N54 Cooling - Temp Control Logic and what are your temps during extended track use?

shushikiary

Sergeant
Jun 4, 2018
304
173
0
Ride
335xi
SPAL does sell some thinner fans at the proper diameter that would likely do the trick, and then we could see if we could find a champion radiator that would work with a led foot racing fan shroud for said SPAL fan.

Obviously that requires custom mounting of the radiator, but quite doable.

A Champion or Griffin 3 core radiator (though 3 inches thick) would work great, and the SPAL fans go as thin as 2 inches.
 

RSL

Lieutenant
Aug 11, 2017
937
501
0
Not to mention, removing the shrouded fan should free up air flow through the rad at speed, but haven't really looked at how much anything in front might be blocking it (supports, crash bar, etc) .

Doubtful any more of came of max cool. Nice thought 7 years ago, but stat still controls the flow and pump/fan running full blast won't change that. Essentially, would be running everything 100%, but not run much/any cooler due to stat regulation. KFT has to be active and targeted appropriately to keep the stat open and KFT should run the pump (and definitely the fan) up to 100% PWM. The temp spread vs. fan PWM can be changed and I'd imagine for the water pump too (i.e. target higher PWM on lower temp diffs).

BTW, some things in the roms do tend to indicate max flow on the water pump is 9000 lph.
 

barry@3DM

Corporal
Jun 4, 2018
155
264
0
Not to mention, removing the shrouded fan should free up air flow through the rad at speed, but haven't really looked at how much anything in front might be blocking it (supports, crash bar, etc) .

Doubtful any more of came of max cool. Nice thought 7 years ago, but stat still controls the flow and pump/fan running full blast won't change that. Essentially, would be running everything 100%, but not run much/any cooler due to stat regulation. KFT has to be active and targeted appropriately to keep the stat open and KFT should run the pump (and definitely the fan) up to 100% PWM. The temp spread vs. fan PWM can be changed and I'd imagine for the water pump too (i.e. target higher PWM on lower temp diffs).

BTW, some things in the roms do tend to indicate max flow on the water pump is 9000 lph.


Very good point, I fully agree! As always, everything is a system. In addition to bigger radiator and running the pump at full speed, I think the thermostat is part of the issue.

Lets talk pump and thermostat. Lets set aside component availability and fitment for now, that can be sorted later, anything is possible. Lets just talk concept.

In a racing application for a mechanical driven water pump you want to run a much cooler thermostat than equilibrium. Just because equilibrium is say, 190F doesn't mean a lower temp thermostat is useless. A lower temp thermostat is more "open" allowing more flow to happen when RPMs are low (think MAP mode of a MAP thermostat). This is one of the drawbacks to a mechanical pump. This is why BMW sells a motorsport thermostat for the e46 M3 that is 55C (131F) even though the cooling systems equilibrium is much hotter.

If we look at an electrical water pump, we don't have the flow problem at low RPMs like a mechanical pump. Therefore we do not need to run an extraordinarily low temp thermostat because the flow will be constant regardless of RPM. Having said that, flow still does matter. The faster the flow through the radiator the more cooling you will get. We are dealing with a thermostat that was designed for a street car. If we found a better thermostat that restricted the flow less, then we would get better cooling.
 

9krpmrx8

Lieutenant
Nov 5, 2016
508
290
0
San Antonio, Texas
I looked at option in the aftermarket, no one sells anything that would fit that will out flow the stock fan. As for the fan no throwing a CEL, I can confirm, my fan failed during cool (well South Texas winter) and it took a few days for me to find out when it warmed up a bit and i was in the drive thru.
 
  • Informative
Reactions: barry@3DM

RSL

Lieutenant
Aug 11, 2017
937
501
0
Very good point, I fully agree! As always, everything is a system. In addition to bigger radiator and running the pump at full speed, I think the thermostat is part of the issue.

Lets talk pump and thermostat. Lets set aside component availability and fitment for now, that can be sorted later, anything is possible. Lets just talk concept.

In a racing application for a mechanical driven water pump you want to run a much cooler thermostat than equilibrium. Just because equilibrium is say, 190F doesn't mean a lower temp thermostat is useless. A lower temp thermostat is more "open" allowing more flow to happen when RPMs are low (think MAP mode of a MAP thermostat). This is one of the drawbacks to a mechanical pump. This is why BMW sells a motorsport thermostat for the e46 M3 that is 55C (131F) even though the cooling systems equilibrium is much hotter.

If we look at an electrical water pump, we don't have the flow problem at low RPMs like a mechanical pump. Therefore we do not need to run an extraordinarily low temp thermostat because the flow will be constant regardless of RPM. Having said that, flow still does matter. The faster the flow through the radiator the more cooling you will get. We are dealing with a thermostat that was designed for a street car. If we found a better thermostat that restricted the flow less, then we would get better cooling.
I get what you're saying, but colder stat would shift the temp curve down in an attempt to offset only IF there's enough capacity in the system to reduce it to the lower temp. Much like a 30psi boost target on stock turbos...you can ask for whatever you want, but you'll only get as much as they can do.

Flow does matter, but it's a double-edged sword. Through the radiator faster also reduces exchange time, but it will also go through the block faster. That would spread heat out more in the coolant and, again, just delay how long it takes to reach an equilibrium. Absent any controls (stat), temps are going to be determined by heat picked up vs. exchange capacity no matter what is done, everything else is just a matter of how long it will take to reach that point.

