Track day #2=Overheating + Code 2?

[R13]

If you plan to replace it anyway you might want to give this a try before you do.

Use a sharpie to color the tank neck sealing surface. Then take a sanding block with fine sandpaper(start with something like 400 grit, you may want to finish with 600 or 800 grit) and carefully sand the neck surface. Any low or high areas will show up as you sand off the sharpie marker. Once the surface is perfectly flat, give it another try.

Not a bad idea. I'll be sure to plug the hole to avoid getting grit in the tank

This would work on the upper seal, but might be more tricky on the lower seal. However, i think the problem is likely with the upper seal so I may get somewhere with this

 

[LJSB]

A bad thermostat can open and close fine and then, stick sometimes and not always the same way and then again, go back to working fine. It can trick you. If you do decide to change it go OEM.
Be careful. It's not going to take much overheating to warp those heads.

 

[R13]

A bad thermostat can open and close fine and then, stick sometimes and not always the same way and then again, go back to working fine. It can trick you. If you do decide to change it go OEM.
Be careful. It's not going to take much overheating to warp those heads.

Since I'm going to be bleeding the system anyway after replacing the coolant tank, I'll do the thermostat at the same time. I just paged through all of the invoices I have from the car as well as from before I bought it and haven't found anything receipt for or evidence of replacing the thermostat (I know I haven't done it), so it could be a gagillion miles old at this point.

I've done the lifted-head thing once with my MR2 so I'm super paranoid about running a car hot. The temp gauge has only hit the top once for about 5 seconds at the track and as soon as I came off the throttle it dropped immediately to normal again so I think there was a bubble that was responsible for that. Otherwise it's never gone above 3/4 on the gauge and I watch it like a hawk.

 

[R13]

Ok, so I swapped the thermostat today with an OEM Honda part. I don't know what was in there, but it wasn't Honda anything. It doesn't look in poor condition although it was not holding a seal even before I pulled it out.

Here's the one that came out:

 

[R13]

The new coolant tank should be here tomorrow. After that it'll be fill, bleed, & test.

 

[LJSB]

Vacuum filler.

 

[R13]

Ok, new tank installed & system bled. I didn't have access to a vacuum thingy, so I did this manually (i.e. messily).

While I was fiddling, I also performed an actual compression test and got consistent, in-spec numbers as well, so apparently the chemical test wasn't lying.


I also had a thought (grant it WAY after the fact here).


I believe it is possibly the symptom of venting coolant vapor could be due to simply having the system over-full. Logic follows:

-The coolant expands and compresses the air remaining in the expansion tank.
-Once the air compresses to a certain point it will exceed the pressure of the cap and vent air/coolant vapor
-SO, if there is insufficient air-space in the expansion tank, venting could occur even under normal operating conditions.
-Also, since no measurable amount of coolant is lost in this process AND since the pressure cap allows air to come back into the system this could happen consistently.

It makes sense to me, anyone who can shoot some holes in that theory, please do.

 

[R13]

Update:

Car pretty consistently would get hot 12-15 minutes into a track session at VIR with 40-65 degree ambient temps two weekends ago, so the issue persisted (also go the Code 2 CEL again).

I got ahold of a vacuum filler yesterday, stuck it on the expansion tank and pulled a vacuum until the hoses were pretty flat both in the engine bay and near the ratiator. I left it like that for a bit to confirm that it was holding pressure (it was). However, I noted that the two large hoses connecting to the thermostat housing where flat, but the hose going to the heater core was not.

It drew about a quart of coolant in, about half of which I siphoned back out to keep the expansion tank filled to the correct level. I suspected that the heater core may not have been in the loop for whatever reason (I had run the car with the heater on max the last time I'd driven it) so I took the car for a drive with the heat all the way on. It would blow hot under throttle, but would only be luke-warm at idle. I parked the car again and burped the heater loop via the bleed nipple in the front compartment.

