DIY_EFI Digest Monday, 22 March 1999 Volume 04 : Number 187 In this issue: Re: injector manifold design Re: injector manifold design Re: Ceramic Coatings and EGT Re: 02 sensor - temprature compensation. Injector Math Re: Ceramic Coatings and EGT Re: Ceramic Coatings and EGT Re: Ceramic Coatings and EGT Re: EEC V (was 8051 programming) TBI Parts Wanted and 2.8 Parts for Free Re: Injector Math RE: injector manifold design RE: injector manifold design Re: Injector Math Re: Ceramic Coatings and EGT Re: TBI Parts Wanted and 2.8 Parts for Free Re: Ceramic Coatings and EGT Re: Best Place to Start Re: Injector Math Re: FI fuel line Re: Injector Math RE: injector manifold design See the end of the digest for information on subscribing to the DIY_EFI or DIY_EFI-Digest mailing lists. ---------------------------------------------------------------------- From: "Tom Parker" Date: 22 Mar 99 21:08:23 +1200 Subject: Re: injector manifold design Stowe, Ted-SEA wrote: >I am having trouble finding a good throttle body with a single injector for >my mgb, (the gm is hard to physically fit). outside of the irritating fact >that my head has siamesed intake ports, could I create an intake manifold >that mimics an injector per port, and have 2, (or one larger one), aimed at >each mgb 'dual' intake port ? this might sound screwy but I have the >injectors and OEM throttle body and I was thinking it was a shame not to use >them. I have a friend who has just efi'd his Mini. I haven't seen it yet, but he says that he took two SU carbs and butchered them to make two throttle bodies and then put one injector per port in the manifold. The software (Link Engine Managment, I think) had to be custom written to accomodate the siamesed firing order. Last time I spoke to him it was in the idle and rev in nutral stage, but had no throttle linkage, so it wasn't drivable. - -- Tom Parker - tparker@xxx.nz - http://www.geocities.com/MotorCity/Track/8381/ ------------------------------ From: "Tom Parker" Date: 22 Mar 99 23:58:21 +1200 Subject: Re: injector manifold design Mike Morrin wrote: >At 08:50 pm 21/03/99 -0800, Stowe, Ted-SEA wrote: >. outside of the irritating fact >>that my head has siamesed intake ports, could I create an intake manifold >>that mimics an injector per port, and have 2, (or one larger one), aimed at >>each mgb 'dual' intake port ? >I think you might have trouble getting mixture balance between the >cylinders on each port, as they are not 180 degrees firing from each other, >so how would you time the injection pulses to give each cylinder its fair >share? I have thought a little about this, and it shouldn't be difficult to design a circuit to take the 4 injector output from an OEM ecm and drive two injectors in the correct order. The Mini fires 1-3-4-2, with 1 & 2, 3 & 4 siamesed. You would just put the 1 & 2 cyclinder output into a logic OR gate and drive one injector, and the same for 3 & 4. The only issue would be overlap between the cyclinders. Is it normal for two injectors to be firing at the same time in a sequential injection? I have a feeling that a standard OR gate will switch off if both inputs go high... there is another gate that I can't quite remember that doesn't The simplest from an electronics point of view is to mount two injectors, you could even aim one at each valve in the port. The trouble is that on a Mini at least, The exhaust is right next to the inlet. You would really have to put one injector above the port and one below, with two fuel rails... - -- Tom Parker - tparker@xxx.nz - http://www.geocities.com/MotorCity/Track/8381/ ------------------------------ From: Frederic Breitwieser Date: Mon, 22 Mar 1999 12:27:34 +0000 Subject: Re: Ceramic Coatings and EGT > at idle went from around 900 F to 1200 F. The mixture is > the same and the cam is larger than the last engine which > would open the exhaust valve earlier, so that would bump > the EGT a little. ceramic coatings reflect or isolate I guess is a better word the metal that's coated from the explosions that occur in your engine. Considering these explosions are a few thousand degrees, an exhaust temperature of 900-1200 is not unreasonable, especially if the compression is high. At idle, our radical twin-turbo Buick V6 displayed 1100-1200 deg, typically towards the higher side, and at much higher RPMs the temperature actually went down 50-100 degrees. It also might be that with the modifications you have made, larger cam, more displacement, your engine especially if you have not modofied your carb or EFI, might be running lean. Engines under a lean condition tend to display a hotter EGT as well as higher water temp. - -- Frederic Breitwieser