BMW E46 Fanatics Forum banner

1 - 20 of 34 Posts

·
Registered
Joined
·
9 Posts
Discussion Starter #1
Hi Guys,

Nick here from the BuildaBeemer channel on YouTube.

Here is our track car:
915651


My father and I have transformed my ex-daily driven 2000 BMW 330i (M) into a dedicated Track Car in which we compete in the local South African BMW Car Club Racing Series. We've just completed the 2020 season, and while it has been an incredible experience, we have been unable to resolve an overheating issue that has plagued us since the beginning of the year :cry:

I'm reaching out to see if anyone can spare some suggestions to try and resolve our overheating. Perhaps some of the SPECE46 racing guys have some thoughts.

The first thing to understand is the context in which the overheating occurs. This is a dedicated Track Car, operating in South Africa where the ambient temperature is often 30 - 40 degrees Celcius and the track temperature far higher than that. The car does not sit in traffic, or idle for any significant period of time. The car lives its life at full load for a 15-25 minute session on the track.

The motor and cooling system is able to maintain our target coolant temperature of 75 degrees Celcius (I have programmed this in the MS43 maps) very easily at idle and while stationary.

It is once the track session starts, and after a lap or two that the temperatures rise well over 100 degrees Celcius and reach 120 degrees Celcius if we sustain the vehicle pace on track. This is obviously far from desirable.

What we have done so far:

Motor: BMW M54B30

Note: there are no leaks whatsoever. We do not need to top-up coolant level at all between track sessions. Also, our oil temperature is completely manageable. It hovers around 90 degrees Celcius without an oil cooler.

1. Tried removing the thermostat from the housing and blocking off the port to the head (Welsch Plug) to ensure flow without the thermostat - no improvement (See video)
2. Installed brand new OEM thermostat
3. Installed brand new OEM Water Pump (Composite Impeller)
4. Installed new Expansion Tank and water level sensor
5.Heater core deleted
6. Waterpipe going from head to old heater core is plugged at the head
7. Waterpipe coming from the old heater core to the expansion tank is plugged at the expansion tank
8. Removed oil-cooler from in-front of the radiator - no improvement
9. Replaced front timing cover after identifying cavitation damage to the water pump housing / Front timing case cover (See video)
10. Rebuilt the motor! (See video)
11. Replaced radiator after identifying that we may have had a variant for an automatic (see video)
12. Bleed the coolant following Mango's guide, although we don't have a heater core
13. We run a recommended concentration of Redline WaterWetter with distilled water to improve the heat transfer of the fluid
14. Built an aluminum plate thing that "boxes-in" the radiator and forces air to flow through the radiator and not around it. This interfaces between the bumper and the radiator.
15. Replaced the coolant temperature sensor in the head
16. Replaced the radiator outlet temperature sensor
17. Replaced the Water Pump again
18. Removed the headlights to provide a little more airflow to the engine compartment and allow for better heat dissipation


I know most are going to suggest an air-pocket as a result of improper bleeding, but we've replaced the coolant probably 10 times over as a result of a replacement unit being installed, where the slow fill process waiting for a constant stream coming out of the bleeder screw was followed.

I look forward to the discussion and any suggestions! :LOL:

Thanks,
Nick
 

·
Registered
Joined
·
3,550 Posts
With thermostat valve removed from the housing and plugged up the head with Welsch plug, this is the same as thermostat at wide open. If this configuration still results overheated then there’s no point to test different thermostats, as this configuration already gave you max flow through the radiator.

So let’s look at the pump and any restrictions. Let say the gauge displays 100C, which is the same at the radiator top hose, what is the temperature at the radiator lower hose? If it’s lukewarm then it means not much flow through radiator for a few reasons.
 

·
Registered
Joined
·
3,550 Posts
What tank cap pressure rating? Lower than stock 2 bar can cause overheated due to lower heat transfer rate coolant.
 

·
Registered
Joined
·
1,159 Posts
I race on the 24hr of lemons circuit in the states, so I get your concerns on heat. here are a couple of ideas. I would do a range of things.

I don't know how much water wetter increases the boiling temp of water, but I would make sure you have the highest boiling point you can within race restrictions. Coolant stops absorbing heat when it boils. We run distilled water on the track

100 degrees Celsius is okay, 120 is probably not, but if you could drop it 10-15 degrees I think that would be pretty good.

By blocking off the heater core, you are blocking off another opportunity to add cooling capacity. Could you run the pipes meant for the heater core to a secondary (smaller) radiator mounted somewhere appropriate ?

I would assume you have adequate air flow over the radiator. What about an electric puller fan mounted on the engine side of the radiator?

