Do higher NO_x numbers explain Eco's much superior fuel efficiency?

My assumption is it does not, and lean burn engines are rare because of the NOx concerns.

 

I took all of the 2016 Toyota models from the "Test Vehicle Database" and selected the "FTP" tests in the configuration with the highest MPG. Then using the roll-down coefficients, calculate the drag HP at 65 mph (104 km/h) and plotted the results:

Higher MPG models have the lowest drag HP - well that was Captain Obvious but how much?

• far right - Mirai fuel cell vehicle with a surprising high drag - but at 4250 lbs, it is heavier than the Level 2 ECO at 3250. IMHO, it may have pieces hanging in the breeze so one wonders if it has a cooling drag issue?
  
• next group - Prius family with the Level 2 ECO on the far right and Prius c on the left
• 3d group - Prius v, Camry and a couple of Lexus with drag HP of the smaller gassers
  
• 4th group - RAV4 on right, highlander on left, and rest of Lexus hybrids with middling drag
  
• last group - GS 450h, Corolla, Yaris, and the rest, none in the Prius range

	model (veh_#)	MPG	Drag HP @65 mph
1	MIRAI (16-JD1F_0)	94.1	21.3
2
3	PRIUS Eco (16-ZV2H_0)	84.1	16.4
4	PRIUS (16-ZV1H_2)	78	16.6
5	PRIUS (16-ZV3H_2)	77.3	16.6
6	PRIUS c (12-NP2H_1)	75	16.5
7	PRIUS c (12-NP1H_1)	71.4	17.6
8
9	PRIUS v (12-ZW2H_3)	61.5	21.4
10	CT 200h (11-ZA1H_2)	59.8	20.8
11	Camry Hybrid LE (12-AV1H_4)	58.9	18.4
12	ES 300h (13-AV2H_1)	57.4	19.1
13
14	RAV4 HYBRID AWD (16-AA2H_5)	48	26.8
15	NX 300h (15-AZ3H_3)	47.5	26.2
16	CT 200h (11-ZA2H_0)	45.5	23.1
17	NX 300h AWD (15-AZ2H_5)	45.1	27.4
18	NX 300h AWD (15-AZ1H_0)	44.2	27.8
19	RX 450h (16-GL3H_0)	42.7	26.7
20	RX 450h AWD (16-GL2H_1)	41.7	28.2
21	RX 450h AWD (16-GL1H_1)	41.3	27.9
22	HIGHLANDER HYBRID AWD LE Plus (14-GU2H_0)	39.1	27.7
23	HIGHLANDER HYBRID AWD (14-GU1H_0)	38.2	27.7
24
25	GS 450h (13-GW2H_1)	40.6	22.5
26	COROLLA LE ECO (14-ZE1C_0)	40.3	19.9
27	YARIS (12-NP1M_1)	40	19.2
28	YARIS (12-NP1A_0)	39.5	19.8
29	COROLLA LE ECO (14-ZE2C_1)	39.4	19.9
30	COROLLA LE ECO (14-ZE2C_1)	39.4	19.9
31	COROLLA (14-ZE3C_1)	38.8	19.5
32	COROLLA (14-ZE3C_1)	38.8	19.5
33	COROLLA (14-ZE1M_2)	37.6	18.4
34	COROLLA (14-ZE4C_0)	37.5	19.7
35	iM (16-ZE2C_1)	37	21.4
36	COROLLA (14-ZE1A_0)	35.4	18.2
37	iM (16-ZE1M_0)	35.3	20.9
38	CAMRY (12-AV1A_1)	32	19.4
39	RAV4 (13-AA3A_0)	30.1	25.2
40	tC (11-AT1M_4)	29.5	21.9
41	tC (11-AT1A_4)	29.3	22.1
42	NX 200t (15-AZ4A_3)	29	26.0
43	NX 200t AWD (15-AZ3A_4)	28.9	26.5
44	RAV4 AWD (13-AA1A_2)	28.5	26.2
45	IS 200t (16-AE2A_0)	28.3	21.9
46	NX 200t AWD F SPORT (15-AZ1A_2)	28.1	27.4
47	NX 200t AWD F SPORT (15-AZ2A_2)	28	27.2
48	CAMRY (12-GV1A_1)	27.1	20.0
49	ES 350 (13-GV1A_2)	26.4	19.6
50	GS 200t F SPORT (16-AE1A_1)	26.3	24.0
51	HIGHLANDER (14-AU1A_0)	25.4	27.8
52	RX 350 (16-GL3A_0)	25.2	28.1
53	GS 350 (16-GR3A_0)	25	23.1
54	LS 600h L (08-UF2H_0)	24.7	28.0
55	TACOMA 2WD (16-TN3A_0)	24.7	34.8
56	IS 350 (14-GE3A_0)	24.3	22.2
57	RX 350 AWD (16-GL2A_1)	24.3	29.4
58	GS 350 AWD (16-GR2A_1)	24.2	25.8
59	TACOMA 4WD (16-TN1M_0)	24	35.5
60	IS 350 AWD (14-GE2A_2)	23.8	24.8
61	HIGHLANDER (14-GU3A_0)	23.6	28.3
62	TACOMA 2WD (16-GN3A_0)	23.5	33.8
63	TACOMA 4WD (16-TN2A_1)	23.4	36.1
64	IS 350 AWD (14-GE1A_2)	23.2	25.8
65	TACOMA 4WD (16-GN1A_0)	23.2	34.9
66	SIENNA (11-GL3A_3)	22.9	28.4
67	HIGHLANDER AWD (14-GU1A_0)	22.8	29.6
68	TACOMA 4WD (16-GN1M_1)	21.6	36.3
69	4RUNNER 2WD (13-GN5A_1)	21.5	31.8
70	4RUNNER 4WD (13-GN3A_2)	20.8	34.2
71	RC F (15-UC1A_1)	20.7	25.3
72	LS 460 L (07-UF1A_1)	20.3	24.4
73	SIENNA AWD (11-GL1A_1)	20.3	30.4
74	LS 600h L (11-UF1H_1)	19.9	31.1
75	GS F (16-UL1A_1)	19.6	27.0
76	LS 460 AWD (09-UF1A_3)	19.5	26.9
77	GX 460 (14-UJ3A_0)	18.6	34.5
78	TUNDRA 2WD (14-UK3A_0)	18.4	38.6
79	TUNDRA 4WD (14-UK2A_3)	17.4	40.7
80	RC F (15-UC2A_1)	17.3	28.1
81	TUNDRA 2WD FFV (16-SK1A_2)	17.2	40.4
82	TUNDRA 2WD (12-SK3A_1)	16.7	38.8
83	GS F (16-UL2A_1)	16.7	30.0
84	TUNDRA 4WD FFV (12-SK4A_6)	16.6	42.2
85	TUNDRA 4WD (12-SK2A_2)	16.4	44.5
86	LAND CRUISER WAGON 4WD (16-UJ1A_2)	16.3	39.4
87	TUNDRA 4WD (15-SK2A_0)	16.1	45.3
88	TUNDRA 4WD (15-SK1A_0)	15.9	45.3
89	SEQUOIA 4WD (12-SK1A_1)	15.5	40.6