Bridgeport, CT 06606 http://www.xephic.dynip.com 1993 Superchaged Lincoln Continental 1989 500cid Turbocharged HWMMV 1975 Dodge D200 Club Cab (soon to be twin turbo 440) 2000 Buick GTP (twin turbo V6) ------------------------------ From: Wen Yen Chan Date: Mon, 22 Mar 1999 07:33:07 -0500 Subject: Re: 02 sensor - temprature compensation. Hello, I was thinking of the scheme that I've used to measure the impedence of NiCads before charging (to find out the condition of the cell). First the open circuit voltage is measured, then the cell is loaded slightly. Based on the voltage drop a rough estimate of the impedence can be obtained. Wen On Sat, 20 Mar 1999, Clarence L.Snyder wrote: > > Just a thought... Why not use a cheaper 1 wire o2 sensor and monitor its > > temperature by measuring the sensor's impedence. > > > > Wen > Likely because you can't. The "impedence" of a battery is hard to read. > > > > ------------------------------ From: Frederic Breitwieser Date: Mon, 22 Mar 1999 12:38:16 +0000 Subject: Injector Math Greetings, as most of you know by now, I'm building an efi-based mopar stroker engine using a turned down 440 crank combined with a .030 over 383 block, for a total displacement of 431 cubes. With the below math, I'm trying to caluclate several things. Approximate HP/Torque to be expected from the motor, both N/A as well as boosted (10 lbs of boost max), the appropriate injector size for a sequential-based injection system, as well as cfm of air required at 5000 RPM naturally aspirated as well as with 10lbs of boost. Yes, I'm having fun. Anyhoo: Using several different sources (boogs, mags, websites) I once again found this formula: fuel flow (lb/hr) = HP x BSFC --------- cyls BSFC would be .45 for a N/A engine, .50 for a Nitrous engine, and .55 for a forced induction motor, and 1.1 for an alcohol motor, and 1.76 for a nitromethane motor. Cyls is easy, I have eight. So I have flow (lb/hr) = HP x .55 -------- 8 Now, I have two variables left. Flow, which I'm trying to calculate (gives me injector size), and HP. How does one calculate the math for the HP without testing the engine? The engine has these stats: 1969 383 cid stroked to 431 cid using a 1970 forged hipo crank turned down for the appropriate journal sizes of the 383 block. Bore: 4.25" + .030 overbore (4.28" bore) Crank Stroke: 3.75" (1970 forged 440 crank) Rods: 6.358" (B rods) or 6.768" (RB rods) I have chevy rods but I'm prolly going with Mpar 440YJ rods and having Ross or Wiseco make pistons with the pin higher into the head of the piston, with a final compression ratio targeted at 9.0:1 before turbocharging, which I think I have sized appropriately for 10lbs of boost as a maximum with a water-based intercooler system. Also, the cam will be an RV cam once I find something appropriate, as this engine will be in a low RPM Torque motor for my Dodge truck, for towing purposes. The goal is to make all the power on the low end for towing in the 1500-3000 RPM range, with a 5000 RPM maximum redline determined arbitrarily. As far as the intake goes, I am using an aluminum Edelbrock 383 Streetmaster intake, which I have modified by welding on injector bungs so the fuel injectors are flush with the end of the runners right at the heads, and where the carb normally would have gone, I've welded together a bolt-on plenum that is 10" wide, 15" long, and 12" high, minus a 3" transmission cooler core configured to become a water-based intercooler. The volume of the plenum therefore is 1575 cubic inches including the volume of space taken up by the intercooler, and should be more than adequate. The front of the plenum has two Ford throttle bodies that I have to interconnect mechanically so they are effectively ganged, each connected to a turbo. TE-34 would be my first guess for a turbo. For horse power calculations, I used Corky Bell's PLAN formula, which is Power = bmep x stroke x bore area x putts. Breaking this down further, I ended up with this: power = bmep x stroke x area x putts (cyl * maxrpm/2) power = bmep x 3.75" x 57.5489 x (8 cyl * 2500) power = bmep x 4316167.5 How do I resolve this with two variables? Same problem as before. I could merge the above two formulas, however then I'd end up with this: So I have flow (lb/hr) = HP x .55 and HP = bmep x 4316167.5 -------- 8 Merged I have flow = bmep x 4216167.5 * .55 ---------------------- 8 flow = bmep x 2318892.125 ------------------ 8 Still, I have two variables, and can't resolve either of them. So, I ventured onto more simplistic formulas. Corky has a range formula like this: powerlow = .052 x 431cid x (10lbs of boost + 14.7) = 553.5764 hp powerhigh = .077 x 431cid x (10lbs of