3 BAR MAP Sensor

[GeoffSmith]

Not trying to but in chaps, but how did Costas do it ?

If I recall, his technician burned a special chip for the standard computer. I saw a thread somewhere and will add a link here if I can chase it down.

You're right. How well (or otherwise) he developed it, I'm not sure (and neither was Costas). I was going to ask to talk to the Sh4rk for a couple of minutes if Costas had made Donny this year!

[tigerdog]

I was going to ask to talk to the Sh4rk for a couple of minutes if Costas had made Donny this year!

That's a great idea! Alan is trying hard to convince me of the folly of this path (and with some success) but if we have some solid empirical data from which to extrapolate, it might be slightly less of a daunting task.

[GeoffSmith]

Actually TD, I'd be more interested in Alan's interpretation of the PROM. Mine originally had a Superchips PROM costing a few hundred quid and only a few bits were changed to suppress the fuel cutoff. I'm not saying that is what has been done to Sh4rk, but Alan would be able to accurately assess what had been done.

[tigerdog]

Yes indeed, Geoff. Amongst us all, only he has the requisite skills to make sense of the data. I certainly don't! And to everyone following this thread, please be aware of what a massive undertaking it would be. To accomplish what he's done, Alan has obviously done a HUGE amount of work in disassembling the code. To help us all appreciate the herculean task, here's a thread that speaks to some of the challenges in an undertaking like this. Be forewarned, the main contributor is a bit forceful in his opinions and some of his info is subject to debate, but you get a sense of the magnitude of the mountain Alan has climbed for us all.

[GeoffSmith]

Nice link TD - lots of $ex to look at!

One of my summer projects was to work out what was going on in the 8 pin communications link to the ROM module in the back of a Texas TI58C to enable external data to be loaded. I'm please to say that after 6 weeks work, armed only with a primitive logic analyser, a pad of paper and a few sharp pencils, I succeeded. Should have put me off but might explain a warped mind.


Calculators as they used to be!

[tigerdog]

I hope the other community members appreciate how wonderful it is for our (relatively) small community to have such talented software and hardware people in our midst!

[dapinky]

Ideally, we'd find someone who makes 487.5cc injectors (1/3 more than 325) but RC offers only 440 and 550 options.

Firstly, I must explain that i know NOTHING about the actual construction of injectors - but i do understand the theory of Carburettor jetting (I'm an old fart!)

In days of old, when i have been ttrying to achieve precise mixture settings on Webber or Del'Ortos i have drilled out smaller jets to give exact flow rates, is this approach possible with an injector?

If the flow rate is atributable solely to the size of the hole, then a 325cc unit could be drilled out to give an exact (ish) 487.5cc by using simple mathematics to decide which drill bit to use!

I think there is probably more of a scientific method of construction these days, with shapes/slopes and curves inside, but Ihope someone will have an answer to this question - even if it is just to shoot me down in flames

[Ashaman]

As a simple answer, you NEED a Haltech to do this.

As others mentioned, a 3 bar map will allow for up to ~30Psi of boost, but you have to have an accurate way of taking the readings of the 3 bar map and controlling injectors capable of delivering enough fuel at those levels. Of course, timing is another issue...fun.

Point is, I know of no way of precisely or accurately doing these things aside from a standalone ECU. All others are band-aids and/or courting disaster, IMO.

The cheap way of getting around this is by playing with fuel pressures - but to me that is scary. Its a pretty crude way of doing it, but you might get away with it. Sure it can be rising rate, and you can "tune" with it to some degree. The trouble is, what happens when you wind up rich at one range and lean at another?

By contrast, the Haltech will allow you to quickly and easily adjust fuel or timing for any RPM or load point.

Of course, I am not saying other methods won't work, just that they are "hope for the best" type deals. I am not the expert on this stuff, but I have sure spent enough time to know it is complicated enough without flaky fuel/timing methods to contend with.

'Course, if someone has the ability to come up with a way of re-programming the stock ECU on the fly with a nice GUI, then great! I think I was told a while back that would be quite a task though...

[Dico]

Geoff can you make that piggin picture smaller

[GeoffSmith]

Geoff can you make that piggin picture smaller

Wow! That was big.

Seeing as it's annoying you Paul, NO!

[Dico]

[GeoffSmith]

'Course, if someone has the ability to come up with a way of re-programming the stock ECU on the fly with a nice GUI, then great! I think I was told a while back that would be quite a task though...

I believe certain people are. In fact, I think Charlie is doing it over a wireless link on his Elise.

[tigerdog]

As a simple answer, you NEED a Haltech to do this.

'Course, if someone has the ability to come up with a way of re-programming the stock ECU on the fly with a nice GUI, then great! I think I was told a while back that would be quite a task though...

The 3 bar sensor and 550cc injectors are working now, with the stock ECU program, in closed loop mode. Maybe not perfectly, but close enough that the engine seems to run very well and pass California tailipipe emission testing. You surely need some kind of programmability to optimise all of the curves in real time. I found a few 'net references to people working on similar projects with GM ECUs and PROM emulators, but haven't had time to research further.

[Ashaman]

Yeah, it will run - it has been done a few times before. I'd love to see a wideband log from a good, hard, run though

You could probably make more power with a 2 bar, stock injectors, one of the aggressive chips.

[tigerdog]

I'd love to see a wideband log from a good, hard, run though

Anyone care to loan a wideband for a good cause?

[tigerdog]

Anyone care to loan a wideband for a good cause?

That certainly killed this thread in a hurry.

[lotusgirlmarie]

You'd be amazed what can kill a thread lately...

[Ashaman]

Why not schedule a dyno run? That should answer any questions, they usually plot AFRs.

[tigerdog]

Why not schedule a dyno run? That should answer any questions, they usually plot AFRs.

Good idea. Smaug comes home from the shop tomorrow with a new timing belt and associated rotating bits. I'll no longer be afraid to drive the car, so at a minimum I'll produce some ElanScan files soon.

[tigerdog]

OK, still no dyno run. Life is just too busy and Smaug is too much fun to drive. Maybe in the spring but for now, my next project has GOT to be the wilwoods. Marie has also fitted her 3-bar and 550cc, apparently. Marie, are you using a Haltech or the stock ECU?