Going back to the original question, the Level 2 ECO does not have a significantly lower HP drag at 65 mph even without the rear windshield wiper. The drag power incorporates the tire and wheel effects. This suggests something else is at play so let's look at the emissions profiles of the 2016 Prius and see if there is a clue:

	NOx (g/mi)	THC (g/mi)	CH4 (g/mi)	CO2 (g/mi)	CO (g/mi)
1	0.00295	0.01172	0.00195	105.40951	0.06442
2	0.00187	0.01438	0.00224	113.86017	0.08763
3	0.00089	0.00988	0.00192	115.26325	0.05622
4	51.7%	32.6%	31.8%	31.5%	30.9%

The percentage is ECO / sum of the three, 2016 Prius.

This data suggests the Level 2 ECO is tuned for a leaner burn with some NOx left over but still well under the EPA limit. Thus it burns more of the fuel reducing the carbon and hydrocarbon emissions with a slight increase in NOx. This is probably due to the improved, fuel-air mixture sensor and slightly reduced vehicle drag.

The tested, Level 2 ECO is 125 lbs lighter, 3250 vs 3375, than the two other 2016 Prius, further reducing the inertial load. The only fly in the ointment is the EPA tested the lower ranked Prius while Toyota tested the Level 2 ECO. But given the early owner reports, it sounds like both the EPA and Toyota 'sandbagged' the tests . . . to the owner's benefit.