boost + 14.7) = 819.7189 hp Then, I opted to extrapolate the same formulas to work without boost, and ended up with the following: powerNAlow = .052 x 431cid x 14.7 = 329.4564 hp powerNAhigh = .077 x 431cid x 14.7 = 487.8489 hp Now I am truly confused. However, I continued the math for giggles, and took the original formula using the highest value I ended up with (820 hp): fuel flow (lb/hr) = HP x BSFC --------- cyls and substituted seemingly random numbers like so: fuel flow (lb/hr) = 819.82 HP x .55 --------------- 8 fuel flow (lb/hr) = 56.362625 lb/hr Bendix/Siemans has injectors that flow 55 lbs/hr as well as 62 lbs/hr. MSD has 50 lb/hrs (2012) that work with stock ECMs (high impedence) and fuel rail pressure can be increased to compensate I would imagine so there is more flow. Injectors are rated based on certain fuel rail pressures, so increasing the rail pressure would increase the flow proportionally, within reason. 50 lb/hr injectors when I need 56.x lb/hr injectors could be close enough if the rail pressure goes up 13%, which I can do easily enough. How off am I in my math? Any ideas? - -- Frederic Breitwieser Bridgeport, CT 06606 http://www.xephic.dynip.com 1993 Superchaged Lincoln Continental 1989 500cid Turbocharged HWMMV 1975 Dodge D200 Club Cab (soon to be twin turbo 440) 2000 Buick GTP (twin turbo V6) ------------------------------ From: "Gary Derian" Date: Mon, 22 Mar 1999 07:52:47 -0500 Subject: Re: Ceramic Coatings and EGT If your exhaust temps went down as rpm went up, it is a good indication that you were detonating. As long as temps are going up you're OK. This is one of the key tuning factors in 2 stroke racing engines. I would expect it to hold true for other types as well. Gary Derian >At idle, our radical twin-turbo Buick V6 displayed 1100-1200 >deg, typically towards the higher side, and at much higher >RPMs the temperature actually went down 50-100 degrees. >Frederic Breitwieser ------------------------------ From: "Clarence L.Snyder" Date: Mon, 22 Mar 1999 09:34:08 -0500 Subject: Re: Ceramic Coatings and EGT All the heat NOT absorbed by the piston, valve, and head (and therefore the cooling system) is doing one of 2 things. It is producing more power (which you want - that's why you coated the engine) and going out the exhaust (which is normal - 30% normally goes out the pipe). If half of the heat that normally went to the cooling system stayed in the engine, and half of what does not go to the cooling system turned to power you would have a 7% increase in power, more or less, and the same increase in exhaust heat-loss. If you get less than 7% more power in this case, you would get more than 7% increase in exhaust heatloss ------------------------------ From: Frederic Breitwieser Date: Mon, 22 Mar 1999 09:45:22 -0500 Subject: Re: Ceramic Coatings and EGT > If your exhaust temps went down as rpm went up, it is a good indication that > you were detonating. As long as temps are going up you're OK. This is one That's interesting, because the knock sensor didn't trigger that much if at all most of the time, though occasionally it did light a few LEDs on the 10 LED scale. Typically two LEDS were light most of the time, except during rapid deceleration of the engine, in which more would light up. The LM3914 being what it is (the chip we used) one LED was always lit. Thanks for the tip! - -- Frederic Breitwieser Bridgeport CT 06606 1993 Supercharged Lincoln Continental 1989 HWMMV w/turbocharged 500cid Caddy 1975 Dodge D200 Club Cab soon to have 431 stroker + turbos 2000 (I hope) Buick GTP (Mid-Engined Sports Car) ------------------------------ From: "Ord Millar" Date: Mon, 22 Mar 1999 09:59:43 -0400 Subject: Re: EEC V (was 8051 programming) I want to be able to adjust the injector size on a 96 Cobra, without having to "trick" the mass air. I also want to disable the IMRC check - either bypass the routine that checks it's position, or put the opening RPM up above the rev limit. I have an autologic module, but hate the idea of paying a tuner to adjust such simple parameters. Ord - -----Original Message----- From: Joe Dzura To: diy_efi@xxx.edu> Date: Sunday, March 21, 1999 11:00 PM Subject: EEC V (was 8051 programming) >I am Ord, What are you trying to find out? I'm part way through a 96 >Cobra EEC V cal. > >Joe D > >Ord Millar wrote: >> >> Are you doing anything with EEC-V's? >> >> -----Original Message----- >> From: Tom Sharpe >> To: diy_efi@xxx.edu> >> Date: Sunday, March 21, 1999 4:48 PM >> Subject: Re: 8051 programming >> >> > >> >Just a thought. I'm working on EEC-IV Fords which use a similar 8061, now >> >replaced with a 8096 (intel). If you pick a 16 bit chip, I'll jump