I also had a section of clear tubing added in-line on the section to the throttle body, and that was solid water as more expect it should be (after the air got out of there that was introduced from popping the line off), so the difference between the two lines was quite apparent.

IMO the TB is too high and this is where air pockets normally develop when 'System' is not bled correctly when coolant is renewed..

2500-3000 is the Minimum RPM to run test, not below, I would go even higher, so If you are getting a constant solid block of water moving with a fine string of pin bubbles @ 4500 RPM, then I think you have coolant entry/blow back leak than only can be rectified with a rebuild.

The majority of returned hot engine coolant comes via your top Radiator Hose (therefore bubbles), ideally this is where you should check the return inside Coolant Bottle from top tank.

So, you have said with your system engine warm you can not start car with Cap Off and correct level of coolant inside (half full)?, quickly bring revs up to 4000 observe inside and look...
You keep your revs high and maintain it there, so it should not overflow...you don't let it idle, once you let it idle coolant will overflow from bottle as system is trying to pressurise (without a cap).

Have you done this?

I think I have done what you are saying, but not sure. Here is the sequence of events, all detailed out:

State: Car off, sitting in the garage, everything ambient temp at about ~80ish. Has not been started in several days
1) I disconnected the upper throttle body hose and add a length of clear tubing to the end of that hose down to the throttle body. Basically just extending the hose with a section of clear that I can watch for bubbles in. This added air to this hose obviously
2) I disconnected the upper coolant bottle return hose and added a section of clear tubing to the end, extending it for visibility just like the throttle body hose.
3) I plugged in my bluetooth ODB2 dongle, fired up Torque on my tablet and got a connection to the ECU for the gauges (coolant temp being the one I was interested in)
4) I removed the cap to the coolant bottle. Water level was at the FULL line
5) My wife started the 8 and immediately backed it out of the garage (catless, expecting overflow, etc...)
6) As soon as she came to a stop outside the garage, she brought the RPM up to 2,500 and held it there for the 2-3 minutes (or more?) while the engine coolant was heating up
7) during this period there was lots of air through both sections of clear hose, as I expected from removing the lines temporarily. Upper coolant bottle return hose was pushing some fluid, but not much, and so there was a pocket of air still sitting in it without going anywhere. The clear hose on the throttle body went completely solid, no air at all.
8) The coolant temp passed the 170F thermostat open point, and nearly immediately I got solid coolant flow from the upper coolant bottle return hose, pushing the existing air out into the coolant bottle.
9) This air disappeared for a short time, then returned as a stream of tiny pinprick bubbles attaching themselves to the walls of the clear tubing as they could.
10) The volume of these pinprick air bubbles continued to increase, RPMs still held at 2,500. The coolant bottle itself started to overflow very slightly.
11) I had my wife drop the revs to idle, and immediately started getting greater bubble flow and the coolant overflow out of the open top increased.
12) I had my wife increase the revs back to 2,500, and the high bubble flow continued, and did not drop off. Coolant overflow almost stopped, but not quite.
13) My wife let the revs fall back to idle, coolant overflow resumed
14) I got my tablet's video camera going to take the video posted above.
15) My wife increased the revs to 2,500 again, and got a gushing increase of bubbles in the hose.
16) I stopped the video and released my wife from assistant duty (she went back to making dinner :D:)
17) I made the shifts needed to get my 8 back in the garage, and pulled it in, shutting it off. The coolant level was still sitting at the top of the bottle.
18) Checking this morning, the coolant level was halfway to minimum level from the cooldown.

At no point could I see bubbles coming up from a lower hose in the coolant bottle itself, as the water there was pretty much constantly cloudy from the "Aerated water" effect.

Is your water pump OK? I had an engine that was burping coolant all over at idle....what it turned out to be was the water pump wasn't pumping enough at idle...and it would overheat and burp....

Get one of Paul's waterpumps..it is worth the price of admission :)

The water pump is a new(?) unit with the reman, so only 10,000 miles on it, and it's the updated one that is much closer to Mazmart's pump. It was not transfered from my old block.

The water pump failing wouldn't explain all this air in my coolant system.

