A different thought, maybe dumb, but different.

From other posts, the stock water pump cavitates at high RPM. Maybe more with older coolant, I have no idea. Water Wetter some have also said caused foaming at high RPM. What are the chances that the coolant is foaming coming out of the pump, the foam is getting into the bottle, and the foam cannot support the weight of the float. Might explain the high RPM issue. I only see it at high RPM.

Have the people who changed the bottle taken that opportunity to change the coolant as well, giving fresh coolant, that is less likely to foam?

Or, does all of this sound like crap?

I think it sounds like interesting crap. :)

Mark..........If you wanna come sit in the engine bay, I'll run it up to redline in 3rd and you can see if there is any foam in there? :)
I think you might be able to hang on long enough to get a look.......might be hot though!



I'm just kidding.
I guess anything is possible.
I put in a new bottle and destroyed the old one for this thread.
I did do a coolant flush/change when I put the new one in, so who knows?????

Dave, at RX-8 Production Start...April 2003, the Coolant Bottle was an 'E'.
Yes, the latest is now 'M'...and possibly the last.

:)

YOUR PM???

Yes sir........I know. Did I write something that lead one to believe I didn't know?
I must go back and re-read..........especially since I'm not aware I might have mis-spoken????

I edit my post above that said "The new revision is "M"" to read "The latest revision is "M'" maybe that was it?

The problem been, if the light doesn't come on when they turn the key into the on position, they'll look to see why not. I presume it comes on with the rest of the warning lights in that position. Must check. I've heard stories of cars failing before they even hit the ramp, on airbag lights.
Welcome to the nanny state.

HUH....do you Really believe they are going to look for a Coolant Warning Light??, at Ignition ON, after all it is NOT a Safety Issue like Air Bags, beside NOT every car has the same warning lights/sensors.

WHY do you not "try" the suggestion first, that is unplug, Ignition, Start.

I can assure you it will not flash once started, as far as On when Ignition is ON, you can check that one.

The low coolant light has started to come on at odd times,but always with high rpm and 7000rpm to clear carbon.Reading here seems to be the sensor.The question is,any danger to the engine if you coolant level is normal and can you go to higher rpm with the light on.Thanks

No problem. It's just a bad sensor, does nothing except put the light on.

Mine comes on at random times, mostly when just cruising along, not necessarily at high rpm. If a passenger sees it, I've got a couple of stories I tell them, depending on whether they think it looks like a Rolls Royce radiator shell or like a Greek temple.

Ken

^...........think most of us would agree with Ken.
As long as you have proper coolant level and mixture, it shouldn't be any issue what-so-ever.

It usually starts out just coming on at high rpm's and slowly gets worse where it will stay on more and more frequently, even at lower rpm's.

My Coolant Bottle OEM Overflow TUBE MOD.
 

Thought I would add this to Dave's excellent thread...:)

My own Overflow TUBE Mod..with a Genuine Mazda Coolant Bottle Hose from the FD RX-7.

It is well documented that the "original" RX-8 overflow tube has its limitations, that being it is small short and known to direct Coolant down and around Power Steering cabling and plugs...NOT GOOD.

Now some guys have simply bought a length of tubing from Autozone or wherever and connected to Coolant Bottle overflow and re-directed it.

I decided on another option...

With this Genuine Mazda Part you get a stronger black overflow tube of around 12 inches long in total, but, the ends are heat molded with the original 90 degree bend and another longer 90 degree bend at the other end of this RX-7 tube...so it looks like a long U shaped hose.

What to do....

Connect the smaller looking 90 degree end to Coolant Bottle OUTLET, bring the tube around to the front left of Coolant Bottle and rest between your painted Radiator Support Panel, then around the Air Conditioner High Pressure Switch and then position the larger end of black Tube OUTLET DOWN so it is located just in front of the Headlight Seal area, there is a gap and the top corner of Radiator can be seen..

So basically the Tube or Coolant Bottle Overflow is now positioned top far left of radiator as you look at it (Passenger Side L.H.Drive)...So if it does overflow coolant should not get anywhere near the Power Steering Wiring.

Just to add, you do not need to push this new tube through any holes, it just sits snugly on the top between your Coolant Bottle and Painted Radiator Support Panel.