If you're not on a locked/OTS bin, you can try to target some stupid low head temp and KFT threshold to guarantee stat is wide open and pump is working at high/full %. At that point, temps will pretty much rest solely on cylinder temps (+ everything else on the coolant loop) vs. exchange capacity (at some ambient temp). I'm thinking it's just not going to be anywhere near as cool as we hope it might.

One thing to keep in mind during the quest for cool temps is modes. I don't know if cool ECTs alone would keep it in or toss it back to any cold or warm-up tables. That would need to be looked into and limits changed if it's a problem. For that matter, not sure if there is any active bottom we don't know about yet where the DME will undo anything we've done to keep it from running colder than some minimum, regardless of what we're telling it to do elsewhere. Would be annoying if we setup the system to target say 150F and 100% water pump only to find out the DME turns the water pump off below 160F because it thinks the motor should be warming up lol There seem to be some parameters/cutoffs to various heating strategies other than temp though (request, speed, etc.), so something like that may not be an issue, it just could be.

If it's just unable to reduce temps to an acceptable level with full open stat and water pump, need more rad or might consider adjusting tuning if you're tuned (boost, EGR, timing, AFRs, etc...anything that might help reduce heat without giving up a lot of torque). I might even run on stock bin + track KFT just to see how much tune/more boost might be driving temps.
 

barry@3DM

Corporal
Jun 4, 2018
155
264
0
I looked at option in the aftermarket, no one sells anything that would fit that will out flow the stock fan. As for the fan no throwing a CEL, I can confirm, my fan failed during cool (well South Texas winter) and it took a few days for me to find out when it warmed up a bit and i was in the drive thru.

Your post was great feedback. I've noted it a couple of times in some of my earlier posts. Knowing it does not throw a check engine light was great info. I'm not too concerned with something that "fits" or has the stock flow rate. If we can increase the radiator size, keep the thermostat open and increase the flow of coolant with the waterpump then we don't need the CFM capabilities of the stock fan. Now we can use whatever fan we want. Something thinner, lighter, and gives us more room at the front of the engine.

I get what you're saying, but colder stat would shift the temp curve down in an attempt to offset only IF there's enough capacity in the system to reduce it to the lower temp. Much like a 30psi boost target on stock turbos...you can ask for whatever you want, but you'll only get as much as they can do.

Flow does matter, but it's a double-edged sword. Through the radiator faster also reduces exchange time, but it will also go through the block faster. That would spread heat out more in the coolant and, again, just delay how long it takes to reach an equilibrium. Absent any controls (stat), temps are going to be determined by heat picked up vs. exchange capacity no matter what is done, everything else is just a matter of how long it will take to reach that point.

If you're not on a locked/OTS bin, you can try to target some stupid low head temp and KFT threshold to guarantee stat is wide open and pump is working at high/full %. At that point, temps will pretty much rest solely on cylinder temps (+ everything else on the coolant loop) vs. exchange capacity (at some ambient temp). I'm thinking it's just not going to be anywhere near as cool as we hope it might.

One thing to keep in mind during the quest for cool temps is modes. I don't know if cool ECTs alone would keep it in or toss it back to any cold or warm-up tables. That would need to be looked into and limits changed if it's a problem. For that matter, not sure if there is any active bottom we don't know about yet where the DME will undo anything we've done to keep it from running colder than some minimum, regardless of what we're telling it to do elsewhere. Would be annoying if we setup the system to target say 150F and 100% water pump only to find out the DME turns the water pump off below 160F because it thinks the motor should be warming up lol There seem to be some parameters/cutoffs to various heating strategies other than temp though (request, speed, etc.), so something like that may not be an issue, it just could be.

If it's just unable to reduce temps to an acceptable level with full open stat and water pump, need more rad or might consider adjusting tuning if you're tuned (boost, EGR, timing, AFRs, etc...anything that might help reduce heat without giving up a lot of torque). I might even run on stock bin + track KFT just to see how much tune/more boost might be driving temps.

I was working with a friend to solve a cooling issue on their e46 M3 racecar earlier this year. Their gigantic carbon intake compromised their radiator. I linked them to this article as it explains it way better than I ever could. 9 Rules for Improving Engine Cooling. Give it a read and it will explain why increasing flow increases cooling with calculations to back it up, and everything else cooling system related.

Here is an excerpt from Stewart Components website that will also help explain it....

"A common misconception is that if coolant flows too quickly through the system, that it will not have time to cool properly. However the cooling system is a closed loop, so if you are keeping the coolant in the radiator longer to allow it to cool, you are also allowing it to stay in the engine longer, which increases coolant temperatures. Coolant in the engine will actually boil away from critical heat areas within the cooling system if not forced through the cooling system at a sufficiently high velocity. This situation is a common cause of so-called "hot spots", which can lead to failures."
 

9krpmrx8

Lieutenant
Nov 5, 2016
508
290
0
San Antonio, Texas
Yeah with a larger capacity radiator and more flow to the radiator (or radiators) you might get away with a 14" fan that flows 1500CFM or so, especially on a track car.
 

barry@3DM

Corporal
Jun 4, 2018
155
264
0
I get it, the N54 has been around for 11 years and this topic has been beat to death. I started this thread with the hope that people would get engaged regardless and I really appreciate all who have and I hope you continue. At this time I think it's time to push this thread towards a new path. Hopefully it turns into something we can all get something out of and help us as a community.

