Padlock

This has been an enjoyable read as a OEM powertrain cooling system design engineer.... lol

*sips coffee*

please continue....

Arca_ex

iTrader: (2)

You're right. I've seen the errors in my ways. He should totally do it.

Gee Emm

Like I said:

Do it.

Document it.

Everybody learns something.

HalalBuilt

*pours some more coffee*

Spill. What's the word?

RASKAR

Yeah do your research first... That's how you avoid all of... this. But yeah, you should definitely do the opposite of what everyone has been saying, we don't know anything here

Padlock

These are some VERY high level statements and none of this should be surprising.... There's ALWAYS exceptions to the rule depending on exact application, but here we go:

I'll start by saying, A2W and A2A can both work very well if designed properly. You are naïve to be thinking otherwise. From an OEM business's perspective, costs matter. From a styling designer's perspective, looks matter. From a racer's perspective, weight matters. From a mechanic's perspective, simplicity matters. From an engineer's perspective, packaging space, proper layouts, and effective heat transfer all matter. It's essentially impossible to win at all of these, and there are more stakeholders than what I've just mentioned. The compromise is real...

Speaking solely from my engineer brain after working with multiple OEM sponsored motorsports teams, curly hit the nail pretty firmly on the head in his initial post. A2W systems are heavier than their A2A equivalents, they are more complex (more failure points), and they are more expensive. In the racing/aftermarket world, simplicity/weight/costs win nearly every time, so A2A is always my (and their) first push.

Where A2W systems really shine is in applications when I have no space to easily package A2A systems. Think of just about any OEM A2W system you can think of? I'd bet all of my MT poscats that those OEM's landed on that design because they ran out of space in some way, shape, or form for a variety of reasons. I'll use the CTSV example just because I'm familiar with it personally and it was mentioned above. There's no reasonable way to design an A2A system on a roots style blower like that one. Just think of how complicated of a system that would be to have compressed air coming out of a blower mounted in the "V" of a V8, running large diameter piping to the front of the car to a large intercooler (which there isn't really space for without other compromises), and then running that same size piping back to the center of the V to stuff into the cylinders. Remember, GM also would ideally want to reuse as many parts from the base model trim package CTS as possible to keep MSRP down. The A2W system that GM implemented on that vehicle makes PERFECT sense for what it was designed to do, and it does perform very well at OEM power (heat) levels.

A2W systems are not remotely close to as easy to DIY as an A2A system. You are transferring heat twice as you are going from ambient air to water, then that same water to intake air, so from a thermal efficiency standpoint A2A wins. With A2W, you have 2 heat exchangers to package/size and a water pump to spec out. If you aren't getting the performance you want out of an A2W system, without fancy heat transfer software and understood performance curves you are really stuck estimating what variable will help fix your problem. Examples: Are your water lines too small/restrictive for your pump? Is your pump too small and that hurting flow? Did you put too large pump on it and actually hurt flow? What air fin densities should I run on each heat exchanger? What heat exchanger is limiting my performance more and is it because of water restriction or air side restriction or just the sheer size of it? Oh, and if you oversize any of these to "kill it with fire", you are just needlessly adding more weight and costs to your system for no reason

So the question to the OP is, if you have package space for A2A (which you do), why not just keep it simple? If your reason is "to be different", then go for it and let us know how things work out. I'd even be willing to provide some level of advice, but I wouldn't personally be heading down that path myself due to the unnecessary complexities and tradeoffs associated to it.

Ted75zcar

iTrader: (2)

If/when I ever develop a gen 3 of my design (unlikely, but not impossible), it is highly probable it will incorporate at least 1 A/W circuit, if not 2.

Dealing in absolutes with complex systems is a losing approach.

Sith

My A-W on my NC. Daily driven. 100* IAT turn into 150* when doing a WOT 1st- 3rd pull when ambient is 95*+ and boost is around 15lbs. It's only about a half gallon of fluid. Definitely needs an ice tank/reservoir if I was going to track it.

Fireindc

iTrader: (13)

Cool post, thanks for sharing.

I just gotta say, I loved reading this entire thread even though I have zero interest in running an A2W setup (thanks for the bump, I didn't see it until just now).

Mt.net has always been such a great place, where I can casually bounce cooling ideas off a friggin' HD powertrain cooling engineer. Or bounce suspension ideas off some of the top miata suspension developers, etc.

