Test Results on a Lean-Burn Four-Stroke Opposed-

Piston Sleeve-Valve Engine

June 12, 2015

Michael “Tony” Willcox

Director, Controls and Simulation

Pinnacle Engines

engineexpo 2015 Presented: Tuesday 16, June

Concept Overview

Opposed piston – 4-stroke

• Reciprocating sleeve valves

• Made with diesel dry liner tooling

• Circumferential ports

Low surface area

• Low heat loss

• Low knock

Long stroke

• B/S: ~0.42

Stable combustion (high turbulence)

• Very dilute (LTC) combustion

• Low NOx

1. Expansion

2. Exhaust

3. Induction

4. Compression

CONFIDENTIAL 2

Valve stem seals

isolate oil circuit

from ports

Circumferential

port allows for

optimized

breathing

Traditional

metal to metal

gas seal

Sleeve Valve CONFIDENTIAL 3

Pinnacle Design History

Phasers

1.0L 2-Cyl

1.2L 3-Cyl

Scooter “Beta”

110cc, b/s 39x(46x2) Tested: 2014-present (~1500 hrs)

Durability underway in customer

vehicle – at ~100 of 200 hrs

Scooter “Alpha”

110cc, b/s 39x(46x2) Tested: 2011-2013 (~4000 hrs)

Passed 400hr durability

Multi-

Cylinder

Designs

Proof of Concept

250cc, b/s: 51/(60x2) Tested: 2009-2011 (~1000 hrs)

400 hrs

CONFIDENTIAL 4

FEV

Simulated

(1.2L)

Low

High

1.2L 3-cyl BSFC (with VCR)

Standard vane phaser

Miller/Atkinson + VCR

Fuel agnostic

• CNG, Ethanol,

Naphtha, Diesel

Optimum CR

•Ambient conditions

•Cold start

•Emissions (lightoff)

•Limit pmax & MPRR

2000 RPM, 2 bar BMEP (Source: AVL 2010)

Pinnacle 1.2-L

VCR 314 [g/kw-hr]

CONFIDENTIAL 5

Crankshaft Phaser

Low Heat Loss

Traditional single piston

•26:1 CR

•CNG HCCI (Ref: Koga and Kiura, 2013)

6.4

Surface/Volume Ratio

(110cc Pinnacle)

Pinnacle Engine (TEST DATA)

•15:1 CR

•SI PFI Gasoline

(247cc Single Piston HCCI)

Same I.Eff as 247cc HCCI on CNG at ~1.4 b/s Ratio

(Significantly lower heat loss results in higher exhaust energy - catalyst)

s/b: 1.0 1.3 1.8 2.1 s/b: 2.4

CONFIDENTIAL 6

HCCI @41.5% I.Eff

• ~1.4 b/s

• ~7.5 S/V

110cc: Engine Dyno Test Data

2000 RPM, 2 bar BMEP

(Source: AVL 2010)

Pinnacle 110cc 15:1

2k2bar 360 [g/kw-hr]

110 cc

Pinnacle 1.2-L VCR

314 [g/kw-hr]

224 [g/kw-hr]

On a 110cc!

Test Data

Bore/stroke: 39x(46x2)

CR: 15:1 (fixed)

Fixed valve events

PFI @3bar. ~68um SMD

Haltermann EC-2005 (95 RON)

CONFIDENTIAL 7

Vehicle Testing – March 2015

2010 Honda Elite • 108cc (50x55)

• 11:1 CR

• Water Cooled

• Keihin EFI

Pinnacle • 110cc 39x(46x2)

• 15:1 CR (fixed)

• Oil Cooled

• Lean Burn

• Pinnacle EFI

Northern California Diagnostics Labs • EPA & CARB Certified Tests (bag emissions)

• Test Cycle: WMTC Reduced

• ARAI road load coefficients

BS IV

Pinnacle 110cc Engine in ‘10 Elite Chassis

NCDL rider, same day, test cell, emissions, cycle, road load, etc.

Steady State:

• Run “WMTC” ignoring

speed targets

• 2 points per 20 min cycle

Steady-state fuel economy

improved 35-60%!!!

CONFIDENTIAL 8

Vehicle Testing – March 2015

BS IV

Pinnacle 110cc Engine in ‘10 Elite Chassis

NCDL rider, same day, test cell, emissions, cycle, road load, etc.

ARAI Regulations & Test Results [g/km]

Regulated

Species

Pinnacle Lean

Test Results

2015 to

2020*

2020* to

2025*

Apr 2015 BS IV BS V

CO 0.38 1.403 1.14*

HC 0.58 - -

NOx 0.21 (No Cat!) 0.39 0.225*

HC+NOx 0.79 0.79 0.45*

*Expected

WMTC Drive Cycle Fuel Economy

34%

Drive

cycle

CONFIDENTIAL 8

HC catalyst not functioning properly

Feb ’15 Customer (India) Test Results

Steady-State (road load) • 20-45% improvement

WMTC Results (from cold) • >30% FE Improvement

• NOx: 2025 w/o cat

• CO – 40% of regulation

• HC – Not well catalyzed (needs close cat)

Acceleration (test track) • 20% down-speeding same 0-60 km/h

CONFIDENTIAL 9

ARAI Regulations & Test Results [g/km]

