Charge air cooling

[MrT]

Hello wonderful forum members. I'm steadily building up to having no more excuses or uncertainties in my quest to supercharge my 1zz and I have a concept to run by your more experienced minds regarding cooling charge air.

1st principal, our rear-mid engine layout doesn't lend itself to air-to-air cooling. No worries, a water-to-air cooler will be more efficient/effective anyway, great!

Secondly, oh, a water-to-air cooler is more complicated a system, ok no problem they still work and well.

So how do we simplify this for our rear-mid engine layout? This is my proposal:

- are we duplicating a system already present? Yes, the ICE cooling system does the same job...
- why would we duplicate the system? The ICE cooling system does not have capacity for both requirements.
- can we increase the capacity of the IVE cooling system? Yes, large performance aftermarket radiators can be bought.
- would this provide sufficient capacity for both systems? Theoretically yes, depending on circumstances. Traffic light racing in Dubai possibly would struggle, but the UK climate and daily or even track driving should provide sufficient cooling flow and capacity to both systems.
- but how do you supply / control fluid flow and delivery effectively between both systems?

My proposal:
- plumb the water-to-air charge cooler in parallel with the ICE within the engine bay.
- plumb a second 12v water pump on the parallel circuit pushing into the water-to-air charge cooler.
- both pumps prevent reverse flow, but must be sufficiently matched (12v pump) to one another.
- the combined output from both circuits returns to the performance radiator with combined flow rate.
- both pumps draw back cooled fluid at equally sufficient rates to satisfy their demands.
- ECU mapping and auxiliary cooling fans compensate for any variations in system effectiveness such as low speed (cooling) or high power (heating) changes to maintain ICE operation and performance.

So, this looks and sounds great in my head and I'm theory on paper but have I missed anything or been naive or anything or is there merit to this idea?

Please post experience and /or empirical evidence to support arguements for or against this concept. Thanks in advance.

[MrT]

I deliberately kept this separate to allow quoting in your responses.

My second concept is that in average daily use and load, a supercharged application doesn't suffer significantly without charge air cooling. So with sufficient supply, would a water (no methanol etc) injection system provide sufficient cooling to the charge air to deliver similar performance when required and (based on control parameters) not operate during light loads or cruising/idling at lights etc. With the caveat that the volume of water required for any reasonable range is not under review, only the engine operation and principal concept.

I will attempt the mathematical calculations to determine delivery volume required and capacity needed on board to match anfuel tank range when I have time on the PC but your input now is appreciated.

Again, please only post experience or empirical evidence to support the discussion, not opinion or hearsay. Thanks again.

[lamcote]

Question 1, I suspect (but don't know) the ICE radiator does not cool the water in the engine sufficiently to work effectively with a charge cooling system. The ICE radiator is designed to keep the engine at 70-80 degC whereas you would want the coolant in a charge cooler to be way cooler than that. Running the charge cooling water mixed in with the ICE coolant seems likely to result in the charge cooler becoming a charge warmer! Perhaps you could measure the temperature of the water at the radiator output?
Question 2, water injection can certainly be very effective at intercooling, the amount of water per tank of fuel would depend entirely on how much of the time you would be using boost, rather than cruising, as long as you include a supercharger bypass system for part throttle operation. Otherwise you would probably need to be injecting water all the time, if you use enough boost to warrant needing it.

Edit. If it was possible to combine ICE cooling and turbo intercooling together into one radiator it would absolutely definitely be done in Formula 1 cars as the weight and particularly aerodynamic benefits would be massive. The fact that it isn't strongly suggests it doesn't work.

[shnazzle]

Patrick summed it all up very nicely!
Can't plumb it into the ICE. Might not turn it into a charge heater but it will certainly make it less efficient than an air-to-air intercooler.
I'm sure it's possible somehow but it'd require a lot of plumbing, thermostats,pumps, valves, etc.

Water injection is great but it's as much a temperature control as a knock control. Mapping it with water means being able to advance a bit more to reach that maximum brake torque across a wider range.
But if water runs out or there's an issue with the water system? All fine but you'd need to factor in a LOT of protection and very quick protection at that.
Running out of water at max boost would be absolutely fatal to the engine.
Cooling would mitigate that risk greatly.

[MrT]

Thanks guys.

The thermostat controls the amount of water going through the motor to cool it. I'm proposing the cooler pumps its own water so it would flow through a larger radiator and therefore run at a cooler temperature to the engine. IE simply higher flow rate than the ICE and therefore lower temps, plus the larger radiator would cool the coolant more.

