DIY_EFI Digest Sunday, 8 September 1996 Volume 01 : Number 265 In this issue: Injector Flow Variation Re: water injection; and methanol Re: Ancient History Re: Ancient History Re: water injection Water injection Why no feedback under WOT? Re: Injector Flow Variation Re: Ancient History See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- From: Greg Finnican Date: Sat, 07 Sep 1996 09:21:33 -0700 Subject: Injector Flow Variation What an injector's flow rating is and what it actually flows can vary quite a bit. When we ultrasonicly clean them it does not always bring them back in line. How much of this is manufacturing variation? After seeing the variation in injectors I can'nt imagine using an untested set in a performance application - certainly not a used set. I would like to hear others experience. greg@xxx.com ------------------------------ From: rickydik@xxx.com (RD Rick) Date: Sat, 7 Sep 1996 07:52:56 -0700 Subject: Re: water injection; and methanol >... >He also suggested that mixing up to 50% methanol with the water helped >improved its volatility. >Jim Steck Word of warning. I have read that methanol is very corrosive to metal parts, while ethanol is not. I don't know if ethanol will work in that application. Ethanol costs more than gasoline, methanol costs much less; many SoCalif gas suppliers/stations have been accused of diluting their gas with methanol, causing serious damage to untold numbers of cars. Ethanol will counter the corrosive effects of methanol, if added in the right proportion. But now, I have read that the new oxygenated fuel has methanol.. There is an apparent increase in the number of car fires with it; probably due to corroded fuel pressure regulators. It happened to my '81 Vanagon last year. Fuel pressure went over 100psi, and leaks sprung up in lots of places. Post-mortem on the regulator showed it to be badly rusted inside. The input was also clogged with an unknown substance, like maybe rust crud from other parts. RD ------------------------------ From: Todd Knighton Date: Sat, 07 Sep 1996 11:12:35 +0000 Subject: Re: Ancient History > > So, you're telling me these mpg figures don't bother you? > > Tom Tom, Actually, we have a Chevy 3500 w/ a 454 and a 3.53 rear, and the mileage sucks. They're running the thing lean enough that it has quite the misfire at idle. It's leased and will probably be going back, otherwise we'd fix it. It only gets about 12 miles per gallon highway and about 8 around town. I'd almost be better off making 3 or four trips in a honda, and still save on fuel / tires / maintenance. Big motors get bad gas mileage. Period. There's too much friction involved to do much better. I'd bet you could get the motor up to mid 30's in mileage with optimum cam timing, port shape, combustion chamber shape and the like, but it'd be useless as a tow vehicle then. compromises. Oh well. Todd Knighton Protomotive Engineering ------------------------------ From: Darrell Norquay Date: Sat, 7 Sep 1996 14:35:27 -0700 (PDT) Subject: Re: Ancient History At 08:06 AM 9/6/96 -0500, Tom Cloud wrote: >I haven't dug up my stuff about VE and CFM and displacement, but seems to >me that an xxx CID engine is a pump that always pumps xxx CI of air every >two revolutions (assuming a 4-stroke). All that changes is the density >(i.e. throttle closed = lower density A/F charge). This would only be true if the engine is a perfect air pump. Due to a few major factors and a zillion minor ones, the engine's VE changes quite radically with RPM, and is in almost all cases <100%. This means that if the VE is 80% at a particular RPM, a cylinder is only packing 80% of it's true volume of air/fuel mixture into it at that speed. VE is related to, but not to be confused with overall energy conversion efficiency, which is much lower. You can tune the intake runners and exhaust system to give 100% or even greater than 100% VE's, but only over a very narrow RPM range. If you tuned the intake and exhaust paths to give optimum VE (which generally coincides with the torque peak) at, say, 3000 RPM, and chose an overall gear ratio that put 70 MPH right at 3000 RPM, in combination with some ignition advance and fuel leanout, you could get phenomenal (comparatively) cruising fuel mileage with any engine. It'd be no fun at all to drive, though, unless it had a 27 speed transmission... OEM's tune the system for a broad, flat VE (torque) curve, to give the engine driveability, economy, and durability. Aftermarket parts makers (and DIY'ers) attempt to increase the peak output at the expense of driveability, economy, and durability. It should also be possible to increase the economy at the expense of power output and driveability. The area under the VE curve, though, would