i had reached out to Ron davis a while back to get a custom radiator made that would have the same mounting as the oem but inlet/outlet to be on the same side and possibly made it wider than stock so i can close my side gaps while increasing the radiator surface. I dont recall what i was quoted for but it wasn’t prohibitive but it felt that it wouldnt have been a major improvement to the system since everything else were to remain the same. so instead i made the shroud which improved the cooling performance drastically but the overall system is not optimized.

i will have to choose different components so i can redesign the whole system. Everything from mounting the intercooler, the intake to the radiator and fans . The last few replies sparked my interest to address this issue and improve it once and for all and stop saying to myself that its good enough for now.

your kit look’s pretty good Pete.

Here's the design for the radiator—it was made with angled end tanks so i could get the most room for a bigger IC, which would've been deeper than the GReddy. I'll see If I still have that plan too.

https://cimg3.ibsrv.net/gimg/www.rx7...1cba2e008a.png

I believe Race Only in Australia is working to make an FMIC kit and retain A/C while using the chase bays radiator. They race their cars and have made the FMIC work vs Vmount. As you can see below, unless you push an IC past it’s rating (which can obviously still work), you run out of room for a vmount real fast.

im a big FMIC fanboy here is what I’m doing. In both cases I have the rx8 AC condensor between the IC and the chase bays rad.
https://cimg0.ibsrv.net/gimg/www.rx7...ab69ce6c8.jpeg
950hp garret core for the 3 rotor. nothing this big would fit without doing my own end tanks, all premade ones I could find with this rating were wider and ate into the oil cooler space unless I did rear entry end tanks.
https://cimg4.ibsrv.net/gimg/www.rx7...d4cc901f7.jpeg
3 rotor. Will be ducted and prevent air from escaping before hitting the radiator.
https://cimg2.ibsrv.net/gimg/www.rx7...8d2172c95.jpeg
600hp (piston) treadstone intercooler. Only one of theirs that would be the right width, I cut off the inflect/outlet and welded on the elbows to save space.
https://cimg3.ibsrv.net/gimg/www.rx7...6f6fd7139.jpeg
Also has the chase bays radiator and will be fully sealed.

This is all pretty useless without AIT data anyway. I'd test air temp at the current filter while moving, and air temp at any proposed intake location. It might not be as different as we think.

Get an air/liquid thermocouple like this one https://www.fueltech.net/products/ai...d-thermocouple

With the methanol injection already being used, this might all be a moot point anyway. Sure lower AITs are better, but redoing the entire front end of a car just for a few degrees lower really might not be worth it.. Most of the full-on race cars I've seen cut the headlight bucket, or run an open filter.

Imagine how much cheaper it would be to just run a little more meth... The HKS stuff is pretty good, even if the fans are low

its only a matter of time for the low hanging fans to be destroyed. While i did solve that issue with what i think its ideal given the circumstances but if i am going to re-do the intercooler i am planning to do a larger radiator both in width and length with a similar shroud . I no longer have the fan multiple speed provision so i had to stick to one speed fans.

i will see if theres a front clip i could use locally and design an intercooler kit that would be my vision of the ideal setup that will be loosely based on the hks kit

I popped the HKS intercooler on a GTX4294R turbo at around 21-25psi. It really wasn't doing a great job anyway, IATs were sky high.

https://cimg4.ibsrv.net/gimg/www.rx7...71021b033a.png

given the available nose space, unless tremendously re-engineered NO intercooler is going to properly get it done.

the Greddy V mount core is 647 cubic inches. my core is 784.

if we relate power to core displacement using an average relationship from Garrett's site, then correct for rear wheel power and then correct for rotary V piston we get .62 rw rotary hp per cubic inch.

