Still overheating

[RacerX-21]

Wow.... Great news!! That sounds like a do-able setup for 300rwhp! How much did that cost?

3.2 head gasket on a 3.0? I heard the 3.2 head gasket is a better make than the 3.0 but do you need to do any modification to the gasket or head to make it fit?

Thnx.

Henry.

Gary rebuilt the entire engine with bearings, valve job, new rings, etc. So I don’t know a price for just doing heads and cams. When and if you do call around for quotes, I suggest that you ask if they can degree the cams. Unfortunately the NSX engine requires some additional tooling and some learning, but if the shop is already familiar with doing cams it will be an easy adjustment, and they could definitely benefit from repeat NSX business. Very few shops Degree NSX cams because of the difficulty of (4) cams and V-tec… it’s also the (4) cams that make the NSX one of the best candidates for HP gains. You will be surprised how many of the well-known NSX tuners/builders don’t do this yet.

Maybe I can do a demo this year at EXPO.

Regarding the 3.2 head gasket on a 3.0 engine; nothing is needed, it’s a direct swap that will bring your compression from 10.2:1 to 10.7:1 or from ~200 PSI to ~230 PSI. This change is best done with a programmable ECU, but I can say that I have run this configuration with a stock ECU without getting codes… unfortunately I have no Idea what the OE ECU is doing internally to react to any changes, or what changes have been made with the OE unit over different years. In other words I can’t guarantee you wont get codes.

Speaking of codes- This engine with different fuel pressure, headers, exhaust, No variable runners, revised crank case evacuation, more compression, and cams was test driven a minimum of three times before I replaced the ECU, and it never through a code. In fact I disconnected an O2 sensor to clean up it’s wire routing and it reacted with a code, after I reconnected the sensor the ECU corrected it self after short amount of driving.

Honda made the gasket type change to address a head gasket failure issue, but frankly I have seen just as many failures with the new design, the crux of the problem is related to only having 4 head bolts per cylinder vs. 5 or 6. This is likely why we never saw a factory boosted NSX.

Oh- the gasket change gives more power throughout the power band, but is definitely noticeable at lower RPM.

 

[RP-Motorsports]

I am curious, could you "up" the torque say with ARP bolts? or will any additional foot-lbs just cause a warp?

 

[RacerX-21]

I am curious, could you "up" the torque say with ARP bolts? or will any additional foot-lbs just cause a warp?

Yes, defiantly- all my engines get ARP studs. But this only helps. Many other brand engines have the same problems with only 4 bolts per cyl. Some engines have kits to add more bolts, some are iron blocks so larger studs can be used, and many engines just come with 5 or 6 bolts per… the NSX engine doesn’t benefit from any of these choices. On hi hp boosted engines I O-ring the blocks now, and T-sleeve to eliminate the flex that adds to the problem. I still think heat is a large contributor so hopefully with the new R&D I can keep everything together. I can tell you shortly if things work out- It should be on the track in the next month with 700 RWHP, I’m trying to make the February VIR Honda Challenge event. It’s about time we show the other Honda cars that the NSX is #1.

 

[mystican]

Any updates...I thought this was a good thread.

 

[L8apex32]

Sorry I'm late to the discussion. My car is stock, but 141k miles and MANY track events. It never used to overheat. Then, we moved to Colorado and at high altitude, it did overheat at the track. Never on the street.

Now back in Kansas at normal altitude, I hoped it wouldn't overheat. At a track weekend this spring, with 70 degree days, it slowly overheats at the track. (Yes, I drive hard and always shift at 8k.)

I am in the middle of a radiator, overflow tank and coolant hose replacement project right now. My radiator has so many dents and nicks in it, it probably isn't flowing much air anyway.

My coolant tank has a small hole in it. By small, I mean so small that it only leaks coolant when coming in after a session and if I let it sit with the engine off. Somehow, the coolant can then start to spray out. I never had to add water, though, as it only mists. The tank is getting replaced now.