on... >> Tom >> >They are a lot easier to program .. >> > >> > > ------------------------------ From: "Kurek, Larry" Date: Mon, 22 Mar 1999 09:00:41 -0600 Subject: TBI Parts Wanted and 2.8 Parts for Free Guys: A couple of quick parts wanted/available...first the parts I need: - - An old LT1 stock cam. - - An old TBI intake, preferably aluminum, and preferably off an angle bolt head. - - An old V8 computer controlled remote coil distributor - - An old set of cast or hyperpathetic .030 over pistons for a 350. I'd like to get these as cheap as possible, so if you have any of the above, I would REALLY appreciate hearing from you. OK, now things I have available, and in the spirit of the above request...these parts are FREE to a good home. Shipping cost upfront only, all are for an 86 2.8 GM RWD engine: - - Aluminum Carb intake - - Exhaust manifolds - - Carb air filter housing - - Carberetor - - Aluminum front cover - - Crank (late model large journal...needs a regrind due to a spun bearing....) - - Flexplate - - Complete HO heads (cast iron...NOT the played valve FWD heads, but these do have the larger valves) - - non-computer remote coil HEI Please let me know ASAP on all of the above! Thanks!!! ------------------------------ From: Barry Gibson Date: Mon, 22 Mar 1999 10:08:08 -0500 Subject: Re: Injector Math - --------------6E9EED6F8E29D91D19A3DEA0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit simplest method i can think of to "get in the ballpark" w/ estimating horsepower is to ask the cam grinder. it's been my experience that for "standard" combinations of carb/intake/heads/exhaust they can often tell you exactly what power to expect -- indeed, a dyno sheet is frequently available, as they actually test their cam(s) to determine what works best with what. some pieces work lots better than do some others ... do avail yourself of this expensive and hard-won knowledge! as to the bumpstick itself, i'd suggest asking about a "negative overlap" cam. the cam grinder you want to do biz with will know what this is and how to tailor it to your situation. the minor added expense of getting a true custom cam will be more than offset by the results. IIRC, the latest tt porsche awd turbo used a neg overlap cam. once you are in the ballpark for the n/a condition, calculate the power in the boosted condition, eg: HP == hp * ( (14.7 + 10 ) / 14.7 ) == hp * ( 1.68 ) yes, this will produce a number that is very much higher than what will actually obtain (no losses due to intake heating, etc) but you _want_ a high estimate to avoid any chance of a lean condition. next, do the bsfc to pounds conversion per your formulas, then correct for injector duty cycle. Frederic Breitwieser wrote: > Greetings, as most of you know by now, I'm building an > efi-based mopar stroker engine using a turned down 440 crank > combined with a .030 over 383 block, for a total > displacement of 431 cubes. With the below math, I'm trying > to caluclate several things. > > Approximate HP/Torque to be expected from the motor, both > N/A as well as boosted (10 lbs of boost max), the > appropriate injector size for a sequential-based injection > system, as well as cfm of air required at 5000 RPM naturally > aspirated as well as with 10lbs of boost. Yes, I'm having > fun. Anyhoo: > - --------------6E9EED6F8E29D91D19A3DEA0 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit

simplest method i can think of to "get in the ballpark" w/ estimating
horsepower is to ask the cam grinder.  it's been my experience that
for "standard" combinations of carb/intake/heads/exhaust they can
often tell you exactly what power to expect -- indeed, a dyno sheet
is frequently available, as they actually test their cam(s) to determine
what works best with what.  some pieces work lots better than do
some others ...  do avail yourself of this expensive and hard-won
knowledge!

as to the bumpstick itself, i'd suggest asking about a "negative overlap"
cam.  the cam grinder you want to do biz with will know what this
is and how to tailor it to your situation.  the minor added expense of
getting a true custom cam will be more than offset by the results.
IIRC, the latest tt porsche awd turbo used a neg overlap cam.

once you are in the ballpark for the n/a condition, calculate the power
in the boosted condition, eg:
  HP == hp * ( (14.7 + 10 ) / 14.7 )
        == hp * ( 1.68 )
yes, this will produce a number that is very much higher than what
will actually obtain (no losses due to intake heating, etc) but you
_want_ a high estimate to avoid any chance of a lean condition.
next, do the bsfc to pounds conversion per your formulas, then
correct for injector duty cycle.
