Mazda Ice Academy: Trial By Ice And Snow CX-5, CX-3 And MX-5

I've had a verified 32mpg highway with my 2016 CX5 AWD 2.5L with OEM roof rails and OEM mud flaps. This was on numerous 50 mile round trips.

Around town I get 25mpg

Good to hear. My car is the same without the OEM Mud Flaps, but was considering getting them. I was worried about it changing the aerodynamics and losing 1mpg or so because of it. Maybe it's silly, but I just didnt know if the benefits of having the mud flaps since they are fairly small on the OEM one were worth the mpg loss.
 
'Just because they lowered the car doesn't mean the cornering is automatically better.'

Just because you lowered the car doesn't mean the cornering is automatically better either

" Did they get an alignment? did they check oem sway bar angles after lowering (if equipped)'

"Sway bars" or more properly anti-roll bar angles aren't locating the wheels. They might bind a bit or break if the mounting position moves too far. An alignment can correct for static camber and caster but can't change the fact that you are operating in the wrong part of the camber curve. The suspension is designed for the static setting to be at the normal ride height and a lowered car shifts them to the end of the travel. If you stiffen it up so it barely moves and run on smooth surfaces it may stick pretty good. Driving such a set up on real roads is no fun.

"The stock motor is calibrated for the oem intake, even if your aftermarket intake has the correct sizing to satisfy the oem maf (if equipped) if it's not tuned to flow more air then it can't flow more air... Period."

I don't know what you mean here. Changes to intake and exhaust are attempts to tune for more airflow. If you mean to say it won't flow more air unless you run modified maps in the ECU you are all wet. Flowing more air won't mean more power unless you add fuel but the engine is an air pump. Changes to the intake tract can make it flow more air regardless of the air/fuel map. The main reason intake mod's don't add much if any horsepower is that the factory engineers have already done a pretty good job. Mostly those intake kits just make more noise.

"And while it's certainly something people do, it's not my case, and frankly it's kind of hard to do with the new gen mazdas because suspension wise, their aren't many "cheap" parts out."

You can pay a million dollars for your springs and shocks but unless you change the length's of the control arms and/or the suspension mounting points you are still running sub optimal geometry.
 
People can pull over 1g with a mazda 3 sedan and far, far less $ in suspension; but then again you are talking about a 1992 ford. People at my local SASCA chapter pull oever 1g easy, and they autox pretty much all the tuner cars you would expect, from 1989 miatas, s2000's to brand new subies evos ad the occasional audi r8.. there even a guy who autox's *the* turbocharged (he's a texas local) Mazda 5 and he get's some damn good times. I seriously doubt that the majority of them have spent 10k on their suspension. You don't even have to spend that much to sup up the suspension of an wrx for example. And about being traction limited by the mods, funny you bring that up because there is a guy on the mazda3revolution forum who tracks his 3 bigtime; he has made his own aero, and splitters (professional quality, some of which he now sells) to steadily decrease his times. And most of his time is modeled aerodynamically. I guess the point being, don't just take a show like top gar for face value because that's not exactly typical. Of course I don't need functional lips and aero because I don't have wrong wheel drive, nor do I drive at speed on a track that I would need them.
The 70mph plus slalom was what impressed me for a fox body
 
Except for me it is because my experience is almost 100kmi of daily driving vs 1 episode of top gear with a bunch of questionable variables on a different car with who knows what parts. Not only is my CG lower (I'm guessing that show had sedans/hatchbacks and not a 8-9 inch CUV) but in lowering my car I've added stiffer springs and in my case adjustable struts as well... i.e not some cheap s*** form ebay with a questionable spring rate that might be less than stock, or stock with a lower height.. They are matched for my ride height and then I can adjust the dampening as desired/needed. My car isn't significantly better at cornering simply because it's lower. Also my front camber is adjustable directly on the shock body.

