NB cooling issues

[Jeo]

OK, medium-long story to get to this point but here we go.

Mild off track excursion last December saw the front of my NB8B into a ditch and fun new hole on the bottom tank of the radiator. I kept in it for probably another 30 seconds or so at full noise after the crash as I hadn't noticed the extent of the damage, but also assumed most things were ok as the factory temp gauge didn't move from normal. Was obviously the end of that day's racing and grabbed a tow truck home. I had another event the following day so was suitably excited when I managed to track down a lightly used aftermarket OEM style NB radiator around 8pm that evening, and after some light percussive persuasion to get the mounting bracket back to where it was supposed to be, I got the 'new' radiator in around 11pm. Filled it with fresh new coolant, all appeared well, then noticed a leak coming from the crankcase directly below the water pump. There endeth the evening. No racing the next day.

Realised I had no spare time between then and christmas so asked local club sponsor ACDC to replace the water pump for me, and while they were in there throw a timing belt kit at it too. No dramas there, I picked the car up about a week later and had a two or three uneventful ~50km round trips to work and back over the next few weeks.

Jump to mid-January and on the drove home from work on a particularly hot day, I noticed it smelled rather hot when I got home. Gauge read normal but was definitely hot when I opened the bonnet, and at that point noticed that neither radiator fans were running. Assumed no fans was the cause of my overheating, so cue this thread where I went troubleshooting over the next few weeks. Fans themselves worked with direct 12v power source, relays worked, wires between fan/relay/sensor all had continuity, so all signs pointed to a broken coolant temperature sensor. Bought and fitted a new coolant temp sensor last week, went for a drive, got up to temperate and the passenger side fan came on. Never saw both fans working but also that was further than we got with the old sensor so I assumed that meant we were good.

Went for a drive yesterday while it was 34 ambient, got home and both fans are running but also coolant is boiling. The drive wasn't particularly vigorous, so even with the high ambient I wouldn't have expected it to boil. Conveniently I had ordered an aftermarket multi-gauge thing a couple of weeks ago which included coolant and oil temp sensors and readouts. That also arrived last week so installed the aftermarket coolant temp sensor probe in the top radiator hose.

Went for another drive just now and took the IR thermometer as well for science. No idea how well calibrated either the IR gun or the aftermarket sensor are but figured they'd be relatively consistent individually, even if the raw numbers off the two readouts wasn't the same. Ambient was again low/mid 30's.

I was picturing this as a table of course but that's somewhat hard to do on a forum so you'll just have to cope with my 5 star formatting here.

Factory gauge - Aftermarket gauge - IR gun at top radiator hose - IR gun at bottom radiator hose - Fan status - Overflow status

From cold start we had...
L - 35 - 33 - 33 - Both off - Nothing
1/2 - 50 - 45 - 34 - Both off - Nothing
1/2 - 65 - 60 - 35 - Both off - Nothing
1/2 - 80 - 73 - 70 - Both off - Nothing
3/5 - 95 - 85 - 75 - Passenger on, drivers off - Nothing
3/4 - 97 - 90 - 80 - Passenger on, drivers off - Nothing
3/4 - 101 - 90 - 82 - Passenger on, drivers off - Nothing
H - 105 - 90 - 80 - Both on - Overflowing
7/8 - 106 - 93 - 82 - Passenger on, drivers off - Overflowing


Fans now work which never did before, so that's an improvement. Factory gauge never got too hot before either so I'm assuming that was also part of the faulty old sensor. Factory coolant temp sensor is of course at the back of the head while the aftermarket one is in the top radiator hose, so I would expect them to see slightly different temperatures. Both radiator hoses get hot so the thermostat must be working(?). Doesn't appear to be any oil in the coolant or coolant in the oil.

And now to the point of the thread - suggestions as to where to go from here? I'm tempted to just go whole new radiator as I can't vouch for the quality/integrity of the one currently fitted, but also an upgrade probably isn't a terrible idea anyway as the car see's motorsport use about once a month. Local guy I got it from said it was fitted to a road car of his for a few months before upgrading when he threw a turbo on as well. I trust him, but also don't have a whole lot of other ideas at this point.

[greenMachine]

Head gasket?

Seems like the thermostat is working, that is one variable out of the way. Oil and coolant in their proper places is good, but no guarantee of anything.

I'd run the car up to get the thermostat open, with the radiator cap off. When it opens, have a look at the coolant, it should be flowing strongly across the top of the core from the radiator inlet. If not, the water pump may be/is cactus - obviously, shut it down and start planning for a replacement. There is a possibility that the radiator has blocked cores (have a look before you top it up), or the fins are in bad shape, or clogged with insects, any of which will reduce its effectiveness, but your readings seem to suggest there is at least some cooling happening.

If it is flowing, plug in one of those big funnels, top up the coolant, and rev the engine. You may get a bit of air from the recent top-up, but if you are getting regular/continuous bubbles and they don't look like going away, it is time for a compression test to confirm the blown HG.

I have one of those funnels if you want to borrow it, and I happen to have a spare (used) SE radiator if you want to drop that in and see if it makes a difference.

[RS2000]

Those temps don't seem high. You might need a new radiator cap.
Boiling point is raised by 2 deg for every 10kpa of pressure. Don't know what pressure your cap is, but if 110kpa (as in my NB8A), then boiling point would be 100 + (2x11) = 122 deg

[greenMachine]

New radiator cap never a bad idea, especially here.

However, the car should not boil idling in the drive, with both fans working even if the cap is a dud, even if it is a hot day. That is, if both fans are pulling, not pushing, although stationary with the bonnet up that should not make a great difference.