stay pretty much the same for a given size NA engine, no matter which approach you take. TANSTAAFL. >And I thought that running R-E-A-L lean meant burning things up (I KNOW it >does!). So, it takes a certain amount of fuel to just coast, much less make >any power. This is only really true at WOT. Running lean, while trying to make maximum power from the engine is a bad idea. Different story at light loads or cruise. I have seen industrial methane powered stationary engines running at AFR's of 28:1, and they last a longggg time. Once you get the engine running, you can lean out the mixture until the engine is just making enough power to balance the power required by the load, (overcoming drag, friction, etc). Ask any pilot, he will tell you that as a plane reaches cruise altitude and speed, you lean it out until the engine is just making enough power to maintain cruise conditions. This gives maximum fuel economy. >So, how does one take a 400+ CID engine and get that kind of mileage >unless he's running a 1.7:1 rear end (i.e. L-O-W r's) -- even if he had >100% VE ? Fuel consumption is pretty much directly proportional to RPM (as is HP) for a given displacement. Fuel consumption is also tempered by the shape of the VE curve. Maximum efficiency occurs at the VE peak, which coincides with minimum energy required to produce a given amount of power. So, you're right, a low numerical rear end ratio or overdrive does increase mileage, but lowering RPM would also decrease HP and VE, and thus overall engine output. If you set it all up jesssstt raaht, so that VE, torque, and HP all peak at a low enough RPM, which also coincides with your normal cruising speed, voila`, mega-MPG. regards dn dnorquay@xxx.com ------------------------------ From: Darrell Norquay Date: Sat, 7 Sep 1996 15:06:19 -0700 (PDT) Subject: Re: water injection At 08:26 AM 9/6/96 -0600, Steve Cicoria wrote: >>O.K., My daily driver (w/ 160k miles) is starting to knock unless I use >>premium. I think I'll put in a water injection system, but have a few >>questions. Can you dump in too much water? I plan on a winshield washer >>pump, and to have it vaccum operated. I will monitor the map signal and >>start pumping water when the engine gets below a certian vaccum. What >>I could do with another op-amp and a 555 is have the water pump come in >>at 50% duty cycle at one vaccum, and full on when it drops below another >>vaccum. I don't think I'll get carried away with using a pic chip, >>because I really don't know the relationship between engine load and >>knock. Yep, you can dump in too much water. This may result (although unlikely) in hydraulic lock, which bends rods. More likely you'll just lose power instead, and/or have to fill the tank every 10 miles.... The Edelbrock Varajection system used 2 inputs, RPM (from the distributor) and manifold pressure (internal sensor). It uses a standard OEM type windshield washer tank with built in pump, and a couple of different nozzles that you can install in the top of the aircleaner. It has a small check valve in the water hose to prevent water from draining back to the tank. It has 3 pots for setting it up: RPM threshold - keeps water off until RPM setting is exceeded. This keeps water injection off at lower RPM's. MAP threshold - keeps water off until manifold pressure drops below setpoint. This keeps water off at part throttle + cruise. Volume - this controls the maximum pump speed, and thus the max volume of water injected. This was all done with a single LM324 quad opamp chip, a Motorola pressure sensor, and a handfull of r's, c's + t's. No PIC's. BTW, I have a complete one sitting in my garage collecting dust, if anyone's interested... And At 01:57 AM 9/7/96 EDT, Jim Steck wrote: >A good treatise on water injection can be found in Sir Henry Ricardo's fourth >edition of "The High Speed Internal Combustion Engine" pages 165-171. His >He also suggested that mixing up to 50% methanol with the water helped improved >its volatility. I used to use stright windshield washer antifreeze in mine. The Varajection is basically all plastic, so corrosion was not a problem. I read somewhere that using Acetone/methanol/water mix worked even better, the acetone acts as an octane boost. Of course, with this, you may have problems with the plastic parts. As far as using Ethanol, who wants to dump in a bottle of vodka with every fillup? D'rather park the car, drink the vodka, and dream about going fast... regards dn dnorquay@xxx.com ------------------------------ From: Jody Shapiro Date: Sat, 7 Sep 1996 19:57:19 -0400 (EDT) Subject: Water injection [This message is being sent to both the Do-It-Yourself Fuel Injection mailing-list and the F-body mailing list] On the