I'll correct a part of what I wrote above: The Gen III Prius engine apparently uses an EWP. (Don't confuse this EWP with another Prius EWP, which has been subject to lots of problems, that runs the cooling system for the battery). One might search for articles about its operation in a street car. I'm guessing it makes more sense for a hybrid since the engine is on and off a lot, at stop lights and so on. Makes one wonder if engine-stop technologies might eventually force a shift to EWP's to keep stopped engine temps under control.

I agree that hybrids have a particular use for EWPs, but I think the rotary is still in need of such technology. However, for the crazy people looking to implement this on their own, I would recommend doing an electric thermostat first.

The EWP offers the advantage of increased flow at low speeds and flow when the engine is off, both of which are great. However, with the current thermostat in there, as you increase flow with an EWP, the coolant will cool off and thereby close the thermostat, which means you're just pumping coolant around the engine. Not running a thermostat is nice in theory but has too many disadvantages to be practical IMO.

To me, the best starting point is an electric thermostat with a MWP. You won't have control over the rate of coolant flow, but you would be able to control exactly how much goes through the radiator. With more advanced fan control, this offers the possibility of making the cooling system very dynamic and able to adjust to a wide variety of ambient temperatures and operating conditions. With this, the days of blocking your radiator in cold weather would likely be gone.

Also, an electric thermostat is much less invasive. No need to mess with readjusting the belts to account for the MWP being gone and design a motor to be able to fit in the available space. Also, if a diverter valve is implemented, a rotary encoder could be placed on the rotation axis of the valve to ensure the thermostat is operating and provide a substantial amount of failure detection.

Of course, a controller would be needed, but the algorithm implemented could be relatively simple. Start with just coolant temperature and adjust the diverter valve based on that. Then implement an adaptive diverter valve position algorithm to analyze ambient temperature and coolant temperature and predict what the valve position should be, then use direct feedback from the coolant temperature to adjust the prediction based on if the engine is too hot or too cold, thereby providing system feedback if your predictions are wrong as well as data for the adaptive algorithm to adjust its future predictions on actual driving data.

While this sounds complicated, I have implemented such algorithms, specifically in the area of real-time noise cancellation, and can attest that it is very realistic to build such a controller.

rjon----that may be another option? the coolant sensor could drive the fans. Matter of fact some aftermarket gauges have that capability and of course the Cobb unit you can do whatever to the fans. The S2 model has updated and more efficent fan package with 3 speeds.
Hmmmm --for street driving it may be best to change one thing at the time?
1st--thermostat idea with more precise fan control?
2- use ewp as acessory pump for low speed use only?
3- computer controlled ewp?
WHat do yall think--other than I am completly crazy and need to get laid more often?

Getting laid more often cannot hurt :lol:


As for the EWP, I think for the RX8 it adds unneeded complexity and another unnecessary failure point. I like the idea of an electric thermostat but i will have to research that bit. I can say that changing the fan controls VIA the Cobb seems to have worked for me but we will see once the hot weather comes.

I like this plan. Honestly, if my 8 wasn't my DD, I would be doing this with you.

I like this plan. :smoker: Though it's generally more expensive than modding cars. :eek:

One thing I realized as I was researching Evans coolant was that in an ideal situation one of the tasks of coolant is to even out temperatures in the engine, not merely shed temperature from where it's hottest. This is one reason I wish that our engines flowed coolant from the bottom to the top (kind of like what you have in a reverse coolant flow piston engine). That being said, I would want coolant flowing full time, whether the thermostate was open or not.

Additionally, on cold winter days, even if the coolant is only 170F I want coolant flowing so my heater works!

On the other hand, having noticed heating issues last Summer when I was stuck in stop and go traffic after hours of highway travel, and again last Autumn when I was in a funeral procession that hopped on the highway, forcing me to pretty much stay in third gear or lower the whole way from Brooklyn to the Long Island, keeping the car in front of me in the procession close by to keep some jerk from cutting me off despite the Funeral stickers in our windows and our lights and hazard lights being on.

In the Summer situation, my engine temperature reached 245F according to my ScanGuage II. I had already converted to Evans coolant, which I think is the reason my engine didn't die right then and there. Coolant temperatures had been fine while I was on the highway.

Please note that I have the original OEM pump, and I think it is perilously vulnerable to cavitation, which is another advantage of a constant speed electric coolant pump.