 
 

Frederic Breitwieser wrote:

Greetings, as most of you know by now, I'm building an
efi-based mopar stroker engine using a turned down 440 crank
combined with a .030 over 383 block, for a total
displacement of 431 cubes.  With the below math, I'm trying
to caluclate several things.

Approximate HP/Torque to be expected from the motor, both
N/A as well as boosted (10 lbs of boost max), the
appropriate injector size for a sequential-based injection
system, as well as cfm of air required at 5000 RPM naturally
aspirated as well as with 10lbs of boost.  Yes, I'm having
fun.  Anyhoo:
 

<snip> - --------------6E9EED6F8E29D91D19A3DEA0-- ------------------------------ From: "Stowe, Ted-SEA" Date: Mon, 22 Mar 1999 07:10:55 -0800 Subject: RE: injector manifold design thanks Tom. with the Bosch LH jetronic even though it has 4 separate injectors, by the design of the schematic it would appear that they all fire at the same time, there is not a separate driver for each, just one line out of the box wired up to all 4. so at this point, I could use a nice tbi with a single injector, (but I can't find a nice one), or simply use the existing injectors I have, or maybe better yet, use 2 larger ones instead of 4, which appeals to me. so the issue is not really the electronics but the aiming (I think), of the injectors. I have basically 2 inlet ports on the head where you can almost see the intake valves. thanks, Ted Stowe - -----Original Message----- From: Tom Parker [mailto:tparker@xxx.nz] Sent: Monday, March 22, 1999 3:58 AM To: Mike Morrin Subject: Re: injector manifold design Mike Morrin wrote: >At 08:50 pm 21/03/99 -0800, Stowe, Ted-SEA wrote: >. outside of the irritating fact >>that my head has siamesed intake ports, could I create an intake manifold >>that mimics an injector per port, and have 2, (or one larger one), aimed at >>each mgb 'dual' intake port ? >I think you might have trouble getting mixture balance between the >cylinders on each port, as they are not 180 degrees firing from each other, >so how would you time the injection pulses to give each cylinder its fair >share? I have thought a little about this, and it shouldn't be difficult to design a circuit to take the 4 injector output from an OEM ecm and drive two injectors in the correct order. The Mini fires 1-3-4-2, with 1 & 2, 3 & 4 siamesed. You would just put the 1 & 2 cyclinder output into a logic OR gate and drive one injector, and the same for 3 & 4. The only issue would be overlap between the cyclinders. Is it normal for two injectors to be firing at the same time in a sequential injection? I have a feeling that a standard OR gate will switch off if both inputs go high... there is another gate that I can't quite remember that doesn't The simplest from an electronics point of view is to mount two injectors, you could even aim one at each valve in the port. The trouble is that on a Mini at least, The exhaust is right next to the inlet. You would really have to put one injector above the port and one below, with two fuel rails... - -- Tom Parker - tparker@xxx.nz - http://www.geocities.com/MotorCity/Track/8381/ ------------------------------ From: "Stowe, Ted-SEA" Date: Mon, 22 Mar 1999 07:15:14 -0800 Subject: RE: injector manifold design yes I wondered about that, but I was considering that since by design the Bosch system fires all 4 injectors like a bank at a time, 2 big injectors per port, both firing all the time, seems like it would have a lot of overlap. I can't immediately see the value of having 4 injectors, which would be physically hard to do with 2 intake ports, but 2 bigger ones instead might logically work. ?? - -----Original Message----- From: Mike Morrin [mailto:mikem@xxx.nz] Sent: Sunday, March 21, 1999 11:25 PM To: diy_efi@xxx.edu Subject: Re: injector manifold design At 08:50 pm 21/03/99 -0800, Stowe, Ted-SEA wrote: . outside of the irritating fact >that my head has siamesed intake ports, could I create an intake manifold >that mimics an injector per port, and have 2, (or one larger one), aimed at >each mgb 'dual' intake port ? I think you might have trouble getting mixture balance between the cylinders on each port, as they are not 180 degrees firing from each other, so how would you time the injection pulses to give each cylinder its fair share? regards, Mike ------------------------------ From: CEIJR@xxx.com Date: Mon, 22 Mar 1999 10:36:40 EST Subject: Re: Injector Math Math with only two variables? Not near enough giggles. What you need to do is program your laptop to prompt for input of intake runner length, diameter, minimum port area, taper, header diameter, primary length, collector length, static compression ratio, rod length, bore, stroke, maximum piston speed (in furlongs per fortnight), number of cylinders, number of valves, four