Hello,

I've recently finished building an RX-8 with my Dad. Project consisted of 2 broken cars (one had a good body, the other had a good engine).

Pulled the broken engine out of the first car and dropped the engine from the other car in and reconnected everything back (means I didn't do an ECU or wiring harness swap).

It runs fine but just too damn hot for my taste. Minimum is 93 degrees Celsius, max was 101 (values taken from OBDII via Torque Pro).

I've changed the expansion tank (sensor included) with a new one but I get the same values. The Mazda coolant in the radiator is cold, even after the car has run. I have a Mishimoto thermostat that opens at 76degrees Celsius so I was expecting a maximum temp of around 80-ish.

The fans don't kick in as soon as I'd like (will fix that soon) but no matter the speed I'm running, the revs, the gear etc the temp is always between 93 and 101. Checked the temps again after mounting a new CAT (old one was empty) and mounted new oxygen sensors.


Is it normal for it to run this hot? Is the temperature reading false?

I'm thinking I may need a new radiator + oil coolers because I'm assuming something is stuck somewhere.

Oil I'm using is 10W40 semi-synthetic with 1 or 2% difference in properties compared to Miller's Trident. No pre-mixing yet.

 

The sensor in the tank is just coolant level. Your temperature sensor is (IIRC) on/near the tstat housing.

That's on the high end of the reasonable if you're stationary because the fans don't kick on to high speed until around 97 C.

When cruising at motorway speeds under low/light load it should hover around the thermostat opening temperature.

If the coolant in the radiator is cold even after running the car for a while then it sounds like the tstat isn't opening up.

 

Damn...and the thermostat was brand new. But yeah, it really makes sense now that you've mentioned it. If it would have been circulating, it would have been warm. Duh, basic basic stuff. I got so caught up in thinking something more serious was wrong that I forgot the basic stuff. Now I feel really dumb )

Unfortunately, no. Even if I accelerate extremely slowly and get to around 120KM/h / 75MPH, the temp is still at 98degrees / 208 fahrenheit and only increases when I lift off the accelerator.

I can thus assume that the oil coolers are working otherwise, with a closed thermostat, it would have overheated long ago.

Thank you, NotAPreppie! That really helped.

 

No problem.

Fresh eyes have saved my bacon more times than I can count.

On the bright side, a new tstat is relatively inexpensive and easy to change. I wouldn't start it up again until you do this.

 

Ok, it's not the thermostat. It opens at 76 just as intended. But the car just doesn't cool down.

I did notice that when the fans kick in the temps drop but other than that, they stay pretty high.

With the engine up to operating temperature, the coolant in the radiator is cold, even though the thermostat is opened. Could be the pump but not sure.


So far, it only cools down when the fans kick in. That wouldn't make sense, considering the coolant in the radiator is cold so I'm assuming it's being cooled down by the fans sucking in the surrounding hot air. Gotta check the whole circuit.

 

Yah, something is odd. Once the engine gets up to operating temp nothing in the engine bay should feel cool.

 

did you put foam around the radiator to force more air through the radiator?


Another thing you can do is put a custom duct on the radiator, I did this with my FD.


A lot of cooling comes from the oil, perhaps you can make your oil cooling more efficient as well or something isn't running as well with the oil cooling system.

 

I didn't mod the cooling but it doesn't cool at idle and it doesn't cool if I go 100MPh and then let it idle until I come to a stop.

The weirdest thing of all is the thermostat works (put in hot water for testing) but the coolant and the radiator are cold, even though the engine is at 208. But the engine cools down when the fans start.

I have ordered a new radiator + fans + thermostat pulled from a running RX8 with 60,000 miles so fingers crossed for a direct swap

 

The foam idea isn't a bad one.

There should be foam baffles around the bottom and sides of the radiator. These work to keep air from going around the radiator.

When I swapped my radiator for a CSF unit, I used 1" pipe insulating foam cut in half. I glued it to the bottom and sides and it works great.

The fact that the radiator works with the fans running (and 207/208 is when both fans turn on full blast) means the radiator and tstat are working but there isn't enough airflow.

Also, is your undertray intact? That's critical to proper cooling.

 

This car gets more complicated with every issue. I love it!

This car will see a lot of city traffic (bumper to bumper, mid-day, middle of summer driving) so I'm not sure that even with the foam it will help.

The engine shield (undertray? it protects the engine from anything that would get under the car) is intact. I could take it off but again, with no airflow I'm not sure how it would help.

We're still investigating as to why it is not cooling but I'm down to three options (for now) :

1. There is space to add another stock radiator in parallel so I would double the cooling;
2. Add a fan switch in the cabin so I can start the fans at full blast whenever I want, not when the car wants;
3. Get the Koyo V2695. Not sure about this though since I'd rather have 200% cooling with 2 stock radiators instead of 130% cooling from the Koyo (from what I've read, it's 30% more efficient at cooling than stock radiator).

