Theory: CVT WOT Power, and how Honda "trick" EPA
 

In case some of you guys who doesn't know, you don't get maximum power on the CVT Fit when you floor the gas pedal(regardless which mode you at.) You can tell easily as the engine rarely rev pass 6000RPM, and the peak power is 6500RPM. If simply floor the pedal, you at ~5000RPM, about 115HP effective. You only get close to maximum power if you repeatedly double tap the gas peddle as you accelerate(first pioneer by Myxalplyx here). The difference is about 1.5-3 second in 0-60 times(HUGE difference).

I been wondering why Honda choose to cripple their CVT this way, now I think I found the answer. According to EPA high speed cycle testing rule, "SFTP US06 is a high speed/quick acceleration loop that lasts 10 minutes, covers 8 miles (13 km), averages 48 mph (77 km/h) and reaches a top speed of 80 mph (130 km/h). Four stops are included, and brisk acceleration maximizes at a rate of 8.46 mph (13.62 km/h) per second. The engine begins warm and air conditioning is not used. Ambient temperature varies between 68 °F (20 °C) to 86 °F (30 °C)."
Basically, EPA accounts for hard acceleration gas consumption at their highway mileage rating.
FTP-75 - Wikipedia, the free encyclopedia

Of course, the Fit cannot accelerate at 8.46mph/sec(equal to 7sec 0-60), so the tester will likely just floor the car during testing. Guess what, Honda artificially limited the power output during WOT, so it would appear to EPA that the CVT uses LESS gas during supposedly "maximum" acceleration.

Bottom line, the Honda fit EPA mileage rating is legit, but the 130HP on paper is rarely, if at all, realized for a Fit CVT driver. I think that is abit of "cheating" and set a bad example. In the future, manufacturer will sell you a 200HP car, but limit the power during WOT to 100HP, claiming that they are selling a 50MPG car with 200HP. Well, I think I exaggerated abit there :).

Addendum
If you practice the "double tap" technique alot(harder than it sound if you turning and looking for traffic at the same time), the Fit CVT is a 7 second car. See video below from Myxalplyx:
www.youtube.com/watch?v=6cT-j8l1Klc

I haven't seen any dyno charts for the '15, but every chart I've seen for the '13 has the power peak of 95HP (at the wheels) between 5500 and 6000RPM.

Honda's specification for the '13 is 117HP at 6600RPM. The real-world dyno charts don't seem to have their peaks where Honda implies that they should be!

I don't doubt that the power peak for the '15 is at a similarly lower RPM versus the advertised specification.

So, is Honda fooling us? Perhaps or perhaps not. It is possible that the transmission is less efficient at high input RPMs which would move the wheel power peak lower than the flywheel power peak. In that case it might actually be better to have the engine sit at 6000RPM under WOT rather than at redline.

It would be instructive to put a '15 Fit on a dyno to see what the power curve truly is. Not my car, though!

The Fit can't average 8.46 mph/sec going from 0 to 60, but at lower speeds it certainly should be able to accelerate at that rate. Acceleration is not linear with speed as friction (mainly air resistance) increases dramatically at higher speeds. If you're accelerating 0-120mph, it will take a lot more than twice the 0-60mph time.

Edmunds lists a 3.5 second 0-30mph time (with the CVT).

Not for very long. 8.46MPH/sec (a ridiculous unit, pretty much to be expected from a government agency!) is equivalent to 3.85m/s^2 or 0.40g A Fit might be able to sustain this out of the hole, but likely not beyond 20MPH. In a FWD car, the wheels will spin at about 0.5g.

Just tried this today when I went to work and around locations (one of two IT staff for a 700 user Clinic with 6 buildings in 3 towns with an interstate between each!) this trick made my lil fit fly up to 80 way before the onramps end, It totally changes the feel of the car!

Every 3rd gen owner needs to know this!


Now, the next step is finding out exactly what sensor is telling the computer what to do, to produce the (bursts of power)- and over-ride it for permanent use if possible :D

I don't really know if it would even be advisable to do this after the warranty expires...