I connected a clear tube to the nipple and hung it (with the end in a coke bottle) from the hood striker. With the car off the fluid climbed the tube to about 3 inches above the weather seal after spitting out some small bubbles. I started squishing hoses in the front compartment as well as poking the hose in the engine bay and kept getting bubbles every time. With the thougth of speeding up the process, I fired the car up and blipped the throttle just a bit (2500-3krpm max) and that really got things going. There was a LOT of air in there.

After I got to a point where I was no longer getting huge bubbles, I turned the car back off, closed the bleed nipple and then repeated the vacuum bleed/fill process now that there was no question that the thermostat and heater core passages where open. Another quart of coolant found a home somewhere even after siphoning the expansion tank afterwords.

I'll be autocrossing this weekend so we'll see if any of the sympoms still persist.

 

[EdinKali]

I have a feeling you found your problem.

 

[FelonyRuckus]

My cousin had some of the same symptoms with his '93 with about 80,000 miles, it ended up being a blown head gasket. He took the car to an "open track" at Willow Springs on the big track. He would do a few warm up laps and it was fine till he started really pushing the car, it wouldn't get hot during regular driving though.

 

[R13]

I'd feared a HG problem, but I can't seem to validate that concern with any static tests. Also, the problem hasn't gotten any worse between the two track events where I've seen the problem.

I hadn't been to the track prior to replacing a cracked expansion tank, so I have a feeling that's where my problems started. I would almost suspect a flaw in the heater core before a HG, except that I don't have any of the classic symptoms for that either.

I suppose with the system bled, it'll either go away, I'll end up parked somewhere when the HG completely goes, or it'll get worse and fail a compression test later on and I can make an expensive appointment with Barney.

There are still some x-factors in the equation though, namely a radiator of indeterminant age/mileage (potentially 18 years and 228k miles).

 

[StmpoRaceProducts]

Two pieces of knowledge for you... not mentioned yet.

If your head gasket is blown... you would be able to tell by running the motor with the coolant tank cap off. If it spits out water and air... it is the HG

The plastic radiator shroud is also important for our car... is it still installed? If not.. it could be the problem as well.

I hope you find a resolution soon.

Regards

 

[ChopsJazz]

The plastic radiator shroud is also important for our car... is it still installed? If not.. it could be the problem as well.

This is an excellent, and often overlooked point.

 

[R13]

Thanks for the suggestions.

If I run the car with the cap completely removed, there is some swirling of the fluid, but no significant bubbling that I can tell, and certainly no massive overflow even after the car is up to full temperature.

Also, If you're referring to the plastic ducting that goes between the front bumper and the front of the radiator, that is intact on the car.

I plan to really dig into this in the coming weeks and hopefully get to the bottom of the issue.

 

[R13]

Yet another update:

Did a charity autocross on 11/20 and played taxi duty for local law enforcement & beneficiaries. The car was doing fine through my 6 competition runs (i.e. hard launch,~40seconds of 6-8k rpm, near the limiter in 1st, and ~50-60mph max followed by ~10 minutes in grid). I'd been checking it after each run. However, after that, I was giving rides to some local law enforcement and some of the donors for the charity and the car started overheating after probably another 5 runs taken more-or-less back-to-back (<1 minute down time between runs). The Coolant level started the day (cold level) with the Dali tank ~50% full, but overflowed that by at least 20 ounces (I have an....improvised...overflow bottle that's 20 ounces in an effort to avoid leaving puddles until I figure this situation out).

As in my previous update, I had vacuum-filled/bled the system as well as manually bled the heater loop less than a week before this event.

This past Saturday I again opened the bleed nipple on the heater core attached a long tube and ran the car at varying RPM with the water valve open for ~45 minutes and was still getting bubbles of various sized. After the initial purge of air, it would not bubble at idle, but if I blipped the gas to ~3krpm and then let off, I would get another flurry of bubbles.