I will either convert to a REmedy coolant pump or an EWP when the weather warms up, to try to keep those troubles at bay.

With no flow at all, another problem is that the WT sensor will not respond in a timely manner. Any heat would have to reach it primarily by conduction, which is a rather slow process. This is one reason why OEM thermostats have the burp hole --> there is *always* some flow with the engine running.

Uh ----no. That burp hole doesnt allow any flow. How do I know this?
Mount a coolant sensor in your top radiator hose using the oem thermostat. You will be surprised how long it takes for that sensor to start recording any warm temps. Now the oem sensor is in the thermostat housing and records the by pass temps also so it reacts much faster at warm up.
Remedy ---- drill two 3/16 holes in the thermostat lid. I discovered this a couple years back when I was figuring out the best way for me to use the second radiator set up.

Point made earier about the fans. That was a good thought and I highly recommend the 09 fan upgrade to the S1 cars.

Wow Evans allowed you to survive with coolant temp of 245F!!
OD

Just saw an article today that the new N4 BMW 4-cyl turbo engine uses an "on-demand" water pump, presumably, electric. Humm, maybe the trend is underway.

Well, drilling holes will allow more of course. I was more concerned about the OBDII sensor reading in the engine block where it is much closer to the heat source. Either way, I'd be a bit concerned about a home-engineered EWP setup that included "off" as part of its running cycle.

Off cycle? No i would never want to do that period--totally agree.
OD

Yep; but the AC didn't work at all at that point. I think I need to seal the sides of the radiator to prevent back-flow of the hot air, and I'm seriously looking at converting to S2 fans, if I can figure out the wiring.

The fan they are using in the race rx8s blow the S2 fans away. When I get the time I will take a photo of my old fan vs the new one. It restricts maybe 8% of the air flow vs the incredibly mismanaged stock system.

What fans are they using?

I have a photo of it someplace lemme see... Yes heres the photo from Eric Meyers. Its a SPAL fan.

https://www.rx8club.com/attachment.p...1&d=1298075769

vs the stock option

http://a1.sphotos.ak.fbcdn.net/hphot..._2573270_n.jpg

I use 2 of these:

http://www.cbsonline.co.uk/product/1...D_BLADE_FAN16C

Shift some serious air!

look at how much area is blocked with the stock fans though. Thats a good amount of real estate.

Oh that not even the worst of it. Stack 2 boxes on top of the fans--then you have a good picture.
Better fans are great but you have to clear the stuff out from the back of them for them to work well.
Do that and then the hot air is not redirected under the car---its shot right into the engine bay. At low speed/normal city stop light driving--thats a LOT of heat. To help with that low speed problem you will need a properly vented hood.
So instead of all that --just add the secondary radiator up front in a position to get good air flow.
It works and i think with the mentioned additions it just may be the best coolant temp control we can get.

I do run a vented hood which helps (even though a lot of people say it doesn't). But I have been able to keep the engine temps down one mod at a time to much more reasonable temps. With one radiator, improved fan, and by opening up the back of the oil coolers.

I think the most inmportant thing is to keep the temps in the correct range...and have the block as uniform temp wise as possible to minimize stress from different expansion rates.....

Bingo!
Great minds think alike:)
OD

Wonder if you put a 7 blade fan on (same as the other 7 blader, so you have two 7 blade fans), would help..

AS far as EWP, no experience there for me..

If one did that, don't expect 2x the airflow. First, you're blocking air with the front fan mechanism. Second, from the aviation world anyway, the corkscrew rotation started by the first fan tends to make the second less efficient. Here too, that twist will cause the airflow to be misaligned with the straight radiator fins which will increase flow resistance. Overall, it might help some, but there are better solutions IMHO.

you know after pondering on this for a little while (he) the best set up may be:
1- a more sensitive electronic thermostat with multistage fan control
2- axillary electric pump
3- stay with antifreeze water as the coolant.
Here are my thoughts.
let me preach on it!!!!!! Can I hear an "Amen!!"
I hear something but it is not that:)
Ok--- the weakness in our oem cooling system seems to be at the low speed, stop and go situations for most of us, or for those unlucky bastards living in the desert states the extreme temps. I choose not to speak about the demands of tracking the car as that is a different world.
AIR FLOW is king when it comes to cooling. Right? So number 1-- better control over high output fans will certainly be of help.
To take advantage of the increased air flow available with the fans, the idle and low speed coolant flow could stand to be increased---so the small ancillary exp that operates only during those times.
Stay with water antifreeze for the simplicity and familiarity. Evans is great stuff but on the street i dont think it will have that much advantage once a better cooling system is in place.
From looking at the options that I am aware of, it looks like the cost would be around $500-$600?
Worth it--for some yes, for all---dont know.
OD

Who suggested it would???