[RS2000]

Jeo had been for a drive on a 34 deg day. If the aftermarket gauge is correct, the coolant reached 105 deg. which is at the upper level of 'normal' running temp. It's getting up, but not quite overheating.
Either the temp is much higher than 105, or the radiator cap is cactus.

[Jeo]

Cap on the new radiator claims 110kpa, same as the one on the old/broken radiator. I still had the broken radiator sitting beside the garage waiting for a tip run so have grabbed the cap from it and will see if that makes a difference. Unfortunately(?) it's not forecast to hit 30 again in Canberra for at least another week so I'm hoping we can still replicate and ideally solve the problem before the next motorkhana on the 5th of March.

Will check that coolant is flowing across the top of the radiator as greenMachine suggested and probably do a compression test for good measure as well. I'll report back at that stage but may well take you up on the offer of another radiator to test as well depending how we go between now and then.

Thanks all!

[Jeo]

More troubleshooting!

Thermostat opened at 81c according to the aftermarket gauge, fan turned on at 96. Coolant is definitely flowing across the top of the radiator with revs, couldn't really tell at idle. Then did a compression test after that and all appeared well - with each cylinder reading ~1150~1200kPa.

I wasn't keen on trying to get it to boil again at 10pm, so will let it run for a good while at lunch tomorrow and see if the new/old radiator cap has solved it. From there I guess it's PM to GM time for borrowing of a radiator and then give the whole thing a good flushing.

[greenMachine]

That's pretty good news - the water pump is working, the thermostat seems OK, cylinder pressures even (I've heard of KPIs but I wouldn't know a kPa if it bit me on the bum ).

It does seem to be pointing at the radiator, perhaps the cap. One mod I have to make in my car is to install the radiator pressure sensor that is currently sitting on the shelf, that will highlight a leak somewhere (but where?), including the cap.

If the HG is good, there are no puddles under the car, water is flowing in the radiator, it looks like radiator cap or the radiator fins are mutilated/clogged, blocking the air flowing through it and taking the coolant heat with it - you should be able to see that.

It also could be that the radiator tubes inside have been blocked with crud (maybe the PO used water, not coolant?) and a back-flush might help. Drain and save the coolant, remove a hose and stick the garden hose down there so that the hose water is travelling in the opposite way the water pump is pumping coolant. Preferably remove the hose on the pump inlet side, and see what comes out of the radiator. It seems that water is flowing inside the radiator, but this should answer the question 'is enough flowing'...

If none of that shows up a problem, send me a message ...

[The American]

My 2c guess is dead radiator cap.

[Jeo]

And putting the old cap on the new radiator seems to have fixed it! Haven't road tested it yet so no test under load, but fans are now able to control the temperature when stationary so that's definitely an improvement. Decided to grab a bottle of coolant flush and run that through it for good measure. Cue what seemed like 17 flushes of the system yesterday while gardening in the front yard, followed by new coolant to finish with and I think/hope we're now right.

Started by letting it run in the driveway until the fan turned on to see if the new/old cap had improved anything. Success right off the bat - fan came on at a reported 96°, temperature came down to 89, fan went off, slowly climbed back up until we hit 96 again at which point fan came on, and again turned off once we dropped back to 89. Couldn't do that before so I'm taking that as a win. Shut the car off to let it cool down a little before dropping the coolant. Now I'm pretty sure I went a little overboard here, but also, whatever, was mostly gardening near the car while waiting for things to either cool down or heat up anyway... but the rest of the day went something like this.

Dropped existing coolant
Refilled with tap water
Ran until warm
Waited until coolish
Emptied radiator
Refilled with tap water and coolant flush cleaning stuff
Ran for 45 minutes as described on the bottle (fans still doing their thing!)
Waited until coolish
Emptied water and cleaner (holy crap did a lot of brown come out with this)
{Refilled with tap water
Ran until warm
Waited until coolish
Emptied radiator}
Repeated bracketed steps like 5? times until the water came out clear. The cleaner stuff just kept coming
Garden hose up the 'from' heater hose to run the system backwards
[Filled with demineralised water
Ran until hot and fans cycled a few times
Waited until coolish
Emptied radiator]
Repeated square brackets again
Filled with coolant
Ran until hot and fans cycled a few times


Pretty sure this was overkill given how the first cycle went, but also with the amount of crap that came out with the flush cleaner solution I'm not upset that I went that far. Surprised at how many goes it took to get aid solution out of the system, but also wasn't a bad day by any stretch of the imagination. I can see how that would have been really crap if that was the thing you were doing for the day, but going back and forth to either turn the car off or on wasn't a huge inconvenience wasn't a huge impact on my productivity of gardening. Most of the time was spent waiting for the car to either warm up or cool down.

Also I feel my "security BMW" parking int he MX5 was a nice touch.

PXL_20230225_000006754.jpg

Thanks all!

[greenMachine]

Good result!

[Lokiel]

:
Also I feel my "security BMW" parking int he MX5 was a nice touch.

:

- people going past probably thought "Typical BMW driver, think they can park wherever they please and have blocked that poor guy from getting out!"

PS: With those rims on a white car, I think red side stripes would look great:

[KevGoat]

Ha! Lokiel, as soon as I saw Jeo's image of his car with the red wheels, I thought of Billybunter's SE! When he painted it white and added the red stripe with the red wheels, it was the best combination he achieved. Still like that look.

On the cooling problems... I can't count the times I've had overheating issues caused by either a faulty radiator cap or a faulty thermostat, especially on my old V8's!

Sent from my SM-G781B using Tapatalk

[Jeo]

I had the same stripe on my old NA and had considered it many times, but keep coming back to thinking it might be a bit too much. Dunno, does cycle in my head many times though.

[Lokiel]

... it'll keep running through your head until

... you give in!