Whats the problem i'm trying to address? The problem is the N54 as a "performance hot rodded" motor runs too hot and the coolant temp fluctuates too much. (This fluctuation as a STOCK street car for the masses is totally fine.) Anyways, personally, I think the cooling system in the aftermarket is severely being overlooked. Everything I have seen/read is about how to prevent these temp fluctuations from causing limp mode or power reduction mode from kicking in. I believe we should be trying to figure out how to get it even cooler and most importantly, CONSISTENT (no fluctuating) like a hot rodded motor's coolant system should be functioning.

Since we have electric water pumps and MAP thermostats, we have the DME added to the equation of cooling the motor. When you are punching out 1.5x to 2x the WHP, you are putting a ton more heat through the motor which the mechanical cooling system was not designed to control nor was the DME programming designed to control. I'll quote Stewart Components again...

"A common misconception is that if coolant flows too quickly through the system, that it will not have time to cool properly. However the cooling system is a closed loop, so if you are keeping the coolant in the radiator longer to allow it to cool, you are also allowing it to stay in the engine longer, which increases coolant temperatures. Coolant in the engine will actually boil away from critical heat areas within the cooling system if not forced through the cooling system at a sufficiently high velocity. This situation is a common cause of so-called "hot spots", which can lead to failures."

So why am I pointing this out? When I was racing e36 M3s back in the 90s our engine builder used to drill out the holes in the #s 4, 5, and 6 cooling channels of the headgasket in order to create more flow to keep them cooler for this very reason of hot spots. On the N54 the coolant enters the exhaust side of block between cyliders 3 and 4 and exits at the front of the cylider head. It doesn't surprise me that the N54 has predominant bank 2 detonation problems when we start pumping 1.5x to 2x power (heat) through them.

So this is where i'd like to see the thread progress. How we can control the waterpump and thermostat better with the DME. We know we can change targets via XDF files (or MHD "race mode"), but do we know what the water pump and thermostat is actually doing when we do that? I personally want the waterpump to be running full blast so that coolant is flowing at it's max around those bank 2 cylinder walls.

Someone on another forum mentioned the Carly or Torque app had a fan speed PID and could log fan RPM. It would be awesome if there was a PID for water pump % we could log. Otherwise we might have to log it with a datalogger such as the AIM. Not sure how to go about that. Its a DC motor with it's own internal module that is controlled via PWM from the DME.

Anyways, again, thanks for everyone's input so far. I really appreciate it!
 
  • Like
Reactions: RSL and fmorelli

RSL

Lieutenant
Aug 11, 2017
937
501
0
Yep, two-way street on the flow rate and pretty succinct: closed system so no matter what you do, will eventually wind up where in and out balance. How long it takes is essentially what's being changed. I dig math, so I'll look at that page more closely when I have time.

INPA could/should be an option for seeing PWM, short of that, maybe Jake can add the IDs to the MHD logger so it can all be viewed in one file. I'd really like him to add EGT. You can hear when the fan is full blast or at a high level (800w sounds like a GIANT vacuum), so if we had to pick, I'd think water pump PWM should be the focus of getting data. I'm moving over the next few days, but when I get time, I'll break out INPA and see what can be seen in it.
 
  • Like
Reactions: barry@3DM

barry@3DM

Corporal
Jun 4, 2018
155
264
0
Interesting, didn't even think of INPA. That would be awesome if you would check that out. I'll give it a look myself but odds are you are WAY better at INPA than I am! I've had no need to use it much so not that well versed in it.

Agreed that water pump should be the focus. It would also be nice to know if MAP thermostat was triggered or not, too. Simple log of 1 for yes and 0 for no. Fan speed I don't really care about, my goal is to ditch it anyways.

As for Jake adding the IDs in MHD, that is why I stated in post #48 that since this topic has been beat to death, I imagine many have skipped over this thread with roll eyes. Hopefully people will see where I'm coming from with the cooling around Bank 2 and bring this back to the forefront of concern. It honestly concerns all N54 hot rodders... street, drag, and track guys. Big market there. Hopefully Jake and/or Martial get engaged in the topic again from a DME standpoint.
 

barry@3DM

Corporal
Jun 4, 2018
155
264
0
I'm going to try something to hopefully capture more people to this thread. I am going to link directly to this post from the original post so newcomers can jump straight to the new direction of the topic of discussion. So for those that have been following along, please bear with me for the repeat information in this post.

Refocused Topic of Discussion...

I get it, roll eyes, the N54 has been around for 11 years and the cooling topic has been beaten to death. However, I want to bring to light some issues that I still see with our N54 cooling system. To date, everyone has been trying to keep the engine from hitting limp mode, which is great, but more needs to be done than that in my opinion.

Lets step back and look at the big picture. The N54 was developed as a street car, not a performance hot rod like the M3. It was also the first new era BMW turbo motor. Fast forward several years and BMW themselves address the lacking cooling system so they added a PPK such as that found on the 1M and the 335is. In their own literature they discuss how it was challenging developing the cooling system because the radiator frontal surface area had to be shrunk drastically to make room for the FMIC. We have to remember, the design parameters were for a motor they considered a street car motor where efficiency and emissions were high on the priority list. To accomplish this they ran the motor hot to lower emissions and to heat up the oil for better fuel economy (less drag).

Now... here comes the aftermarket world of hot rod turbo motors. This creates a WHOLE different set of parameters for the cooling system. Unlike the original intent of the cooling system, we don't care about running hot for emissions nor do we care about fuel economy. Now, the design logic/parameters built into the stock cooling system no longer apply.