For water injection I'm saying under lighter loads, not no boost, the engine would run fine without water injection. Theb like you say Shnazzle, under high power demand the water injection would kick in to provide cooling capacity to allow making more power.

Ok I need to figure out some math to determine the feasibility of either route.

[lamcote]

How do you get the ICE coolant and the charge cooler coolant flowing through the same radiator at the same time but each at different temperatures?
Also, you seem to be proposing to have exactly the same system as a normal charge cooler setup (including pipes, pumps, coolant etc) so everything other than using a separate charge cooler radiator is the same as usual. But since the ICE radiator has to be bigger and therefore heavier than normal, the benefit of sharing the radiator is presumably pretty minimal anyway isn't it? Is it worth the effort?

If you are using a Roots supercharger you can only have full boost (supercharger engaged) or no boost (supercharger bypassed), there isn't really a light load low boost option (ignoring low rev leakage) as you get with a turbo.

[MrT]

Hi Lamcote

I'm not saying they flow through the rad at different temperatures, I'm saying the coolant will be colder entering the engine now due to the larger radiator but the thermostat will manage that. The engine will also likely run warmer due to forced induction etc so benefits from cooler coolant.
The big factor that is different is that the engine coolant flow rate is room dependent however the charge cooler will be independently controlled, either variable by ECU control, or full speed all the time so majority of the time very effectively cooling because the ratio of coolant flow Vs charge air flow will be greater during light/normal running and during high power running the fuelling will richen up anyway to deliver performance but also help mitigate higher charge air temps and inversely the higher charge air temps will facilitate greater fuel vapourisation (?) and better fuel burn.

Again, in practice this has to be proven but does the premise seem reasonable or out there?

I agree it isn't saving alot of hardware in the system but the 2 biggest challenges are getting coolant to the front of the car and having space and effective mounting & airflow for an additional radiator. Having a MR2 specific large capacity radiator and retaining stock coolant runs combats these and in majority of running conditions I believe it benefits both systems also.

I'm not planning a race engine build, it's a daily driver. So the target optimum running is not full power but low-mid power and when I stomp the throttle at 2k rpm to have the same punch (torque) as at 5k...

[Call the midlife!]

Just as an aside to the main topic, re your charge cooler water, driving home today at a steady 70-80 mph the outside ambient temp was 26 degrees or more.
Stopping at the McDonald's drive through the temp shot up to 31 plus. The probe is mounted on the front slam panel facing the inside of the front bumper.
Just wondering where you would be planning on mounting your water tank?


Sent from my iPhone using Tapatalk

[lamcote]

I'm not quite sure I understand your proposal sufficiently to be able to add any further comment.

[shnazzle]

Just re-read the above and saw a crucial point that may have set you in the path you are going down.

The thermostat does not control flow of coolant into/out of the engine. It controls the flow of coolant from the engine into the radiator.
There is a secondary circuit through the heater core that is always running and allows the water to keep moving even if the thermostat is fully closed.

On that logic; the engine wants the coolant to be 90deg. All of it.
So a massive radiator (or freezer, just to take it to the extreme to make a point) would mean that chances are the water would be too cold when the water got to the thermostat. So it would close and wait for it to reach 90.

Only option is a separate water circuit.

As for water/methanol. Well, a 70/30 mix has a definite impact on cooling, but not enough to stop you from having to severely retard timing and lose heaps of power.
So just water basically is wasting time.
Great for detonation prevention though.

I know you like breaking the mold but the coolant system  is pretty well-established

[MrT]

Hi Midlife
Ironically, I'm thinking of mounting it above the subframe or behind a hot exhaust manifold... Will have another think. I wasn't keen on losing the trunk through I don't really use it. And didn't want such a long run of pipework.

[Call the midlife!]

Hi Midlife
Ironically, I'm thinking of mounting it above the subframe or behind a hot exhaust manifold... Will have another think. I wasn't keen on losing the trunk through I don't really use it. And didn't want such a long run of pipework.

How big a tank are we talking? There's a bit of a void between the standard expansion tank and the airbox but will that be full of something else? Failing that I reckon you could mount something inside a rear wing as long as you balanced the mass either side, only problem with that is you're going to be losing mass as you use the system and have unstable fluid sloshing around.


Sent from my iPhone using Tapatalk

[MrT]

Hi Shnazzle
That's a good point, I hadn't figured how the thermostat would work with ranges of temperature like that. But theoretically with a colder coolant you'd reduce the flow to manage the cooling rate on the engine. So I was wondering about an ECU controlled flow valve to manage the flow rate rather than turn it on or off (thermostat). I agree though, the typical systems used are well proven.