What about getting the largest radiator that will fit? I think I would take some measurements and then research radiators from M3, M5, X5 both i6, V8, and diesel variants and see if you can stuff a bigger radiator in there. Maybe an Aluminum racing radiator..

Here is a post about using an OEM radiator from an M roadster (Z3). According to link it has a thicker core than the M3.
 

·
Registered
Joined
·
3,550 Posts
The hotter the coolant going to the radiator, the higher the heat transfer rate the radiator can dump out. So to have hotter coolant in the in liquid form at over 100C, it has to be pressurized, higher is better subject to material ruptured.
 

·
Registered
Joined
·
102 Posts
Sounds like you haven't replaced the fan yet. I would start there. When I first got my car it used to slowly overheat when driving but be fine when idling. Turned out the fan needed to be replaced.
 

·
Registered
E46 '04 330Ci convertible 750K km
Joined
·
1,142 Posts
The hotter the coolant going to the radiator, the higher the heat transfer rate the radiator can dump out. So to have hotter coolant in the in liquid form at over 100C, it has to be pressurized, higher is better subject to material ruptured.
About the hear transfer rate - it increases not with increasing of coolant temperature, but with increasing of temp difference between coolant and radiator. You want dump more heat, you increase capacity of radiator, or amount of air to blow trough the radiator, to increase temperature difference between coolant and ambient air. You do not increase engine temperature in order to increase heat exchange rate, you decrease temperature of radiator to increase temp difference (and therefore heat exchange rate, if you really wanna talk about such).

Even at racing engine temp should not go that far. OP says 120 C - it is truly critical level, and that is where 2 bar cap suppose to drop pressure. If we assume they use 1.4 Bar , it would do its work at 110C, although I doubt they use anything but standard 2Bar one. So cap threshold cannot be cause of overheating, as it is just a valve preventing catastrophic consequences of risen coolant temperature in case cooling system fails.
 

·
Registered
Joined
·
2,192 Posts
Yeah, hotter water = greater temperature differential = greater rate of heat transfer.

It's why you get to 120- and then stabilise.

I agree with Eff- you need to figure out if the radiator is heat- soaked. Frustratingly,
there's no way to get a real temperature out of the lower hose temperature sender,
because it can tell you if you ARE under- radiatored. An aftermarket sensor in the hole
works, but then you lose the fan controller function.

You've done most of the right things- what brand radiator did you use when you replaced it?

Three things generically cause the problem you're having;
a.limited radiator capacity
2.limited water pump capacity
iii.improper engine heat output.

a. includes all sorts of things like dirt in the fins, a path for air to bypass
the radiator, junk in the radiator itself, or the damned thing is too small.
It's indicated by the lower hose temperature rising unusually.
In 40c ambient, you should see a lower hose temperature that
stabilizes at least 20c below the thermostat temperature. Or
add radiator to get there.
You've done all the right things there, generally, but:
DO put the headlights back in- you want to get as much air pressure on the
front of the radiator as you can, and as little behind,
and taking them out MAY actually reduce
flow through the radiator. Likewise, the undertray and other
parts of the associated ducting need to be there to move the air from
the lower part of the front bumper through the radiator. Most of the radiator's air
comes from the bumper, as you've already figured out.

2. is often a water pump problem, and it can be caused by
pump cavitating, not working as efficiently as it should,
junk in the cooling system ('Where'd I leave that blue towel?')
or hoses that are somehow restrictive. I agree- for a track car,
a 2 bar cap is a good idea as it can reduce cavitation and
hot- spot boiling. Harder to diagnose, but if the radiator is working
well, this is the next thing to go after.
Maybe try an underdrive pulley to slow it down?

iii. is probably not your problem, but if the car's running even a little
lean, it can overwhelm the cooling system. Likewise, slightly early timing
can do the same thing. It's worth confirming the real- time AFR and timing
numbers on the track, just to rule this out.

Pragmatically, in your shoes, I'd strap a sender on the lower hose,
and add radiator capacity when it turns out the radiator's overwhelmed.

I totally agree about adding the heater core back in- the engine's designed
for that flow, so why not let it have it?

not much help,

t

edit-
About the hear transfer rate - it increases not with increasing of coolant temperature, but with increasing of temp difference between coolant and radiator.
Actually, it increases either way. But because aluminum and water are pretty good
thermal conductors, the water coming out of the head is within something like 10c of the
head, by the casting design.
Air, on the other hand, is a crap conductor with very few molecules per cubic
meter, so there's going to be a significant temperature differential between the
radiator and the air.

The ONLY problem with running a higher coolant temperature is that the vapor
pressure of water makes it more and more likely that localized heating
(occasional detonation, maybe) or extreme agitation will cause the liquid
in the coolant to try to become gas- and that's almost always catastrophic to the head.
So we run cooler outlet temps to provide an insurance policy.
Ideally, we'd run something well over the boiling point of water- less loss as heat, more
powah out. But the current production metallurgy doesn't like that so much...

t
 
  • Like
Reactions: Zound

·
Registered
Joined
·
3,550 Posts
About the hear transfer rate - it increases not with increasing of coolant temperature, but with increasing of temp difference between coolant and radiator. You want dump more heat, you increase capacity of radiator, or amount of air to blow trough the radiator, to increase temperature difference between coolant and ambient air. You do not increase engine temperature in order to increase heat exchange rate, you decrease temperature of radiator to increase temp difference (and therefore heat exchange rate, if you really wanna talk about such).

Even at racing engine temp should not go that far. OP says 120 C - it is truly critical level, and that is where 2 bar cap suppose to drop pressure. If we assume they use 1.4 Bar , it would do its work at 110C, although I doubt they use anything but standard 2Bar one. So cap threshold cannot be cause of overheating, as it is just a valve preventing catastrophic consequences of risen coolant temperature in case cooling system fails.
I will address each BOLDED sentence below.
1. To increase heat transfer we need to increase the temp difference between coolant and radiator as you said, but then you disagree why I wanted to have higher coolant temp.

2. Again, the objective is to have higher coolant temp (I didn't say we want to increase the engine temp).

3. For sure a 2bar cap will give higher coolant temp compared to a 1.4bar cap, give the engine temp is 120C. Water boils at lower pressure at x temp and same water will boil at higher pressure at y temp, where Y > x.
 

·
Registered
E46 '04 330Ci convertible 750K km
Joined
·
1,142 Posts
I will address each BOLDED sentence below.
1. To increase heat transfer we need to increase the temp difference between coolant and radiator as you said, but then you disagree why I wanted to have higher coolant temp.

2. Again, the objective is to have higher coolant temp (I didn't say we want to increase the engine temp).

3. For sure a 2bar cap will give higher coolant temp compared to a 1.4bar cap, give the engine temp is 120C. Water boils at lower pressure at x temp and same water will boil at higher pressure at y temp, where Y > x.
Well,
here we go again.
We do not want to increase coolant temperature in order to increase heat transfer (coolant temp and engine temp is pretty clearly correlated values, so increasing coolant temp we do increase engine temp, and that is what we trying to avoid by any means, don't we). Instead, we do all possible to decrease the radiator temperature.
Increased coolant temp to 120 C or whatever is not an aim in order to cool engine effective - it is given condition we fight with, in order to keep engine in its working condition.

Lower than stock 2 bar can cause overheated due to lower heat transfer rate coolant.
You didn't reply my question, how lower than stock cap can cause overheating due to lower heat transfer rate coolant.
 

·
Registered
Joined
·
1,307 Posts
What sort of fan are you using? Is it possible the blocked heater hoses are preventing proper circulation?
 

·
Registered
Joined
·
3,550 Posts
Well,
here we go again.
We do not want to increase coolant temperature in order to increase heat transfer (coolant temp and engine temp is pretty clearly correlated values, so increasing coolant temp we do increase engine temp, and that is what we trying to avoid by any means, don't we). Instead, we do all possible to decrease the radiator temperature.
Increased coolant temp to 120 C or whatever is not an aim in order to cool engine effective - it is given condition we fight with, in order to keep engine in its working condition.


You didn't reply my question, how lower than stock cap can cause overheating due to lower heat transfer rate coolant.
OK I will try my best to explain using thermodynamic.

1. "We do not want to increase coolant temperature in order to increase heat transfer"
Well, want or not it was already done this way in billion cars on the road. This is main reason why we have pressurized coolant system in our cars. Without pressurized the coolant will become gas at 100C and the engine will overheated due to lower heat transfer rate.

2. "so increasing coolant temp we do increase engine temp, and that is what we trying to avoid by any means."

Not quite. The head has many hot spots which have much higher temp than 120C even if the head sensor reported at 96C. These spots are around the EX valves and the EX ports. In order to carry enough heat from the head to the radiator in order to balance the amount of heat generated by the combustion, the coolant must be capable of holding heat at higher temp before boiling. Un-pressurized water only can move 20 liters/minute of 100C coolant, while pressurized coolant can carry 20 liters/minute of 110C or higher, to the radiator. The point is we don't need to increase the engine temp in order to increase the coolant temp; In the OP case for racing engine, his engine already generated much higher temp than the normal street cars.

3. "how lower than stock cap can cause overheating due to lower heat transfer rate coolant."
Street cars maybe OK with 1.4 bar cap instead of the stock 2 bar cap, but this race car generated more heat and so 1.4 or anything lower than 2 bar can cause problem. How low the cap rating can cause the problem? It depends on how much heat generate by this engine, assuming it still under the factory spec in worst case condition as the factory had selected 2 bar cap in the design and test cycles. I just wanted to point out to the OP that this is one factor to look.
 

·
Registered
Joined
·
9 Posts
Discussion Starter #17
I race on the 24hr of lemons circuit in the states, so I get your concerns on heat. here are a couple of ideas. I would do a range of things.

I don't know how much water wetter increases the boiling temp of water, but I would make sure you have the highest boiling point you can within race restrictions. Coolant stops absorbing heat when it boils. We run distilled water on the track

100 degrees Celsius is okay, 120 is probably not, but if you could drop it 10-15 degrees I think that would be pretty good.

By blocking off the heater core, you are blocking off another opportunity to add cooling capacity. Could you run the pipes meant for the heater core to a secondary (smaller) radiator mounted somewhere appropriate ?

I would assume you have adequate air flow over the radiator. What about an electric puller fan mounted on the engine side of the radiator?

What about getting the largest radiator that will fit? I think I would take some measurements and then research radiators from M3, M5, X5 both i6, V8, and diesel variants and see if you can stuff a bigger radiator in there. Maybe an Aluminum racing radiator..

Here is a post about using an OEM radiator from an M roadster (Z3). According to link it has a thicker core than the M3.
We have not experienced boiling of the coolant yet, and I think this is attributable to the coolant additives we use like Redline WaterWetter and Motul Mocool with the distilled water.

I like your idea about the secondary radiator fed from the ex-heater core pipes. I will look into this.

Yes, airflow over the radiator has been prioritised, but we also have the original OEM Electric cooling fan pulling on the engine side of the radiator.

The larger radiator is certainly an option, but it is starting to feel like that will also not resolve the problem if we cannot guarantee that it will get adequate coolant flow through it.

Many of our competitors are running similar if not identical cooling systems and are able to maintain coolant temperatures under 100 degrees Celcius throughout the race. With this in mind, anything additional to the cooling system (Bigger radiator, or 2nd radiator) is going to be a "band-aid" over the real problem as there is plenty of evidence that supports the notion that the stock cooling system has sufficient cooling capacity to maintain coolant temperatures for a NA M54B30, even at slightly higher ambient temperatures and in racing conditions.
 

·
Registered
E46 '04 330Ci convertible 750K km
Joined
·
1,142 Posts
Street cars maybe OK with 1.4 bar cap instead of the stock 2 bar cap, but this race car generated more heat and so 1.4 or anything lower than 2 bar can cause problem. How low the cap rating can cause the problem? It depends on how much heat generate by this engine, assuming it still under the factory spec in worst case condition as the factory had selected 2 bar cap in the design and test cycles. I just wanted to point out to the OP that this is one factor to look.
OK, you know your thermodynamics:cool:
Now, back to the question.
Lower than stock 2 bar can cause overheated due to lower heat transfer rate coolant.
As you mentioned heat transfer rate, all I wanted you to describe the way you connect those two things.
Cap holds certain pressure, and after exceeding that level valve suppose to open and reduce pressure, in order to protect engine (more exactly, to protect cooling system elements from bursting).
Now, what happens when cap releases pressure? Temperature of coolant remains the same, but as pressure dropped, it causes coolant to immediate boil. So, valve saved, let's say, ET or radiator from bursting, but we have air in a system now, which leads to spikes of temperature in certain areas of engine, cavitation on water pump impeller, etc. So, we do not want any of that happens, and it is absolutely right and logical to use 2 Bar cap (holds coolant pressure upto 120C), rather than lets say 1.4 bar cap (110C).
So, not questioning your comment about stock cap, yet I cannot see why heat transfer rate was mentioned.
 

·
Registered
Joined
·
3,550 Posts
Cap holds certain pressure, and after exceeding that level valve suppose to open and reduce pressure.


Now, what happens when cap releases pressure? Temperature of coolant remains the same, but as pressure dropped, it causes coolant to immediate boil.
No, pressure wasn’t reduced after cap released pressure.

As the pressure building up and exceeded 2 bar, for example, the relief valve started open to release the excess to maintain a Constant pressure at 2 bar until the system pressure dropped below 2 bar due to less heat generated or more cooling air at the radiator. IOW, the moment the valve started releasing P, the system P did not drop below 2 bar for a 2bar cap.
 
1 - 20 of 34 Posts
Top