Regulated

Species

Pinnacle Lean

Test Results

2015 to

2020*

2020* to

2025*

Feb 2015 BS IV BS V

CO 0.46 1.403 1.14*

HC 1.07 - -

NOx 0.17 (No Cat!) 0.39 0.225*

HC+NOx 1.24 0.79 0.45*

*Expected

32%

68.4 km/L

~52 km/L

Bag Emission

Test Results

Flat Crown CCW Light Swirl

CONFIDENTIAL

Hardware aligns with simulation

Swirl visible (~1.3 at -20 deg)

Piston crevices return HC to cyl

End of

Combustion

End of

Combustion

Inlet Piston

CONFIDENTIAL 10

Combustion burns to

end of bore

Simulation Case Study

Hourglass Squish Flat Crown

CONFIDENTIAL

91.5% fuel conversion at 90 deg

Simulated: 31.6% I.Eff

Most of HC is from hourglass ends. Some

flame penetration into conical squish ends

with bias in direction of swirl.

94% fuel conversion at 90 deg (reduced HC)

Simulated 34.6% I.Eff

Radial flame propagation unrestricted by crown,

higher retained swirl. Flame propagates into

conical regions

10% Test

9.5% Simulated

CONFIDENTIAL 11

Simulated Beta Piston Crowns

June 12, 2015

HG (E040366/367)

•Hourglass Squish •\\192.168.12.119\share\work\E04\20140623A

FC (E040401/402)

•Flat Crown •\\192.168.12.119\share\work\E04\20140620A

ES (E040602/608)

•Elliptical Squish •\\192.168.12.119\share\work\E04\20140618A

DD (E040643/644)

•Dual Deck •\\192.168.12.119\share\work\E04\20140630A

FCTB (E04646/647) •Flat Crown Through Bore •\\192.168.12.119\share\work\E04\2014702A

FCDD (E040662/663) •Flat Crown Dual Deck •\\192.168.12.119\share\work\E04\2014710B

FCHG (E040664/665) •Flat Crown Hour Glass •\\192.168.12.119\share\work\E04\2014717A

More Turbulence =

More Efficiency

CONFIDENTIAL 113

27

AFR!

LEANER

TPS

24

AFR!

LEANER

TPS

Lean Burn Test Process

Best Fuel

Economy

5% COV

4k RPM

5% COV

Standard Induction (Scooter) High Turbulence Induction

CONFIDENTIAL 13

Defines dilution

tolerance vs. load

Lean NOx: SAE 2013-32-9064

27

AFR!

LEANER

TPS

24

AFR!

LEANER

TPS

Lean Burn Test Results

Map targets

on scooter

5% COV

4k RPM

<50ppm NOx!

Standard Induction (Scooter) High Turbulence Induction

CONFIDENTIAL 13

Defines dilution

tolerance vs. load

Secondary Throttle “Swumble”

Hardware tests in Alpha & Beta hardware sets (110cc)

Enhances light load dilution tolerance

Test results

• Lambda >2.0 with standard spark ignition (5% COV IMEP)

CONFIDENTIAL 16

Vo

l. E

ff / 1

00

[]

Engine Speed [RPM]

110cc PFI gasoline

• 2.7 bar fuel pressure

2 spark plugs

Single double-ended coil

• 25mJ total energy

~32 deg spark advance

CONFIDENTIAL 14

<5% COV IMEP

Why

Stability increases due to:

1. Bulk Motion: Swirl increased by ~3x

2. TKE: >2x increase at -20 deg

3. Mixing: nearly homogeneous – HCCI-Like

Swumble

4K, 3 bar IMEP Standard “Swumble” Units

Tumble @-20 0.25 0.2 []

Swirl @-20 1.2 -3.8 []

TKE 14 29 [m2/s2]

Stdev(Lambda) 0.22 0.04 []

Standard Port

(Swirl)

Lambda Distribution

Lambda Distribution

CONFIDENTIAL 15

2x

As run on

scooter

CRASI: Compression Ratio Assisted Spark Ignition •High expansion (20:1 - 25:1)

•Ultra-Dilution via CR (HCCI/SACI Levels)

•>100% at light loads

•EGR or Lean

•Flex-fuel optimized

•CNG/E85/Gasoline

•Knock-sensor controlled CR

•Full Load - CR Reduction

•Avoid DI (PM Challenge)

•Knock & mega-knock mitigation

Pinnacle Roadmap

Time

E

F

F

I

C

I

E

N

C

Y

Beta:

110cc 15:1

Lean Burn

2011 2015

Downsized,

DI Turbo

Downsized Turbo,

VCR, VVT, EGR

MPFI

30 bar BMEP

1.2L 3-Cyl

Naturally

Aspirated, VCR,

VVT, EGR, PFI

2009

Alpha:

110cc 15:1

Lean Burn

1. Mega-Knock

2. DI: Particulate (Soot)

3. Poor light load combustion Stability

4. Low Dilution

5. No fuel flexibilty

250cc

HCCI, GCI,

SACI,

Stratified Lean

Homogeneous

No Soot, low NOx

Diesel

VCR, VVT,

EGR CRDI

Fixed CR

CONFIDENTIAL 16

30%

Advanced

Combustion

(GCI, HCCI,

SACI)

Thank You

Questions…

CONFIDENTIAL 17