or five temperatures in Kelvin degrees, fuel octane, timing advance, air cleaner pore size and your age at last birthday. Then, the program should combine all of those in a multi-step algorithm which returns the answer of 500 HP, which is what you have been thinking all along. (Then, have it display to six decimal places and you can market the program as a laptop engine analyzer.) FWIW, fuel injector flow isn't directly proportional to rail pressure. It varies as the square root of the increase, so to get from 50 to 56, you need to increase pressure about 25%. And that, of course, doesn't match what is really happening, because it assumes wide open constant injector flow. The rates of opening and closing are also affected by rail pressure, and the percentage of total pulse width consumed by opening and closing affects the average flow, so the actual fuel delivery has to be measured at each rpm and pulse width if you want an accurate number. Then, of course, you still have to assume the horsepower before you can decide if the injectors are big enough. Safest is to go big, just in case your engine really works. Having destroyed more than one turbo engine with lean mixture at max power output have begun to suspect that the injector assumption of .55 BSFC may not take into account enough internal losses or necessary cooling effects when the system is really working. Charlie Iliff ------------------------------ From: Darryl Snover Date: Mon, 22 Mar 1999 10:50:18 -0500 Subject: Re: Ceramic Coatings and EGT Hi All, This talk of ceramic coating sounds really interesting, but where would I find more information about this type of thing, as well as who in my area (Easton, PA) would do this? It really sounds like the ticket for turbo engines. Best Regards, Darryl Snover 1980 Porsche 924 Turbo (with diy-efi augmentation ;-) >All the heat NOT absorbed by the piston, valve, and head (and therefore >the cooling system) is doing one of 2 things. It is producing more power >(which you want - that's why you coated the engine) and going out the >exhaust (which is normal - 30% normally goes out the pipe). >If half of the heat that normally went to the cooling system stayed in >the engine, and half of what does not go to the cooling system turned to >power you would have a 7% increase in power, more or less, and the same >increase in exhaust heat-loss. If you get less than 7% more power in >this case, you would get more than 7% increase in exhaust heatloss ------------------------------ From: "Peter Fenske" Date: Mon, 22 Mar 1999 07:55:09 -0800 Subject: Re: TBI Parts Wanted and 2.8 Parts for Free Hi Larry I got a few olde tbis manifoldes here larry but I live in kanadada.. Let me know if you turn up a blank. Petered out. ------------------------------ From: Frederic Breitwieser Date: Mon, 22 Mar 1999 11:22:16 -0500 Subject: Re: Ceramic Coatings and EGT > This talk of ceramic coating sounds really interesting, but where would I > find more information about this type of thing, as well as who in my area > (Easton, PA) would do this? It really sounds like the ticket for turbo > engines. I've done my own coatings at the recommendation of an internet buddy, and found it to be fairly easy. You need "tech-line" from Summit Racing or a similar product, and a toaster oven. Mask the pistons from the compression ring down with masking tape and newspaper, and apply the techline to the piston top and above the compression ring on the sides. I used a windex bottle, but techline is fairly thick, so take your time. Once you have coated your pistons the way you like, cook the pistons in the oven at a temperature indicated on the techline package (450 deg I think) for 20-30 minutes, then let cool. you might have to sand the coating on the sides of the piston if it globbed or ran. I did this on two engines and the difference was astounding. Its a real cheap solution to a detonation problem, and keeps more heat in the chamber where it can do useful work rather than enter your water jackets. I did my twin-turbo Buick V6 as well as my supercharged 3.8L Continental. Valves, head chambers, piston tops. - -- Frederic Breitwieser Bridgeport CT 06606 1993 Supercharged Lincoln Continental 1989 HWMMV w/turbocharged 500cid Caddy 1975 Dodge D200 Club Cab soon to have 431 stroker + turbos 2000 (I hope) Buick GTP (Mid-Engined Sports Car) ------------------------------ From: steve ravet Date: Mon, 22 Mar 1999 11:28:02 -0600 Subject: Re: Best Place to Start I'd say you've got 3 choices. adapt a gm or other OEM style EFI system, or build from scratch either the efi332 project or the 8051 project. Details about all three can be found on the WWW pages. - --steve > Justin & Julie Carroll wrote: > > I am a electrical/mechanical engineering student at the University of > Alaska, Fairbanks. I want to build a ignition system as part of a > project and maybe use it as a senior project in a year or two. I plan > to build a dyno and use either a Subaru EA-82 or an aircooled > volkswagen as a test platform. > > Where is the best place to start? What components should I plan on > buying? I realize that I could just buy everything but I really want > to learn how everything works so I would like to build and program as > much as possible. > > Any suggestion are appreciated. > > Justin > > - -- Steve Ravet steve.ravet@xxx.com Advanced Risc Machines, Inc. www.arm.com ------------------------------ From: andy quaas Date: Mon, 22 Mar 1999 10:24:01 -0800 (PST) Subject: Re: Injector Math Oh my god. You must keep us up to date on your progress. You are going to use the eec-4, no? Andy - ---Frederic Breitwieser wrote: > > Greetings, as most of you know by now, I'm building an > efi-based mopar stroker engine using a turned down 440 crank > combined with a .030 over 383 block, for a total > displacement of 431 cubes. With the below math, I'm trying > to caluclate several things. > > Approximate HP/Torque to be expected from the motor, both > N/A as well as boosted (10 lbs of boost max), the > appropriate injector size for a sequential-based injection > system, as well as cfm of air required at 5000 RPM naturally > aspirated as well as with 10lbs of boost. Yes, I'm having > fun. Anyhoo: > > > Using several different sources (boogs, mags, websites) I > once again found this formula: > > fuel flow (lb/hr) = HP x BSFC > --------- > cyls > > BSFC would be .45 for a N/A engine, .50 for a Nitrous > engine, and .55 for a forced induction motor, and 1.1 for an > alcohol motor, and 1.76 for a nitromethane motor. > > Cyls is easy, I have eight. > > So I have flow (lb/hr) = HP x .55 > -------- > 8 > > Now, I have two variables left. Flow, which I'm trying to > calculate (gives me injector size), and HP. How does one > calculate the math for the HP without testing the engine? > > The engine has these stats: 1969 383 cid stroked to 431 cid > using a 1970 forged hipo crank turned down for the > appropriate journal sizes of the 383 block. > > Bore: 4.25" + .030 overbore (4.28" bore) > Crank Stroke: 3.75" (1970 forged 440 crank) > Rods: 6.358" (B rods) or 6.768" (RB rods) > > I have chevy rods but I'm prolly going with Mpar 440YJ rods > and having Ross or Wiseco make pistons with the pin higher > into the head of the piston, with a final compression ratio > targeted at 9.0:1 before turbocharging, which I think I have > sized appropriately for 10lbs of boost as a maximum with a > water-based intercooler system. > > Also, the cam will be an RV cam once I find something > appropriate, as this engine will be in a low RPM Torque > motor for my Dodge truck, for towing purposes. The goal is > to make all the power on the low end for towing in the > 1500-3000 RPM range, with a 5000 RPM maximum redline > determined arbitrarily. > > As far as the intake goes, I am using an aluminum Edelbrock > 383 Streetmaster intake, which I have modified by welding on > injector bungs so the fuel injectors are flush with the end > of the runners right at the heads, and where the carb > normally would have gone, I've welded together a bolt-on > plenum that is 10" wide, 15" long, and 12" high, minus a 3" > transmission cooler core configured to become a water-based > intercooler. The volume of the plenum therefore is 1575 > cubic inches including the volume of space taken up by the > intercooler, and should be more than adequate. The front of > the plenum has two Ford throttle bodies that I have to > interconnect mechanically so they are effectively ganged, > each connected to a turbo. TE-34 would be my first guess for > a turbo. > > For horse power calculations, I used Corky Bell's PLAN > formula, which is Power = bmep x stroke x bore area x > putts. Breaking this down further, I ended up with this: > > power = bmep x stroke x area x putts (cyl * maxrpm/2) > power = bmep x 3.75" x 57.5489 x (8 cyl * 2500) > power = bmep x 4316167.5 > > How do I resolve this with two variables? Same problem as > before. I could merge the above two formulas, however then > I'd end up with this: > > So I have flow (lb/hr) = HP x .55 and HP = bmep x > 4316167.5 > -------- > 8 > > Merged I have flow = bmep x 4216167.5 * .55 > ---------------------- > 8 > > flow = bmep x 2318892.125 > ------------------ > 8 > > Still, I have two variables, and can't resolve either of > them. So, I ventured onto more simplistic formulas. Corky > has a range formula like this: > > powerlow = .052 x 431cid x (10lbs of boost + 14.7) = > 553.5764 hp > powerhigh = .077 x 431cid x (10lbs of boost + 14.7) = > 819.7189 hp > > Then, I opted to extrapolate the same formulas to work > without