Here are my thoughts. These are forward looking statements, but I can report how some of them turn out in the near future.
1) Your radiator needs to not be all banged up or it won't flow. No flow, no cooling. I expect the replacement (stock) radiator to be better.
2) The cooling system works on pressure. If you have a leak anywhere, then you likely don't have enough pressure to cool adequately. My tank has the pin-hole leak, I likely was below optimal pressure.
3) Then engine isn't putting out much more power at 8k than at 7k. Changing shift points is not decreasing work much at all. To check, what HP is produced near 8k and what at 7k? I guess there isn't much increase. That means it isn't "cooler" to shift at 7k. My theory is that the water pump may be cavitating. We may need to analyze water pressure and flow at varying RPMs.
4) You may not want to go to a higher pressure cap unless you have increased the pressure holding ability of the entire cooling system. Can all the hoses handle more pressure? I don't know. Also, if your cooling system isn't boiling over, a higher pressure cap won't change anything. I haven't read where anyone boiled over (but that doesn't mean it doesn't occur.)
5) I did not check my system for air. (Sounds stupid to say now, but if you saw my radiator and knew your tank had a hole and had hoses that are 15 years old...) If I continue to overheat with the replaced tank and radiator, I'll definitely check for air bubbles carefully before considering anything else in the future.



Q: In a healthy cooling system, if we continue to add heat, behond its maximum heat dissipation, which measure should "red line" first - coolant temperature or coolant pressure? What would it take for the coolant pressure to exceed the cap's limit?

Q: The fan and shroud came off pretty easily. Has anyone tried removing it at the track for extra cooling at high speed? I am sure we need the fan for street use.

Hope any of this helps.

 

[Hrant]

This topic is very dear to my heart and wallet :biggrin:

I think we may have finally found a solution to the overheating at the track, though not a fix - but then again, the test was done with ambient temp of low to mid 80F. Hoses were changed last Fall and the mix was 50/50 with water wetter. I do have an Accusump so the extra 3qts should help too.

As always, some of the ideas were already floated by Wolfgang, Rob, Tom, Chris and others but perhaps not tested independently. At the track last Friday, we had the luxury of having Shad with us. As I came to the paddock complaining that my temp gauge was just 1.5 bars below the red, he suggested taking the fan and its shroud off entirely to let more air flow through the back of the radiator; as well as he removed the rubber seal under the rear edge of the hood. Next session, the water temp gauge was dead at center or a bit below it just like street driving

Well, the third session of the day, I was duking it out with one of our local fast NSXers and the gauge went about one bar above the middle, still pretty good as I thought we may have solved the problem. The final session of the day, I had a passenger with me and duking it with a Porsche 911/Carrera behind me. The temp went up to about 2 bars below red after several laps.

Since my 98-T is not dedicated track car only, this leads me to believe that:

(1) air flow out of the radiator and a vented hood will help. Cutting the shroud around the fan is a compromise given that it is also street driven - actually Kip's car has the shroud cutout and it looks very clean and almost OEM look.

(2) increase mix to 25/75 with water wetter. I have a call to Red Line to see if adding two bottles of water wetter helps

(3) I agree with Chris that once it overheats, it is more difficult to keep it from overheating again. A continuously switched fan with the shroud cut out might do the trick at the track - not sure if this will be enough for higher HP cars. My 98-T is 3.2L NA - at least for now:wink:

(4) Oil cooler will be a definite plus especially on hot days. Can't comment on using 20/50 viscosity oil that entails more engineering than I can explain, but I hear Motul is touting their engine oil as capable of withstanding more heat. Don't know if they are willing to put in writing with some independent lab tests.

(5) Bigger radiator should help theoretically, though Rob noted it hasn't helped much. Shad has designed a new radiator and I will let him post the data download. Plugged radiators will cause overheating though my overheating occurred after we cleaned the old radiator and spent 2+hours trying to straighten the fins ..... yes I had free time while watching Shad work ...... we still have the option of changing the radiator with an almost new OEM one. One step at a time.

As always, we keep experimenting, but it is obvious this happens to some cars but not all - unless of course the others aren't pushing as hard :tongue:

YMMV.

Edit: We have a few more track events, one in Reno-Fernley so the impact of the altitude would be interesting to see, then in July at Infineon Raceway.

 

[Ponyboy]

...unless of course the others aren't pushing as hard.

This is my thought. And in the case of the stock NSX brakes as well.

 

[OLDMNSX]

There is also a product made by the people who make ROYAL PURPLE synthetic oil products that is an additive to the coolant. I don't remember the name, but they say that it will lower the temp. Has anyone tried this product?

Brad

 

[L8apex32]

Got a quick "where does it go" question.

I'm in the middle of the coolant hose replacement project. I just completed replacing the two hoses going to the oil cooler.

Part way through replacing these two hoses, I noticed that the oil pressure switch wire, that runs from the oil cooler to the wiring harness, is missing a connection point.

Specifically, the wire has an attached eye that connects to the oil cooler itself. I removed the 10mm bolt when taking the oil cooler off and I have not put it back in. Check.

About 6 or 8" down the wire from this eye is another eye. I do not remember taking this eye's bolt out. Where does this eye connect?