Regarding water injection, the methanol acts as an octane booster for high compression running does it not? Methanol has a low vapourisation point and low Specific Heat Capacity so you are actually getting more like 90% of the cooling in a 70/30 meth mix from the water meaning 100% water will be significantly more effective at cooling the charge air.

In relation to water consumption, during light load running/ cruising it seems perfectly acceptable to me to retard timing and running conservatively without water injection since the power requirements and risks are well within safe margins.

PS, my Mercedes-Eaton M45 supercharger has a servo electric bypass valve to manage boost etc so give even more tuning flexibility. I'm considering these things alongside either a standalone ECU or piggyback like DET3 that had a number of extra control outputs for servos etc to allow running the bypass valve, water injection pump control and possibly fancy coolant flow valve or fuel pressure regulator.

There's a lot of factors to consider.

I'm considering a supercharger mounting location that would allow me space to try the installation without a charge cooler but be able to fit one if the tuning and performance aren't working out.

I'm thinking aloud so to speak to bounce ideas off you good folks and take advantage of your experience and knowledge so forgive me. No point making someone else's mistakes all over again...

[MrT]

Hi Midlife
Ironically, I'm thinking of mounting it above the subframe or behind a hot exhaust manifold... Will have another think. I wasn't keen on losing the trunk through I don't really use it. And didn't want such a long run of pipework.

How big a tank are we talking? There's a bit of a void between the standard expansion tank and the airbox but will that be full of something else? Failing that I reckon you could mount something inside a rear wing as long as you balanced the mass either side, only problem with that is you're going to be losing mass as you use the system and have unstable fluid sloshing around.


Sent from my iPhone using Tapatalk

No I want a custom back box that hangs low and doubles as a diffuser so I'll have a lot of space above that and carry up to 40ltrs, same as fuel tank. Well it's a theory anyway.

[Call the midlife!]

Hi Midlife
Ironically, I'm thinking of mounting it above the subframe or behind a hot exhaust manifold... Will have another think. I wasn't keen on losing the trunk through I don't really use it. And didn't want such a long run of pipework.

How big a tank are we talking? There's a bit of a void between the standard expansion tank and the airbox but will that be full of something else? Failing that I reckon you could mount something inside a rear wing as long as you balanced the mass either side, only problem with that is you're going to be losing mass as you use the system and have unstable fluid sloshing around.


Sent from my iPhone using Tapatalk

No I want a custom back box that hangs low and doubles as a diffuser so I'll have a lot of space above that and carry up to 40ltrs, same as fuel tank. Well it's a theory anyway.

40kgs of weight hanging out behind the rear wheels, that'll make for some interesting handling characteristics [emoji38]


Sent from my iPhone using Tapatalk

[MrT]

No no. I could fit a tank between the headers and the triangular bracing in front of the silencer. Like I said, above the subframe. A tall, possibly triangular tank quite skinny.

[Call the midlife!]

No no. I could fit a tank between the headers and the triangular bracing in front of the silencer. Like I said, above the subframe. A tall, possibly triangular tank quite skinny.

A 40 ltr tank is a fair old size, together with the associated pump etc, I'll be impressed to see you squeeze it in that space. [emoji1303]


Sent from my iPhone using Tapatalk

[MrT]

Oh that's just a foolish guess and blind hope.

[Call the midlife!]

Oh that's just a foolish guess and blind hope.

I'm sure it'll all be much clearer for you once you start your build or at least build your list of parts and equipment, then you can start mocking up with cardboard boxes etc to get a rough idea of available space.


Sent from my iPhone using Tapatalk

[1979scotte]

Give you your due Tyler you are think outside of the box with this.

[shnazzle]

There's unfortunately more cleverness to the coolant system.

The radiator used for engine coolant is built to maintain water temps from 100-110 down to 85-95.
It's not as effective at all in maintaining water temps from 60 to say 30, for a charge cooler.
The difference in input temperature between engine and charge cooler water is massive.

Now imagine the best case scenario;
1) water output from rad is 90, as it needs to be for the engine.
2) you have a second radiator after the engine rad to feed colder water to the charge cooler.

At optimum engine temps, the input to the charge cooler rad is always going to be 90. You'd use a 2nd thermostat so that the chargexooler uses the heat exchanger to continue flow before the engine is warm.

So, the minimum input to the charge cooler rad is 90. Best case scenario when the engine is warm. So the charge cooler rad is seeing water about 50deg warmer than it normally would. Also its using coolant, which dissipates less heat than water (used in charge cooler).
That means the charge cooler is circulating water at about 50deg at best. Maybe 40.
That makes it far too hot to cool air coming through it from 70+ to 30 tops.