Just because you lowered the car doesn't mean the cornering is automatically better either
You can still adjust your camber into spec with the appropriate part. Just like you can reset your sway bar (should and sometimes 100% have to) into the correct angle with the appropriate part. If you feel so inclined, you can adjust your toe with the appropriate part as well; something I highly doubt was done for this episode. That alone are 3 arms/link that you can get for the CX-5; " OEM upper arm" (aka camber arm) "Lateral Link" (aka toe arm, also camber adjuster), 'Stabiliser link" (aka adjustable end links).. not to mention adjustable tie rods. All of those are readily available items I could order right now from Hardrace and or Autoexe, JBR Corksport, Megan Racing etc.. And really, when you are talking aftermarket, you really are free to use whatever angles you like since the OEM angles probably aren't giving the best cornering for your new set up anyways. Having a poor alignment can affect your cornering especialyl camber and that's usually the first thing to become off when you lower. Since this was an instrumented test comparing aftermarket mods to a "proper" oem set up, how is it even reputable if they didn't at least try to set the alignment to oem specs? It's like when you look at these AWD comparison and the winning car has winter tires and the competitor is on all seasons or something. Doesn't sound to me like they were actually looking for a legit outcome aside form production value of the next episode. For roads, my car sits about the height of a mazda 6. If I wanted to I could raise it higher but I don't need to because basically, I don't take my cx-5 anywhere I couldn't pass with a sedan like a mazda6. I bought it for it's many qualities, but ground clearance wasn't one of them. If there was a mazda6 wagon with AWD here (like in Japan), I would probably be driving that. Although VERY subjective I do fine on real roads and have fun; I avoid potholes always regardless of my ride height. Feeling the road is like feeling the flight controls in that it's part of the feedback I use while deciding what to do when driving.


"Sway bars" or more properly anti-roll bar angles aren't locating the wheels. They might bind a bit or break if the mounting position moves too far. An alignment can correct for static camber and caster but can't change the fact that you are operating in the wrong part of the camber curve. The suspension is designed for the static setting to be at the normal ride height and a lowered car shifts them to the end of the travel. If you stiffen it up so it barely moves and run on smooth surfaces it may stick pretty good. Driving such a set up on real roads is no fun.

Keeping in mind this is an NA car. Also ECU controlled. On our engine, I don't think of intakes and exhausts pshhysically tuning for more airflow but more as tools to allow the ECU to be tuned for maximum airflow. Considering the OEM equipment is made for the OEM calibration with noise, reliablilty emissions and other things in mind.
Most people already know for an NA engine an intake doesn't do much by itself. I mean when you look at a dyno from the manufacturer and see a 5hp gain it's obviously not doing much. For a show like top lol *gear to feel they need to point out that 5hp is not a lot is pretty dumb and shows what they think of demographic watching their show. Of course it's not simply where the peak hp/tq was it's the change in the power curve that usually comes along with one of both of these mods. But I don't think you've ever data logged a motor before let alone one in the process of tuning. You can easily read directly from our ECU the total g/s or mass airflow. I personally have logged the 2.0 and the 2.5 and for sure, even though my g/s increased simply by adding an intake, it increased even more by the arrival of my last/latest (tune) revision.

The engine is flow limited not only physically by something like an intake but also physically by intake and exhaust timing. Our engine reads a certain volume of air, based off the dimension of the OEM MAF and determines how much fuel is needed. If you go add an intake with an incorrect sized maf, chances are you are running rich or lean. In our cars (most modern cars) this pretty much will throw a code and it will negatively affect the performance of your engine. This is not some 1975 carbureted engine we are talking about. Pretty much every part of the combustion cycle is controlled or can be controlled through the ECU on this car. There are alot of thing to manipulate to get the most out of a powerstroke (hundreds of pids) that don't necessarily involve adding more fuel (I've recorded lea conditions as much as 1:30 if I remember correctly).

I don't know what you mean here. Changes to intake and exhaust are attempts to tune for more airflow. If you mean to say it won't flow more air unless you run modified maps in the ECU you are all wet. Flowing more air won't mean more power unless you add fuel but the engine is an air pump. Changes to the intake tract can make it flow more air regardless of the air/fuel map. The main reason intake mod's don't add much if any horsepower is that the factory engineers have already done a pretty good job. Mostly those intake kits just make more noise.

Fact is, there is so much information available on the many mazda forums and outlets stating otherwise.. and considering how many mazdas are actually competition raced... Pretty much every push/pull arm on the suspension can be swapped as needed for an adjustable version. The LCA is just one part of the suspension, and one that is rarely changed out.. in fact he term LCA (lower control arm?) applies to a different assembly on a WRX for example than it does on our car.. I've seen an aftermarket LCA for the mazda 2 (which is a winning bspec racer BTW) and it's roll adjustable just sligthly..They aren't even allowed to use it and they race for cash money. That's not even a part people usually touch so if you want to say technically the set up is subpar, ok, but that certainly doesn't mean that everything else is moot and somehow automatically gives less performance than OEM. I'm sure if you go to a local sports car club of america autox the cars getting the best times do not have a stock suspension, tune or intake/exhaust. If so you have a weak showing in your area lol.

You can pay a million dollars for your springs and shocks but unless you change the length's of the control arms and/or the suspension mounting points you are still running sub optimal geometry.

Bottom line is, a questionable test on a questionable TV show is trumped by real world feedback and racing and personally, over 100000 miles in AWD cx-5's with said mods. I pay to get the max performance from my vehicle, anything less is removed and resold, or modified or if needed another part purchased to complete it.
 