 

The foam and undertray work to channel air through the radiator at speed (since at highway speeds, the car will produce more airflow than the fans ever could). If you're missing the foam or if the undertray isn't fastened adequately at the back, air will pass around the radiator and reduce its effectiveness.

The temperatures should drop to the tstat opening temperature pretty quickly if you accelerate up to 60 mph and then coast to a stop. If that isn't happening then you may have an air or water flow issue.

If you're seeing temps in the 208-region while stationary, then this is normal because that is when the fans turn on high. More surface area will not help with this because even God's own radiator can't work without air flow.

Trying to sandwich two radiators in the space meant for one seems kind of silly. Assuming you can even make the plumbing work, you aren't going to double your cooling because due to air and water restrictions. If you're looking to maximize surface area, the is probably the thickest one you can fit without removing the airbox and battery trays (I have it and it's tight but a direct drop-in).

Honestly, I think you need to go back to basics and verify everything because some stuff isn't adding up.

You know the thermostat works.
You know the fans work.
Since the temps don't rise much above 208°F, you know that coolant has to be making it through the radiator. If it wasn't, the fans could run full blast but the temps would rise until it blew a coolant seal.

What is the coolant temp reported via OBD-II when cruising at 60mph/100kph?
What is the coolant temp after idling for 10 minutes?
Find a way to measure the temperature of the coolant overflow tank (below the level of the coolant) and the radiator end tank(s). Do this both before starting it up and after a long drive.

 

I'm really frustrated that this thing doesn't cool down the way I want it to. I'll take your advice and cool down myself before the car

The car seemed to run fine (before I noticed the high temps) but I wanted it checked anyway. I took it to Mazda who said that the expansion tank and sensor need to be changed. I did that and now when I accelerate, most of the time the "check coolant" light comes on and the car goes into limp mode until the RPMs drop to idle.

Unfortunately, the temps idle at 93C / 200F with idle revs, no longer the duration. This is monitored via OBD-II. They get to 100C~105C / 212F~221F when city driving.

On the highway I couldn't get higher than 100KPH/60MPH because the car would go into limp mode. That's 6th gear at around 3,000RPM

I hit 200KPH/125MPH and the temp shot up to 110 and went into limp mode (did this when engine was still at 93C / 200F. From 200KPH/125MPH put it in neutral and monitored the temps. It stayed at 110 until I hit 160KPH/100MPH and then dropped down
93C/200F but didn't go below that until the car reached a complete rest.

When I said I'd put two stock radiators in it, I was thinking of using both fan shrouds (one front and one back) but considering airflow, I think that's a bad idea.

I'll change the expansion tank back to original to solve the limp mode issue. I'll keep "cool" until the other radiator arrives and until we can do a detailed inspection of the cooling circuit.



Thank you for your reply, I'm getting ahead of myself here. This is my first build and also my first encounter with mechanical stuff. Before I started this project, my other and only car which I am still driving is a Smart ForTwo 700cc

 

You have a problem that's not related to cooling off you're limited on power or going into limp mode.

I would start with checking out the catalyst, if it's clogged it will restrict flow, which in turn will cause heat to build up at the exhaust ports and get dumped into the coolant. It would also explain the power loss and limp mode.

Do you have any CELs?

 

Car had a blown oxygen sensor and a deCAT. It was running at the same temps. I had to buy new oxygen sensors and a new CAT to register the car and also have it pass emissions testing. The temps have not changed.

CAT temperatures with the new CAT are at 630C / 1166F at idle and max was 930C / 1706F at the 200 KPH / 125MPH run.

CAT temperatures with the old, gutted CAT were 660C / 1220F at idle and 880C / 1616 during a couple of minutes of 8,000RPM shifting.

Not a lot of difference between the two. The only thing I did notice was that with the new CAT my RX8 is actually behaving like an RX8 and the cupholders inside the center console are getting heated, since the CAT is directly beneath them. Didn't used to do that with a gutted CAT.

Going back to the old CAT is not an option for me. I have to keep the new one. Emissions regulations are very strict here. The car has to pass once every two years. If police also see you do some more...energetic driving they can also pull you over and decide if your car will be impounded for some checks. And if it fails, bye-bye car. Same goes for different headlights, tire sizes, tinted windows, different paint and the list goes on.

I had solid CEL when I had the broken O2 sensor and the gutted CAT. Also a couple of blinking CELs when it was misfiring due to one of the coil leads arcing the spark to the body. All fixed now. I continuously monitor as much info from the car as possible and decide if I run it gently or hard.

The limping didn't start until I followed Mazda's advice and changed the expansion tank (the one you put the coolant in). I check oil and coolant levels every stop and they are fine but even so, when the "check coolant" light comes on, the car goes into limp mode until it hits idle RPM and then it goes away.

It did not have this issue with the old expansion tank and there has been no temperature difference change between the two tanks so I will switch back to the old one.

 

What are you running for coolant? Do you have a way to verify that your cooling system is actually holding pressure?