Now that we are pushing 1.5x to 2x the power through it, we have to follow "performance" engine cooling parameters, which is simply keep it cool at all times. But there are details involved with this. We want to do things like keep the motor at volumetric efficiency (VE) which BMW claims is 194F, any higher than that is pointless. We also don't want a drastically fluctuating temp, again, absolutely pointless. lastly, and most important of all, we want max coolant fluid flow through the motor especially at high RPMs and load. That last parameter right there is what I think has been overlooked and what I want to discuss.

Cooling systems have been studied and perfected for decades, Stewart Components sums it up best with regard to coolant flow:

"A common misconception is that if coolant flows too quickly through the system, that it will not have time to cool properly. However the cooling system is a closed loop, so if you are keeping the coolant in the radiator longer to allow it to cool, you are also allowing it to stay in the engine longer, which increases coolant temperatures. Coolant in the engine will actually boil away from critical heat areas within the cooling system if not forced through the cooling system at a sufficiently high velocity. This situation is a common cause of so-called "hot spots", which can lead to failures."

So, we want lots of coolant flow at high RPMs where we are putting the most heat into the motor. We don't want hot spots that temporarily distort cylinder walls, cause detonation, list goes on.

This is where we start talking coolant targets built into the DME. Most of us know there are several different targets for coolant temps. ECO (226F), Normal (219F), High (203F), and High + Map thermostat (194F). And most of us know we can change this with XDF files or MHD. What we don't know (or at least I dont know) is the logic behind these modes. I'm not talking about the logic of ambient vs coolant temp vs throttle position. I'm talking about the logic of the duty cycle of the water pump.

What I want to determine are the times when the waterpump is running at a low duty cycle creating slow coolant flow. Lets imagine we are doing some spirited driving and are running along at 210F. Then we decide to do a pull and take off. We now immediately start putting heat into the cylinders but what we don't know is if the waterpump is at 100% duty cycle to dissipate that heat. Lets remember, the DME logic was designed around a 300 BHP street car. It may not go to instant 100% duty cycle. BMW may have determined it wasn't necessary. But we have 1.5x to 2x the power of stock. If the pump is not at full duty cycle our cylinder walls may start to heat up which aids in detonation causing the timing to get pulled back. Loose all around.

I'm not an engine builder or tuner (or DME coder) so I am not an expert and will never claim to be. I will say I've built racecars both in club and pro racing for over 20 years and have dealt with the lingo and the concepts enough to have a clue. I'm hoping some engine builders, tuners, and more specifically DME coders will chime into this thread. Or anyone who might know the logic for that matter.

I was discussing with @RSL about logging waterpump duty cycle within MHD. It is currently not an ID we can log but in theory should be easy to add. Again, not a coder so don't know. Logging Map thermostat actuation would be nice too. @RSL had a great idea to try INPA. Anyways, once we get that information we can start understanding the logic of the DME and perhaps make changes for the better.

Again, sorry about the repeat info to all who have been following this thread.
 

RSL

Lieutenant
Aug 11, 2017
937
501
0
INPA appears to have everything: water pump PMW, fan, coolant temp in/out, thermostat heater activation, etc. It's INPA though, so, naturally, all are on different pages lol I'm not sure if custom pages can written to move all into one, but logging ability would be ideal.

If anyone knew the actual logic for the modes, it would be @jyamona.
 

gmx

Specialist
Dec 8, 2017
69
29
0
Ride
BMW 1M
Nice thread, glad to see "it has begun". I was shocked with an electric waterpump these parameters weren't even in consideration until a handful bugged Martial about it :)
Now we're getting to mechanical aspects and hopefully, actual DME logic applied with values in table(s).
 
  • Like
Reactions: barry@3DM

barry@3DM

Corporal
Jun 4, 2018
155
264
0
INPA appears to have everything: water pump PMW, fan, coolant temp in/out, thermostat heater activation, etc. It's INPA though, so, naturally, all are on different pages lol I'm not sure if custom pages can written to move all into one, but logging ability would be ideal.

If anyone knew the actual logic for the modes, it would be @jyamona.

This is fantastic! At least it is something to work with immediately. Can be as simple as having someone sit in the passenger seat and watch it as it happens. Or... use screen video capture software. I hope to give this a try sometime in the next couple of days.

Nice thread, glad to see "it has begun". I was shocked with an electric waterpump these parameters weren't even in consideration until a handful bugged Martial about it :)
Now we're getting to mechanical aspects and hopefully, actual DME logic applied with values in table(s).

I agree. The cooling system... such a foundation piece of the motor where you want that foundation to be rock solid.

Again, not a coder, but seems like figuring out all the logic applied to all the tables will be time consuming. So I could envision this being a two step process.

Quick Fix 1st Release:
Leave all current settings of Stock, Sport, etc. Add and additional mode that would do the following...
Run the pump at 100% duty cycle at all times. Remove MAP thermostat from the equation so it doesn't throw codes or go into limp mode. Then hardware change would be modify thermostat (by replacing internals) that runs at 190F instead of the stock 210F.

2nd Release:
Dig into the logic and work with the community on what would work best. There could be multiple settings, just like there are now in MHD. Examples...

StockMode: don't change anything for those who don't care

Mode 1: Target X temp with Y ambient. When throttle goes over 75% run pump at full duty cycle. This would get the coolant flowing at full speed through the motor for max cooling of cylinder walls BEFORE you blast the motor with heat from doing a full throttle pull to redline. Great for everyday street driving.

Mode 2: Target X temp by using MAP thermostat, remove ambient from equation. Run pump full duty cycle at all times. Great for drag racers.