As for the rest, you are ONLY adding 2 extra blades , your analysis is a little WOT in this case.
Comparing to aviation is not a comparison..really RX-8 don't fly...do they.

The extra blades are disrupted by the airflow from the rest of the disc so the gain would probably be marginal anyway, if mazda stuck a 5 bladed one in there it was probably for a reason - same as the blades being so wide and thick will be there for more pressure/chord speed to overcome the blockage behind the radiator at the expense of outright airflow.

Yeah it is called noise..reduction

Doesn't really wash, more blades is usually less noisy for the same design perameters. Usually you'd just change the strut spacing or the rpms if you were getting an audible harmonic from the blades passing the casing.

Actually, it makes perfect sense from the standpoint of harmonics and resonance. If the fans have the same number of blades, then they have the same tone. Two sources of the same tone, or a multiple thereof, can cause a resonance that is drastically louder than the sum of the two sound sources.

Unless the motors or airflow were perfectly matched the chances of the tone being exactly the same would be fairly slim anyway, although the overlap could be close enough for it to be a consideration, but then, if the fans are running then outright cooling would probably be a priority over a little noise?
Anyway, as I said, it would make more sense just to move a support strut to alter the tone from one fan if the 7 blades really did have an airflow advantage. I think it's probably more likely that it was simply that a 5 bladed fan suited the airflow at that side of the engine bay on a standard car.

I guess where MAZDA have stated that they REDUCED fan blades to REDUCE Noise means NOTHING????

The blades in question are NOT RX-8 specific and are used on many other Mazda models, Tribute for one...and the 5 blade fan is about $7 cheaper for Mazda to use.

The Series 2 use the exact same two Electric motors with the S1 5 and S1 7 blade fan, , so what about the harmonic distortion and load because it is spinning another 2 blades with the same motors...OOh! watch out..!...your flux capacitor will blow out.


"Oh no its not.." :slap:

PIA...;)

What's the same motor got to do with anything? Doesn't mean the load or chord speed will be the same.

Ah it's clear now, I thought you were suggesting to add another fan in front of the radiator; I stand corrected. Air is air, to cars, airplanes, and living room fans. Do you know if the rpm is the same between series I and series II fans? If noise reduction is the goal, and the fan diameter is the same, adding blades will allow, up to a point, the same volume to be moved at a lower rpm hence less noise. The fan+shroud comes close to what's called in a wider sense, a ducted-fan. If the design is clever enough, the fact that the edge of the opening is close to the blade tips, plus forming somewhat of a cylinder, the gap can be adjusted so that the "induced drag" of the blade moving through the air is reduced. Almost without exception, fans work more efficiently the larger their diameter, so a single large fan may be more effective (depending how it fits in the space available) than two smaller ones.

Where did I say about adding another fan IN front of radiator??

Change the existing 5 blade to a 7 that is all, nothing more nothing less.

Series 1 has same FAN Blades as S2, but different size motors (S1 see pic), do they spin at same RPM??

Series 2 has upgraded (all new not used in any other Mazda's like S1 Motors) same two fan motors, do they spin at same RPM?, one would think so.

So, my suggestion was ONLY to use the same 7 blade fan and put on the Left where the 5 blade fan is...see S1 pic

https://www.rx8club.com/attachment.p...1&d=1298327090

Plenty of people--but lets keep this thread uncluttered.
od

I'm going to suggest a variation:
1. REmedy pump to minimize cavitation in stop and GO traffic where you have higher revs than average but lousy air flow
2. small ancillary EWP to boost flow under low rev, high heat conditions where you have more STOP than go

I'm sticking with Evans coolant so I don't have to worry about corrosion and I can even out the heat load if there's a defect in my deduction. Also, I can install it and forget it, without the hassle of changing it out and finding a legal disposal for it.

Understand your thinking.

I THINK I MAY HAVE IDENTFIED SOMETHING.
I am just a backyard shade tree work on car fool- Ok.
But last night as I was installing a super secret, black opps, completly unauthorized thermostat housing and thermostat, I noticed something.
Its hard to explain but I will try.
Currently I am measuring the coolant temp at the top radiator hose. My gauge is accurate to 2-3 degreesF. Now also I have the Pettit set up and we have an idle speed at 1.1K
I no longer have a coolant bypass system and I have primary coolant flow all the time.
I noticed something at warm up that may have been masked before by the by pass system?

At idle the engine was warming as expected. NP. But while keeping a close eye on my coolant gauge i would give the engine just a little gas--maybe to 2K and then immediatly let off the throttle so it would go back to idle speed.
When I did that I notice that my coolant gauge would bump up about 2-3 degrees. That little extra surge of coolant flow would hit the sensor and that coolant would be hotter than the coolant being normally circulated. This is NOT the result of momentarly increasing the engine heat by higher rpm--- the coolant cant react that fast and the increased heat would be minimum to none.
So that got me to thinking AGAIN---Oh God here i go--right?
Is the mechanical water pump circulating enough coolant at idle and low rpms? Is the increased temperatures during stop and go driving a result of poor air flow to the radiators, or inadequete cooling flow during idle and low rpm operation? Or is it both?
Interesting.
My 1st impression was and is at this time, the pump is not flowing enough at low rpm to properly cool? I have nothing to prove this other than my own observation, but it kinda makes sense. Stop and go driving is where most people have trouble with engine temps. Right? IF the pump is not flowing enough at idle and low speed (less than 2K) and we have poor air flow through the radiator its the perfect set up for overheating/hotspots.
The poor air flow has already been discussed and recognized. The possibility of poor coolant flow at low speed has not really been identified as a concern. My oridginal idea/thoughts did not include this. I was thinking more about proper and stable temps.
IMHO this is more casual evidence toward a anxcillary low speed operation only EWP and a better thermostat?

you stll get some corrosion with the Evans coolant. I was a bit surprised at the amount with the most recent change out. Otherwise this has once again become another RX8Club thread on arguing pointless minutiae ...

How often do you change out your Evans coolant? I'm not asking to start or continue an argument. You have more experience with it than I do, so I want to benefit from your observations.

What is pointless about trying analize as best we can how to obtain proper and stable coolant temps?
Who is arguing?
You know I am not in a real good mood this morning and your comment is just not resting well with me--not that matters to anyone.
Some of us try things, see some things, try to figue out the best we can how to address some issues, share our observations/thoughts/ideas in an attempt to correct or improve our surrent situation and then you come with an asinine remark about how once again this thread has become pointless.
IMHO there is something basically wrong passing judgement on others thoughts and ideas without explaining why those thought and ideas are not the way to go. Its almost sadistic.
One time you gave me some helpful information concerning not posting unless you really had something to contribute. I appreciated that . The Evans coolant finding you did share was great. But then you had to add the other didnt you?

I think you're onto something. I know there are other cars where low engine speed coolant flow is so bad that the owners' manual will say not to let the car idle for more than 5 minutes (I believe it was the early 1990 something Ford Escorts that were notorious for catching fire if left to idle too long).

Do you have the OEM coolant pump or something else, OD?

OD, Interesting observation, but I think you're missing something here: there's a thermostat. Plus, consider this for a moment, the cooling circuit is not only moving heat, it's moving information. In any condition other than the one where the thermostat is fully open, it adjusts the water flow based on its target full-open temp of 180 deg F. (Ideally, the feedback loop would look at the OBDII engine temp and keep that at 180 F.) What's your information flow look like? {radiator} --> {lower hose} --> {engine heat added (OBDII sensor)} --> {thermostat} --> {upper hose}. Increasing the rpm increases water flow, which pulls more heat from the engine, sending a burst of warmer water to the thermostat, then the upper hose and to your sensor. The thermostat reacts in a few seconds by reducing flow - and then you see your temp sensor drop down again. Note the phrase "pulls more heat from the engine" is judgment neutral. The goal is to keep the OBDII at 180 F.

