I have been watching a number of these issues go by on the list for some time, and I am moved to offer a few comments. First, I am stunned at what often passes for technical fact on the list, but I am pleased to see such discussions in contrast to vendor bashing, character assassination and vegetable oil reviews. I want to enter some material here for archival evidence (should any newbie avail himself of such). I want to address several points: A. Longevity of engines as a function of size and vehicle they are mounted in. B. Longevity of engines as a function of mean cruising rpm. C. Displacement, torque and engineering poweress. D. Engine modifications vs. power - fact and fiction. A. Longevity of engines as a function of size and vehicle they are mounted in. The lifetime of a modern engine. intelligently maintained (Manufacturer's specs mixed with common sense - oil and air filter changes, timing belt renewal, coolant and belt maintenance) is already impressively long. Taking the same engine (the VW I4 for my purposes), the useable lifetime (time to engine output reducing to 75% of its as delivered state) is related primarily to the total work generated and secondarily to the total distance traveled by the piston rings. Of course, there are many other mitigating lesser factors, but when the statistics of real numbers work out, engine failure is metallurgical and is directly related to the work done. Work done is torque applied for unit time. Minimal torque situation: Launch the Westfalia down a 10 degree hill onto the interstate in Indiana, set the cruise control to 35 and put a thousand miles on the prairie. This is a low wind resistance regime, and the land is sufficiently flat the primary torque load on the wheels is just the requirement to overcome drag. Acceleration -- makes the hot air engine develop more torque. Sustained high speeds -- even more of it. As I have discussed in previous posts, there is a minimal torque required at the wheel to maintain a given speed on a level stretch of road. Any grade change, any acceleration has to come out of the engines reserves. This is why the Westfalia Diesel requires 3 minutes to accelerate to 60 and a supercharged waterboxer can do the job in 15 seconds or less (between crankshaft repair stops). Now as I have said before (echoes, echoes) the required torque at the wheel is the product of the engine developed torque and the torque multiplication through the transmission. I have (see the archives) shown the difference in required engine torque for a given speed vs. final drive ratio (most common difference in factory transmissions). Fact: I4 driving 5300 pound Westfalia through the diesel DZ transmission will have to develop less engine torque to sustain any highway speed than if it was driving through the DK (Air-cooled) transmission. Therefore it will last longer in the case of higher engine speeds than in the case of lower ones. Ah - yes say you. But piston speeds! Fact is that a modern VW I4 will travel more than 300,000 miles at 5000 rpm with synthetic 20W50 oil at temperatures of 250F or less. Of course it may well run for 375,000 miles at 4000 rpm. But so what? No other system on the vehicle is remotely as reliable. As to work produced, a big impact on lifetime is the instantaneous peak load achieved and the connecting rod loading at low piston velocities (near TDC). This translates as "the worst treatment for a small displacement engine is lugging or demanding high torque output a low rpm". So if you want to compare the lifetime of an I4 (or A/C boxer) in a Vanagon as compared to a rabbit, golf (or beetle), just look up the required road torque to maintain your favorite velocity (I've given you the numbers in the archives for the Vanagon based on Martin Jagerstand work), determine the gear multiplication ratio for each case and calculate the value. This is a starting point. Relevant to the gentle driver. The second term is the work expended in accelerating. Here, a good approximation is given by dividing the vehicle weight by a given gear (times the final drive ratio). Generate a table for first through fourth and compare vehicles again. If the numbers are close, the work expended will be similar. By my numbers, the top gear comparison of the diesel Westfalia and the 1.8L VW GTi (1990) is the same within 2%; and less relative work has to be done by the Vanagon engine accelerating the vehicle from rest than in the case of the golf. So, lifetime of any modern engine is directly related to the work you require that it generate during your ownership period. Period. Corollary: Do manufacturers fine tune an engine for a particular car. Yep, you betcha. They choose gearing, gearing and gearing! For some they mill move the engine tuning compromises around for more power or better mileage at a driving model. Nothing more. Can't afford to build custom engines at this price point any longer. B. Longevity of engines as a function of mean cruising rpm. See above. Within limits, the engine will perform superbly (more than 200,000 mile lifetime) when operated within its design