I spent at least an hour looking for where this eye can connect. There are no open threaded holes for another bolt and I don't have any extra bolts. The ones I have I know where they go.

Here are the possibilities I have come up with:

1) The eye connects to a drive shaft shield bolt (one of 3 I removed). However, I don't think that the wire is long enough to reach up that high. This is doubtful.

2) They eye connects to a 10mm bolt that holds a black housing against the engine. The cable reaches there - but this 10mm bolt has never been removed. I put a socket on it and it was tight - like some gasket material or something is glueing it in place. I am sure I didn't remove this bolt and put it back so this can't be it.

3) The eye was never connected. It does not look "really" clean like a bolt has been keeping the eye clean. It looks dirtier like it might have been just hanging there.

4) ???

Anyone know where this thing connects???
Thanks.

 

[Hrant]

OK, a shot in the dark after focusing on Tom's posts again ........... those having overheating with OEM radiators at the track, how many of you are still using the OEM milk bottle overflow coolant tank?

Those not overheating at the track - and focusing on advanced drivers with OEM radiators, what type of overflow tank do you have? At what level do you fill the tank.

Red Line noted that unless the coolant capacity is 18 qts, adding a second bottle will not help. " The best heat transfer and the greatest temperature reduction would be with just water and WaterWetter. In a vehicle with air conditioning we recommend 15% antifreeze to prevent icing of the heater core." And they were surprised that by removing the fan and the shroud around it it help cool the temps.

YMMV.

 

[Gerard van Santen]

I don't know if the overheating of Wolfgang's NSX is solved, but I have overheating problems since a few months myself. The overheating occurs only with track use (after a few laps hard driving). And long standing still while the engine is idling.

Yesterday I did a coolant leak test and set the whole system under pressure with 15 lbl. At first the pressure didn't show a noticeble drop. So I thought there was no leak in the system. After two hours I saw the pressure was dropped to 14 lbs. Not much but enough to be worried about. After a thorough search I saw that the upper white bleeding plug of the radiator was leaking small drops of cooling water on the fan housing. Not much but enough to get some pressure out of the system. I took the plug out (after depressuring the sytem) and saw that the rubber ring (costs $ 0,01) was the cause of the leak.

Later this evening I will test the system with the new rubber ring. I hope this will solve my problem. The lesson for me is that you have to give the coolant leak tester enough time to find the smallest leak possible.

 

[RacerX-21]

Hi guys,

Problem solved 100%.

Two important details and a few contributing details that cant hurt:
1) After exhausting everything related to blown head gaskets, I went back to retrace some of my assumptions. After talking with the ARP Engineers I found that the torque specified by the sellers of the NSX head studs was below the minimum needed to stretch that spec of stud. They added that when ARP actually specifies the torque, they usually give you the minimum torque to get the proper stretch and added that normally the bolt or studs yield is well beyond that number. We assume that the person who originally came up with the NSX torque number didn’t really know much about the engineering of studs, and may have assumed that the torque needed to be similar to the original bolt. I was also told that the turbo diesel HP builders regularly torque higher then the specified number from ARP, and that is the only way they can hold the pressures. I spoke with another builder who was one step ahead of me and was already torquing NSX head studs at a higher number. Armed with this info I tested how much the block would take using a trashed block I still had out back.
Anyway Get ARP Studs, Clean the threads in your block, use Molly lube on the threads and the washer faces, torque your studs in at least three increments before you reach the max, work from inside out with the bolt pattern, and end up with 85 foot pounds of torque. Other details that can help but I credit the torque- If you get the block or heads machined for flatness, get as smooth a surface as your shop can do. Spray your sheet metal gaskets with Hylamar according to the directions. This will coat the surface for better water sealing.

2) Make a clear-ducted path going to the radiator and out the hood. Airflow is key, the biggest, most expensive radiator cant cool your engine if the air flow is limited. Other details that can help but I credit the air flow- Bleed your cooling system properly. Make sure you have no leaks. Raise your coolant tank as high as possible. Fill it and let it overflow to find it’s natural max. Build a secondary overflow tank that will catch and recover the expansion of fluid… like every other car on the market.


Result- I have a car that I have beat the crap out of since the beginning of the year. Depending on the event and pulley I run, it makes between 500 and 600 at the wheels (700 crank). It just came in 2nd place in the Grassroots Ultimate Track Car Challenge with no heating problems with 97-degree air temps. And most impressive- Last weekend I tested with its new owner at Road Atlanta running 1:34 with two people. The ambient paddock temps were 112 degrees and the track temp was a record breaking 146 degrees.