boost, and ended up with the following: > > powerNAlow = .052 x 431cid x 14.7 = 329.4564 hp > powerNAhigh = .077 x 431cid x 14.7 = 487.8489 hp > > Now I am truly confused. However, I continued the math for > giggles, and took the original formula using the highest > value I ended up with (820 hp): > > fuel flow (lb/hr) = HP x BSFC > --------- > cyls > > and substituted seemingly random numbers like so: > > fuel flow (lb/hr) = 819.82 HP x .55 > --------------- > 8 > > fuel flow (lb/hr) = 56.362625 lb/hr > > Bendix/Siemans has injectors that flow 55 lbs/hr as well as > 62 lbs/hr. MSD has 50 lb/hrs (2012) that work with stock > ECMs (high impedence) and fuel rail pressure can be > increased to compensate I would imagine so there is more > flow. Injectors are rated based on certain fuel rail > pressures, so increasing the rail pressure would increase > the flow proportionally, within reason. 50 lb/hr injectors > when I need 56.x lb/hr injectors could be close enough if > the rail pressure goes up 13%, which I can do easily enough. > > > How off am I in my math? Any ideas? > -- > Frederic Breitwieser > Bridgeport, CT 06606 > > http://www.xephic.dynip.com > 1993 Superchaged Lincoln Continental > 1989 500cid Turbocharged HWMMV > 1975 Dodge D200 Club Cab (soon to be twin turbo 440) > 2000 Buick GTP (twin turbo V6) > _________________________________________________________ DO YOU YAHOO!? Get your free @xxx.com ------------------------------ From: andy quaas Date: Mon, 22 Mar 1999 10:38:30 -0800 (PST) Subject: Re: FI fuel line I just use 3/8 steel brake lines with double flares on the ends with brass brake fittings. Hasn't leaked with 18 psi. Andy - ---Jim Davies wrote: > > > > On Sun, 21 Mar 1999, Shannen Durphey wrote: > > > Double flare ok. Single flare leaves sharp edges, good for cutting > > line. I've seen single flare work many times, but ya might as well do > > it right. Main point was to have something to prevent the hose from > > working back down the steel line under pressure. > > Shannen > > > If you do the first part of a double flare and stop at the right point, > you will wind up with a pretty good fascimile of a factory bubble-type > flare as found on low pressure gasoline and other lines on OE > applications. Dont know whether I would use it on 40 psi plus systems, > though... > > For hose, I have always had good results with weatherhead H100. Available > everywhere and good for pretty high pressures. IIRC, 5/16 ID is rated at > 1000 psi or thereabouts. > > _________________________________________________________ DO YOU YAHOO!? Get your free @xxx.com ------------------------------ From: Frederic Breitwieser Date: Mon, 22 Mar 1999 14:23:06 -0500 Subject: Re: Injector Math > Oh my god. You must keep us up to date on your progress. You are > going to use the eec-4, no? I'm not sure if I can get this speed density ECM to work appropriately, but I'm going to make a good go at it. yes on the EEC-IV thing, out of a 88 Grand Marquis. I am pretty sure I have to go MAF, however the local machine shop I use often keeps pushing me back into the speed density system swearing its the best solution for a twin-turbo setup like I will have. I'm still not convinced. If its not successful, I do have a GM TPI setup from 1993, and might just slap that on. - -- Frederic Breitwieser Bridgeport CT 06606 1993 Supercharged Lincoln Continental 1989 HWMMV w/turbocharged 500cid Caddy 1975 Dodge D200 Club Cab soon to have 431 stroker + turbos 2000 (I hope) Buick GTP (Mid-Engined Sports Car) ------------------------------ From: Mike Morrin Date: Tue, 23 Mar 1999 07:33:55 +1200 Subject: RE: injector manifold design At 07:15 am 22/03/99 -0800, Stowe, Ted-SEA wrote: >yes I wondered about that, but I was considering that since by design the >Bosch system fires all 4 injectors like a bank at a time, 2 big injectors >per port, both firing all the time, seems like it would have a lot of >overlap. I can't immediately see the value of having 4 injectors, which >would be physically hard to do with 2 intake ports, but 2 bigger ones >instead might logically work. ?? I don't know if you saw my meaning.... Looking at one of your ports, the firing order is something like 1-gap-gap-2-1-gap-gap-2....... Now if you are squirting fuel at 80% duty cycle, what is going to stop cylinder 2 from sucking in some of the fuel which was intended for cylinder 1? You could go for a sort of modified-sequential injection, but you would then be limited to a 25% maximum duty cycle per cylinder, which would require timed injection, large injectors, and difficult idle control, not to mention custom ECU software. regards, Mike ------------------------------ End of DIY_EFI Digest V4 #187 ***************************** 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".