If it's not holding pressure (could be as simple as a bad coolant bottle cap) then it could be that the coolant is boiling over, which is sinking the float switch and turning on that light. If you can verify that, just unplug the sensor and see if the limp mode problems go away.

 

I'm running the Mazda coolant which is FL-22 if I'm not mistaken. Bought 5L of the stuff straight from the dealership.

Ah, pressure... hadn't thought about that. Was only thinking of coolant flow. I don't know how to check the pressure, other than splicing a pressure gauge in the cooling system.
But I have two coolant tanks and they should both be working.

I was thinking that it is boiling over so that's why the sensor is triggered and the car is going into limp mode but the thing is, I checked the coolant tank regularly and I found no marks or traces of coolant or anything. So the cap is fine.

The coolant level does drop though but slowly and I can only assume that is because it is boiling and is eliminated as steam.

I know the boiling point of this coolant is about 120C / 248F . So I can't understand how it would boil with an engine temp between 93C / 200F and 105C / 221F .

I've checked and checked and found nothing wrong with the cooling system. Most likely the radiator is done for. I'm awaiting delivery on the radiator which is still in the UK. Should arrive on the 11th of April

 

the Koyo Hyper V-core is a good choice.

 

Two schools o'thought, coolant flow in any given channel or coolant is finding the path of least resistance.

If you have a coolant flow issue in any area like as in one or more of the block areas - that coolant will see higher than BP skin temps on the engine side and create pockets of "steam" forcing liquid to flow out of the expansion tank. Time will "wound" that area and cause the block to warp or crack.

If your new expansion tank isn't sealing well at the cap or the cap is/has failed, that would be another area to check.

 

Which v-core and how was fitment in your S1?

 

Koyo V2695. Fits all series1 RX8s. Shop around online, many reputable sellers carry it. Around $265-$275. Varies, some offer free shipping too.

http://koyoradracing.com/about/technology.asp

HYPER V SERIES 36mm Core Thickness with Hyper Fin and Tube Pitch

The Hyper V-Core Series is the newest to the Koyorad | Koyo All Aluminum Racing Radiator line up. An engine's cooling system in many newer vehicles are suffering due to limited space and crowding which affects air circulation. Our new ultra efficient Hyper V core was developed specifically for vehicles which see extreme heat, are limited in space, and where the factory inlet/outlet locations eliminate the option for a proper N-FLO design. The Hyper V Series radiators have a specially designed fin and tube pitch which significantly increases its surface area while still allowing for excellent air circulation through the radiator. The Hyper V-Core series is the perfect solution where space savings and high performance is a must. Unless noted otherwise, all Koyorad Hyper V-Series All Aluminum Radiators are Nocolok brazed, designed as a factory fit, direct bolt in with no engine modifications necessary, and retain mounts for the stock fan shrouds and fans. See notes in individual item descriptions for details.


Fitment was perfect. It aligns with all oem mount points. I have it in my RX8, and as well as being an absolute beauty of workmanship, it works great. Better cooling than oem. A real bargain with top quality and performance. Recommended to me by several RX8club top folks. Glad I bought it.

I bought it for $275 from these folks, but I have since seen it for as low as $265.
http://www.importtuned.com/6512-koyo...zda_rx_8_04_08

 

^ "Better cooling than oem" . Do you happen to have the temps before and after?

 

I saw a drop of between 5-10 degrees F, from my stock oem radiator to the Koyo, depending on the engine load , and speed of airflow over radiator. With stock configured engine bay and undertray in place, you get more airflow to the radiator as you drive through the air faster. this could be better or worse depending on the ambient temps and other factors.

BTW I also had other cooling upgrades done ,but saw the temp change immediately with the radiator upgrade first. Other longtime RX8club members who live in year-round hot climates have gotten good results with their Koyo radiators and recommended it to me.

 

Thanks for the info. Was expecting a bit more to be honest. What about this stuff? Would it help? http://www.engineice.com

 

I totaled an even higher total temp drop when my other cooling upgrades were added also: Remedy lower temp thermostat, lower temp cooling fan switch, and oil coolers lower temp thermostats. Suffice to say they all made a real total difference, as noted by my Torque Pro app temperature readings.

Since I am just one owner here, I won't make any outrageous claims that I don't have loads of test data for. That would be a disservice to you. For your engine, your climate and your driving habits you might see more or less of a difference, but physics is physics. So these mods and others like creative ducting that can increase the airflow to the radiator have been shown to help.

I have no experience with or knowledge of this "engineice" coolant. You must use a coolant compatible with the rotary engine's bi-metal construction, or risk serious damage. Fl-22 or equivalent is what is needed and safe for your engine.

 

Wait...so the oil coolers have thermostats also? This just gets better and better! Can those thermostats be removed to allow for a free-flow oil circulation?

I'm suspecting the oil coolers might not be working well on my car. I say this because I've done almost 5,000 miles with the car since it got an oil change and I check the oil at every stop. The oil level does drop, which is good. But it still has the same color as when I first filled her up. I was expecting something darker? Is this normal?