Keep on keepin on guys.

Gee Emm

Is that a radiator for the AWIC?

Padlock

I just want to point out here that an ice tank or additional reservoir won't really fix your issue here for track use, unless you are referring to a very brief burst like in drag racing. I had an ice tank on my modded CTSV. Mind you, that car was pushing 820whp on a cooling system largely designed to manage 500whp, so it was taxed, but it wasn't uncommon to see 80* or more temp increases from IAT1 to IAT2 during a simple 10 second quarter mile pull if I wasn't using ice in the tank. When using the tank with ice, it was possible to get below ambient IAT2 temps for a couple runs (until the ice melted). Forget about even considering to bring that on a road course... it wouldn't survive as the ice would simply melt out quickly (but it wasn't built for that). My point being here is that ice and extra water capacity is only a temporary solution that extends the time before you will have the same problem that your current smaller water capacity system has.

If you are seeing 100* turn to 150* IAT's in such a short time that's telling me that your heat transfer capability of the system isn't high enough. If you monitor the temp of the water and IATs during a log, it should help paint the picture to where you are at a shortcoming for your system. If IAT is climbing rapidly and water temp is climbing rapidly with it, that points towards the front mounted water to ambient air heat exchanger not being utilized well enough (ducting of air into it maybe?), the size of it being too small in any of the 3 directions, or you not pumping water fast enough which could be due to line size or pump sizing/selection. If IAT is climbing rapidly and water temp is climbing at a more gradual/slow slope, then that points towards the water to charge air heat exchanger not being big enough. If neither IAT nor water temp is climbing rapidly then your system is likely sized big enough, but now you can question if it's too big/heavy/expensive for what you really NEED.

This is really just a balance of absorbing heat and then dissipating heat with the water at the same rates. The ideal point to be at is where water can dissipate all the heat it absorbs so the water temp stays fairly constant, but achieving that can be easier said than done without a lot of experimenting.

Sith

Yes, a heat exchanger if you will.

Sith

Did your CTS-V have a larger heat exchanger?
Did it have an upgraded pump?
What was the capacity of the system?

We have a CTS-V here too! We upgraded the pump, 3 gallon reservoir, bigger heat exchanger, increased line size, and added a controller. But it too is a daily with minimal drag passes.

On my setup for street driving, 50* rise in temps on a full pull in high ambient is more the fine. Given my capacity is only half a gallon I have no data on how well the exchanger is really working. I'd assume the coolant doesn't spend a lot of time in the heat exchanger itself. I also don't monitor that coolant temp. I agree that more data would determine what is needed for track use. My original statement was not an end all for what works. Just a statement of how mine works with hopes toward future use.

I went A/W for the way I could package the system in the car, and to not block airflow to the radiator in the same way a full size intercooler does.

OptionXIII

In a previous life I was an engine calibrator on one of the generic 2.0L turbo 4s that have taken over econboxes. The radiator for the A2W intercooler system had the same face area as the radiator (not the same thickness though) and was just another heat exchanger in the stack. Another nice feature about A2W intercoolers is that by controlling the pump speed, they give you a lot more control over the charge temps. For cruising in normal ambient temps, we'd have AITs in the mid-40s. Go WOT, and the pump picks up up to give as much intercooling as possible. Winter in the Upper Peninsula? Of course the pump will work less instead of constantly giving you full blast cooling like an A2A cooler out in the airstream would be doing. That's not really a big deal for Miata homebrew performance focused turbo setups that don't expect -40 starts or care about fuel economy and emissions, but it's great to have that additional layer of control for fuel economy and emissions for an OEM.

Sith

Ha that's awesome. Knowing actual flow and being able to control pump speed would be nice. But yes, maybe too much for these generic setups.

Padlock

Most of the cooling system work on the car was done by the previous owner that I bought it from. There was a Vadder dual HX kit put on the front, ice tank in the trunk, and I forget the size/brand pump now that I'm trying to think back 6 years ago. I never measured it prior to the call being totaled (thanks hit and run drivers), but capacity was quite high for the system. All I can say with confidence was, the limitation felt like it was in the size of the single heat exchanger on the blower itself. Looking at hellcat blowers, you can note they run dual heat exchangers on the superchargers (I used to work in design at the company that makes them). Dodge did that for good reason to gain heat transfer capacity, but it's also why that blower is a lot physically larger.