What I also did (OPTIONAL Mod) was to place a very small plastic catch bottle (about one or two fluid ounces in size, that is all that will fit) and the new black outlet tube/hose sitting inside of this small plastic catch bottle, so basically any minor discharge from my Coolant Bottle is caught...Frankly, if you are spewing more than an ounce or two of Coolant out of Bottle you have other issues that need looking at.

So, How does it look...VERY Neat and original looking..
So, How does it work...so far no problems, I was not expecting any.

Where to buy the RX-7 molded factory OEM Coolant Hose??, from a Mazda Dealer.;)
Montgomery's, vendor here is just under $7.00.

Or Paul from Mazmart will get it in for you.

Don't delay your purchase as stocks are limited before a price increase.

Part Number..
N3A1-15-383 Hose - SU Made in Japan

https://www.rx8club.com/attachment.p...1&d=1292978565

I have the latest Mazda EPC (December 2010), and the M Coolant Bottle revision is still the latest used in production and IMO the last...there has been no change since May 19,2009.

Thanks! Good to know!

I also like your overflow hose mod, but as you might be aware, I had some blue hose left over from my catch can install, so I used that and provides a better match for me. I had enough left to run it down beside the fan all the way down to one of the holes in the undertray.

Recently I'm experiencing the coolant light issue too.
Here's what happened:

About a month ago it came on the first time around 4k rpm, figured it's low, so I filled it up (just around F mark) with regular coolant. And it never come back.

I did my coolant flush few weeks agot and replaced my coolant with 20/80 coolant and distilled water plus water wetter. And I also over filled the F mark, but not all the way to the top of the bottle. Since then, my coolant light comes up from time to time, usually after highway spirit, around 4 or 5k rpm. It goes away about a min or so, did NOT lower the rpm.

One time I poped the hood right after I parked the car and saw coolant/water boiling inside bottle. Some came out of the overflow hose too.
Has anyone experienced this?

Would overflow cause the light to come on?

Thanks in advance

Overflow wouldn't cause the light to come on IMO, but shouldn't overfill and better extend that overflow tube so it doesn't get on the power steering connectors!
Keep and eye on it, you may have other issues like a bad thermostat. I'll have to see if mine boils in the bottle, but I know I have popped the hood and looked right after a good drive and not seen it. You have any way of checking your actual temps?

No, I don't have water or oil senders....but planning to get them in the future tho. I'll extend the tube for now, i guess.

My other guess is that, cuz' it's overfilled, while in high rpm/temp (boil?), it disturbed the magnetic ring and trigger the reed. Just a guess...

I'll definitely keep my eyes on it...

Quick update, after I leveled the coolant down to full mark. The light hasn't come on so far.
I'll keep my eyes on it in the next few days and see how it goes.

If my theory is correct then that means light go on simply because the heat pressure from the coolant created enough turbulence to bounce the magnetic up and down to trigger the reed. Maybe that's why the top of the bottle is a distance away from the F mark? To keep certain amount of pressure/heat which also stabilized the magnetic ring.

Then that'll explain why light come on when coolant is low or overfilled.

We'll see...

It's good data, but we pretty much know that the float lost buoyancy over time in the old bottles. You may however have caught another interesting phenomenon in your case.

I had to run back and forth to the airport in Orlando yesterday cruising around 82mph for a round trip of 125 miles and as soon as I got home, I popped the hood and no boiling. My water temp never did get above 200F the entire run over and back though.

Ok, so it came on again...and this time after I parked the car and poped the hood, there are smoke coming out, which smells like coolant and the coolant are below the F mark in the bottle. Which leads me to overheat issue rather than simple sensor issue...

I'll do more search on it for now....

Yep...........I was afraid of that............good luck!
You have any puddles under the car? Hoses look okay?
Have any in the exhaust? (Worst case scenario)
Here's hoping it's just the thermostat!

Fill it back up with car off and cool. Pop the hood and wipe all the hoses down, then start her up and let it get to normal temp. Shut it off and see if you can tell where steam(coolant) is coming from. Of course be careful though, it will get mighty hot and don't let the car overheat. A lot of times it can be difficult to see a leak with car running, but once you shut it off, it will show up!
Oh................look around your cap too, some have reported cracks in the lip of the bottle where the cap goes on.

