to lower cylinder head temps, thermal rejection coatings in combustion chamber and/or exhaust ports, vented hoods and so on.

 

Yo, thanks for chiming in Emilio! Dang, I should really measure my inlet vs radiator area. I've got most of my other bases covered aside from the really fancy stuff like thermal coatings...

So after our Sunday morning canyon run ritual today, the car started developing a driveline clunk when transitioning on and off throttle. I was really bummed on this initially, as driveline NVH can be both difficult and time consuming to diagnose. Luckily for me, the issue was pretty glaringly obvious as soon as I jacked the rear of the car off the ground.


Axle nut on the left side was just barely snug when I pulled it off. Driver side rear bearing is toast. Bummer. I could've sworn I hit both the axle nuts with a 1/2" impact just a couple months ago, but maybe it was more like six given how my memory works. I put a dimple in the nut too, but not a big enough one apparently. Oh well. I ordered another Timken 510051 bearing from Rockauto and will swap it in this week. Only bummer about these Mazda Motorsports rear hubs is that the bearings are kinda expensive. I paid $102 shipped for this one but they're generally around $100-$130 before shipping and tax everywhere else, unless you want to get a Detroit Axle or knock off bearing. I already pulled the knuckle off the car and will swap the bearing once it comes in later this week.

I've also torn a new page in the pursuit of the last few horsepower from this engine platform. Went into the MZR/Duratec book of knowledge and ordered a kit to build a longer intake and squeeze that last bit of torque out of this thing. Started off with a longer straight 3" MAF tube and some couplers.



Then added an apparatus to make sure intake temps stay at ambient no matter what.



This looks like a restriction to me but we'll do some testing with and without it to see what its effects are.





Fancy new exhaust toooob.



Fancy new paperweight.



And here's an unrelated picture I took of our W.A.S.P. 40th Anniversary Tour banner.



Alright, the gig's up. Or what little bit was left of it. I ordered the complete Kraken kit from Michael back in May before the 20% off first month promotion ended. I couldn't pass it up, ended up paying under $3,600 all in including tarriff fees.

Basic overview of what I got:
-Kraken cast manifold
-PSR G25-660 turbo, Turbosmart 40mm external wastegate, 7, 11, and 14psi springs
-Full 3" exhaust with a 200 cell cat and "superstreet" muffler (the smallest of three muffler options)
-Full intercooler, charge piping, air intake, BOV, oil and coolant lines

Really just the complete A to Z kit minus injectors.

My hope/educated guess is that this kit will make ~280whp at 7psi on E85. My fear is that it might make substantially more and take me into bottom end/transmission danger territory. Michael claims this kit made 280whp at 10psi with a PTX2860 on a stock 2L running pump gas (note that MBT on these engines is pretty low, easy to get close with boost even on pump gas). Even if that claim ends up being a little inflated, I think it puts me on track to be right in that realm with a few less PSI. Bigger motor, bigger cams, bigger turbo, E85... blah blah blah. If it makes too much power, I can always limit it via throttle opening, but maybe I'm getting too ahead of myself.

Current plan is to survive the 3-day track/race weekend on 10/17-10/19, then start tearing the car apart that Monday and get the turbo system installed, tuned, and hopefully ironed out for Global Time Attack at Buttonwillow three weeks later. Is it the most conservative plan? Probably not, but it'll be exciting and will get the blood pumping. I should note that I'm still waiting on the manifold, cat, and BOV, but those are slated to land in under two weeks from what I'm told.

Sorry to everyone who answered that I should stay N/A when I asked about power-adders. I hope you can find it deep in your hearts to forgive me

 

Oh my God, I knew it was coming - but didn't expect this update

Yesssssss. Really excited to see this unfold.

 

I knew it was coming!

Looking forward to the results. That thing is going to RIP!

 

It was only a matter of time but I also didn't expect it so soon! I'll be very happy seeing you smash more egos and all will be well in the world.

 

Me boosta.

 

I saw the intercooler and thought you were insane to run an intercooler on an NA setup

Can't say I'm surprised, but I'll be really interested to see where this goes. Any reason for the 660 over the 550? Also, just curious, what clutch are you planning to run? I know one of our local NC guys said he had a clutch that would handle a crazy amount of power (north of 350 WTQ if I remember correctly) and still felt like stock, but I can't remember what it was.

 

Haha, thanks all! Hopefully this'll keep the thread exciting (but not too exciting).

Sim, I did a little research (had some guys in the know explain both turbos' compressor maps to me) and asked around and basically got the answer you might expect. 550 will spool faster (really fast on a 2.5 especially with a low A/R), but the 660 should produce less back pressure and lower EGTs even if I'm only running it at 7-8psi on this engine. I did some spitball math from my current MAF sensor readings. 147g/s at redline x 1.25 (to account for the MAF scaling I did to fool the ECU into thinking the 2.5 is a 2.0) = 184g/s, converted to lb/min = 24.2 lb/min naturally aspirated. Multiply that by 1.47 to loosely account for 7 extra PSI and you get ~35.6lb/min airflow at redline. This puts the 660 in a higher efficiency island than the 550, even at lower PSI. Also, I hope that the slower spool will keep my bottom end and trans happier as I don't immediately want to start dealing with that can of worms.

I think I did this kinda right. Now we wait for one of the smarter guys to tell me why I'm blatantly wrong hahaha.

Top two maps for the G25 550 and 660 at 7 and 10psi. With the 550, I'm actually off the map a little bit at redline if my math is correct. Bottom map is a GTX2860 with the same airflow at 7 and 10psi, just for sh*ts and giggles.

Edit: I didn't move the upper bar to the right to account for the extra airflow at 10psi Disregard that, it's not the boost level I'm going to be running anyway.







And here we have 7 and 10 psi on a GTX 2860. Uhhhh....

 

Makes sense given the compressor maps. I guess with your ambient pressure being a bit higher that would keep the PR lower as well. I'm going to be pretty jealous when you're making 300 WHP at 160 KPA...

 

Bearing manufacturers say to not use an impact to tighten them. It can damage the ***** and races.

 

Haha.. couldn't stay away from the boost! That sounds like a heckuva deal from Kraken so makes sense to jump on it while it was available. Definitely can't blame ya!

 

Hide yo porsches, hide yo 'vettes

 

Dang, we're going full circle now haha. I was torque wrenching all of my hub nuts until I mentioned it and Curly said to just send 'em with the 1/2" gun. I just popped the new one in at home so no gun used.

Quote:

Originally Posted by SimBa

I know one of our local NC guys said he had a clutch that would handle a crazy amount of power (north of 350 WTQ if I remember correctly) and still felt like stock, but I can't remember what it was.

Oops, glossed over this. I've got a Supermiata clutch in the car right now with an organic disc. Emailed Ed and he said I could get a 4-puck if I wanted, but that they're holding upwards of 360whp on their organic disc in Kaiko currently. I think they were hitting ~260wtq. Not sure where my torque figures are gonna land, but I think they'll be close to that. Great news because the organic disc/SM clutch combo feels indiscernible from the stock clutch. Lots of guys seem to run the ClutchMaster FX350 or Uniclutch on these things, but there's quite a few complaints of the FX350 being juttery and the Uniclutch... exploding. Wonder if your NC guy has a Uniclutch on his car or what.



Manifold showed up yesterday. Purdyyyy. I was informed the catalytic converters are on back order so it's gonna be a bit till I see mine. My kit shipped with a test pipe included, though, so I'll just rock that in the interim.

Final piece of the puzzle, installed a boost gauge in one of the middle air vents. My windshield wiper lever was slightly in the way so I popped the end off of it to create better visibility. I think the move is gonna be to trim the end off and cut out a piece of ABS to epoxy onto the exposed hole. I probably won't end up doing that soon or ever, though, with the amount of work I'm gonna have on my plate after next weekend lol.

Cammed motor is funny. Only pulls 16 inches/40kpa of vacuum at idle. Vacuum increases fast once at cruise, though. Just a bit weird after watching my old BP's pull 26in/hg at idle or something like that.



Ah. Also got my hub bearing replaced. Pulled the knuckle off Sunday night and brought it to work with me yesterday. Didn't realize RockAuto has a warehouse 45 minutes from me, so the replacement bearing showed up quicker than expected. Had the bearing sent to my work and popped it in today, then installed the knuckle when I got home. Other rear axle nut is tight still. Car's gonna get another bolt check this week or weekend to make sure I didn't miss anything else.

 

Welp, got an interesting and unexpected email this morning from our TT leader...


Super big bummer about Chris, the Buttonwillow shop dyno operator. I didn't know the guy but his on-site shop had helped me out a couple of times, welding my throttle Skunk2 cable bracket back together after it snapped on track (twice) at GTA in 2023.

So that's a bummer, but the rest of the email means I must be doing something right... right? I'm committed to staying in TT5 next weekend. There aren't enough drivers in TT4 for tire contingencies to take effect, and I'm kinda leaning on those to subsidize all the other expensive stuff going on around the car. I booked an hour of dyno time this Wednesday at Advanced Engine Dynamix (the tuning shop from Greg Peters' videos) and will be getting the car on the rollers finally. Might mess around with VVT and ignition advance if I have time, but I think the tune is pretty solid and there's not much more to be squeezed out of it.

Aside from that, holy heck I got a lot of work done on the car this week.



Wednesday, I pulled the valve cover to do a quick valve clearance check. All good there. While I had it off, I reoriented my crankcase breather to point towards the driver side of the engine bay, as its previous orientation had it going directly over where the turbo outlet is going to be. I also found a spot to mount a catch can on the frame rail under the intake manifold. Figured I should get that set up before the turbo goes on rather than after. It's a tight squeeze and I just have it zip tied to the body for now, but it's functional. Will find a spot to put another catch can coming from the hot side of the engine bay. There's a lot of space under the intake manifold runners now...



Then Wednesday night I was doing a quick inspection of the engine bay and noticed the intake manifold gasket was sagging around the #1 runner. Luckily pulling this manifold is only a 10-15 minute job and I was able to get to the gasket easily. The gasket wasn't sagging actually, it was...



Alright, good catch there too. I went to the Fab9 manifold install instructions (which I never read before the install) and found that they direct the end user to seal the manifold to the head with RTV. Dang, that's kinda ghetto, right? After some reading, though, I guess it's what everyone's been doing. I RTV'd the manifold to the head Wednesday night and haven't had an issue.

Then on Friday, I got a head start on more turbo things. Getting the block drilled and tapped for the oil drain. There's a few spots on the Duratec block to do this, and most people drill into the wall directly under where the turbo goes. This wall is pretty thin, however. Also, drilling too far towards the rear will put your hole next to the gear on the crankshaft that drives the balance shafts, which creates a bunch of windage going into your drain. Luckily, there's an unused boss for a halfshaft bracket just a couple inches forward that makes for an easy drain location, and has way more material for an NPT fitting to grab onto. I removed the RH motor mount for access and drilled/tapped that boss for a 1/2NPT fitting.

I greased the drill and tap of course, but this time around I had a buddy blow a little compressed air into one of the crankcase breathers while I was working. It worked gloriously, nearly all of the shavings and oil blew out and hit me in the face instead of going into the pan.



Photo taken right after I finished drilling. Access without pulling the header or subframe is tight, but doable. I was able to miter the hole out to be a little less ovalized and the tap had plenty of material to grab onto still.

Next up, dyno time! So I can finally get numbers on the current setup less than a week before the turbo gets installed