Mode 3 (requires hardware change): Run the pump at 100% duty cycle at all times. Remove MAP thermostat and fan fault (if any) from the equation so it doesn't throw codes or go into limp mode. Then hardware change would be modify thermostat (by replacing internals) that runs at 190F instead of the stock 210F. Replace radiator and fan with whatever you want. Great for track guys.
 

Bnks334

Lieutenant
Dec 1, 2016
524
340
0
New York
mon misconception is that if coolant flows too quickly through the system, that it will not have time to cool properly. However the cooling system is a closed loop, so if you are keeping the coolant in the radiator longer to allow it to cool, you are also allowing it to stay in the engine longer, which increases coolant temperatures. Coolant in the engine will actually boil away from critical heat areas within the cooling system if not

This is fantastic! At least it is something to work with immediately. Can be as simple as having someone sit in the passenger seat and watch it as it happens. Or... use screen video capture software. I hope to give this a try sometime in the next couple of days.



I agree. The cooling system... such a foundation piece of the motor where you want that foundation to be rock solid.

Again, not a coder, but seems like figuring out all the logic applied to all the tables will be time consuming. So I could envision this being a two step process.

Quick Fix 1st Release:
Leave all current settings of Stock, Sport, etc. Add and additional mode that would do the following...
Run the pump at 100% duty cycle at all times. Remove MAP thermostat from the equation so it doesn't throw codes or go into limp mode. Then hardware change would be modify thermostat (by replacing internals) that runs at 190F instead of the stock 210F.

2nd Release:
Dig into the logic and work with the community on what would work best. There could be multiple settings, just like there are now in MHD. Examples...

StockMode: don't change anything for those who don't care

Mode 1: Target X temp with Y ambient. When throttle goes over 75% run pump at full duty cycle. This would get the coolant flowing at full speed through the motor for max cooling of cylinder walls BEFORE you blast the motor with heat from doing a full throttle pull to redline. Great for everyday street driving.

Mode 2: Target X temp by using MAP thermostat, remove ambient from equation. Run pump full duty cycle at all times. Great for drag racers.

Mode 3 (requires hardware change): Run the pump at 100% duty cycle at all times. Remove MAP thermostat and fan fault (if any) from the equation so it doesn't throw codes or go into limp mode. Then hardware change would be modify thermostat (by replacing internals) that runs at 190F instead of the stock 210F. Replace radiator and fan with whatever you want. Great for track guys.

It sounds to me like you are trying to do something that has ALREADY been done...

Race Mode (only for drag race application)
• Coolant normal 85c(104c Stock)
• Coolant high 85c(95c Stock)
• Coolant high + kft 80c(85c / 80c Stock) (This mode is activated at full throttle)
• Amb threshold for high mode 22c(32c Stock)
• Amb threshold for high + kft 27c(37c Stock)

The coolant logic is pretty simple for "high+kft" mode... It's an either/or scenario. Refer to MHDs data above...

STOCK settings:
If throttle = 100% then coolant temp target = 185f (you rarely go WOT on track so this statement is rarely true)
else if ambient > 98.6f then coolant temp target = 185f (it's rarely > 98.6f ambient temp outside so this logical statement is rarely true)

on a 90f day at the track, the stock cooling logic will pretty much never try to target 185f. It will be targeting 203f coolant at all times unless you go WOT.

MHD simple changed the setpoints to:
If throttle = 100% then coolant temp target = 176f (you rarely go WOT on track so this statement is rarely true)
else if ambient > 80.6f then coolant temp target = 185f (it's rarely > 98.6f ambient temp outside so this logical statement is rarely true)

If you flash MHD race settings, and it's hotter than 80.6f outside, coolant temps should not fluctuate at all. They should be pinned at 176f all day. There are no hardware changes needed to achieve this. If you wan't to target 176f at 30f for winter racing, then get a custom tune. Justin @ TT knows his way around these tables and can target whatever coolant temp you want at whatever threshold you want. No need for a "map thermostat delete."

What you seem to be missing in what people are trying to tell you is that the stock radiator just isn't good enough to achieve 176f coolant even when it's 70f outside lol. A radiator upgrade has proven to drop coolant temps dramatically (or add an aux radiator like PPK does). The radiator appears to be the first choke point. This is especially true for AT cars that have reduced capacity.

The scientific formulas for calculating heat exchange are readily available and common knowledge. How much heat you can shed from the system is just as important as how much heat you can wick away from the cylinder head. You can't just upgrade one component to see a result unless that one component is creating a restriction in the system. Move to straight water and you improve the coolant's ability to transfer heat; however, if the radiator is still heat-soaked and can't exchange any more heat to the atmospheric air passing through it you won't drop temps one bit. The radiator is the first place to start addressing the cooling system after coding with MHD to target 176f coolant at all times and realizing you can never achieve that target temp...

I'd argue oil cooling is just as important. If you can keep oil at <250f on track via oil coolers instead of 280f then your cooling system is going to be much more effective.
 
Last edited:

barry@3DM

Corporal
Jun 4, 2018
155
264
0
It sounds to me like you are trying to do something that has ALREADY been done...

The coolant logic is pretty simple for "high+kft" mode... It's an either/or scenario. Refer to MHDs data above...

STOCK settings:
If throttle = 100% then coolant temp target = 185f (you rarely go WOT on track so this statement is rarely true)
else if ambient > 98.6f then coolant temp target = 185f (it's rarely > 98.6f ambient temp outside so this logical statement is rarely true)

on a 90f day at the track, the stock cooling logic will pretty much never try to target 185f. It will be targeting 203f coolant at all times unless you go WOT.