Bob Wilson

 

It's important to remember that the original Insight-I MT had very special hardware to enable it to achieve LB at 22-25:1 levels; it wasn't simply programming the ECM to inject less gas. It had three intake valves per cylinder (and just one exhaust). One of the intakes delivered a richer mixture concentrated near the spark plug gap (stratified charge). Ignition initiated there spread into the leaner parts of the charge, with an OVERALL ratio of 22-25 to 1 in lean burn. A NOX catalytic converter was added to take care of the high levels of nitrogen oxides generated as a result. Periodically this catalyst was "cycled" by running a richer mixture through it, which gave you a period of lower mpg which was hopefully more than counterbalanced by the higher mpg's generated in between these regenerations.

The Insight's ICE had essentially four "modes":
(1) off
(2) on, in lean burn
(3) on, in normal burn but with only one of the two other intake valves opening (VTEC)
(4) full power, all valves in action

To get the best mpg out of it, you spent as much time as possible in the first two modes. Unfortunately lean burn yielded very little power, so staying in lean burn was an art. I had an MT Insight-I for nearly 10 years and so was intimately familiar with this limitation.

Is there something special about the Prius' VVT or Atkinson cycle that allows it run properly at leaner mixtures than a typical gas engine?

 

Not really. However, the September data release mentioned an intake manifold path that improved the "tumble" in the cylinder. This is not a trick to support lean-burn as much as improve combustion. We already know the injectors are longer from the Gen-3 and injects gas closer to the intake valves, reducing manifold surface effects. The New Car Features had one sentence about a 12-port injector head. But unique to the Gen-4 is a fuel-mixture sensor that allows finer control than previous O{2} sensors. Still, you've given me an idea.

In theory, one could have single, individual, rich cycles interspersed with normal operation. But there is no free lunch. All this might do is reduce the amount of NOx trap material needed. In contrast, past systems oscillated between a uniform lean and rich fuel trim which made a detectable change in operational characteristics. Also, it suggests a min/max active area whereas an interlaced, cylinder operation might minimize the extremes at both ends.

Bob Wilson

 

We don't have enough information beyond the lighter weight. The New Car Features didn't have the electrical specifications to tell. Worse, we don't have the test metrics for engine and traction battery power flows. We might get them from Argonne Labs but I suspect it will be some time later.

Hopefully this summer I may be able to rent (or borrow) a Level 2 ECO and any of the other models of the 2016 Prius. Give me a day or two with each and I'll have the details needed.

Bob Wilson

 

Toyota claims it is. They point out that it is lighter weight than the equivalent NIMH battery.

 

This is the Gen4 mystery we await Bob Wilson to solve. Toyota ought to just loan him a couple vehicles right now.

 

It also had a LNT/NOx trap in order to get to the ULEV.

The Eco could be running more on the lean side of stoich, but not enough to be the "secret" for the car's better fuel economy, and one sample is not enough to rule out the NOx results being within norm for all the models.

The Li-ion pack could provide some efficiency in charging and discharging, that could allow regen to have a better impact on the test. The Touring models were tested, and there is a different engine code for the different packs. Unfortunately, the larger wheels and extra weight of the Touring models pulled their MPG estimates to under 50mpg, Prius FE Answers buried in the EPA Test Car Database | PriusChat. So we can't tell if the pack made a difference there.

 

One late thought, Toyota reports the Gen-4 peak at 121 hp and I measured Gen-3 at 118-120 hp using SAE J2908:

If I get a chance to rent 2016 Prius this summer for a weekend, I'll run the same test and compare the results. But my leading hypothesis is the 15" vs 17" wheels.

The reason is the polar moment of inertia, the inertia of a spinning mass. The 17" wheels have moved a significant part of their metal mass out 2". Given the comparative mass of 15" and 17" tires found on Tire Rack, they also increased and moved their mass further out. The polar moment of inertia increases by the 4th power of the diameter.

Our biggest problem are the sales labels, New Car Features, and EPA test vehicle labels:

	sales\EPA	16-ZV2H 0 (Pwr)	16-ZV2H 1 (Nrm)	16-ZV2H 2 (Eco)	16-ZV1H 1 (3_4 17")	16-ZV1H 2 (3_4 15")	16-ZV3H 1 (NiMH sum. tires)	16-ZV3H 2 (NiMH all tires)
1	Level 2 (NiMH)						*	*
2	Level 2 ECO	*	*	*
3	Level 3					*
4	Level 3 Touring				*
5	Level 4					*
6	Level 4 Touring				*

	sale\NCF	ZVW50L-AHXEBA	ZVW51L-AHXBBA	ZVW51L-AHXGBA	ZVW51L-AHXHBA
1	Level 2 (NiMH)	*
2	Level 2 ECO		*
3	Level 3			*
4	Level 3 Touring			*
5	Level 4				*
6	Level 4 Touring				*

FYI, the New Car Features only list 15" and 17" wheels and tires. Not shown are two grades of 15" tires.