No, nothing under the car that i notice, but I'll double check.
And nothing in the exhaust, thx god. lol

I'm sure the steam is coming from the coolant that came out of the overflow hose, because there is a small puddle at the cooling fan "cleavage". lol
I'm looking into cooling fan and thermostat at the moment...something is just not functional....

Ma car needs to be cool! lol

Yeah, make sure both of your fans will free turn.............spin them with car off and see if they spin freely.
Then you need to see if they are coming on when they should.

Any parts that could cause cooling system not functional?
I don't have any CEL, just radiator light from time to time, so i'm guessing it might not be thermostat, but I don't know...

How about water pump? How can I tell if it's functional or not?
I'm no mechanic in anyway, I would appreciate any suggestion at this point, and look into those.
Thanks in advance...

Found this...

In addition...

/\ Sorry to come into this late, BUT, did you bleed out AIR in your cooling system originally??..

By now though it should have worked its way out with a few drive cycles of hot and cold, would have seen your coolant bottle level get lower?..

Just a thought I would throw that in there???..

Hey, that's a really good point!
Glad you finally arrived.........I've been waiting for other ideas! :)

Yeah, already bleed the air.
Just got a new cap, same psi, maybe I should have get a higher one...
The inside of the original cap is a bit loose, i don't know if that's it, cuz' the new cap is pretty tied.

So far so good, and we'll see...

I have been really slack with my coolant and haven't flushed in about 18 months. I started to get the coolant low light when there was plenty of fluid so decided it was time to do a flush and clean the overflow bottle. The problem only really happened when I was at constant rpm for a few minutes, ie highway.

Unfortunately while removing the overflow bottle I failed to read up and note the proper procedure and broke the connector on top of the OEM radiator. I now have a BHR Radiator on the way but thats all another story.

My overflow bottle had a LOT of crap built up. I have flushed with water and removed a lot of it. I have filled it up with acetic acid (White Vinegar) overnight and that really didnt do a lot. I then tried citric acid but again nothing. Guess I will just leave it plugged up with some fluid in it and hope it loosens some of the remaining gunk. Then when the radiator turns up will see how it goes. Also going to get a new cap just in case.

Cheers

Andrew

I thoroughly cleaned my 7 year old reservoir with various bleaching agents and final hose blasts (still slightly yellow). Soaked it in vinegar for nearly a month stirring weekly and the float (although free to slide up and down now) sinks like the Titanic in water. I had bought a new reservoir when I replaced the OEM radiator with a 3 row aluminum so this was merely a test prompted by this great thread. Knowing what I know now I wouldn't have replaced the original reservoir - just cleaned and left it unplugged.

Quick update:

Since I changed to the new cap, the boiling stops, no more coolant leak.
But my sensor eventually failed. lol, so I just unplugged it for now...

Just want to make sure I understand you.................so you are in the camp that believes the float just looses buoyancy over X amount of time as I do?

Absolutely agree with you.
(The floating magnet works the switch perfectly when turned upside down but has no buoyancy.)

Okay, I just wanted to make sure as I still want to compile the results so to speak, so I can kinda keep the thread updated. I went back and sort of put the running results(if you will) in the first post, so those that don't want to read the entire thread can skip to the chase.

I do think cleaning the bottle helps, but that doesn't seem to last very long. The reality might just be that when it's taken out for a cleaning that the magnetic float actually dries out a little bit and then when put back in it starts to absorb coolant again over time and temp, get's heavy yet again............and WALLA. :dunno:

Vindication is so sweet on occasion!

Notice what it says under "Description".


Check it out!

I also spot what I think is a very, very important item. Check your VIN number because according to how this reads, I could have replaced just the bottle and not the sensor itself(the reed switch) since I have an Oct. 05 build date, but I'll have to confirm that against my VIN.
This assumes of course you can actually buy the bottle without the sensor.................and not sure you can. :dunno:
The float(the culprit in most cases) come with the bottle and not the sensor itself.

Also...........this implies Tommy needs to do an update to the finishline site as it doesn't list this bulletin.