Greddy 647 X ..62 = 401 rotary rwhp at 1.5 psi pressure drop

HC intercooler 784 X .62 = 486 rotary rwhp

as you can see, our IC cores are a bit on the light side. of course all of these numbers have a bit of wiggle in them but they are helpful.

according to my 1996 Spearco Intercooler materials a very good intercooler can remove 67% of the temperature rise from ambient. ambient being at the airfilter not outside the car. i do have two Spal puller fans on the backside of my core and my core is sealed to the duct.

here's a recent run showing the cooling dynamics:

https://imagizer.imageshack.com/v2/xq70/922/YVvCIx.png

at the end of this 3rd gear pull at 2625 altitude my actual (thermocouple measured) air out of the turbo was 264.4 F and my IAT was 77 F. my turbo is a highly efficient G40-1150 and was loafing off the wastegate spring at 15.1 psi but was making 539 rwhp. air temperature at my air filter was 105.

of course we are going to do the math:

264.4 minus 105 produces a temp rise of 159.6 F.... if my IC was 67% efficient, which i doubt, the IC would have reduced the IAT by 106.9 leaving me at 105 plus 52.66 or 157.66 F IAT. and that is assuming i have top flight IC efficacy. (probably not)

BUT, my actual IAT was (and again this is a real temp generated by a thermocouple) SEVENTY SEVEN degrees.

157.66 minus 77 =s 80.66 F further cooling from wonderful methanol.

106.9 from the intercooler.... maybe

80.66 from the methanol... probably low.

this is all taking place at 15.1 psi. guess what happens to the temp coming out of the compressor at 20 or 25 psi... it will have a "4" on the front end. then do the numbers as to your real IAT. is it any wonder we can have problems considering gasoline autoignites around 450F?

after fully evaluating 3 other IC cores in 2022 my conclusion is that the answer to healthy engine dynamics lies in the proper application of AI to the system.

i was running two M15 old school nozzles. each is capable of 945 CC/Min at 100 psi. if you note the graph, the green sawtooth (pulse pump, high sample rate) is maxxing around 65 and bottoming around 25 so my average delivery was much less than the 1890. i had tried the ProMeth trick nozzles and my system was showing 150 psi. my IATs were 200... hmm, switched back to the old school.... something was amiss w the PM nozzles. when i get some time i will return them for evaluation.

BTW, porting is a significant factor re IATs... lower the system restriction, lower the compressor charge air temp. i just added a porting tech section to my site.

PORT TECH

Crazy you popped the HKS core, I just spent good money getting one. Seems like a really good piece, not many tube and fin cores on the market.


Every single turbo car has the same IAT struggles we have, and tons of people are making huge power on 600x300x100 cores. Our front space isn't that restrictive if you have a 99 or aftermarket bumper. A heat exchanger (oil cooler, intercooler, radiator,) will only pass 1/3rd of its frontal space in air volume. As long as the opening in the bumper is 1/3rd or larger than the exchanger area, you're good to go. Turbos, intercoolers and intake temps are the same struggle for any platform this isn't new science.

in 1975 Roger Penske wanted to try and use the 917/30 to break the closed course speed record at Talladega.
after melting a few engines, they called Porsche. Porsche knew that they needed to keep the temps under control for the 2 minute lap, and designed an intercooler setup.
they beat AJ Foyt's 217mph, with a 221mph speed...

I am currently looking for a front clip or borrow a car that can stay at my place for some time, possibly during the winter, anyone looking for a free storage spot in up north in jersey in my garage let me know:lol: . I am determined to fix the intercooler issue and redesign something from scratch..
and depending on how well i succeed i might offer it for sale. i have some concepts in mind that they wont only fix the intercooler and the intake issue but also improve the radiator/and the infamous radiator fans that are associated with the HKS kit.

There are some upgrades, all of them are actually in progress but i should have some completed stuff for November and somehow i have managed to be in Europe for the 4th time this year so I am hoping that i will be able to address some things that have been on my mind for a while.

I am switching the OEM waterpump to the Meziere water pump but i needed a mounting solution.
As i like to make my library of parts larger I have modelled the meziere water pump along with a bracket to go with it.
the Meziere provided bracket is a 4140 CRS .12" thick.

https://cimg4.ibsrv.net/gimg/www.rx7...3a979573c0.jpg

I have decided to make this in 3/16" thick 304 CRES. Similar(ish) to the Meziere Bracket.
The mounting point will probably be at the cross bar by the intercooler. I will 3d print the pump and the bracket in Cyprus and confirm fitment first
https://cimg4.ibsrv.net/gimg/www.rx7...905ded38f1.jpg
https://cimg6.ibsrv.net/gimg/www.rx7...6ae700c26a.jpg
https://cimg5.ibsrv.net/gimg/www.rx7...a824ce0eb2.jpg
https://cimg3.ibsrv.net/gimg/www.rx7...2b58307725.jpg


I purchased the sakebomb garage EWP water pump block off plate. That thing is gorgeous. I also bought the thermostat delete kit BUT there's a lot to be desired from that.
The Sakebomb thermostat delete requires you use the oem Spring and plate (for the dual pass) soI had ordered an OEM thermostat but I must say I was surprised with my findings for a few things
I ordered two OEM thermostats from 2 different vendors from the US. Both of them came in the same package and it was the same thermostat

https://cimg3.ibsrv.net/gimg/www.rx7...481bd70fa1.jpg
I thought that was strange because the dual pass plate on the bottom of the thermostat is removable on the OEM thermostat and I needed that part to use with the sakebomb thermostat delete part.
So something changed. Did not exactly know what but whatever i figured i will destroy the one so i can get the plate and the spring. No big deal.
One thing i noticed though is that the dual pass will not even come close to the cavity on the water pump housing. Also, the seal was pretty loose.

https://cimg2.ibsrv.net/gimg/www.rx7...03771c0c5a.jpg

Whatever the seal i can somehow make it work but it is an "OEM" thermostat. The dual pass cavity is not being plugged. So it doesn't serve the purpose as the dual pass thermostat.
I was browsing the YJ and I came across the actual OEM thermostat that I am used to seeing for the last 20 years. So i figured let me order that and see how it is in comparison.

So the thermostat sold by the mazda USA is different. Markings , C-clip on the bottom for the dual pass etc.
I assumed that Mazda perhaps sent me a thermostat for a different model

https://cimg4.ibsrv.net/gimg/www.rx7...998588a637.jpg

I have come to find out that this is actually a Re-boxed Motorad Thermostat. Sorry Mazda USA but you Effed UP on this one ! So anyone looking for an OEM thermostat, be aware that this is what you will be getting from Mazda USA
https://cimg5.ibsrv.net/gimg/www.rx7...4c221b6bf3.jpg

Anyhow, since i did not like the Sakebomb kit i thought let me re-design this and make the dual pass actually work.

I was planning on making the whole thing as kit with a purchased spring and a C-clip and only make the plate. I calculated the OEM and the Motorad spring rates (they are very close) but what was the point. I am not making a bunch of them so i figured i might as well use the parts I have

https://cimg4.ibsrv.net/gimg/www.rx7...02243e8409.jpg
https://cimg6.ibsrv.net/gimg/www.rx7...0c00a8dc2e.jpg
https://cimg5.ibsrv.net/gimg/www.rx7...1c5e1f3c71.jpg
So the initial plan was to machine this and make it a 2-piece construction. I did however change my mind and i will be 3d printing this in 316 Stainless Steel. I do have quite a bit experience with 3d printing but i don't have any experience with Metal 3D printing. This is obviously a paid service but i am curious to see how it will come out.

More updates to follow

​​​​​@Howard Coleman This post is for you. Your words inspired me to finally pull the trigger and make this change. You were right all along... !

https://cimg6.ibsrv.net/gimg/www.rx7...3c3bce8dfa.png
I have decided to address the issue with the poor intake and improve my intercooler altogether.
I have teamed up with BELL for this one. So the idea is to "replicate" the mounting and the basic shape of the HKS V mount. It will be significantly shorter so i can squeeze the intake on the side, pull some cold air from the filter and also cool my charge more with this beastly intercooler. I have removed the dimensions because it is not my drawing and we have many changes, so possibly the dimensions might change until we finalize it, but i kept the 6" core dimension..

Wow, 6" thick??!!
I'm surprised such a thickness can be used in an air-to-air design. Would have thought the air velocity would slow down too much but what do I know.

Coolers lose effectiveness at something like the square of their depth. Would you mount a 3" core, and then another 3" one behind that? No, because the 2nd cooler wouldn't do much. I honestly think you're better off with the larger surface area intercooler and an engine bay intake. The extra heat will bleed out though the larger intercooler surface. Hell, just run a hood scoop isn't it a track car?

its been this way for years, the 9U means Mazda USA is supplying an aftermarket part. in the rest of the world they seem to have some kind of rules, and only sell you the correct part

it is a street car.

To answer your question, it's all relative. If you can get enough air pressure for the air to pass through you can have a 6" core or a 106" core. it all depends what you are capable of pushing. I am planning to increase the pressure with ducting but whether that would be enough it is to be determined. I mean the current HKS core is 4".

Is it easier to increase the surface area instead? Sure. but in this situation I can't do it easily. so i cant have my cake and eat it too so i am trying to figure out a best of both worlds scenario. It might be a complete failure, or a great success. Well one way to find out !

I did not know but thanks for that.. I dont even remember the last time i used an OEM thermostat :lol:

This is what was recommended by Bell after discussing my project to them. We have made several changes the drawing i have posted a couple of days ago but we are ready to start manufacturing i believe.
And if it doesnt work, oh well! I will re-use the HKS until i figure out a better option
I am also planning on using K-thermocouples on both end tanks to check the efficiency.
i have always experimented with stuff and sometimes they work and sometimes they don’t but this is how you learn.

you could also add a fan on the ic, if needed

I actually had this discussion with Bell and with proper ducting and the vented hood and they said that it didn't seem to be necessary since the application is primarily high(er) speeds.
I am not opposed to the idea to add a fan to the intercooler if its needed

There are a few road cars here running fans on the intercooler. You'll definitely want to keep that divider between the radiator and the intercooler so you're not pulling up hot air.

Definitely interested in the results, heat is the enemy here

The divider is definitely staying.
In my next year plans i am planning to design a custom radiator that will allow me to widen the radiator so it reaches both side mounts , extend it so it will reach the bumper (no front gap) but also to redesign the divider so i will get rid of the cutout for the hot coolant in as i am planning to relocate it on the back side. I am debating whether i should get a radiator mount Meziere pump or stick to the one i have .
As for the road cars, i completely understand the need for the intercooler fan.
I do not currently have a provision to power an electric fan (PDM problems:lol:) but after i shuffle my light power outputs around i might be able to have an extra power output and find a fan that wont draw too much power (less than 15A). Thats assuming i will be running the car at slow speeds in the street, which it is a very possible scenario with my car.

Run it in tandem with the radiator fans. No harm in having it run all the time as long as you have the amps to support it.

https://cimg1.ibsrv.net/gimg/www.rx7...b8db91caf.jpeg
https://cimg2.ibsrv.net/gimg/www.rx7...c39f19959.jpeg

New intercooler is in. New vs old setup.

Very Nice! What’s the size? And where did you get it?
Thanks Steve

core size is 16”x10.4”x6”
The intercooler is made by bell intercoolers in a design collaboration i had with Bell. The concept of this intercooler is to replicate the hks intercooler but make it narrower so i can bring the intake to the opening of the bumper.

to make the installation a little easier due to the additional thickness we have welded part of the tubes on the intercooler, hence the longer inlet/outlet tubes

Looks like a drag core. Thick, bar and plate, meant to heat soak down the 1320. Interested to see how it performs on the street / track when it's seeing constant load. Rule of thumb is, air passes thought a cooler at 1/3rd of it's frontal area. How much area did you lose?

Do you have pre and post-intercooler data from the HKS core? Interested to see how it compares.

It's almost moot (the loss in length), because the old HKS basically dictates a hot-air intake that's way less than ideal. The engine bay also dictates a shorter (fewer row) cooler than you'd maybe ideally want, so while there's a diminishing return in thickness, it doesn't hurt either and may help with pressure drop when running high boost levels.