Last edited:
I pay to get the max performance from my vehicle, anything less is removed and resold, or modified or if needed another part purchased to complete it.

I think I understand. Your heavily modded CX-5 could beat a Toyota Prius. (rofl2)

But what's that have to do with the Mazda Ice Academy and the outstanding snow and ice performance of the CX-5? (dunno)
 
I think I understand. Your heavily modded CX-5 could beat a Toyota Prius. (rofl2)

But what's that have to do with the Mazda Ice Academy and the outstanding snow and ice performance of the CX-5? (dunno)

I think I understand. Your heavily stock CX-5 could match a Toyota Prius (loser).
It has nothing to do with op but rather it was a response to another off topic post. (hijack)
 
I think I understand. Your heavily stock CX-5 could match a Toyota Prius (loser).
It has nothing to do with op but rather it was a response to another off topic post. (hijack)

OK done hijacking, However, "My CX-5 will run rings around everything else here" wasn't exactly on topic either. ;)
 
OK done hijacking, However, "My CX-5 will run rings around everything else here" wasn't exactly on topic either. ;)

Wait a moment. Are you saying we should stay on topic. Wow. I hadn't observed that before. I think I am in trouble. Discussing on / off topic is off topic. Gulp. Grin.
 
impressive. Trying to judge by the video, and it seems to me that all vehicles had the same tires. Looked like Michelin X-ice 2 to me...

I wish I had known about them before I bought tires, or I would have bought those. However, live/learn. I plan to see how well/bad my current LX20's do if it ever snows this year here...
 
impressive. Trying to judge by the video, and it seems to me that all vehicles had the same tires. Looked like Michelin X-ice 2 to me...

They are all on Bridgestone Blizzak WS80 tires. And that's what makes this test a fair comparison of the various AWD systems.
 
The woman makes some good human arguments why winter tires are a good investment vs. "getting by" with all-seasons.
Snow tires are like a 5-point racing harness where I live, but some good 40K mile tread-life "compromise" tires like the Michelin's do seem a good investment.
 
Snow tires are like a 5-point racing harness where I live, but some good 40K mile tread-life "compromise" tires like the Michelin's do seem a good investment.

A strong case could be made that AWD is like driver training wheels and winter tires are like airbags. I've driven to some pretty amazing places with FWD and good tires (summer and winter). Driver skill almost always negates the need for AWD but good tires are additive to driver skill.

That said, I've taken my CX-5 places both summer and winter that would not have been possible with FWD (even with good tires). But they are few and far between.
 
Can someone enlighten me what's going on in the most recent video with three SUVs climbing the snow hill from a stop (https://www.youtube.com/watch?v=bLnFHKZLApk)? In the Subaru promotions with rollers/water on steel ramps (https://www.youtube.com/watch?v=RCuvwYd9JuE), all cars showed all of their wheels spun trying to get up the ramp. But in this video, the Subaru and Honda had their front wheels slipping like crazy, yet the rear wheels didn't even kick in.
 
Can someone enlighten me what's going on in the most recent video with three SUVs climbing the snow hill from a stop (https://www.youtube.com/watch?v=bLnFHKZLApk)? In the Subaru promotions with rollers/water on steel ramps (https://www.youtube.com/watch?v=RCuvwYd9JuE), all cars showed all of their wheels spun trying to get up the ramp. But in this video, the Subaru and Honda had their front wheels slipping like crazy, yet the rear wheels didn't even kick in.

The mazda video turned the wheels and most roller videos are done with the wheels straight.

AWD systems have a hard time dealing with "tight corner braking" and they have to reduce the rear wheel torque when the car is trying to go around a tight turn. Mazda likely put a extra line or two of code in their logic..
If(car is stopped and wheels are turned)
do not worry about "tight corner braking"
else
operate like every other AWD system

This PDF for the 2006 rav4 (system pretty much identical to CX-5) talks about it at little:
http://rav4world.com/pdf/2006/2006_4wdsystem.pdf
 
Last edited:
AWD systems have a hard time dealing with "tight corner braking" and they have to reduce the rear wheel torque when the car is trying to go around a tight turn.

Thanks much for the explanation, I was scratching my head when I saw the first video a few days back, why did the drivers turn their front wheels before going over the hill.

At least Mazda was honest in their talk in one of the video that their hardware is not much different from others, it's the software that is smarter.

From a software engineering point of view, with 27 sensors, I wonder if they just have rules/decision-tree deducted from a huge pile of collected data and desired behaviors, or if there's some neural network involved (though unlikely as AI is less predictable for fine-tuning).
 

New Threads and Articles

Back