If you want to check to see if your sensor is bad before attempting to buy the bottle itself, or whether you need the sensor and the bottle, then do this with the bottle still in the car:


1.) Unplug the sensor at the connector located to the left of the bottle.

2.) On the sensor side of the connector, use an ohm meter and check the resistance between the two terminals. With it unplugged, and assuming you have adequate coolant level, and a "good" float, you will measure > 2meg ohms. This means the reed switch(sensor)switch is open meaning............
a.) You have adequate coolant and........
b.) The float is okay and floating like it should be.

3.) If you measure less than say 100 ohms(I forget the actial number and finally threw away my old butcher shaft and magnet, but I think it was 10 ohms), then the float is riding low on the shaft, the reed switch has closed, and the float has lost buoyancy!

Dave, perhaps I can add some information here to simplify things.
You have not or are not reading the Bulletin correctly:) as YES you can Now just Buy the Tank ONLY (N3Y1-15-351) and the Tank with Sensor (no pressure cap) (N3H1-15-351G).

ALL RX-8's S1 and S2 produced before May 19th,2009 Will have the "Faulty Float" -Tank Installed.

If you have had your Tank replaced with N3H1-15-350M (ONLY) you have the latest and greatest.

Or if your S2 RX-8 was produced after May,19th, 2009 you have N3H1-15-350M already.

ALL other "iterations" are Defective.

Also, It appears there is NO Extended Warranty for this Bulletin and Repair.
It Only applies to Customers who Complain that their Coolant Light comes ON.
And..THIS REPAIR Will be covered under Mazda's New Limited Warranty term.


All "Defective" Iterations..

N3H1-15-350E Original
N3H1-15-350F Used from May 19th 2003 PRODUCTION
N3H1-15-350G Used from Sept 18th 2003
N3H1-15-350H Used from May 22nd 2005
N3H1-15-350J Used from July 1st 2005
N3H1-15-350L Used from Oct 1st 2008

Plus, IF you had any of the above Complete Tank Assemblies replaced under Warranty you still have a "Defective" Tank Unit.
That will give you Dash Coolant Light issue...eventually.

BELOW here is the same TSB as Dave posted above, just a little "Clearer" on the eyes..:)

Dave,

This Bulleting is a little bewildering...lets see If I get it correctly..

They are saying that the "Float" absorbs Water and Sinks, OK, and have now modified Tank and Sensor and I guess Float.

However,They also say ..

Up to July 1, 2005, Replace the tank with sensor.
From July 1, 2005, Replace TANK ONLY.

So if you purchase the TANK only, does it come with a new Float?, after all it is the FLOAT that sinks/absorbs coolant over time.
IF the 'TANK only' does not come with a new Float, then you are 'reusing' a Float that is over 5 years old?

This really needs to be cleared up by someone buying the two new parts and telling us what is exactly included.
Edit: Both new TSB replacement Parts Comes with NEW FLOAT.

WARNING
The above Production Dates assumes you still have the ORIGINAL Installed Tank, IF, you have had the tank Replaced over the years that Production Date information is somewhat USELESS.

*rolling my eyes*

if someone willing to buy the bottle later sure. :P

Dave,

I am putting this here in response to the TS Bulletin and to help those who have had their Tanks Replaced on making a decision, if they want to.

Now, this information applies ONLY to those who have ALREADY had their Coolant Tanks Replaced with the following replacement Tank by PART Number.

N3H1-15-350E Repair Existing Tank with New "Tank with Sensor" P/N N3H1-15-351G as per TSB.
N3H1-15-350F Repair Existing Tank with New "Tank with Sensor" P/N N3H1-15-351G as per TSB.
N3H1-15-350G Repair Existing Tank with New "Tank with Sensor" P/N N3H1-15-351G as per TSB.
N3H1-15-350H Repair Existing Tank with New "Tank with Sensor" P/N N3H1-15-351G as per TSB.
N3H1-15-350J Repair Existing Tank with New "Tank Only" P/N N3Y1-15-351 as per TSB.
N3H1-15-350L Repair Existing Tank with New "Tank Only" P/N N3Y1-15-351 as per TSB.