My add-on gauge sensor is in the lower hose, eg., the water feed to the engine. When I'm cruising, the feed water is running around 165 F or 15 F under the target of 180 F. In > 80 F weather, then traffic stops suddenly and I'm then at idle, the input water temp rises over a couple of minutes, but the OBDII temp stays at 180 F until the input temp reaches 175 F, then the two temps will rise together with 5 F or less difference. (An idling engine doesn't generate all that much heat). If there were a flow problem at idle, it seems to me that there would be a bigger difference between the incoming water temp and the OBDII temps. (Note: I do have the Mazmart WP - things may be different with the stock one). When I get around to it, I'm gonna add a second WT sensor in the upper hose. That should give a more complete picture of the cooling system. In principle, with the second sensor and the flow rate were known, one could take a thermodynamic snapshot of the cooling system.

thanks for the info---good observations. If I understand correctly the difference between your post radiator temp and your hot temp is only 15 degrees F? 180-165=15?
So the radiator is only cooling your coolant 15F?
I thought there would be a broader difference between the hot and cool temps?
Wow.
I understand what you are saying about thermostatic function. The example I used of the temp "bumps" was something that I did not expect to happen. I thought it was a smooth gradual increase in coolant temps. Maybe I didnt explain well enough but these little bumps occured multiple times--whenever i would just slightly increase the rpms. I did stop at 185F hot side. It sure looked like when the coolant flow increased just a little the temp would jump--if i kept the rpms up to 2K the temps didnt jump like that--it was a smooth increase. If coolant flow is adequete there shouldnt be a difference--should it?

I have never measured temps from the cold side. Thats good to know.

180-165=15F. yep, silly math error (I edited my post to avoid future confusion). It may be off by a few degrees, but yeah, the small difference surprised me too. I did the "boiling water" check of the gauge sensor and it was right on at 212F, but of course if the calibration slope isn't quite right, it could be potentially off a bit at 165 F. But a clue that the calibration about right though is how the temps connect with the rise in the OBDII which results from the 180 F thermostat being open all the way.

There's no question that the flow rate is lower at idle than at cruise. Ideally, the heat load (rotor housing) should be connected directly to its heat sink (radiator). Instead, it's connected by water flowing through several feet of hose and radiator tubing, so many seconds connect the "information" of heat from one end to the other. Additionally, the sink isn't clean, because heat is lost not just there, but in the paths between. With no flow at all, the radiator will eventually get warm (long after your engine fries). The quicker the flow, the more "firm" the connection between load and sink becomes. You might repeat the test, and note the OBDII temp - is there also a blip there? Does it occur a little before the blip in the upper hose? Knowing the distance between the two, the time offset should give a way to estimate the flow velocity.

Really though, if the OBDII temp isn't showing much variation, I'd conclude the idle flow velocity is adequate.

/\..Wow.you have WAY too much time on your hands...

Well yeah, frankly, you wont find any real revelations for 99.9% of your driving, unless you are stuck in a traffic jam for more than 15 minutes in 90+ outside temps without moving (Air Flow), mods are a waste of time and money.;)

also remember the stock radiator sucks. It doesn't provide much time or the correct angles to allow the fluid to flow great.

The alternatives force the fluid from left to right back to left at 45 degree slopes, where as ours just flows from top to bottom.

Ok new observation.
I installed a custom thermostat housing that I modified to accept the Stewart barrell type racing 180F thermostat. The thermostat as three 1/16 holes to allow flow before the 180F temp is reached. That also allow any air in the coolant a way out.
I no longer have the coolant by pass system.
I am running regular over the counter antifreeze. Oem radiator cap.
WITHOUT my secondary radiator set up.
Ambiet temps right now is 75-78F with humidity of 40-60%

Warm up was only very slightly affected ( that surprised me).

Interstate running ---- i am running too cool. At 140--155F. (My oil temps are ok)

These temps are per obdII and my top radiator hose sensor.

In town/stop and go driving---170--180F.

My radiator is 7 yrs old. I have the mazmart pump.
Now what this tells me several things
1- the OEM radiator is good as long as it gets good air flow.
2- the mechanical pump is pumping too much for ordinary driving on the interstate at those temps. (Wasted energy).
3- the OEM radiator should not be a factor in whether or not the EWP works?
4- a better thermostat will only work much better as part of a cooling package.

Plans--- this thermostat has three 1/16 holes for constant flow, I am tempted to close one at the time to see if that gets my interstate temps where I want them. Goal is 180F--in town interstate and track. It may end up as simple as that and then during the extreme temp season (+90F) just hook up my secondary radiator (takes 5 minutes).
Then during the hot season--see if it holds during stop and go traffic. I have a feeling thats when the axially ewp and better fans will be needed?

Without reading the whole thread I'm not sure if anyone actually has tried an electric pump or not ?

A local here fitted one recently and is a regular race competitor - can find out more details if anyone is interested ?

i am..