I don't have any of the classic symptoms of heater core failure (i.e. wet floorboards, constant window haze, etc.) I've smelled coolant inside the car, but it was very faint and I assumed it to be coming from the engine bay where it seems there is a constant supply of vapor from whatever the problem is. At this point, however, I'm running out of ideas.

I'm in the process of acquiring a pressure tester (i.e. one that will apply pressure as opposed to vacuum) to see if I get anywhere with that.

 

[StmpoRaceProducts]

Yet another update:

Did a charity autocross on 11/20 and played taxi duty for local law enforcement & beneficiaries. The car was doing fine through my 6 competition runs (i.e. hard launch,~40seconds of 6-8k rpm, near the limiter in 1st, and ~50-60mph max followed by ~10 minutes in grid). I'd been checking it after each run. However, after that, I was giving rides to some local law enforcement and some of the donors for the charity and the car started overheating after probably another 5 runs taken more-or-less back-to-back (<1 minute down time between runs). The Coolant level started the day (cold level) with the Dali tank ~50% full, but overflowed that by at least 20 ounces (I have an....improvised...overflow bottle that's 20 ounces in an effort to avoid leaving puddles until I figure this situation out).

As in my previous update, I had vacuum-filled/bled the system as well as manually bled the heater loop less than a week before this event.

This past Saturday I again opened the bleed nipple on the heater core attached a long tube and ran the car at varying RPM with the water valve open for ~45 minutes and was still getting bubbles of various sized. After the initial purge of air, it would not bubble at idle, but if I blipped the gas to ~3krpm and then let off, I would get another flurry of bubbles.

I don't have any of the classic symptoms of heater core failure (i.e. wet floorboards, constant window haze, etc.) I've smelled coolant inside the car, but it was very faint and I assumed it to be coming from the engine bay where it seems there is a constant supply of vapor from whatever the problem is. At this point, however, I'm running out of ideas.

I'm in the process of acquiring a pressure tester (i.e. one that will apply pressure as opposed to vacuum) to see if I get anywhere with that.

You just supplied enough info there man... Dont shoot the messenger. But it's 1 of two problems at this point.

1) The nsx has a complex bleeding system.. you may want to do a full bleed. Check wiki or search prime.

2) Minor blown head gasket issue that turns into a bigger one. Apply throttle while the cap is open.

Sorry bro... I hope im wrong.. but I've been through this before.

Regards

 

[R13]

Quick update:

Santa brought me a pressure tester and venturi bleeder :wink:

I figured before I bothered attempting to bleed the system I'd use the pressure tester.

I found two leaks. One was a crunchy hose between the heater core and the water valve seeping around the connection, and the other was one of the large connector hoses under the middle of the car. I replaced the former. The latter had been installed by a local shop who apparently thought that 1/2" of pipe engagement was sufficient for a 1+" diameter radiator hose....(note: that mechanic is no longer employed at that shop). The latter problem wasn't really obvious until I let the car sit with the system pressurised for ~2-3 hours or so (@16psi per the manual). It's probably been leaking onto the plastic undercover and then either evaporating or dripping out the back of the car.

I re-tested the car after fixing these issues and it held pressure for several hours without problems.

Going to bleed again tonight.

 

[docjohn]

I've been following this thread...nice to see you found some hidden leaks...nothing worse than nagging non fatal mechanical issues preventing 100% performance at the track.

 

[pgilliam1]

Quick update:

Santa brought me a pressure tester and venturi bleeder :wink:

I figured before I bothered attempting to bleed the system I'd use the pressure tester.

I found two leaks. One was a crunchy hose between the heater core and the water valve seeping around the connection, and the other was one of the large connector hoses under the middle of the car.

If you haven't replaced all your hoses - at this age and mileage you should. It's a PITA but don't forget the 2 hoses going to the oil cooler. It's Ok to reuse the OEM hose clamps.
Happy Motoring!