DIY_EFI reading list there's a book written by Jeff Hartman (published by Motorbooks International) titled "Fuel Injection: Installation, Performance Tuning, Modifications". I've found this book to be pretty good in terms of info, including nitrous and turbo stuff. Anyway, Jeff Harman has a new book out, "High-Performance Automotive Fuels & Fluids" that I think is pretty informative as well. Issues of interest to me that he covers are water injection, nitrous oxide, octane & anti-knock additives, racing fuels (ethanol, methanol, etc.). He also discusses lubricants, including a detailed discussions of synthetic motor oil (why it's good) and oil additives (and why most are bad). Obviously, this book is no Heywood, but I found it to be pretty informative... ISBN is 0-7603-0054-2 and price is $19.95US for those interested. - -Jody ------------------------------ From: Thor Johnson Date: Sat, 7 Sep 1996 21:32:19 -0400 (EDT) Subject: Why no feedback under WOT? >From studying my 300ZX Manual (and looking at the ECM tuning program), I understand that the ECM runs open loop in WOT situations. Why? Is it because the ECM program uses the O2 sensors for MPG reasons only, or do the physics change under WOT? Thanks! Thor Johnson johnsont@xxx.edu http://falcon.mercer.peachnet.edu/~johnsont Have you seen the WarpMap lately? http://falcon.mercer.peachnet.edu/~johnsont/warpmap ------------------------------ From: MaxBoost@xxx.com Date: Sun, 8 Sep 1996 00:33:57 -0400 Subject: Re: Injector Flow Variation >What an injector's flow rating is and what it actually flows can vary >quite a bit. When we ultrasonicly clean them it does not always bring >them back in line. How much of this is manufacturing variation? After >seeing the variation in injectors I can'nt imagine using an untested set >in a performance application - certainly not a used set. I would like to >hear others experience. greg@xxx.com Thats why so many injector service/sales company's have sprung up in the past few years. In showroom stock racing it has been the rage to flow the injectors and then install the rich ones on the cylinders that run lean etc. At Cosworth they flow each injector on the bench before running them. This allows them to make sets and to seriel numbereach one to track their life. Nissan rates their injectors in sets, they are 0-5% > spec or 0-5% < spec. When you replace one it has to match the rest of the set and there are usually 2 or 3 part numbers for one flow rate. Max. ------------------------------ From: "John Faubion" Date: Sun, 8 Sep 1996 00:45:48 -0500 Subject: Re: Ancient History > power from the engine is a bad idea. Different story at light loads or > cruise. I have seen industrial methane powered stationary engines running at > AFR's of 28:1, and they last a longggg time. Once you get the engine Whoa! As soon as you changed fuel (methane) you changed the stoichiometric ratio! So your comparison of 28:1 on methane means absolutely nothing when related to gasoline. Its an apples to oranges. In addition you wouldn't want to lean out an engine to reach maximum power. Max power will occur with a slightly richer than stoich mixture. While best efficiency will occur leaner than stoich. > etc). Ask any pilot, he will tell you that as a plane reaches cruise > altitude and speed, you lean it out until the engine is just making enough > power to maintain cruise conditions. This gives maximum fuel economy. True. > right, a low numerical rear end ratio or overdrive does increase mileage, > but lowering RPM would also decrease HP and VE, and thus overall engine Mainly by reducing frictional losses at the lower RPM. > output. If you set it all up jesssstt raaht, so that VE, torque, and HP all > peak at a low enough RPM, which also coincides with your normal cruising > speed, voila`, mega-MPG. Of course the ONLY rpm where torque, horsepower and VE could all peak at the same time is at 5252 rpm. Not exactly most people's idea of a best cruise speed! Now if your cam duration is short enough keep the engine speed down, reducing the overall frictional losses (which are exponential with regard to rpm) and then geared for this lower peak torque (at which VE also peaks and usually BSFC) than you could have great mileage. I know this is sort of what you were eluding to but just not getting there on the same track. :) John Faubion jfaubion@xxx.net ------------------------------ End of DIY_EFI Digest V1 #265 ***************************** To subscribe to DIY_EFI-Digest, send the command: subscribe diy_efi-digest in the body of a message to "Majordomo@xxx. A non-digest (direct mail) version of this list is also available; to subscribe to that instead, replace "diy_efi-digest" in the command above with "diy_efi".