So, next day after, coolant down to "normal" level, how does you car start from cold?..OK? (and yes I know you have have system pressure tested).

If you don't think it is pump, I assumed with new engine you would have a new pump.

I know you are probably mad with me but I would still try a "genuine" cap, and try and get system to normal, as in no air pockets, I don't think that is an issue with you though...overnight settled.... re-check.

I know it is not the same system, but you would not believe the number of owners I used to get when they used non-genuine caps in earlier Mazda's...This was when there were pressurised and non pressurised expansion (coolant) Bottles used on Mazda's, with or without cap seal, small or large spring, small or large lugs, small or large cap....now it is all at the Coolant Bottle.

I would not be concerned about bubbles at low or idle RPM, it is really above the 3000 RPM where the 'check' can work...the system is not perfect...no Water Pump can push the same volume at different RPM's.

The NC has two small returns at top of CB, you can externally see the green coolant flow in over plastic, at idle it stops to a on/off dribble, rev it up and she blows..:)

I haven't had a cold start problem yet at all. Can let you know when i get home from a day of sitting after the test I described.

I might just go ahead and order a new bottle and cap from Mazmart just to appease you :) I'm pretty solid that I'm keeping the 8 regardless of warranty coverage, and the new stuff certainly won't hurt regardless of which way that goes.

And no, I'm not getting mad at you at all. I'm glad you are responding and pushing me on stuff!

Any info from Blackstone?

Not yet.

I'm hoping that the oil in the coolers would have any of the coolant though. I've only got ~80 miles on this oil change :)

Haven't been able to drive it more with it dumping coolant like this.

Dude, call them and make sure they got your sample, they never take more than a couple of days. They are really quick usually.

Wait, did you take the sample during a change?

Hmm. I looked at the instruction paper last night and it says "1-2 weeks", so I wasn't worrying about it.

I'll call them tomorrow.

During, no. I had changed it about 3 weeks and 80 miles prior. I still have the old oil, but it is in my Pela pump, which has also pulled oil from the Miata recently, both the old engine and the new one, and it never empties completely when I pour out the pump container, so I can't trust that the oil in there is an accurate representation of what is in the 8.

Ah okay, they usually recommend the sample be taken mid stream as the oil is emptying from the pan.

I actually pulled it through the dipstick with a the Pela pump (cleaned pickup) and some timely vacuum management :)

It's 4 quarts of 0w40 Mobile 1 with 80 miles on it, I didn't want to dump it all! :lol:

:lol: I bet.

So....

https://www.rx8club.com/attachment.p...1&d=1345495604


I'm ordering a new bottle and cap from Mazmart, and if that doesn't fix it I'm going to hoses, water pump, and radiator. Possibly all at once?


Anyone think of a way that I could have a coolant seal failure and NOT have coolant in the oil? I don't know enough to be sure either way. :dunno:

Hmmmm. I can't think of how you could have a bad coolant seal and not have coolant in the oil.

bottle and cap is a good starting point.

Depending where the leak is it can end up in different places

If you have a small leak in the compression/combustion area you can get hot gas into the coolant under pressure, but no coolant leaking out. Not really common, but does hsppen

True. But I think he said he doesn't have bumbles in his coolant.

When it's cold, does it idle rough?

I just talked to the guy that wrote it up, and he said that even with the 80 miles on the oil, there is no way that if I was getting contamination that it wouldn't show up. So that's pretty conclusive from that end.

Agreed. Order already placed.

Hmm...

No coolant in the oil though means that if no coolant is getting into the engine, no harm being done there from it. The Blackstone guy said the same thing, with the wear metals being so low, there is clearly no damage being done at the moment.

Nope. It's got the typical periodical "puff" like a missed combustion, but like once every 30 seconds or so, nothing unusual compared to what I've seen/dealt with over the last 98k. My coils are headed toward the replacement mileage, so it's likely that. Plugs look fine (for a rotary). No extra gunk on them. Engine interior stays dry when cold. No hard starts hot or cold.

This is what mine sounded like with a cold start condition and a bad coolant seal. You can hear it smooth out after the coolant burns off.

my bet is on bad coolant bottle.