I have a crazy idea...

I have an idea that I've been thinking about for a while. It's a bit crazy, and uses a PCV & Autotune. Bear with me.

OK, the basic function of the PCV + Autotune is to intercept the injector pulse width commands form the ECU, add or subtract the values in the PC map and then apply the trims from the Autotune before sending the pulse commands to the injectors. The main issue with some of the approaches to using the PC is that the assumption is that 8V is underfuelled from the factory. This is a misconception.

The factory sets the map rich and then trims the heck out of it using the narrowband sensor to get it to pass emissions testing. It then runs a bit ordinary and well, underfuelled. The common fix is to pour fuel into it and say 'fixed'. The 8V does need a more rich AFR than the narrowband and factory trim allows and here is where the Autotune can come in handy. As it uses a wideband controller to trim to a set AFR value, we can use this to our advantage.

Here's what I propose: Flash a slightly modified factory map to an ECU with lambda off, and a couple of other tweaks. Install a map in the PCV with values of zero, i.e. the PCV map does not modify the injector values. We then use the Autotune to trim the base zero values in the PC map to effectively trim the ECU map injection pulse widths to produce an AFR value of our choosing.






Well?

I dont know much about this stuff, but would it be possible to use the autotune connected directly to the ecu to manipulate the settings directly?

Your one crazy beetle....

Crazy? Undoubtably! But the craziness works on a level of sense and logic that eludes me and certainly a lot of others who purport to be *Experts*.

I find myself being invited in to the world of a very interesting mind. I also find this far more interesting, challenging and worth while than simply accepting an 'Off the shelf' solution from a large company with a reputation that doesn't really have any interest in OUR bikes.

When you stop learning you begin to die. I fully accept that a lot of the stuff Mark talks about I find challenging. The difference is that acceptance of knowledge and wanting to learn and bare-faced denial of the evidence in front of your face.

I like being alive. I also like being an accessory to the generation of fun! Bring it on! Let me learn more!

Pete

Ahdammit, the answer is no. If my level of nerdiness would allow me to hack the 5AM software to allow direct connection of a wideband sensor, I would do it in a heartbeat. At this time there is no way to connect the Autotune directly to the ECU.

I conjured up this idea in the wee hours one night when laying awake thinking. At first I just considered it an exercise in lateral thinking, but I've just suggested it to a guy who wants a map for a setup that I don't have a map for.

I would build a base map with lambda off and with the same ignition advance as all my other maps, as well as the pressure-temp tables. The main map would simply have a simple across the board reduction, maybe 10%.

Hardware wise, the overall length off the headers and midpipe would still need to be factory. This is critical with regards to the delta-fuel map, but you could stick an open pipe on it and let the Autotune do its thing. Ride for a few hundred kilometres and you'd be sweet. Theoretically.

While I have absolutely no idea what language you speaketh, if some nerd tells you it can't be done, or won't work. I say, "Bullshit"!

It just means they haven't thought about it long enough to come up with the solution.

Keep going Beetle

Hey, here's a dumb arsed question.
Can you adapt telemetry/data logging/memory or whatever to the A/T, thereby getting a list of settings and changes it makes to attain the pre-determined AFR across the rev range.
Then use that data to just write the exact map for other people's bikes.

That way there are less bits to go wrong, as in, Less is More!

Well your proposal is not so different from what the pc5 with autotune does. The only thing is, they offer pre made maps with the values that you try to capture.
Calculating the pc5 values to the 5am map is another problem since break points aren't the same.

I like the logworks route more. That can calculate you a map proposal too, based on the measurments

Ghezzi, how do you think I've been making maps? With regard to the Autotune, what Paul says is 100% correct. All the breakpoints are different and it takes some guesswork interpolation to convert a PC map into a real map.

I don't actually want any 3rd party black boxes in the mix, but this crazy idea is for situations where there isn't a pre-made map that will work for a particular bike.

I found myself with a test case when a chap with an 8V with a very open pipe and running an open airbox wanted a map. He has an PCV and AT already installed, and has the typical case of crap mileage because it will be running rich. I don't have a map that will work for him. This is quite serendipitous as I was thinking about this idea for a couple of weeks, and then voila!

But why is it running rich, in the pc5 you can set target afr I think, why does it run rich then? He has lambda off? Or is this a case where he also has a lambda fooler in the mix?