OR if you do not know and or do not want to go to all the Trouble of Pulling Tank apart and Installing either of the above parts
you can just replace the TANK Assembly again with the Latest and Greatest and last P/N N3H1-15-350M.
For Most Owners it will only costs an extra $63, (at the moment).
Some how I think the 350M will increase in price once existing stocks are sold out.

No problemo.............it was just something I had located.
Yeah, the date is mysterious mate. My car is an Oct 05 build date and my float definitely absorbed and sank. So who knows about that date.....that's what I found strange.

The only thing I was really interested in was..............."it is apparently the float".

Not sure I put it here or in another thread, but I put how to test that sensor before buying both the sensor and bottle now, cause as we both hopefully suspect, the reed switch seldom will fail............. and IF I read it correctly......you may save the sensor and just buy the bottle...........which would have to come with the "new" float. Course if the sensor is like $5.00, then who really gives a damn!

I am now certain the new Replacement Parts will come with a new Float.

If you want to do the repair Job, I would go with the N3H1-15-351G Tank with Sensor, it is ONLY around $24 more than the Tank, all you have to do then is replace it and re-use your old Pressure Cap.

The way to go for all owners....

Yep............totally agree. If you gonna buy the bottle(float comes with it), just buy the damn part number that includes the sensor as well.

Again, just wanted to know that we finally got around to the right answer on the float absorbing being the cause and post it. :)



N3H1-15-351G Tank with Sensor is the way to go!

Yep, Agreed! ;)

I just posted ALL the other information, to try and Stop all the questions, of "I had this 350H fitted then", what do I do...blah, blah, blah.

The 351G IS the Best value and easiest one to do.

I can't understand the factory even making the other option (Tank ONLY), to save them what a few dollars for the few S2's (like mine) that are still covered under warranty?...that is provided their Dash Light comes on, as there is No extended Warranty.
There is an extra Labor cost (time) to them for the Tank only repair Option.

Check this out!
Get'em while they are hot..........Good price people!

$79.95 if you go without the sensor(N3Y1-15-351)........and ............$99.95 if you get the sensor(N3H1-15-351G).

http://www.finishlineperformance.com...cat=336&page=1

Ash................checkout the ordering info selection!
Don't know when it appeared, but had to be recent! (ha ha ha)

Sorry Dave, YES he is a little cheaper than Monts..

Can you PM me on the Ordering Info Selection, can't see what you are meaning...sorry..

For those in Aus, Finishline charge $160 US for shipping, Online Mazda Parts ... $55 US and are only $2 more expensive.

Ash, I think he is referring to the commentary as follows:

• For VINs JM1FE****** 100000 to 157837 (built before July 1, 2005), replace the tank with sensor.

• For VINs JM1FE****** 157838 to 403405 (built on or after July 1, 2005), replace the tank only.

WARNING
The above Production Dates assumes you still have an ORIGINAL Japan fitted Tank, IF, you have had the Tank replaced over the years the Production Date information is somewhat USELESS.

Just go with N3H1-15-351G, which is the Complete Tank Assembly minus the Pressure CAP.

As the TSB Bulletin says, Reuse your existing CAP.

The Finishlines Pic is incorrect as it shows Tank with new CAP attached, the pic is in fact a N3H1-15-350M.

Just order N3H1-15-351G which will fit ALL model RX-8's.

As I said, IF you already have had the N3H1-15-350M Tank installed under warranty, you already have the latest Float, Sensor and Tank....No need to do anything.

It will be interesting Dave and Andrew to see if the 350M and 351G new float will last, it has been in later Series 2 for only 20 months...time will tell I guess.

With this one I really cant understand WHY Mazda took so darn long to rectify this!...if they have.

Ash it's interesting, I wonder how long Mazda hav known the issue is the float. As was pointed out here, when I first took apart my coolant bottle to see what the issue was I also notice the float wouldn't float properly. Mazda have a knack of fixing things when a thread like this comes up.

Well.....here's my theory on that "auzoom".
Mazda must be aware how many coolant bottles are sold and/or used in warranty work.
As the numbers begin to show they must say............."WTF is going on" and start looking into it.
There were many threads on this site(mine is but one) and of course there are numerous other sites as well, plus all the data from dealers.
This thread was started approx in September of 09 and TSB is dated July of 2010.

Dave you are DEAD Wrong....Sorry Right!:naughty: