[decipha]

as some may already know i'm making my own engine computer, im trying to incorporate as many failsafes as possible, i already have a handful of failure scenarios in the programming to prevent engine damage, i'd like to hear more stories of engine failures if anyone has any. This is important as the 'user' has no way of changing any values, so if the computer can't identify a problem there's no way for the user to let the computer know there is one. Thanks in advance.


edit: the user now has the ability to change any values she/he desires using the manual override controls from the menu on the cluster

[dualdj1]

building it from scratch? I'm interested in hearing your progress. How comprehensive are you going to make it?

[decipha]

im writing the program in c++, i'm going all out in depth to the greatest extent of my abilities applying all my experience and knowledge in the programming. It does everything every engine computer on the market now can do and much more. I'm writing it to be 100% universal, so it can virtually run and self tune anything, even a diesel.

I've even concocted a way to bypass the crank input!!! If your crank sensor/cam sensor/ and/or distributor fails and you loose the pip/tach signal, you can still drive it!! Even shut it off and restart it!!! Even WOT it and it won't make any difference!!! Injectors would still fire sequentially!!


I still got a bunch of coding to write but the major calculations are done. As I near closer to finishing up I'll put the full specs and capabilities on my website but until then I need to hear more engine failures so I can be able to hopefully prevent it from happening by putting failsafes in the programming. As of now I feel pretty confident about the failsafes I already have in there, but to prevent more would be even better.

[dualdj1]

Very cool. Maybe you could list your failsafes you already have (just generically), so if someone sees something that's not in the list, they could point it out.

[decipha]

I'd rather not, as its nothing but variables so putting it into context would take a lot of typing out variables to 'scenarios'

but in short though, i shut down the engine on

overtemp ect/act/tot/oil
low oil pressure
internal transmission slippage (w/o use of trans brake)
lack of temp, inferred temperature (a memory based on past engine cycles)
high carbon monoxide levels
fire detected
running lean at high load
excessive knock detected
loss of fuel pressure (fuel leak)
throttle pedal sticks open
over-rev & valve float
overvoltage
brake failure
cam and crank unsync (possible failed timing chain)

thats just off the top of my head, i spent a great deal of time writing just the failsafes, over 200hrs of code and theory on failsafes alone

if anyone has anything to add don't hesitate im open to anything and everything

[85GT]

I think your engine safes, need to be prioritized as second, compared to 'people' safes. Unless you're intending this for 'off road use only.

[decipha]

how are people's "safe's" effected by the engines failsafes?

[85GT]

An example would be like need to make a rapid life saving hard manuver to get you out of the path of a head on with a truck. Last thing you want happen is the engine gets shut off in the midst of it because of a momentary drop in oil pressure. Which could have happen cause you just happened to be just a bit low on oil that day.

It's these kind of stupid 'what if's' you might want to consider.

Sometimes loosing an engine is small price to pay. That's all I'm trying to say.

Similar decision I had to make when I converted to FI. Inertia switch or not? Back in the earlier mustang efi days, I knew a few people that got stuck because the got bumped in the back. Resulted in a tow as they didn't know about the switch. So, get hit and be safe that fuel doesn't keep flowing if there's a fire, OR let the engine keep running if it lets you move out of the way of a second hit? Two choices, neither is best for all cases.

[85GT]

Maybe a G sensor that senses hard maneuver vice impact. Let it override an impending shutdown for a few seconds. Or adding the right time delays for the various failsafe conditions to help avoid the wrong shutdowns.

[decipha]

no faith? I have a 10 second delay on engine shut off with loss of oil pressure, as with all the failsafes as soon as a condition is recognized it displays on the 10"x5" lcd gauge cluster with an override function to bypass it, if the oil pressure loss isn't bypassed within 10 seconds the engine shuts off

[dualdj1]

I think 85GT had some good points there, which I'm glad to see you've covered. Think you've got most of the bases covered though on the engine side.

[decipha]

i think so too, i already have an accelerometer present in the dec, its used to compensate for a bunch of calculations.

I'll leave the inertia switch to being mechanical, I hadn't even considered that as being part of the engine programming.

[byrne331]

This sounds like quite an undertaking, much props to you. You could be on to something great here. Having everything run through a nice LCD touch display will be titties. If you need a beta tester im just right here in Houston, lol... My only engine failure was from a bad harmonic balancer, I don't think that helps much. It was beautiful destruction though, quite an impressive mess.

[decipha]

its not difficult just time consuming

i don't foresee beta testing anytime soon

[2Shaker]

As chronicled in these forums a few years ago I burned 3 pistons on the last pass of the evening. My son was driving and we made a hot lap as the track was ready to close. I had compared CBAZA W4HO to J4J1 (I was basing on W4HO at the time) and noticed J4J1 (Cobra) did not richen the mixture at high ACT's. As I recall we were launching at close to ambient (around 85 deg.) and at the lights it was close to 180 degrees. I also noticed a lot of very black smoke at the lights and without benefit of a wideband I simply removed the enrichment function for the last pass. That's all it took. I saw a puff of oil smoke out the driver's side on that last pass. When my son drove it back to the pits he told me it made a bad noise in the lights and it skipped for a second or so. Well I revved it up and there was no sign of any problem, not even oil out the dipstick although when he got back to the pits the dipstick was pushed out a ways. Anyway we drove it home and subsequently drove it a couple more times before winter set it and then I tore it down for a refresh. I was amazed to find number 2 and number 7 had burned so cleanly there was no debris inside the cylinders even though both pistons had burned about 1 inch at the top of the piston. The molten aluminum simply exited through the exhaust. Number 8 was a different story. It too had burned about an inch but the top ring had also broken. While there was no debris at teardown it was obvious from the dents in the piston and head that bad things had happened.