ARP studs torqued to 85.
Venting for more airflow.

 

[Gerard van Santen]

My problem is also solved. I think:wink:

I did a coolant leak test after I put in a new rubber sealing (ring) on the white radiator bleeder plug. No pressure drop this time. After that I took my NSX for a ride to warm up the engine. Then I stopped the car for about 1/2 hour with the engine running. The fan ran and no overheating :biggrin:

I haven't pushed the car as much I would do on track but the problem looks like to be solved. Fingers crossed.

 

[goldNSX]

My problem is also solved. I think:wink:

I did a coolant leak test after I put in a new rubber sealing (ring) on the white radiator bleeder plug.

There must have been water marks around the white bleeder, no?

 

[Gerard van Santen]

There must have been water marks around the white bleeder, no?

No water marks after replacing the rubber sealing. I had set the cooling system under pressure for more than an hour with no leaking whatsoever.

Perhaps the rubber sealing ring may be the source of overheating for more NSX's than only mine. If I had set the system under pressure for a to short of time, I had never noticed the leaking bleeder.

 

[goldNSX]

No water marks after replacing the rubber sealing. I had set the cooling system under pressure for more than an hour with no leaking whatsoever.

Perhaps the rubber sealing ring may be the source of overheating for more NSX's than only mine. If I had set the system under pressure for a to short of time, I had never noticed the leaking bleeder.

Thanks very much for the input. That's a very good input to keep an eye why my cooling system has bubbles after very hard driving. I had no overheating just the annoying bubbles. I'll keep track of the events after my coolant hose replacement procedure which will include every gasket in the cooling system.

 

[Hrant]

The rubber seal is an interesting observation. But how are you measuring/reading overheating? The OEM gauge or an aftermarket gauge?

After adding an oil cooler in the passenger side vent, and using a new overflow bottle, still about 50/50 mix with water wetter my water temp peaked at 220F (using digital SPA gauge) in 95+/- ambient temp with a 250 lbs passenger and I was pushing the car hard using all that the CTSC can give.

The OEM water temp gauge was disconnected. But I would guess from previous track events, and at the time without a CTSC and with more water mix, my OEM gauge would have pegged at red.

Edit: I must also note that regular and moderately tame street driving with same ambient temp, the water reads 190-195F with the AC on ....... so the track is adding no more than 30F. This tells me perhaps the OEM temp gauge is partly the issue.

 

[Elite]

Hi guys,

Problem solved 100%.


Anyway Get ARP Studs, Clean the threads in your block, use Molly lube on the threads and the washer faces, torque your studs in at least three increments before you reach the max, work from inside out with the bolt pattern, and end up with 85 foot pounds of torque. Other details that can help but I credit the torque- If you get the block or heads machined for flatness, get as smooth a surface as your shop can do. Spray your sheet metal gaskets with Hylamar according to the directions. This will coat the surface for better water sealing.

\
ARP studs torqued to 85.
Venting for more airflow.

Rob,

Do you have a camera in my shop?:tongue: That is the absolute exact procedure I do on all my motors. Didn't miss a detail, nice write up.

 

[Gerard van Santen]

The rubber seal is an interesting observation. But how are you measuring/reading overheating? The OEM gauge or an aftermarket gauge?

My reading was from an OEM gauge. After a view laps hard driving and after that going into the pits, the neadle went straight into the red. This also happened standing still for more than a couple of minutes with the engine running. I also noticed steam coming out of the overflow cap of the coolant tank. After cooling down I checked the water level in the coolant tank where I found the water level was down one or two inches.

I have to find out if the bleeder was the source of the overheating with some track time now. The bleeder was leaking so little that I didn't see it earlier. But the small leak was probably enough to loose enough pressure and water to interfere in a proper cooling of my engine.

 

[Gerard van Santen]

I thought my problem was solved. But after a very quick warm up, my engine overheated again. I checked the rubber radiator hose and it was cold. So I assumed something was wrong with my thermostat and that my thermostat didn't open above 78 degrees Celsius. Resulting in a 'heat soak'.

So I pulled out the thermostat to check if it would open in hot water. I didn't have a thermometer for use with this high temperatures, so I let the water boil in a pan with the thermostat in it. It opened at boiling temperature. That must be 100 degrees Celsius. It should open at 78 C.

I cut the thermostat to leave the the outer ring with the rubber seal and put it back in the engine. The cooling circuit will be completely open from now on. I know I the engine will take more time to warm up, but I hope the overheating will be over now