MHD simple changed the setpoints to:
If throttle = 100% then coolant temp target = 176f (you rarely go WOT on track so this statement is rarely true)
else if ambient > 80.6f then coolant temp target = 185f (it's rarely > 98.6f ambient temp outside so this logical statement is rarely true)

If you flash MHD race settings, and it's hotter than 80.6f outside, coolant temps should not fluctuate at all. They should be pinned at 176f all day. There are no hardware changes needed to achieve this. If you wan't to target 176f at 30f for winter racing, then get a custom tune. Justin @ TT knows his way around these tables and can target whatever coolant temp you want at whatever threshold you want. No need for a "map thermostat delete."

What you seem to be missing in what people are trying to tell you is that the stock radiator just isn't good enough to achieve 176f coolant even when it's 70f outside lol. A radiator upgrade has proven to drop coolant temps dramatically (or add an aux radiator like PPK does). The radiator appears to be the first choke point. This is especially true for AT cars that have reduced capacity.

The scientific formulas for calculating heat exchange are readily available and common knowledge. How much heat you can shed from the system is just as important as how much heat you can wick away from the cylinder head. You can't just upgrade one component to see a result unless that one component is creating a restriction in the system. Move to straight water and you improve the coolant's ability to transfer heat; however, if the radiator is still heat-soaked and can't exchange any more heat to the atmospheric air passing through it you won't drop temps one bit. The radiator is the first place to start addressing the cooling system after coding with MHD to target 176f coolant at all times and realizing you can never achieve that target temp...

I'd argue oil cooling is just as important. If you can keep oil at <250f on track via oil coolers instead of 280f then your cooling system is going to be much more effective.

I realize I updated the original post and said skip straight to post #51. I will say its still worth reading as a fair bit of detailed info has been covered.

I think you are missing what I want clarification on. First things first. I know the radiator is the choke point of the system, likely the thermostat, too. I started this thread to get people engaged, not because I didn't know what was going on and needed someone to tell me. This is why I changed the focal point of the conversation (post #51) after 3 pages of engagement. I'm not miss understanding what people are saying. I'm well versed with the N54. Rebuilt them from the ground up. Helped a friend run and prep a 335i when they first came out in 2007 for track use. Been helping another friend with a track duty single turbo conversion 1M for the last 3 years. Ran an e90 328i in Grand AM which uses the exact same waterpump and thermostat as the 335i. That car never had a problem with cooling but the motor only made 300 HP and we didn't have to run the stock ECU as we were using the Bosch Motorsport ECU. With that we ran the pump full bore. Obviously not an apples to apples comparison there. Having said all of that, I have not been fully engaged on an N54 cooling solution because I was personally campaigning other models of BMWs and it wasn't a priority for me. But now it is.

Here is a snippet from BMWs literature about the cooling system:
The engine control unit of the N54 engine controls the coolant pump according to requirements:
• Low output in connection with low cooling requirements and low outside temperatures
• High output in connection with high cooling requirements and high outside temperatures.


The software in the engine control unit now features a calculation model that can take into account the development of the cylinder head temperature based on load. In addition to the characteristic map control of the thermostat, the heat management system makes it possible to use various maps for the purpose of controlling the coolant pump. For instance, the engine control unit can adapt the engine temperature to match the current operating situation. This means that four different temperature ranges can be implemented: • 108°C ECO mode • 104°C Normal mode • 95°C High mode • 90°C High + map-thermostat mode The control system aims to set a higher cylinder-head temperature (108°C) if the engine control unit determines ECO (economy) mode based on the engine performance. The engine is operated with relatively low fuel consumption in this temperature range as the internal friction is reduced.

Here is MHD Literature:
Race Mode (only for drag race application)
• Coolant normal 85c(104c Stock)
• Coolant high 85c(95c Stock)
• Coolant high + kft 80c(85c / 80c Stock) (This mode is activated at full throttle)
• Amb threshold for high mode 22c(32c Stock)
• Amb threshold for high + kft 27c(37c Stock)

Nowhere does it say in either of those literature pieces that the water pump is at 100% duty cycle or MAP thermostat engaged at all times when trying to "target" a temperature. What does "high output" mean... a range between 50% to 100% duty cycle or just 100% duty cycle? My ASSUMPTION is it is running the pump at full speed and in MAP thermostat mode if the coolant temp is above target temp. I'd rather not assume and have hard proof data. It does say it "features a calculation model that takes into account the development of the cylinder head temperature based on load." This leads me to believe that it's possible it turns the pump down even if the cylinder head temps are high but load is low. For example from 100% down to maybe 75% until it sees load again then ramps it back up to 100%. I reference a throttle histogram earlier in the thread (post #26) where it shows 30% of the time is spent below 30% throttle, in other words a low load input. Basically I have no idea what those calculation models are but I hope someone does! That is the point of this thread.

As for the thermostat, I cover this in post #43. The thermostat is a 210F thermostat. It has to use MAP to open more to go any lower. While the radiator is the biggest culprit to the system, the thermostat may be a restriction to the system reducing coolant flow capabilities (street car thermostat). We just don't know. As for scientific formulas, I reference coolant flow calcs in post #46. If I go and replace it with a higher flow version and one that opens at 190F, I will want to omit the need for the MAP logic of the DME as I will get a check engine light and power reduction mode.