Spending a lot of time in a heat exchanger isn't necessarily a good thing. Heat transfer is a function of mass flow and temperature differential between mediums. The slower your flow is, the more time it has to transfer heat, but then your temp difference between mediums lessens greatly therefore negatively affecting heat transfer. In general, the more flow you get, the more heat transfer you will see, so maximizing both air and water mass flow rates is always a priority. Increasing water volume only helps dampen water temp spikes during transient operating circumstances. I don't see this being an issue on OE systems I've designed so far, so most of the time I push to minimize fluid volume to save money and weight.

OptionXIII mentioned it above, but if you design a heat exchanger that is very thin and large from a length/width perspective, the air side restriction of the heat exchanger is quite low and therefore the impacts to the radiator mounted behind it are also fairly minimal. You see this "stacked" configuration a lot on the OEM side for this reason.

Padlock

Just because of relevancy and cool factor of engineering knowledge dumps, here is the cooling package I designed with Caterpillar for their latest generation of Medium Wheel Loaders (1 of 6 different packages I was working on simultaneously over a 5 year span). From top to bottom, there is a hydraulic oil cooler on top, air conditioning condenser in the middle, and diesel fuel cooler on the bottom. Behind them are a radiator for engine coolant, and intercooler for the charge air system. The size (length, width, and thickness) for all the heat exchangers shown were all selected based on airflow performance curves provided on the hydraulically driven fan which you cant see behind all of this. If you make the hydraulic cooler taller to increase its performance, you have to make the condenser shorter and give up its performance... if you make the oil cooler too thick, more airflow will go around the oil cooler and choose to go through the condenser instead.. air always follows the path of least resistance, so understanding pressure drops through all of the components is key. You can't change one part without affecting another given airflow is a constant. So many iterations.... but damn is it cool (pun intended?) seeing them out in the world now..

Sith

One thing we didn't like about the heat exchanger inside the CTS-V supercharger was that it was dual pass. I know of someone trying to create a single pass style but it also requires a whole new lid.

So, instead of going for larger coolant volume I should start looking at wider heat exchangers? I don't mind that idea at all.

OptionXIII

Like Padlock point out, more volume or starting from a lower temperature just gives you a tiny bit more time before you're too hot, and that's also adding weight. Not a way to go about it if you want to turn laps at a road course. It's the same as trying to fix overheating by going with a lower temperature thermostat.

Padlock

I'd be curious to understand what his logic is in trying to make a single pass core... Unless he's also increasing the size of the core dimensionally, I wouldn't predict that the change from dual pass to single pass would be worth the effort involved. I put a dual pass radiator on my kmiata to dramatically clean up the hose routing as I now don't have an ugly crossover hose that many kmiatas have. When sizing it, I kept the finned area very close to the OEM Mazda radiator and unsurprisingly, I've seen no real cooling differences with the change, which was the whole point. The positive aspect of a dual pass heat exchanger is that you usually can get a bit better thermal gradiant across the part, but the downside is you are forcing all of the fluid through less total passages at one given time. Therefore, it is harder for your pump to push fluid through less tubes (pressure drop is higher) and your flow rate will go down on a dual pass part when compared to an identically sized single pass part. Evenly spread thermal gradient is good, but flow rate going down is bad... what variable is driving your net heat transfer more is what you need to try and calculate. In my experiences so far, heat transfer predictions between single and dual pass parts was never notable enough to make me immediately want one over another on a coolant/water system.

Usually packaging requirements on connection locations or fluid types used drive the decision. On the Caterpillar example above, the oil cooler is single pass (you can see the hard line crossover we opted to use). This was done because thick hydraulic fluid is a pain in the butt to pump through small passages, especially when its cold. Pressure drop and flow made a big deal to heat transfer so single pass won. In comparison, the condenser is a 4-pass part. 2 phase high pressure refrigerant is relatively easy to pump through a part. Inlet passes are gas phase, and as the refrigerant cools, it turns to liquid phase by time it gets to the outlet. Having an even thermal gradient with a multi-pass part during phase changes is beneficial there.

poscat awarded for this

I just had this argument with a well known miata vendor that him trying to sell a low temp thermostat for road course use is straight up not correct.. but I'll let the internet be the internet... I can't win them all...