I took too simple an approach to tuning and failed to figure out how exactly J4J1 deals with high ACT. In that calibration they apply spark retard instead of fuel mix enrichment. Boy have I learned a lot more since then. Hope this helps.

[decipha]

thanks for sharing warren, thats the type of info im looking for

do you think you could have avoided this if u had a wideband?

[2Shaker]

thanks for sharing warren, thats the type of info im looking for

do you think you could have avoided this if u had a wideband?

Actually I mis-spoke above. We have a freestanding wideband from Western Motorsports which has its own software, but it is quite difficult to use and its almost impossible to sync its log to an EA log. It also has a digital display which we mounted near the steering wheel but it is hard to read, especially at 120MPH and with only a couple seconds to do it. I asked my son if he saw the WB values in the last couple seconds of the run and he looked at me like I was crazy so I took that as a "no". So if we had paid closer attention...

It would be cool if we had some watchdog SW running in the background that monitored a WB if present. Maybe we could think of a few other dangerous conditions it could monitor and if any were "true" it would kill the ignition requiring a restart and report a reason code.

[decipha]

that scenario couldn't happen with the dec

[85GT]

So it's got a name already, cool.

[decipha]

d.e.c.

dynamic efficiency cognitive

thats what i got so far

[copcar]

Not sure if this counts but I'll chime in. Long ago, and I mean a LONG time ago, My car was running 24#ers and a C&L maf and blew a head gasket with 5* advanced at the distributor. Funny thing is the thing blew leaving a parking lot at partial throttle. Obviously the damage was done and, cumulative over time. Perhaps a failsafe at WOT for a lean condition using wideband O2's.

[decipha]

"running lean at high load"

she's already in there, appreciate the input though

[copcar]

I need to brush up on my reading skills. any eta on a release? and ofcourse will it plug into the factory harness? would love to run it if it does.

[decipha]

no eta as of yet, yes it will plug into the factory harness simply unplug the stock computer and plug this one in

still writing code, this sucka has a butt load of calculations going on, still got to work out the diesel's and 2-stroke engines


supports all internal combustion engines up to 20 injectors
20 injector drivers control both low and high impedance injectors
injectors do not need to be of the same size, dec will calculate individual injector size (not recommended)
ability to run multiple injectors per cylinder and/or manifold injection (tbi)
supports sequential direct port injection and throttle body injection in sync
flow matched injectors are not required

average maf element in a 4" diameter pipe is sufficent to support most engines up to 1500hp
maf sensor does not need to support maximum horsepower output
dec compensates if maf sensors pegs
'calibrated' maf sensor is not required
maf sensor not needed (maf recommended)

oem type bosch 3-bar map sensor is recommended

oem type bosch bap sensor is recommended (where neccessary)

supports idle air control valves (air bypass) and drive by wire throttle bodies

supports all positive-accel throttle position sensors

supports all automatic transmissions with up to 4 shift solenoids & 11 gear ratios

supports dohc variable cam timing up to 4 camshafts

supports ford/gm/EDIS/COP/TFI/Distributor ignition types

supports all bosch type knock sensors

supports all bosch type oxygen sensors

includes 2 built in wideband controllers

supports power adders and fuel adders, meth injection, nitrous, etc...

supports 2 exhaust gas temperature sensors

supports vehicle speed sensors as well as calculated speed via output shaft speed
dec calibrates speedometer

full lcd gauge cluster, all gauges can be custom adjusted, displays all engine parameters
manual override and performance review (full menu)
estimated hp/tq 1/4 times, bsfc, fuel flow, ve, bmep, the works!!!
and warnings 6x15 lcd touchscreen

supports oil pressure input

supports variable speed fan, dual speed fan, electric water pump

no lift shift

supports return, returnless, multi-speed, and variable speed fuel pump systems

hassle-free simple plug in harness available for most factory vehicles -no wiring required
direct plug in harnesses available for many vehicles

boost control, overboost protection

1/4 1/8th 1/2 and 1 mile desired limited et for bracket racing

launch control/traction control to minimize 60 foot times

tune override manual control

controls glow plugs for diesels


this is all that i've got so far, still working on it, i want to incorporate cruise control and abs into it as well, in all honesty if i keep pursuing it i think it'll be a couple years before im completely done, since im doing this in my spare time for 'fun' i won't be in any hurry to finish up

[copcar]

Sweet! Keep us posted.

[dualdj1]

this is all that i've got so far, still working on it, i want to incorporate cruise control and abs into it as well, in all honesty if i keep pursuing it i think it'll be a couple years before im completely done, since im doing this in my spare time for 'fun' i won't be in any hurry to finish up

I think that's an understatement, even assuming you've made some good headway already.

Still a very ambitious project, and think you've hit on a lot of good areas, if you can pull it off

[decipha]

no "if" about it, even 'if' i stop now i have enough of it done to do everything+ what the factory ford computer could do up to 2004, so either way its not a complete failure

[Cougar281]

Holy Smokes this thing sounds neat!

[vristang]

do you think you could have avoided this if u had a wideband?

Maybe a crankcase pressure sensor?

I had a piston failure before as well.... wasn't 100% what the problem was until I did a compression check...
Could also be handy for guys running evac systems... especially if the crankcase pressure could be controlled in closed loop...

[vristang]

Now that I think about it... it would be nice to have integrated electric water pump control, similar to what we currently have for fan control...