Testing is expensive. I can remember a handful of times in the last couple of years where I went to the track to test something and got no data because of either problems or rain. Entry fee, hotel, travel expenses add up. I would hate to spend a ton of time to make a custom radiator to find out it didn't solve my problem. Then I would hate to develop a custom thermostat to find that it did not solve my problem (and put the DME in limp mode). So I'm starting with the foundation, the piece that controls the water pump and thermostat. Once I determine DME logic and deem what it's doing acceptable, then I'll move on to the radiator and thermostat projects.

If I have 220-230F coolant temps on the track I would rather attack that first. Dropping coolant temps down to 195-200F will drop oil temps. Modern day oil such as Motul 300v can handle higher temps and not be a detriment to the engine.
 

Bnks334

Lieutenant
Dec 1, 2016
524
340
0
New York
I realize I updated the original post and said skip straight to post #51. I will say its still worth reading as a fair bit of detailed info has been covered.

I think you are missing what I want clarification on. First things first. I know the radiator is the choke point of the system, likely the thermostat, too. I started this thread to get people engaged, not because I didn't know what was going on and needed someone to tell me. This is why I changed the focal point of the conversation (post #51) after 3 pages of engagement. I'm not miss understanding what people are saying. I'm well versed with the N54. Rebuilt them from the ground up. Helped a friend run and prep a 335i when they first came out in 2007 for track use. Been helping another friend with a track duty single turbo conversion 1M for the last 3 years. Ran an e90 328i in Grand AM which uses the exact same waterpump and thermostat as the 335i. That car never had a problem with cooling but the motor only made 300 HP and we didn't have to run the stock ECU as we were using the Bosch Motorsport ECU. With that we ran the pump full bore. Obviously not an apples to apples comparison there. Having said all of that, I have not been fully engaged on an N54 cooling solution because I was personally campaigning other models of BMWs and it wasn't a priority for me. But now it is.

Here is a snippet from BMWs literature about the cooling system:
The engine control unit of the N54 engine controls the coolant pump according to requirements:
• Low output in connection with low cooling requirements and low outside temperatures
• High output in connection with high cooling requirements and high outside temperatures.


The software in the engine control unit now features a calculation model that can take into account the development of the cylinder head temperature based on load. In addition to the characteristic map control of the thermostat, the heat management system makes it possible to use various maps for the purpose of controlling the coolant pump. For instance, the engine control unit can adapt the engine temperature to match the current operating situation. This means that four different temperature ranges can be implemented: • 108°C ECO mode • 104°C Normal mode • 95°C High mode • 90°C High + map-thermostat mode The control system aims to set a higher cylinder-head temperature (108°C) if the engine control unit determines ECO (economy) mode based on the engine performance. The engine is operated with relatively low fuel consumption in this temperature range as the internal friction is reduced.

Here is MHD Literature:
Race Mode (only for drag race application)
• Coolant normal 85c(104c Stock)
• Coolant high 85c(95c Stock)
• Coolant high + kft 80c(85c / 80c Stock) (This mode is activated at full throttle)
• Amb threshold for high mode 22c(32c Stock)
• Amb threshold for high + kft 27c(37c Stock)

Nowhere does it say in either of those literature pieces that the water pump is at 100% duty cycle or MAP thermostat engaged at all times when trying to "target" a temperature. What does "high output" mean... a range between 50% to 100% duty cycle or just 100% duty cycle? My ASSUMPTION is it is running the pump at full speed and in MAP thermostat mode if the coolant temp is above target temp. I'd rather not assume and have hard proof data. It does say it "features a calculation model that takes into account the development of the cylinder head temperature based on load." This leads me to believe that it's possible it turns the pump down even if the cylinder head temps are high but load is low. For example from 100% down to maybe 75% until it sees load again then ramps it back up to 100%. I reference a throttle histogram earlier in the thread (post #26) where it shows 30% of the time is spent below 30% throttle, in other words a low load input. Basically I have no idea what those calculation models are but I hope someone does! That is the point of this thread.

As for the thermostat, I cover this in post #43. The thermostat is a 210F thermostat. It has to use MAP to open more to go any lower. While the radiator is the biggest culprit to the system, the thermostat may be a restriction to the system reducing coolant flow capabilities (street car thermostat). We just don't know. As for scientific formulas, I reference coolant flow calcs in post #46. If I go and replace it with a higher flow version and one that opens at 190F, I will want to omit the need for the MAP logic of the DME as I will get a check engine light and power reduction mode.

Testing is expensive. I can remember a handful of times in the last couple of years where I went to the track to test something and got no data because of either problems or rain. Entry fee, hotel, travel expenses add up. I would hate to spend a ton of time to make a custom radiator to find out it didn't solve my problem. Then I would hate to develop a custom thermostat to find that it did not solve my problem (and put the DME in limp mode). So I'm starting with the foundation, the piece that controls the water pump and thermostat. Once I determine DME logic and deem what it's doing acceptable, then I'll move on to the radiator and thermostat projects.

If I have 220-230F coolant temps on the track I would rather attack that first. Dropping coolant temps down to 195-200F will drop oil temps. Modern day oil such as Motul 300v can handle higher temps and not be a detriment to the engine.

Ok the thermostat restricting flow is different.. I didnt see that as being what you were getting at in any of your posts...

The bulk of what you're saying seems to be focused around the map control and the logic on how it works.

Maybe I'm wrong, but I've literally watched my coolant temps drop to the targets in the cooling table posted above purely as ambient conditions changed... I'll have to read back about your throttle comments but it just sounds like youre overcomplicating a very simple thing.

If I used mhd race settings, and its 100f outside, my car is trying to achieve 176f coolant temps whether I am using 10% throttle or 100%. Are you saying that I am wrong on that?