envelope. For the I4 engine, this includes cruising at 6000 rpm day in and day out with proper maintenance. For the waterboxer, the redline is nearly 1,000 lower than the I4 engines related to the second order moments (see the VW design articles for these engines I have posted at Alistair Bell's website). C. Displacement, torque and engineering poweress. There is more religion and snake oil in this topic - exhausting (pun intended). For years people have been promulgating the BS of higher technology Japanese engines over German, American, British and Korean. Now, I do understand that fact probably has no place in this discussion, but tough. A B&W comparison can be made readily for any engine, so long as the same numbers are used in the comparison. The best comparison of any set of engines comes from a comparison of the Brake Mean Effective Pressure (BMEP) for the engine. This is in essence the peak torque normalized by the displacement with a few constants thrown in for good measure. This is largely established by combustion chamber design and the efficiency of hot air generation (combustion and heat transfer to the compressed air charge). Now the very best comparison can be made with the torque curve (torque vs. rpm) but the Japanese are too ashamed of their engine performance to publish these figures, they only release (Subarea) the peak torque and HP numbers. Now as I said, I put this in the archives some time ago, but I reproduce it in full in the following: ========= From: <FrankGRUN@aol.com> To: <vanagon@GERRY.VANAGON.COM> Sent: Sunday, October 28, 2001 11:51 PM Subject: On Engine Efficiency, Comparing I4s, WBs and Subies Some time ago as I was surreptitiously lurking on the subaruvanagon list (before being caught and excised by Warren as an unwelcome alien) I perused a minimalist tome penned by pensioner that waxed enthusiastically about the modern design of the Subaru engines and their great efficiency as compared to the VW ilk. Well, as some of you might imagine, this stimulated my seating area, and I'm moved to respond. Numbers are good. Measurements are good. Mechanical (careful here) feelings are often not real. The theory of the nonlinear tail desperately sensing the positive improvement stimulated by the most recent and stunning infusion of cash into the bottomless cash consumer. o, anyway, I decided to prepare a table that could compare the mechanical efficiency (and therefore engineering prowess) of the various engines that could power these bricks through the ether. The quantity of comparison I have chosen is the brake mean effective pressure developed in the combustion chamber while doing work on the piston. The number is derived from the peak torque developed at the flywheel per unit engine displacement. These numbers are readily calculated from the meager engineering data provided by the reclusive Subaru personnel, and are readily expanded by the addition of your favorite heat pump. I have arranged the values in ascending order of higher pressure (means higher efficiency). Pressure expressed in units of psi. In essence, the torque maximum represents the peak volumetric efficiency of the pump as well as the maximum efficiency of completely consuming the available fuel for air heating. This number is a function of the combustion chamber design, the intake and exhaust manifold flow dynamics, cam and ignition timing, etc.,etc. Bigger is better. (Also often newer.) Engine Displacement Torque BMEP (liters) (ft.lb.) (psi) Vanagon 2.0L A/C 101@3000 126.9 VW GX 1.8L I4 96@3000 132.8 VW MZ 1.8L I4 98@3250 135.6 Vanagon 1.9L WB 106@2600 136.6 Subie 2.5 2.5L WB (<96) 144@2800 136.6 Vanagon 2.1L WB 117@3200 136.7 VW 9A 1.8L I4 113@4400 140.8 VW HT 1.8L I4 105@3000 145.3 AUDI 2.3 2.3L I5 140@4500 149.7 VW RV 1.8L I4 109@3800 150.8 Audi 3A 2.0L I4 121@3200 150.8 VW ACC 1.8L I4 107@3500 150.8 VW AAZ 1.9L I4 TD 107@2500 150.8 Subie 2.5 2.5L WB (>97) 162@2800 150.8 Subie SVX 3.3L WB 228@4400 150.8 VW ABA 2.0L I4 122@3200 152.0 VW RD 1.8L I4 110@3200 152.2 VW AAA 2.8L I4 173@4200 152.6 Subie 2.2 2.2L WB 137@4400 153.0 Subie 2.5 2.5L WB P II 166@2800 157.4 VW PF 1.8L I4 114@3800 157.7 SAAB 2.0 2.0L I4 128@3000 159.5 TIICO (SA)2.0L I4 132@3500 164.5 Subie 2.2 2.2L WB P II 149@3600 166.4 VW AHY 1.9L I4 TD i 149@1900 193.7 VW 1.9 TDi1.9L I4 TDi 155@1900 201.5 Subie 2.2T2.2L WB T 181@2800 202.2 VW 1.8T 1.8L I4 T 162@2200 224.1 SAAB 2.0T 2.0L I4 188@3000 234.3 Some comments: 1. The table is in 12 point Monaco Font. 2. The engine longest in the tooth is the old Type 4, followed closely by the old low compression I4 CIS engines, and Vanagon WB engines. 3. Note that the pre 96 Subie 2.5L engine is just as inefficient as the Vanagon WB's. 4. The 1.8L Digifant engines (RV, RD and PF) are much stronger than the VW waterboxers as well as the 1.8L CIS and CIS-E engines. 5. The Subie SVX and the late 2.5 engines are no more efficient than the post 1988 VW engines. 6. The ABA cross-flow head (2.0L) and the VR6 are at the same generation and are very good. 7. The SAAB 2.0L (very similar to the chipped Digifant or Bosch LH-Jetronic driven Audi 3A or RV/PF engines) is a very efficient design. 