 

[R13]

If you haven't replaced all your hoses - at this age and mileage you should. It's a PITA but don't forget the 2 hoses going to the oil cooler. It's Ok to reuse the OEM hose clamps.
Happy Motoring!

Good advice, but the two straight hoses between the water valve and heater core are/were the only old ones on the car. Barn man replaced the rest of them when I had the T-Belt done within the last year and a half.

 

[WPG_NSX]

Almost does sound like a hg problem, but you should try using that chemical that detects exhaust gases after you notice the temp guage starts to climb after a race.

 

[R13]

Almost does sound like a hg problem, but you should try using that chemical that detects exhaust gases after you notice the temp guage starts to climb after a race.

Yeah, that seems to be a common diagnosis. However, having used the block tester (chemical test), and done a compression test...both of which seem to indicate healthy HGs... if it is a HG problem then it must only occur once things are very hot, at which point it would be very tricky to use the block tester on account of the pressure that would be built up in the coolant system.

In the past month I also exchanged PMs with VBNSX who had seen some similar symptoms, and his problem was solved through extensive bleeding of the system.

Now, I've bled the system a few times, however, you can bleed all you want and not get all of the air out if you have leaks in the system. Which is what I've been trying to address lately.

 

[R13]

Yet another Update:

So....

After testing the system under pressure, and replacing or more-securely-installing a couple of hoses, I moved on to attempting to bleed the system using the venturi tool.

Instructions are pretty straight forward: draw vacuum / make sure it holds a vacuum / fill under vacuum.

Step 1: Check.
Step 2: We have a problem.

The car would, in no way hold a vacuum, even at only 15lbs (tool manual suggests high 20s depending on system volume). I went back to the pressure tester and it appeared that either the hose to the heater core that I had just replaced was still leaking, or the cap on the heater loop bleed nipple was still leaking.

In reality, I discovered what may have been the problem the whole time. To verify if the bleed nipple cap was the culprit, I pressurized the system and then proceeded to gently squeeze the cap on the sides, poke it from the top (I apologize for this sentance, I really do), and twist the cap around. Somewhere between the sqeezing and the twisting, the entire nipple just popped off in my hand....followed of course by a geyser of coolant.

Upon closer inspection it became clear that someone had previously broken the sucker off and then, rather artistically, glued/epoxied it back on. It was leaking/sucking air from around the base of the nipple where a pin hold had developed in the glue.....Wiskey, Tango, Foxtrot...?over.

As a side note, Acura no longer carries that particular heater pipe, and there are actually two styles. The '91/'92 cars have an open nipple (I hate nipples), with a rubber cap on it. The '93 and later cars have a proper threaded bleeder fitting. I managed to source one of the later-style pipes from a Prime member.

After installing said pipe, I have not had a chance to do a thorough bleed/road test because I have had the entire front suspension off the car being re-bushed, but I have been able to hook the venturi tool back up and can confirm that it will hold a steady vacuum now @15psi. I can't get much more than that without draining more cooland than I have, however I'll probably do that just so I can be thorough about the whole thing.

If I get everything back together in time, I've got a track day coming up on 3/5 which will provide a nice final exam to this round of trouble-shooting.

 

[R13]

Just to put some closure to this thread in case I need to reference it in the future, or if anyone else finds it searching.

The problem was a flaw in the Head Gasket that, under load only (i.e. not at idle), was allowing combustion gas into the coolant system.

It would pass a block test at idle, but would push coolant out the overflow under load. As it worsened over time, it would do this with even a moderate pull in gear on the highway.

I went ahead and had Barney (Barn Man) replace the head gaskets earlier this summer. I've autocrossed the car several times since, and have had no repeat performances. I should be able to make a track day in October to completely confirm that this issue is resolved.

 

[stuntman]

If the hissing cap is on the milk jug, and if you didn't have this problem until you replaced the header tank (milk jug) then it could be a bad tank.

0.02