You're absolutely right that it shouldn't be assumed that we are getting 100% out of the water pump. That would need to be varified in inpa. I've raised this question myself elsewhere... I got shot down pretty quickly by almost every who responded with "it's a pid based system" and it does t make sense that pid would be restricted to a certain range in each mode...its assumed 100% duty cycle can be achieved in any mode in order for the dme to meet the coolant set point for the given ambient temp.

Ambient thresholds seems to be the main driver behind what coolant temp you see. Throttle is only in mentioned for high+kft... so, I'd think that's the only mode where load/throttle matters. The literature is generic. Cant apply it across the board. I'm basing my statements on watching my coolant temps as ambient temp changed from one mode set point to the next... coolant temps changed 10* within seconds as I drove down the highway purely because of a 1f difference in ambient temp...
 
Last edited:

barry@3DM

Corporal
Jun 4, 2018
155
264
0
If I used mhd race settings, and its 100f outside, my car is trying to achieve 176f coolant temps whether I am using 10% throttle or 100%. Are you saying that I am wrong on that?

No, never questioned anything about that ever. I simply want to know what the water pump and MAP thermostat are doing when you are at 10% and 100% throttle. That's all i'm asking. My assumption is water pump is at full duty cycle and MAP thermostat is in effect. But I do not know that for a fact.

I'm not trying to over complicate it, i'm trying to understand it.

One thing that seems to a problem people face is Bank 2 has more problems and failures than Bank 1? I don't have enough data points on that from people, please help me out on that.

In an earlier post I quoted the Stewart Components website about coolant flow through the block and hot spots lead to trouble etc. The n54 coolant enters the block exhaust side between cylinder 3 and 4 and It exits at the front of the head. What if bank 1 gets more flow and cooling than bank 2? We don't know because we only see a single coolant temp as it exits the front of the head.

Look at the s14, one of the most successful motorsport engines BMW ever built. It distributes coolant evenly across each cylinder. I circled the coolant pipe in red.

s14.JPG



Now look at the s54, another helluva BMW motorsport motor. It distributes coolant evenly across the cylinders as well. Blue circles are each pair of cylinder's coolant exit on the coolant pipe.

s54.JPG



I've never had a problem with s14s and s54s. When I raced e36 325s and M3s (the m50 based mass production streetcar motor) we had problems with the cooling of bank 2. Coolant enters at the front of the head and exits at the front of the head. My engine builder would drill out the bank 2 head gasket holes to increase flow back there to keep the cylinder walls cooler.

I am NOT saying this is the problem with the N54 but there is a chance it could be. It's open deck so different design but that doesn't mean coolant flow is even across all cylinders. Too bad it doesn't utilize an exit coolant pipe like the s54. Anyway, I just want to verify coolant flow, I want to make sure its flowing as best as it can when there is a ton of heat going through the motor.

So to the point...

I love technology, and what the DME does with the coolant system is awesome for a street car with all the targeting this, mode that. We have modded them and changed them to "performance" motors. It's simply over complicated. It should be simplified down. But I personally do not have the knowledge to do that within the DME so hopefully the community can help. This is why i'm trying to figure out what the pump is doing. If it truly is running at 100% when targeting X temp, then maybe I don't have to worry about it. (However, that still doesn't solve the potential thermostat flow issue). If I can figure all this out then that will sure save me a lot of headache in the future.
 
Last edited:
  • Informative
Reactions: fmorelli
Jan 31, 2017
364
716
0
www.hydraperformance.com
Ride
2010 135i 6MT
Barry,

Keep in mind that the NA M motors have a MUCH lower heat flux than our engines do. Coolant flow into the N54 block is pretty symmetrical across Bank1/Bank2 , and the coolant passages in the HG actually get smaller as you move closer to cyl #1. I would imagine that, with the advent of modern FEA tools and unconstrained by using an all-new design, the NG6 motors should have pretty symmetrical coolant distribution across the cyls despite the lack of external cooling manifolds like in the S14/S54 above. FWIW the older Ferrari "Dino" V8s and V12s use a similar setup, its just the way things were done on race/hi-po engines back in the day before the advent of modern computer aided engineering tools.
 
  • Like
Reactions: RSL and barry@3DM

Bnks334

Lieutenant
Dec 1, 2016
524
340
0
New York
I realize I updated the original post and said skip straight to post #51. I will say its still worth reading as a fair bit of detailed info has been covered.

I think another thing that should be noted in regard to the cooling logic is that there is an "economy mode" that MHD does not cover (this might be what the literature you quoted is referencing). They leave it unchanged in their cooling setting changes.The car rarely goes into economy mode, from what I've seen. It's definitely not applicable at all to a discussion about tuning cooling tables for a track car. It's not like the car is dropping into economy mode on the track as you coast around a corner. If there is a load, throttle, and speed logic applied to all "modes" then it seems to have a relatively insignificant impact. Ambient thresholds seem to be the main driving factor of what coolant temps you see.

I can't speak on cylinder head coolant flow other than it's an open deck. So long as the channels are "tuned" well there shouldn't be too much hot spotting, like you noted. However, BMW did revise quiet a bit of the oiling and cooling channels on the N55 over the outgoing N54. The head-gasket is significantly different. I did notice there is quiet a bit of overlap of the gasket over certain ports. BMW must'v seen improvements to be made to have casted a new block and cylinder head...

No one at Pirelli world challenge seems to have any issues racing the M235iR cars (n55s). You might be able to contact classic BMW or something and see what they have to say in regard to coolant DME logic on those cars.
 
Last edited: