opinions needed for non audio project i'm working on...

wow it's been so long since i visited my favorite forum, i didnt realize how much i missed this place. first off, sorry to those who pm'd me about those pga 2500's, i still have 8 if anyone is still on that project.

To the point, im working on something for a buddy of mine that has a class 8 drag racing/tractor pull truck (semi). the way they class the trucks is to plug a diagnostic scan tool into the ecm and look at the op file to see what horsepower class the truck belongs in. the rulebook says as long as we cant find the modification you can use it. (so no power boxes, power chips etc.).

there are 2 things i would like to try, which is where i need some opinions. the motor has  crankshaft and camshaft position sensors that feed a pulsed signal to the ecm so it can calculate when to fire each injector. diesel engines make more power when the timing is advanced so im trying to come up with a way to reliably get the signal to the ecm sooner.

my thoughts so far:
  1. maybe theres a way to amplify the pulse so i could detect it sooner, im only talking 2 or 3 degrees of crankshaft rotation.
i can feed the pulse back to the ecm from whatever source i want it just has to look like the sensor signal and be accurate. i can insert a circuit between the sensor and ecm, with a stronger voltage so maybe the sensor might "see" the magnet on the crank sooner?? was that a statement?
  2.  or, catch the first pulse and delay it 358 degrees and then show it to the ecm, it would always be nearly a rev behind but i dont think that would matter.

the second thing i wanted to try was to lengthen the pulsed signal to each injector to hold them open longer for more fuel. its dc, but pulsed dc, could i do that with a cap? again, im talking low voltage ecm signal, and only extending it milliseconds.

thoughts, ideas, i praise you in advance oh wise ones.

Well I'm not a complete computer systems hacker...but I'm pretty sure none of those options will work.

But to get a better picture, what kind of crank sensor? Optical? Hall effect? Is there no way of adjusting the sensor position by a couple degrees and keep it looking stock?

Boosting the voltage is inadvisable, computers like only the voltage they're design to see from a signal. Logic low, logic high... if it's a hall effect sensor, it probably has some sort of schmidt trigger circuit, there may be a way to tweak there. If it's optical, then there's not a lot you can do but adjust when the beam is broken.

What type of engine is this? (cummins, international?)

its a c-15 cat.
to be clear, i wasnt suggesting to send a higher voltage signal back to the ecm. i just thought if i amped the voltage to the sensor it would see the magnet sooner (or the magnet would interfere with the signal sooner) and i could lower the voltage before it went to the ecm. that computer wouldnt know if it was a cps or an atari pulsing the signal return line. im not saying it would be simple, just that there should be a way to do it. and unfortunatley, i cant move the sensors. there is 2 wires, 1 source, 1 signal return. assuming the sensor either normally open or normally closed so the magnet on the crank gear is the trigger.

Do you know if the trigger is logic hi or logic low?

Raising the voltage to the sensor still isn't going to do anything unless there IS a schmidt trigger inline to the ecm. Then MAYBE it play off the trigger thresh-hold (whatever that may be) and push the time a bit, but I wouldn't think more than about 1/4 degree.

I gave this some thought back in the '70s (I was going to roll my own motorcycle ignition timing advance).

The first question is how responsive does this need to be? For reasonably constant RPM, the simple delay less than one RPM period to generate advance works, but if the motor is accelerating. You will always be playing catch up that way.

Can I assume you also want to perhaps retard this injection signal for improved economy, when not in a go faster mode.

Caveat: I am not that familiar with how diesels  respond since they don't use conventional spark plugs to time combustion firing.

It seems to me that this might be best addressed with a a digital processor. You could do a lot of math calculations, during the time it takes between ecm pulses. So it seems you could adaptively target a certain amount of advance and then correct on the fly based on last reading.  It would be useful to have a throttle position input to modulate when you want to make more power or optimize for economy.

I know very little about diesels so expect there are other complexities with emissions, turbo boost if exhaust temp/pressure is involved, etc.

Good luck

JR

> retard ... when not in a go faster mode.

I'm going to ass-ume this is a Race-Only engine, always "GO FASTER!!!" mode.

If it hauls beer 6 days a week and races on Sunday, go-fast options are very limited in favor of the daily pay-run.

I also note that the slack economy, and improved trucks, has left a small glut of older trucks sitting idle. The fuel used in 14 hour days can pay-off a 1MPG better truck even though the old one isn't totally clapped-out. What to do with a beloved truck you can't afford to run 600 miles day? Run it 1/4 mile at a time for trophies and glory.

> amplify the pulse so i could detect it sooner

John may correct me; you should also put a 'scope on the trigger.

At least some, your plan won't work. The voltage rises positive approaching the pole, then abruptly swings negative _AT_ the pole, then returns to zero. This is easy to detect very precisely, and the cross-swing is so quick that you can't "anticipate" by looking closer.

Why don't you take it off, fill the holes, drill new holes 2 degrees over? Or get a spare rotor, fill the key-slot, cut a new keyway? (Or cut new keyway 118 degrees around.) They'd have to look mighty close to see that.

A stock Diesel won't gain power with advance, except a wee bit at the top end. Since in racing you only run at the top-end, this may help, but you would only notice on the dynamometer or in an evenly-matched race.

The most likely way to "fake advance", short of mechanical twisting, is to sync an oscillator and take a trigger to give a few-% duty cycle between "normal" and "advance". As JR says, this will lag when RPM changes quickly. IMHO a heavy vehicle's engine will not change RPM quickly in one turn. Simple PLL techniques ought to keep-up. (Note though that PPLs can be treacherous and go un-locked for no reason. This might be disastrous.)

Is your RPM limited? CAT built that engine to run "one million miles" hard work. Raising the governor directly buys power (more air per minute) but wear increases far faster than power. In drag-racing, total engine life can be a few hours (hundreds of 20(?)-second runs) so higher RPM is warranted.

The absolute best way to illegally raise the power of a production Diesel is more fuel. There is a lean ratio which limits NOx production. There is a richer ratio where smoke is still invisible. When smoke appears you are using 95% of the available air.... combustion is over too quick to ever make full use of additional costly fuel. However where fuel costs and great clouds of smog are tolerable, you can go another 10%-20% before the over-wet mix starts to dampen the fire. The excess fuel is lightly toasted and will blacken the pistons, pipes, and sky, and the cost is way out of line with power, but this is a race.

The best way to inject more fuel is to diddle the screws on the injector pump, or go to a larger nozzle. I'm sure modern production Diesels (anything with wires) don't have diddle-screws. You should see if there is a larger model with higher flow-rate in the same size parts. Fer example my plowtruck has xxx-size gasoline injectors, I can buy XXX!! size injectors to fit the same throttle-body. Knowing how these (very low pressure) injectors work, I could open-up and file the end of the solenoid (putting-together would be the hard part).

Lengthening the injector pulse is simple in concept but hard in practice. (Diesels use electric injectors now??) "Pulse stretchers" are a common semi-digital thing. Logic probe senses a 1uS glitch and holds the LED for 1 second so you know it glitched. A fixed-width pulse *may* work here (total fuel/pulse proportional to air per cycle and the air/cycle is reasonably fixed.) However it is a very HIGH-energy pulse. And may "know" if it is diverted to a dummy-load. And you'd have to create your own pulse-driver which does not ever fail. (The stock pulse-driver is probably inside the "ECU" housing, which is probably crammed, so tapping-between is unlikely.)

You could try feeding natural gas to the intake manifold. Diesels, at least lower-compression ones, won't self-ignite natgas. If it did, fuel in intake air would ignite before TDC and reduce power or break something. However one maker is making NatGas Diesels where 90+% of the fuel is natgas. A "pilot" of fuel oil is injected in the normal way, and that small oil flame ignites the natgas. Propane might work; I just do not know.

If you can enrich the fuel, Nitrous Oxide is a classic cheat. It breaks-down to oxygen, which amounts to feeding liquid air (bypassing the intake valve restrictions on air in gaseous forum). It also cools the charge, which a Diesel may not like. It's also a very classic cheat in gasoline and nitro drags, and it is hard to hide.

Given ample fuel, anything you can do to increase the air-rate will help. Working Diesels are sized for long life, which means big slow pistons which wear slowly at design power. Valves cams and manifolds will then be designed conservatively to pass only as much air as the power specification demands. With wear of no concern, you should be porting and polishing and headers and re-grinding cams for a little more cam-advance (high RPM) and as much lift as will clear the pistons.

Diesels love blowers!! Production Diesels will, short-term, take more boost than any single blower can shove. Triple-turbos exist pushing many times atmospheric pressure at the engine. With corresponding fuel, power goes through the roof (or pistons go through the track).