I'm guessing it's something the ECU is doing with the transmission that allows the quicker acceleration (and the motor able to jump rpm so quickly). maybe the transmission instantly hits the clutch as soon as the pedal is released, while the engine can spin up the rpms freely- as brief as it is, and the tranny just picks up where it needs to to keep up when it is engaged again at the now-more aggressive rpm (in the space of a few milliseconds)

anyone?

Honda may not have chosen to "cripple" the CVT in the manner you mentioned as that has been illegal in EPA testing for some years now. Previously there were tricks played in calibration to pass certain tests. Drive-by noise was one of them. Honda likely calibrated the CVT in the manner they did to optimize end-user fuel economy.

Regarding the EPA test cycles and the quoted acceleration rates, something seemed a bit off to me so I dug a little. If you take a look at the US06 test cycle (the one that the 8.46mph/sec is mentioned for), there is only one section that the acceleration is that harsh and it is for 1 second. From 49 to 50 seconds in the cycle the vehicle accelerates from 0.8mph to 9.2mph, there are a couple other 1 to 2 second sprints at 5-6mph/sec, but all under 30mph.

Interestingly also if you take a look at the reported data (EPA Test Car List Data) Honda does list different engines/transmission combinations that were tested. The EPA also replicated on of the tests (lines 2218-2229 of the 2015 excel sheet).

Just a few interesting facts.......also makes me even happier to drive a manual!

Test cycle data is here: Dynamometer Drive Schedules | Testing & Measuring Emissions | US EPA

So hiding your engine's power behind a torque converter and not having real control over the transmission's gearing is a bad thing, and this is news?

Kudos to Myxalplyx for the workaround, that's some great work finessing the black box to ge it to do what you want. But "Hey, it's less crippled now!" isn't all that rousing.

I'm still waiting for some 6MT owner to put their Fit on a dyno to see where the wheel power peak actually is. Experience with earlier generations shows wheel HP actually peaks between 5500-6000 even though the advertised peak is 6500.

It may be that the double-tap is altering parameters other than transmission ratios. It might be providing a better FAR for power, for example.

For those of you who think it's illegal for Honda "trick" EPA this way, you are wrong.
EPA only cares how much gas it consume when you floor it. EPA does not care at all rather your only has 50hp effective or your car accelerate 0-60 in 20 second.

Also, a aggressively programmed CVT beats a MT, or any transmission for that matter, hands down. This is simply because you are at peak power 100% of the time. There is a reason why F1 banned the use of CVT.

This is something I have been trying to capitalize on heavily. Just trying to put the right combination of things together to make the car shine. Thanks for all your help with the acceleration suggestion in 'L' gear. It's totally usable when I know I need to get on a highway from a dead stop.

I've been watching a lot of Civic Si's at the dragstrip with your average drivers doing their thing. They are ranging from 14.8 to 15.8 in the 1/4 mile. With minimal modifications, a CVT is able to hit 15.5-15.8. I don't think it's all about the horsepower since my trap speeds are much less that the Si's but how quickly the car is accelerating up to speed. I honestly believe since it is able to sustain peak hp, it enables the car to get from point A to point B as quickly (Not as fast) as cars like the Si or the Type R (without professional drivers). This car is a breath of fresh air.

This CVT transmission is a special kind of quiet beast. Once people start to realize it's potential, they will be flocking to it. As long as it becomes more reliable. I have been doing a LOT of testing this cars acceleration. Lots of personal videos and a good deal of timeslips. I am truly loving this transmission so far. https://imagizer.imageshack.us/v2/87...540/tR1tOj.jpg


It's key weak point is acceleration from a dead stop. You can turn this to the car's asset if you have just enough power to not break the front tires but keep it accelerating strongly. Once this is achieved and you hit peak horsepower at high rpms, hp for hp with another transmission, it kicks butt.

You guys are aware honda dropped the peak power number for the fit. Same reason the 2nd gen fit makes 120hp to our 117. All in the name of fuel quality in USA. Those number of gain look wimpy but the mid range gains are huge. Which is where cvt revs Regular in JDM/EDM is 95 octane. If u want the correct performance of JDM/EDM cars, run 93. I wonder if they ran a hotter plug in the 3rs gen like they did in 1st/2nd. If so, running a cooler plug with 91/93 would bring u to the real numbers of the jdm model.
.

so for us non-tech types, this thread is all about how Honda might be intentionally crippling CVT and/or engine power potential in order to look good in EPA stats, correct?

That is pretty much it, but thread drift happens.

Myxalplyx, are all those timeslips from your new Fit? If so, you've answered the "how reliable is the CVT?" question!

Yes they are George! It's the many more 'Undocumented Runs' off the track that would probably answer that question though. :rotfl:
If I see the EPA though, I leave the transmission gear in 'D'.

Seriously, this is very interesting information. I'm looking forward to learning and reading up more about it. Thanks for bringing this to our attention Eigenmeat.

Re: Bach
"so for us non-tech types, this thread is all about how Honda might be intentionally crippling CVT and/or engine power potential in order to look good in EPA stats, correct?"

Yes, the EPA will only leave it in D and floor, so it would appear that the Fit use less gas on hard acceleration. Of course, the flooring a FIT in D produce no where near its maximum power potential, but EPA literally give two shit about your car has 0-60 time of 2 minutes.

Not really sure why so much talk about the EPA. The EPA let's manufacturers self-certify. Meaning the OEM submits all the test data, method and application to the EPA for approval. The EPA may also repeat one of the tests, but not always. From what I saw in the EPA docs (link in my post above, tried to paste below), of the 12 tests, the EPA ran 1, the rest were submitted by Honda.



Model Year Vehicle Manufacturer Name Represented Test Veh Model Test Vehicle ID Rated Horsepower Engine Code Tested Transmission Type Code Tested Transmission Type # of Gears Transmission Lockup? Equivalent Test Weight (lbs.) Axle Ratio N/V Ratio Test Number Test Originator Analytically Derived FE? Test Category THC (g/mi) CO (g/mi) CO2 (g/mi) NOx (g/mi) PM (g/mi) CH4 (g/mi) N2O (g/mi) RND_ADJ_FE FE_UNIT FE Bag 1 FE Bag 2 FE Bag 3
2015 Honda FIT EX EFR71C 128 FR72C1 SCV Selectable Continuously Variable (e.g. CVT with paddles) 7 Y 2875 5.44 32.3 FHNX10028933 MFR No FTP 0.01418 0.192 208.26 0.0056 0.00333 42.6 MPG 40.3 42.4 44.8
2015 Honda FIT EX EFR71C 128 FR72C1 SCV Selectable Continuously Variable (e.g. CVT with paddles) 7 Y 2875 5.44 32.3 FHNX10028934 MFR No HWY 0.00421 0.035 161.95 0.0023 0.0015 54.7 MPG
2015 Honda FIT EX EFR71C 128 FR72C1 SCV Selectable Continuously Variable (e.g. CVT with paddles) 7 Y 2875 5.44 32.3 FHNX10028935 MFR No FTP 0.01325 0.188 208.17 0.0051 0.00291 42.6 MPG 40 42.4 45
2015 Honda FIT EX EFR71C 128 FR72C1 SCV Selectable Continuously Variable (e.g. CVT with paddles) 7 Y 2875 5.44 32.3 FHNX10028936 MFR No HWY 0.00118 0.023 161.54 0.0011 0.00054 54.8 MPG
2015 Honda FIT EX EFR71M 128 FR71M1 M Manual 6 N 2875 4.62 48.9 FHNX10028931 MFR No FTP 0.01398 0.322 236.26 0.005 0.00334 37.5 MPG 38.1 35.7 40.7
2015 Honda FIT EX EFR71M 128 FR71M1 M Manual 6 N 2875 4.62 48.9 FHNX10028932 MFR No HWY 0.00354 0.053 171.49 0.0009 0.00195 51.9 MPG
2015 Honda FIT LX EFR71M 128 FR71M1 M Manual 6 N 2875 4.62 49.8 FHNX10028929 MFR No FTP 0.01505 0.313 232.04 0.0053 0.00344 38.2 MPG 38.6 36.4 41.6
2015 Honda FIT LX EFR71M 128 FR71M1 M Manual 6 N 2875 4.62 49.8 FHNX10028930 MFR No HWY 0.00438 0.08 167.74 0.0021 0.00224 52.8 MPG
2015 Honda FIT LX EFR72C 130 FR71C1 CVT Continuously Variable 1 Y 2875 5.44 30.2 FHNX10029500 MFR No HWY 0.00651 0.03 151.87 0.0038 0.0026 58.5 MPG
2015 Honda FIT LX EFR72C 130 FR71C1 CVT Continuously Variable 1 Y 2875 5.44 30.2 FHNX91002142 EPA No FTP 0.0127109 0.1742354 198.3511029 0.0056853 0.0038919 45 MPG 41.4469971 45.0640739 47.6128369
2015 Honda FIT LX EFR72C 130 FR71C1 CVT Continuously Variable 1 Y 2875 5.44 30.2 FHNX10030094 MFR No FTP 0.01409 0.166 197.67 0.007 0.00346 44.8 MPG 42.9 44.6 47.2
2015 Honda FIT LX EFR72C 130 FR71C1 CVT Continuously Variable 1 Y 2875 5.44 30.2 FHNX10030095 MFR No HWY 0.00456 0.04 150.82 0.0029 0.0021 58.8 MPG

