 |
|
| The Suzuki G-10: 3-Cylinder 4-Stroke Engine |
| ENGINE DATA Naturally Aspirated This engine is ideally suited for aircraft use. Its most outstanding attribute is its torqu curve. It is virtually flat throughout its operating range. The Suzuki's 4-stroke, 3 cylinder, one-liter engine is called their model G-10. Naturally Aspirated (NA), the G-10 at 3500rpm has 60ft-lb & at 5800rpm it 55 ft-lb. The question everyone is asking is, "Can it replace my Rotax 582 two-stroke?" Since HP = Torque X RPM / 5252, we can compare the useful power of both engines very nicely. So as not to be comparing apples to oranges, we'll use % of maximum operating rpm. At 100% throttle: The Rotax 582 has 6400rpm/52.5ft-lb/64hp; & the NA G-10 has 5800rpm/55ft-lb/61hp. At 75%: The 582 has 4800rpm/45ft-lb/41hp; & the NA G-10 has 4350rpm/60ft-lb/50hp. At 60%: The 582 has 3840rpm/38ft-lb/28hp; & the NA G-10 has 3480rpm/62ft-lb/41hp. The bottom line is: The G-10 has a 32% more power @ 60% throttle; 18% more power @ 75% throttle; and 5% less power @ 100% throttle. So att full throttle, the 582 will slightly outclimb the G-10 but at cruise, the 582 is left in the wake! The above formula works amazingly well. If you take my computations & compare them to the factory you will see they match perfectly. At 2gph in cruise, The G-10 has less than 1/2 the fuel consumption of the 582. On a 4 hr trip that amounts to at least 48 lbs of fuel, which more than compensates for the slight increase in engine weight. As per issue 37 of CONTACT! magazine, Steve Parkman had a NA G-10 in a Jenny replica. Because of the prop diameter his engine was limited to 4800rpm. He dyno'ed the motor in 3 configurations & got 51 hp @ 4800 with a carburator and 53 hp @ 4800 with Throttle Body injection. He got 64 hp with port injection, but did not specify @ what rpm (probably at 5800). TurboCharged The G-10 Turbo is were things really get exciting! I have not found any published performance curves for the turbo version. All manuals agree that it has a peak torque of 107 ft-lb at 3500rpm. For max HP I've seen: 70hp; 80hp; & "Not Available". The HP ratings don't jive. As we seen above, torque & HP are mathematically interchangable. So we know for sure, that at 3500rpm & 107ft-lb of torque, it is producing 71hp. From another approach, we know the NA port injected version gets at least 64hp at 5800rpm & the turbo G-10 has 8.7 lbs boost @ 5800rpm which is 59% above one atmosphere, and 1.59 X 64 = 102hp. The conversion aftermarket rates this engine from100-115hp. If we use the rule of thumb that an engine's peak torque is close to its max HP, then it would make about 107hp. If we extraplolate off the NA G-10 torque curve we get the following for the Turbo G-10: 60% = 3480rpm/95ft-lb/63hp; 75% = 4350rpm/104ft-lb/86hp; 100% = 5800rpm/95ft-lb/105hp. If we compare this to the $12,500 912S we get: 60% = 3480rpm/84ft-lb/56hp; 75% = 4350rpm/91ft-lb/75hp; and 100% = 5800rpm/89.2ft-lb/99hp. Translation: This obscure little engine outperforms the $14,500, premier, "100hp" Rotax 912S by 13% @ 60% throttle; 15% @ 75% throttle; and 6% @ 100% throttle. Obviously, the standard "80hp" 912 is not even in the same league. This Turbo G-10 a torky little beast! But that's not all... The Turbo G-10 can do this at a comparable weight (170 lb vs the true all up weight of 168 lb for the 912); it has better fuel economy; can cruise at a lower/more comfortable rpm; can deliver full hp on a hot Florida summer or at 10000ft; and best of all, you can go to your local AutoZone and rebuild it for $500. You can not bore a 912 jug. Guess how much money one of those four Rotax jugs cost? Hint: More than a G-10 rebuild. What about reliability? How does it compare? Well its a no brainer (In my opinion) compared to the two-strokes. There are hundreds of G10's flying, many with over 300 hours. I know personally of one with over 500 hrs used as a trainer. The owner loves it & said all he does is check the oil & turn the key. If you compare the Turbo G-10 to the 912S, well there are maybe ten of the Suzuki's flying, and many hundreds of the 912S's. The 912's certainly have a good record and a long history. However, if the "it takes a beating & keeps on ticking" track record of the hundreds of thousands of G-10's in abused auto's (that routinely surpass 150,000 miles with turnpike speeds that are comparable to aircraft rpms) are any indication, then an 800-1000 hour TBO is quite reasonable. 500 hours is 5-10 years of flying for most of us, and if just changing the oil and plugs during that time is all that's needed, then for the cheap price of an overhaul, even 500 hrs would be acceptable. Versiosns The Suzuki G-10 engine comes in at least 5 versions: 1) A marine model, which I know nothing about; 2) A Chevy Sprint naturally aspirated; 3) A Chevy Sprint Turbo; 4) A Geo Metro; 5) And a Canadian (GM Canada) Sprint Turbo.. The Chevy Sprints have a Hemispherical head using rocker arms with the valves perpendicular to the cam. This early head is at least 15lbs heavier than the Geo Metro & Canadian Turbo Sprint heads which do not have rocker arms (hydraulic valve lifters ride directly on the cam lobes. The Chevy Sprint is carburated, the Geo Metro has throttle body injection. The Chevy Sprint Turbo & Canadian Sprint Turbo have port injection, bigger wrist pins & rods, and a lower compression ratio. Heads are interchangable. Only the Geo & Canadian heads will adapt to the Raven Redrive system, which are very popular. The SPG-2 redrive will bolt onto either configuration. My Engine's Story I did not know the Canadian Sprint