Does that make sense?

A separate circuit is far far easier, more efficient, less maintenance, and if you get a leak in your rad you don't end up melting your engine

As for water/meth. Remember that it's sole purpose is to cool and prevent detonation. The water increases the density of the mixture. Any gains made by cooling are offset by the increase in non-oxygen density.
This is why you actually need to run more boost on water meth. It's a give/take.

[MrT]

Hi Patrick
I'm afraid I don't agree with your understanding of the cooling system of the ICE, but perhaps it's terminology. The outlet temperature of the ice coolant is maintained at the target temperature of ~89°C but to maintain that the input temperature is significantly lower. I'm not going to disagree that the input temperature would still be relatively high for a charge cooler potentially but this is why I'm proposing a larger more effective radiator. There is no temperature management on the radiator, the coolant pump is engine speed dependent so only proportionally increases cooking effect by flowing more coolant through the radiator and I believe the thermostat is not an on/off switch but expands with temperature to meter the flow rate proportional to engine operating temperature.

Therefore decreasing the temperature of teh coolant should only cause the thermostat to reduce the flow rate through the engine in order to manage the cooling effect in the engine and maintain the target engine running temperature required.

By running a separate parallel pump through the charge cooler I can push most of the coolant through the cooler (allowing the engine to only draw how much coolant is required depending on engine speed) and therefore linked to already greater cooling efficiency of the aftermarket radiator, achieve a lower running temperature in the charge cooler than engine.

On the subject of water, I believe the added volume of water is negligible in comparison to the significant heat absorption the water provides, the SHC of water being very high even compared to methanol. The reason higher boost isnrun is specifically because the cooling is so great that it is possible to boost significantly more, as boost creates heat no matter how you achieve boost. An N/A engine will benefit significantly from light water injection as you can advance timing and run a bit leaner and the water expansion when it is super-heated by the combustion also offers a small boost in performance.

But I don't want to argue with you sir. My knowledge is still lacking and so I will do some more research and feed back what I find.

[lamcote]

The input temperature, as you refer to it is only going to be about 6 degrees C cooler than the output ie about 83 degrees, way too hot for a charge cooler. Any greater temperature differential would run the risk of warping the block due to constant differential thermal expansion/contraction across it.

"As a general rule, cooling systems are designed to operate with a coolant temperature of about 190 degrees F at the radiator inlet and have about a 10 degree F temperature drop through the radiator at rated power and rated coolant flow. This will result in a bottom tank temperature of 180 degrees F."
Engine Basics website quote.

The thermostat has no impact on coolant flow rate in the engine whatsoever, it simply directs the coolant flow into/away from the radiator exactly as shnazzle described. The speed of flow does vary but the speed depends entirely on the water pump, however the engine has the same quantity of water flowing round it all the time, otherwise the water galleries would get air bubbles in them. Lots of YouTube videos show all this very clearly and should give you all the information you need.

[lamcote]

Give you your due Tyler you are think outside of the box with this.

The point is that the ICE has now had over 100 years of development by some of the cleverest engineers in the world and they tend to be pretty good at thinking outside the box themselves, so to come up with something worthwhile, that hasn't been thought about already and actually works is not too easy.

If it's beneficial and it works, the chances are you already have it/know about it.

[lamcote]

After giving a lot of thought to this I think I am beginning to understand your idea. I think you are saying you want to cool the ICE coolant down sufficiently to a temperature that would mean it was then suitable to be used as the charge cooler coolant and you will then manage the ICE cooling system to mean that the engine can cope with the lower coolant temperature?
If so, I would suggest that the coolant will need to be coming out of the radiator at between 30-40 degrees C in order to be any use as a charge cooler ie to try to get the inlet air down to somewhere near a normal ambient temperature.
This means that you will need a radiator capable of reducing the coolant from 90 degrees C to say 35 degrees C. Since the normal radiator only reduces the temperature by about 6 degrees C, I would think any radiator capable of reducing the temperature by a further c.50 degrees C is going to be the size of a sail on a small boat. You are then going to have to swap the ICE water pump for one that pumps the water incredibly slowly in order to allow the coolant to heat back up to 90 degrees C whilst it travels round the engine so you will then have to deal with the fact that half of the engine is constantly running far too cold for the oil to work effectively, and risk possibly warping the block due to the significant temperature differential between the water input and output.
Maybe I still haven't got it, but if this is the correct understanding, it seems to me to be an awful lot of trouble just to avoid fitting a charge cooler radiator?