I had to make a similar decision. I opted to make the radiator slightly shorter to give space for the filter and used a Garrett GT 24x12x4 bar/plate core. Probably resulted in a bit more fab work.

That being said I am not 100% convinced keeping the filter behind a properly sized IC core is as big of a problem some make it out to be. I could be wrong but I do not think the air Temps from the inlet, outlet of the turbo and inlet, outlet of the IC are completely liner. For example, ingesting 30 degree higher temp through the filter is not going to result in 30 degree higher Temps on the outlet of the IC if the IC core is sized properly.

Actually the intake the way i currently have it, is an inspiration from your setup. Well its not like i am not using a few of your components anyway :lol:.
But i figured if it had worked for you i don't see how it shouldn't work for me even as a temporary solution.
The only difference is that i stuck the largest filter i could find since the flow in the area is somewhat restricted.
I also never really had an issue with the intake temps with the filter being where it is. My intake temps were always pretty low. They were low enough that i had other priorities to keep an eye for.
Could they have been any better? Sure, but like with everything, there's always room for improvement.

I can't make the radiator any smaller actually. I have replaced the radiator with a brand new of the same unit (Koyo N-FLOW) I will be going back in 3 weeks to test whether my overheating issues were resolved. I will make the decision then to switch to the Electric waterpump or stick to the OEM one.
I am actually planning to do a new radiator for next year, it will be much wider and several inches longer .

It is a bar and plate intercooler. whether that alone classifies it as a drag core i dont know. I had a very long discussion with Bell about the setup. They said that this is the right core for this configuration.
I dont have the numbers handy but if i recall, this intercooler is 8 inches shorter from the original HKS unit, so 8"x10",= 80in^2? something like that

Ooooo shiny, custom, new parts :djcelebra

Looking forward to seeing how this next iteration of your cooling system performs.

I talked to Bell a decade ago about a custom v mount similar to your project. Brilliant group of engineers to work with. True enthusiasts. Ultimately did not follow through as I hooked up with "Speed of light" and he built a few kits.

Take their word over mine. I think they might know a bit more about it haha.

Very interested in the results!

I cant wait to be honest. 3 weeks is not coming fast enough haha.
There's a few more things coming related to cooling. I am expecting the thermostat delete kit i designed to come in the mail today.
As for the Bell engineer i worked with, his name is Justin.
I had called in with my request after sending an email with some pictures. He was on board within seconds. Very interested in the whole project and very enthusiastic about it.
I wasn't exactly sure what to ask for and i had told him that. I told him I am in the design and manufacturing business but i don't know the nitty gritty details for the intercooler design and manufacturing.
He did not disappoint. I am not 100% sure that this intercooler will work but i would much rather try it and fail than never try it and go lecture people how things should and should not be done :lol:.
I had reached out to other manufacturers about this intercooler design they either turned me down or didn't even respond. Bell was the only one seriously interested in this.
Once we agreed on the drawings and all the changes, i was given a lead time of 3 weeks. it was done within the week...
Very impressed with Bell so far. Most likely i will have them manufacture the radiator i think i will be using for next year.





I will update the thread with the results once i have done the testing.

Thermostat delete kit has arrived. 316 stainless steel 3d printed. I think this will be great
https://cimg1.ibsrv.net/gimg/www.rx7...3e0aeda12.jpeg
https://cimg2.ibsrv.net/gimg/www.rx7...a6f767680.jpeg
https://cimg4.ibsrv.net/gimg/www.rx7...5cd7a6c14.jpeg
https://cimg7.ibsrv.net/gimg/www.rx7...0fe523204.jpeg

How do we go about getting one of those bad boys? I also have a hole to fill :)

https://cimg2.ibsrv.net/gimg/www.rx7...25f718d37.jpeg

i had it made at shapeways. I think there is a way to make it public in shapeways so you can order it from there. I ll get back to you on that but i would rather test it first and confirm the functionality first in about 3 weeks.

but DAMN that twin turbo system will kick ass once completed

thanks. And no worries, we’ll see if it ever gets completed haha.

I also thought about just filling the bypass section with Devon and gutting an old thermostat to use to keep the rubber seal functioning.

what are you doing for a thermostat/bypass? I just got these. Took me forever to find something like this.

https://cimg2.ibsrv.net/gimg/www.rx7...7aa00e499.jpeg
https://cimg3.ibsrv.net/gimg/www.rx7...c47da74cf.jpeg

the idea is to control the speed of the pump to control the temperature. i am not planning to use a thermostat. but this will block the cavity to let air out.
i am thinking to slow down the pump to 10-15% until it reaches the operating temperature. The beauty of having a pdm :nod:

interesting....was 3D printing cheaper than having it machined? Looks easy enough to machine from bar stock.

Machining it would have probablybeen slightly cheaper but doing it properly i would need to mill and turn the part. Its not a difficult setup but it definitely adds setup time since i was making only one piece, but i was mostly interested to see how it would turn out printed because i do not have metal 3d printing experience.

even though i liked the overall results in the printed part , i still have to passivate it and one small insignificant section came out crooked.

I'm use you thought of this but just to make sure (and for other people in the future), the pump always needs to be flowing at least a little bit, or you can suffer from hot spots and you won't get an accurate temperature reading. That's the benefit of keeping the thermostat with normal flow, no hot spots but a quicker warm up without the radiator in the loop.

I noticed you're running a different trigger wheel with A/C and PS and all the stuff? I need to read through your thread and find the details. I have a Hall trigger kit but gave up on it when it wasn't an easy fit.

correct the pump always needs to run when the motor is running. The plan is to slow down the pump enough to reach the operating temperature quickly. I currently do not have a thermostat and reaching the operating temperature is not an issue but now with the new radiator it might be. I will have to experiment.
Also worth noting, since i have replaced the battery with a 60Ah (previously 19Ah) i am planning to run the pump for a certain amount of time after the car is off to prevent heat soaking. Whether thats 2 minutes or 20 minutes it is to be determined.

as for the trigger wheel i am actually running the stock one. Yes, I still have the A/C and P/S . I am not sure whether it was mentioned previously in the thread or not.
It is one thing i wanted to keep stock to avoid potential headaches and i am very glad i did.

Ok so that pic was just a mock-up with a different trigger wheel. I've seen at least one person make it work, but I haven't found the details yet.

i dont recall what picture that was. I did not know that using a different wheel will create some kind of an issue. I only wanted to keep the oem trigger wheel to avoid the issues that other people had. I do like the oem design. It makes it fool proof

LMAO I'm so blind. TwinChargedRX7 posted that pic, I thought it was yours seeing as this is your build thread hahaha.

Carry on with the good work, I'll go visit an Optometrist

i just did the math, and the HKS IC is ~961 cubic inches, and the new one is 1018cid, about 6% bigger

the HKS is 600x255x103mm, its 23.6x10x4 inches

correct. It is similar size in volume. The idea was to keep the volume roughly the same while being able to bring colder air from relocating the intake.
i did not want to compromise the performance with a smaller intercooler. It is to be determined whether this is the case or not due to the thickness.
so if this works i will have my cake and eat it too and if the improved efficiency of the new intercooler can help push past the 700whp i will be very happy but at high boost the car is frightening as is

i'm cautiously optimistic, you had an intercooler engineer help. its bigger too, which is good.
the intake setup is reminding me of the old Mazda race cars, in the 1st gen the radiator is over to the right to make room for the battery, in the race cars they shove it over to the left and then the intake feeds to the bumper.
they called it the cassette style bumper... you can kind of see the ducting
https://cimg6.ibsrv.net/gimg/www.rx7...d40bac1f58.jpg

i know i'm coming along late to the party, but it looks promising!

This is exactly what i am looking to do but with a different opening on the bumper.
I could have manufactured the intercooler myself but like you said, get the help from the people who do this every day and they specialize in this. I am sure there is a lot of tribal knowledge that i do not know. If it doesnt work, oh well , at least i tried