Bob Wilson

 

Mass moved out 1", not 2".
And I think the polar moment increases by the 4th power of the radius, not diameter, no?

 

Per wiki, (pi/32) * ( D**4 - d**4) for a hollow cylinder. They call them diameter.

Now the hub and spokes ca be reduced to a pair of cylinders but I have assumed a minor effect.

Bob Wilson

 

Polar moment of inertia - Wikipedia, the free encyclopedia
Further up it defines the moment of inertia as pi/2 times radius to the 4th power

If one uses diameter, which is 2 x radius, the fourth power of diameter is 16 times radius to the 4th power, so one must divide by 32 rather than 2 to correct for that.

I don't know why Wiki did that. Maybe somebody in Wiki likes larger numbers?

 

No problem on the terms. It has been a long time since I got to play with those terms in College. And we agree about the maths enough we don't have to get measurements and do the integrals. But we'll disagree on the regeneration effect.

Regeneration is far from 100% efficient even in these controlled, EPA benchmarks. Fortunately, I have recording OBD instruments and can gather some metrics. I can easily run a series of controlled tests and bring the numbers back.

I have access to:




GPS data is attached to: Efficient driving for a 1,000 mile tank | PriusChat

The proposed protocol will be:

1. Pre-test: oil change, tires, air filter, spark plugs (both cars) and measure weight with driver.
   
2. OBD recording: ICE rpm, ICE torque, Traction Batt V, Traction Batt A, velocity
   
3. Clockwise, three loops to normalize temperature and battery SOC @25 mph
4. Test A - no regeneration
   1. At a fixed point at the apex - shift into "N" and let car continue downgrade, no braking, until motion stops.
   2. Repeat five times
5. Test B - regeneration
   1. At a fixed point at the apex - leave in "D" and let car continue downgrade, no braking, until motion stops. The upgrade side is steeper than 'creep'.
      
   2. Repeat five times (may interlace with Test A)

Toss out the highest and lowest and average the three middle runs. Then calculate the regeneration energy minus the engine 'creep' and 'drag'.

Do the same tests for the Gen-1 and Gen-3 Prius.

Bob Wilson

ps. Many years ago, someone once posted that the regen energy capture was about '30%' but the metrics and protocol were not defined. BTW, I also have access to a larger, faster, and shallower loop.

 

While regenerative braking can recapture some of the extra energy used by the extra weight, it is far from 100% efficient.

The 3 Touring is just pounds heavier than the 3 and a mere 5 pounds heavier than the 2 in the listed curb weight. When normalized to what we'd see on a window sticker, a Touring model gets 49.2mpg city and 46.3mpg city. The weight and larger wheels hurts it about the same extent as the Eco has in improvement.

The Eco is 65 pounds lighter than the base 2. While the weight savings isn't in the wheels where it would have a larger effect, the tested car had lower rolling resistance tires than the base that were pumped up to higher pressures. In light of how the Touring test results dropped, there is no reason to look for more reasons for the Eco's improved fuel economy beyond what Toyota has stated; lower weight, slighter better aero, and lower rolling resistance.

 

easier to hyper mile and get 70 - 80 mpg.

 

I did a quick practice run tonight:

• 25 mph just 20 yards south of the East end, access road, continue in a clock-wise direction
  
  • In "N", at 26-27 mph at the access road on the South side
  • In "D", at 10 mph at the access road on the South side

In effect, the regeneration is so power hungry is slows the 2003 Prius from ~26 mph to 10 mph. This is a big chunk of energy. The beauty of this segment is the gravitational assist, the potential energy change is enough to sustain the initial kinetic energy. It is in effect, very close to a ballistics path. The traction battery energy gain will be the result of the change in kinetic energy.

In both cases, there is not ICE power involved so no need to monitor MG1 torque, rpm, and ICE rpm. For this study, just MG2 rpm, torque, traction battery voltage, current, and velocity. This will be enough to measure the best-case, regeneration efficiency.

Bob Wilson