8. If the TIICo numbers are to be believed, this 8 Valve head with Motronic management outperforms the ABA crossflow head with Motronic. 9. The latest phase II engine design of the Subie 2.2 is a competitive WB, as is the phase II 2.5L Subie. 10. The 1.9L and TDi engines are very efficient in this comparison. 11. The turbocharged gas engines lead the pack, with the highest efficiency. The volumetric efficiency achieved by the pressurized inlet far exceeds pumping losses due to inserting the turbo in the exhaust stream. 12. So, the VW WB's and the old A/C engine are really from an elder time as are the same vintage Subies. The I4 engines are right there with the strongest. Only the turbos do better. Go to the turbo if you can. Frank Grunthaner =========== As you can see, the latest Air-cooled VW is technically the most inefficient loss leader here. Similar comparisons can be made for AC Porsches vs. watercooled Porsches, BMW to BMW, AC Turbo Porsche to water cooled turbo porsche ... the results will follow. Watercooled, more efficient. Subaru engines up to the Phase II type are no more advanced than the 2.1L WBX. The I$ holds its own with the best of them. The real torquers are the 2 valve per cylinder engines. The engine efficiency roughly breaks into three classes (for gasoline engines): low compression or aircooled, modern high compression N/A engines and turbocharged systems. So if you want more power (read more torque) choose your displacement! I would suggest staying below the point that the Vanagon trans will turn into metal shavings! (right around 200 foot pounds). If you want a barn burner, put a Subie H6 (3.3L) behind a DZ trans and smoke em. Before you bolt up the trans, add a set of Aussie super tuff gears, but I figure 10 mpg anyway so ... Otherwise, choose a 2.0 L I4, a 2.5L subie, etc. All good efficiency. Or respond to the more enlightened call and add a turbo. I'm still preparing a turbo charged 3A 2.0 L Audi engine with the maximum torque tuned to just under 200 foot pounds. For maximum efficiency, I'll bolt it to the DK trans, but then I'm just after an optimum performance vehicle and have no fantasies about tire smoking. American engines? Get the numbers, put them on the table, contemplate the additional mass and prepare for the comfortable experience of trying to get Kennedy to bless you with one of their adapter kits. Frankly, for the cash, go full TDi. D. Engine modifications vs. power - fact and fiction. Finally, I can't handle the Lilly (SP?) number acceptance game. Horsepower numbers are torque measurements! Unless a set of modifications are given with before and after numbers they are nothing more or less than a good animated fantasy. The devil is in the details. I've watched the high lift rocker game lead to the same or lower power depending of the cam tuning. Statically balancing an engine that never sees the high side of 6,000 is an exercise in philanthropy for the participating machinist. I've seen amateurs port a head along with intake and exhaust manifold and lose 22% of the factory power. On the balancing issue: VW stock is impressive. I have weighed the piston and connecting rod assemblies in my 3A engine. The difference from heaviest to lightest was just under 0.30 grams. Most American amateur and stock car engine builders will try to balance these components to 1.5 grams. 0.5 gram span is professional Nascar stuff for engines designed to spend a lot of time the other side of 7K rpm. Want to effect power output. Follow dyno or performance results. For example just comparing 30 to 70 times on your ride in top gear in the same stretch (knowing your mass) will give you a more realistic number than any dyno can. Or buy a g-tech. $130. It will give good numbers, and show just how hard it is to get good numbers. Again for more power, add cold air, fine tune the engine management system, add a turbo (exhaust driven or electric) or supercharger, add a windage tray or oil film scraper, etc, etc. For economy: control oil level and temperature, add oil pan baffles, add cold air, intercooler to turbo, clean and match fuel injectors, high power repetitive spark, etc. Oh well, too long anyway. I'll go back to the Atacama. Frank Grunthaner |
Please note - During the past 17 years of operation, several gigabytes of
Vanagon mail messages have been archived. Searching the entire collection
will take up to five minutes to complete. Please be patient!
Return to the archives @ gerry.vanagon.com
The vanagon mailing list archives are copyright (c) 1994-2011, and may not be reproduced without the express written permission of the list administrators. Posting messages to this mailing list grants a license to the mailing list administrators to reproduce the message in a compilation, either printed or electronic. All compilations will be not-for-profit, with any excess proceeds going to the Vanagon mailing list.
Any profits from list compilations go exclusively towards the management and operation of the Vanagon mailing list and vanagon mailing list web site.