It's a benchmark. Under X circumstances, car Y can expect to get Z mpg. Just because none of us ever experience those exact circumstances doesn't make it a useless benchmark, and just because the EPA doesn't conduct the tests themselves doesn't make the reported numbers less valid- if they conduct one test and it's in line with the factory's reported results there isn't much sense in duplicating every other test.

I find it interesting that, seeing as how the CVT is the first transmission that can theoretically keep the engine at its power peak continuously, some have accused Honda of cheating because the engine doesn't stay at exactly that peak.

It seems to me to be prudent engineering to have the engine get near but stay under the supposed peak. Otherwise they risk overspeeding and hitting the rev limiter if any wheel slips in the least.

I wonder if the EPA cycle for the manual transmission demands that the car be accelerated to redline in every gear?

First, my 08 will exceed max hp rpm in any lower gear and go right to right up to limiter.
second, acceleration is not linear, meaning straight line, but instead a curve. if you take acceleration rates in small tangent increments you will find the rate (slope) decreasing steadily. that's due to the greater effort required to increase the speed and to air resistance.
third, there are lots of things that will cause an engine to 'run out of horsepower' short of the redline. tire size/weight, aero, intake, exhaust, timing, and even gasoline, Individual engines also matter (not all engines have the same torque and hp curves even for like brands and models.. In many cases careful rebuilding engines can result in hp increases up to 40%. (why do you think showroom stock racers have their engines 'refreshed' just before the runoffs).
Production limitations result in differences in hp and torque from even adjacent engines off the production line, mostly the engine parts and assembly on the engine assembly line though exhausts and even ecu's will be just a little different. Then there is differences in the friction between moving parts, like gears crankshafts, pistons, and the like.
There's a reason SS racers want cars off of certain days - and times. Its also why racers from manufacturers have an advantage.
I'd say your car just doesn't have the horsepower to reach redline; if it does in lower gears your complaint is incorrect.
As for advertised hp and torque that's for engines set up with SAE test procedures which sometimes aren't possible on the street. In the old days exhausts were hooked to vacuum pumps, the fuel and air was pressurized, jets doctored, and special but barely legally in spec, cams were used. At least SAE finally got some uniformity in testing though there are still some tricks, including downright lying by marketing ads.
Welcome to the automotive world.
good luck

I think the GE JDM Fit is advertised as 120PS, which = 117HP?

IIRC the GE also does not open its throttle plate all the way when the accelerator is floored. Not sure if related, just saying.