Turbo existed when I began building my engine. I wanted a 4-stroke with at least 75hp that weighed under 185lbs. The Rotax 912 is unaffordable. The Chevy Turbo Sprint was the only engine I found. It is rated at 70-80hp, but is closer to 100+, as it has 107lbs of torque @3500 & its almost flat all the way to 6K rpm. I am projecting a final, all up wet weight, including cooling system & redrive, of 170-175lbs. And that is with the heavy flywheel, extra quart of oil, a crank scraper, & beefy engine mounts. My version of the engine ended up being almost identical to the Canadian Sprint Turbo. I started with a USA '87 Chevy Sprint Turbo. Put a Geo Metro head on it. Modified the head to recieve the Sprint's port injection. Originally my intent was to use the USA Chevy Sprint oversize pistons & flycut new eyebrows into the top to match the new valve location. When I went to buy oversize pistons & rings I found out to my shock that GM lists them but does not & never has had them available. I did a long & frantic internet search looking for aftermarket options and discovered that GM Canada made a G-10 based turbo that used the new head! I managed to find the part numbers for the 0.5mm over piston & rings and when my local dealer punched them in, they not only came up but were cheaper! The cylinders were cleared to the maxmum recommened clearance, to allow for the higher expansion due to the heat of high continuous hp draw. The stock turbo (water cooled) & intercooler are used. The stock computer, distributor & airmass sensor were replaced with Tracy Crook's (Real World Solutions) electronic fuel & ignition system. This allowed me to get rid of the distributor which was in the way if I later wanted to go with Raven's heavyduty redrive; it got rid of the heavy, bulky & difficult to plumb airmass sensor; & it gives me in-cockpit ontrol of ignition timing & air/fuel mixture. The stock intake was thrown out It's long tuned runners cost about 10 lbs, without them the peak torque would go up a few hundres rpm, but since I'm turning it faster anyway it actually should work out better. The new intake was made up of a severely modified (lightened) a Geo Metro intake plumbed to a severely modified (lightened) throttle body from the Sprint, using brass elbows (from a kitchen sink) floating on rubber. The heavy cast iron exhaust manifold was replaced with a 304 stainless steel log manifold, moving the turbo from the right side to in front of the fan belt side, of the motor. This moved the cg forward & made for more compact & easier plumbing to the intake. The exhaust/wastegate turbo housing was significantly lightened & welded to a 2 1/2" stainless steel elbow exhaust with a oxegen sensor fitting on it. In order to keep the engine height down (engine is in a pusher configuration on top of my wing), the stock 6" deep oil pan had to go. I cut off & tossed all but the gasket flange & about 3/8" of the original sides & put a shallow sloped pan there. The block of the engine is 2" above the Kolb flatbed motor mount, so the pan sits quit low. Since I needed to move the prop back from the wing at least 4" for better prop feed & quieter performance, & since I didn't want to use a prop spacer, I moved the whole engine 4" back. The rear of the engine now hangs 4" past the flatbed engine mount & this space allowed me to put an oil tank to collect the oil draining off the sloped oil pan. These 2 pieces were welded together. The factory, screened oil pickup was welded to the bottom of the tank & plumbed to an outside fitting. From there the oil goes thru a hose to the auxillary input of the oil pump located on the right side/bottom of the block. The original oil feed inlet in the oil pan was blocked off. The tank has 2 sight glasses to mark "full" & "1qt. low". This setup has the added advantage of not letting oil build up near the crank during those long steep Kolb climbouts (stock oil pan has resevoir on front of engine) & the tank near the prop should help with cooling. I increased the oil capacity one qt to help compensate for the higher engine loading & hoping to preclude the complexity of an oil cooler. The cooling system will use a Cadilac air conditioner condensor with 234 cubic inches of fin area. Rule of thumb is you need 1cubic in per1 hp. That will give me over 200% cooling capacity. I heard that this deep radiator may not work well at speeds below 100mph, I need to explore this mor. Since heat is the demon of continous load conditions, I mean to tame it on the 1st go around. I'm doubling the factory intercooling capacity & cutting the restriction in half by using 2 OEM intercoolers in parallel. Automotive intercoolers really are used as a heat sink rather than a true continuous radiator,eg, because boost is typically for a matter of seconds & the aluminum mass just absorbs the heat for that time, so I lincreased their capacity. Reducing the air-charge 10 degrees increases the hp 1%. For every 10% increase in air-charge density, you get an 8% increase in hp. Reducing the air-charge temperture 200 degrees, reduces the exhaust gas temperature 200 degrees,and reduces the load on the cooling system. A typical intercooler allows an additional 3-4 lb of boost without increasing octane, reducing compression or retarding timing. So they are an excellent investment of weight. I replaced the stock 7 lb altenator with a 3 lb, 20A, John Deere permanent magnet altenator, saved a couple pounds with a smaller diameter aluminum crank pulley, & shaved 2 lb with a lighter starter. That's it for now. I'll update this as I go. .Last Updated: 02-02-05, |