> no power boxes, power chips etc.
> i can feed the pulse back to the ecm from whatever source

? ? ? There's many ways to fool ECMs for fun and profit. Pre-processing sensor signals is very standard. If "power boxes power chips etc" are banned, the inspectors presumably will be looking for any funny-bits in the factory sensor wiring.

Cat® C15:
Cylinders: In-line 6
Bore/Stroke: 5.4 x 6.75 (137 mm x 171 mm)
Displacement: 15.2 L (928 cu in)
Weight: 3090 lb (1402 kg)
Truck and Bus Ratings: 435-625 hp @ 2100 rpm
RV and Fire Truck Ratings: 600-625 hp @ 2100 rpm
Torque: 1550-2050 lb-ft @ 1200 rpm
King of the Hill Horsepower: 600-625 @ 2100 rpm
King of the Hill Torque: 1850-2050 lb-ft @ 1200 rpm

Hi bachevelle52

You might like to try how European diesels work nowadays.  A small burst of fuel is squirted in before the main burst.  This pre-heats the chamber so when the main shot goes in it burns instead of explodes/knocks.  I had a Renault Clio diesel a few years back that ran just like a petrol engine.  Not only was it quieter but its torque curve was wide, not narrow like an old style diesel.  The injectors were electric operated by computer controlled injection.

best
DaveP

Note that this engine has ratings from 435HP to 625HP. This is set by limiting the fuel rate (and turbo and possibly EGR and other trims). A cost-sensitive operation will use a bigger engine with a low max HP for fuel economy and low wear rate. Obviously a fire-pump needs all the HP which is safe for a few hours per fire. Learning how these tunings are gotten will be clues to how to tune further.

I later noticed the C-15 has a turbo blower. Fiddling the boost pressure is classic way to raise air-rate at the same RPM. Because a centrifugal blower's output does not match a displacement engine's need, there is usually a blow-off to limit boost at high airflow/RPM. Washers under the blowoff spring is a classic trick.

> no power boxes, power chips etc.

The ECU "chip" begs to be re-programmed. ROM is dead, it's all PROM, and maybe Flash. Flash you can just re-program. Even one-time PROM, you can get the same chip, sand-off the generic ID and screen a CAT part-number graphic, looks "stock" visually. Now you read a 625HP's fuel codes and a 435HP's ID codes into one image which reports itself as 435HP class but squirts 625HP of fuel. I do not know how much cross-checking the ECU reader can do, if you must also gimmick things like injector duration to put 20mS to the injector but 14mS to the code-reader report. (I've been pondering similar issues on my plowtruck ECU to run more to my liking.)

One thing which strikes me: this mill turns FAST. Yeah, 2,100RPM is 1/3rd of my Honda, but the stroke is 6.75". Comparing two engines of different sizes you should allow for how far the parts move to get rubbing-speed and air velocity.

Just cuz I know the numbers, compare to a 1967 Mustang:

C-15: 2,100RPM * 6.75" = 14,175
Ford 302: 4,400RPM * 3" = 13,200

(Engine linear speed is normally cited in Mean FPM; divide by 12" then times 2 for two strokes per turn.)

The C-15 is suggested to cruise (spend most of its life) below 1,400; the 'Stang would cruise ~~2,200. 9500 vs 6600. The Cat cruises much faster. True the bore is not scaled-up as much as the stroke so the piston mass is modest; however they are steel pistons and beefy to support Diesel detonation. They have surely improved cylinder rings finishes and oils since the 1967 'Stang was slapped-together.

The mean piston speed of the C-15 is 0.175 times the speed of sound, peak piston speed is 1/3rd SoS. If the head were wide-open, air would flow this fast and fill the cylinder. However two poppet valves tend to give a total intake opening like 1/10th the piston area. They are really asking the air to move fast. (However the torque/HP numbers suggest torque and cylinder-fill falls-off to 75% at 2100.) It needs that turbo to fill the cylinders at high speed, and improvement in valves cams and manifolds will improve air-rate (and with added fuel, HP).

I said older trucks may be a glut. The C-15 appears to be not that old (2007) and far from obsolete. I see rebuilds offered for $15K-$20K with a $5K core charge.

> A small burst of fuel is squirted in before the main burst

Inject cold liquid fuel into hot air. There is "ignition delay" before the first fuel breaks-up and starts to burn. Meanwhile more fuel has come in. Once fire starts ALL that early fuel burns at once. BANG.

A classic trick is a pre-combustion chamber, very solid, allowed to run very hot, with turbulent flow. Combustion starts faster in the hot turbulence, and the bang sound is reduced by small sturdy chamber, but it is ugly for production and somewhat wheezy for power.

The pre-squirt idea is tricky on classic mechanical injectors but may already be in CAT's computer-controlled injection. If the pintles will move that fast, a pre-squirt is just a few lines of code.

Most Diesels' torque curves are wide. But moreso on the low end than high RPM. Diesel combustion is slow. Diesels will run slow but not fast. Diesels are fundamentally costly (due to high peak strain) so are normally tuned for long life rather than high-RPM power. While that C-15 probably has 80% of torque down to 200RPM (ignoring turbo-boost and ECU programmed to avoid lugging stress), if you are well below top RPM and wanting more torque, the best thing to do is select a lower gear. Road engines don't need a lot more than 1.5:1 of RPM range to meet high-demand loading. They run 2:1 to extend down to low demand with low losses. Narrower in heavy-duty use with 16-speed gearing (C-15 offers 1.75:1).

And you thought we were all just Tube Guys!!

Some of us were hot rodders in a previous life!

best

DaveP

me> Diesels love blowers!!

Cat knows this. The C-15 has _two_ turbochargers. It is hard to see on the 2007 model but shows on the 2004 model:



Anyway: 2,000lb/ft torque from 900 cubic inches strongly suggests intake pressure above two atmospheres. That's possible (with tolerable loss) in one turbo at one engine RPM; the two turbos extend this to full RPM range.

Centrifugal blower(s) feeding a displacement engine makes not-enough boost at low engine RPM and/or too-much boost at high RPM. (This is even favorable to spark engines which knock more at low RPM.) The "natural" blower for a piston engine without knock problem is a displacement blower such as piston compressor or Rootes blower. But these have large friction or large leakage at high boost, and are heavy. Jet-engine technology has made the turbocharger the tool of choice for many engine applications.

For wide range of engine RPM they should be throttling the turbo. I think I see a waste gate on the #2 turbo turbine. At high engine RPM the large gas flow over-excites the turbines and makes more boost than the engine can use (within life and economy goals), so the waste-gate bleeds-off exhaust before the turbine. Jamming this linkage (and adding fuel (if the ECU MAP/MAF won't do so)) is a sure path to more power and less life/economy.

The engine is rated to 10,000 feet altitude which I guess means that the turbo will hold sea-level performance in thin air. This is reasonable.... low-altitude blown military piston engines were tuned slightly-rising from sea to 10,000 feet (hi-fly engines were peaked around 40,000 feet and had to be throttled to stay-together at low altitude).

The plots show considerable goal-fitting:



Dead-constant torque up to a point, then constant HP as far as valves and blower will support. The precisely defined torque allows closer specification of gearbox strength. The constant HP allows prediction of high-output fuel consumption. The rise in torque as RPM falls (i.e. constant-HP) is important on over-the-road trucks because it is forgiving on hills.

I am disappointed that no fuel curves are given. They are probably available after you select which of the many-many tuning options you desire. IAC, lb/HPh is fairly constant among competitive engines of similar duty.

Maybe you can pick up some cheap go-fast parts from their junkyard...? Probably not diesel, but ferrari did start out making tractors.

JR

> but ferrari did start out making tractors.

I think Ferrari is the one which DIDN'T make tractors. Lamborghini made his money in tractors, and used the tractor gearbox in some cars. Aston Martin was owned by David Brown who made major advances in tractor technology. Porsche turned out a few tractors.

Lagonda fooled with flame throwers.

Enzo was always a race-guy, as driver and as team leader. Early he re-designed small trucks into passenger cars. In WWII his shop made tools and airplane parts. Later he sold some road-cars to cover his racing expenses. Can't find any mention of tractors.

I thought there was a classic anecdote about using a tractor clutch in an early ferrarri, but who knows or cares that much (they probably do). There is/was a ferrari tractor company but probably a common name, or intentional play on the famous cars. 

That car crash is painful to look at the aftermath, lots of high Yen iron, trashed in a moment of poor judgement. Boys and their toys.

JR

I'll be darned: Ferrari tractor company

Company founded 1954, tractors a few years later, merged with big co in 1970s but still offers a range of sleek nimble machines. Main US op is in California. I think they get a lot of use in vinyards.

I can not find any connection to Enzo's Scuderia.

I don't doubt a tractor clutch, specially in post-war Italy. It was Fiats and tractors. Anything he didn't want to make himself, and needed something bigger than a Fiat part, tractor supply would be the obvious place for beefy parts.

Okay, first thanks to everyone for responding. I've been thinking about this on and off for a while now.

1st:  i cannot move the timing sensors in the block unfortunately.
2nd:  the inspection is visual, but also electronic. i cant make the 625 hp op file look like the 475. there is a 680 hp heavy haul package that cat offers if it were possible.
3rd: i've accepted the fact that i cant advance the timing without hacking the timing and fuel curves.
      (as an aside, i know a few things about diesel engines as ive been a mechanic and driver for years. there are 3 things that make a diesel run: air, fuel and timing. to make real power, you need them in the right amounts. i asked here because of the electronic nature of what i wanted to do)

PRR: you are right about advanced timing only really helping at higher rpm, however, in the past 15 years as the epa has tightened the nuse on emissions the big 3 engine manufacturers have slowly, systematically retarded their engine timing.
Also, the c-15 made it debut around 1998, with the best pre-emission motors being manufactured around 2000-2001. the engine you showed us was likey what they called the accert. those 2 turbos look great on there till you realize its a series configuration, rather that a compound. 

There is a place in Utah called PDI that can custom program the ecms with any curves they like, just not sure how they do it. That would be a racket, we'd just run unlimited class. any insight on how they cracked cats programming? just to flash the ecm to higher hp rating takes a key code from cat. i thought of a microcontroller to lengthen the injector pulse, dont know how much delay that would add. could add all the fuel i wanted that way.

In fact, the more I think about it the more I like that option. Catch the ecm injector pulse, use it as a trigger. 6 injectors, 1 crank position sensor, code it to calculate rpm and act on injector pulse accordingly. Nice and neat, and it would piss off cat.

Those microcontrollers have clocks, no? Shouldn't be that hard to have it calc rpm. You really could map your fuel then, and you could turn it off. The more I think about it, advancing the timing would probably help lower cyl pressure and exhaust gas temp too. Is this possible? I usually dream big and I'm not afraid to learn something new in the name of horsepower.

:Bump:

I was thinking about buying a data acquisition system so i can get a better idea of what the fuel and timing maps look like on this motor. Anyone have any DAQ experience? I was thinking the Picolog 1216 would work. I just want to be able to graph the timing sensors and injector pulse against each other. I'm thinking if i get enough data i could get a microprocessor to add the fuel i want when, and under the conditions i want. Any ideas, advice?

> data acquisition system so i can get a better idea of what the fuel and timing maps look like

All that stuff (and turbo boost!) is surely already known to the ECU.

For Hondas and Chevvies, there's data-taps and readout programs to expose all this data to the mechanic/tinkerer. http://www.tunerpro.net/screenshots.htm

For a CAT, I just do not know.

Additional complication because evidently CAT semi-locks their ECUs so that authorized personnel can change the duty-rating but mere users can't.

Yes, Cat does require a code to update (flash) the power ratings. I'd like to look at the data so i can better understand the injection timing with relation to the crankshaft position. I cannot trick the ecm directly, however i can alter the low voltage trigger signal to the injectors themselves.
There are already products like this available to be sure, however they are pricey to put it mildly. Beyond the economic issue, they are not very user friendly. You can add power in 25-35 hp increments but the additional fuel is added across the rpm band. From idle to 2100, it's overfueling. This dramatically increases exhaust gas temps, and without altering the injection timing, increases peak cylinder blast pressures.

In a perfect world Paul, i'd use the ecm fuel and timing tables for stock operation. Code a microprocessor to look at the tach signal from the ecm to the guage to check rpm, then look at the throttle position sensor. (something below 20% throttle would just bypass.)  Given a high/low signal from and on/off switch for bypassing, and a high low switch for additional fuel, i could add the fuel i want where and when i want. In fact, if i could code it to sample and calculate the crank position sensor, i could advance my injection timing.

High low fuel operation would look like this: over stock, in low setting, fuel would increase starting at approx 1000 rpm, rising thru and peaking around 15-1600 rpm and ramping down thru 1800 or so. That would better match the fuel needs for spooling the turbo sooner, and having additional power where needed in common driving conditions and engine speeds. High setting being noticeable increase ramping from 1300 to peak and level off at 15-1600 and remain constant thru 2100 rpm giving plenty of fuel for hard pulling heavy loads. This is where you would benefit from advanced injection timing, as you are lengthening the injector pulse effectively changing the dynamic compression cycle.