Paul, the Autotune has a limit on how much it can correct. I think it's 20% if I recall correctly. If the PC map or the ECU map are set too rich, the Autotune can't correct enough to meet the target AFR. The Autotune is a good concept, but the maps have to be right to begin with.

Ok but if basemap is to rich it will end up as -20, so there are the ranges you know should be at least 200% lower in the map, before continuing. But 20% is pretty much to be wrong.

Correct. That's why I sent him a map for the ECU and a PC map with achievable AFR values.

The only reason I'm so anti PCV/AT is because of the many, many cases I have observed of bikes over-fuelling, sometimes massively, when one is used. While my experience of this is limited to Guzzis tales of other vehicles fitted with the system behaving the same are legion.

While some of it may be down to it simply not being able to correct enough on an over rich map I can't see this always being the case. I don't have the requisite skills or ability to actually work out thewhy and the how but given that even simple O2 sensor foolers can trigger disastrous over fuelling at certain points under certain conditions I have been convinced that it is far better to try and avoid relying on a self-tuning interface or anything that modifies the map of its own volition as I don't think we know enough about how the various sensor inputs are interpreted within the tables in the ECU.

This is not being flung out there to muddy the waters, this is you guys' area of expertise not mine but having seen engines destroyed I urge the greatest caution.

Pete

Well Pete, self tuning looks like an interesting feature for people who don't have the knowledge, but problem is they never look what the system is doing, because A) they don't know how to interpret the values and B) they rely on the system to do a good job. And if you see what it costs, I don't blame them.
From what I hear, tuners who do have a dyno love the pc5 because you can put a bike on it and dail it in while running. Doing the same with a 5am and making maps takes much more time. And when time is money...

I'll freely confess to no knowledge of fuel injection computers, but all this additional hardware gives me the hives !
Basically, you're proposing a sensor (the Autotune) to modify the calcuations of a computer (the Power Commander) that modifies the output of another computer (the OEM 5AM fuel injection computer) ??
Isn't this what is wrong with carburetors ?? They're basically a box of corrected mistakes. Jets and tubes to correct errors introduced by the other jets and tubes etc and repeat
(Pete or some grizzled veteran can explain this better than me)...

Less is definitely more....

And - can the processors keep up ?? Am I right in calculating ?
6000 RPM = 100 Revolutions per second. So there'll be 50 firing pulses/second.
Each one is 1/50second = 20 milliseconds for the whole sensing/calculations/modifications process plus actually generating the spark and propagating the flame front.

Bloody Hell, when you look at it this way it's a miracle it actually works.....

Jeremy, this would be the last resort for folks who have drunk the Power Commander Koolaid and installed ridiculously loud open pipes and rock strainer air filters with an open lid. They want to keep the pipes and rock strainers and yet want a map that will work. Sorry, no can do. I can provide a solution that may prevent overfuelling and keep fuel consumption at sane levels.

If I had my druthers, I'd have all these black boxes tossed in the skip. But if I can help some peeps get a reasonable result due to the blinkered vision of others, I'm happy to have a crack.

Oh, and 20 milliseconds is an eternity to a computer. They operate in nano seconds.

I am starting to come to the conclusion that the PC/Autotune/open air box/open pipe setup is the "full race" setup for our bikes, where ultimate power is the goal and mileage and longevity are no longer a concern. Because a race bike/track bike doesn't have to worry about stranding its rider miles from the nearest fill up, and will get rebuilt regularly. Am I off base with this observation?

TalkingGriso wrote:

I am starting to come to the conclusion that the PC/Autotune/open air box/open pipe setup is the "full race" setup for our bikes, where ultimate power is the goal and mileage and longevity are no longer a concern.  Because a race bike/track bike doesn't have to worry about stranding its rider miles from the nearest fill up, and will get rebuilt regularly.  Am I off base with this observation?

well, don't see what pc5/autotune has to do with a full racebike?

The thing is it isn't that simple. There is no *Magic Bullet* for 'Ultimate Power'.

All an engine is is a self propelling air pump. Getting it to produce its best depends on three things, mechanical efficiency, thermal efficiency and volumetric efficiency.

Mechanical efficiency can be improved by reducing weight and friction, thermal efficiency is dependent mainly on the motors ability not to detonate, you want it as hot as possible. The greater the heat utilised from burning the fuel the greater the expansion/pressure of the gas in the combustion chamber and the greater the force exerted on the piston. Torque is force times distance. Power is torque times revolutions per minute!

The third one, volumetric efficiency, is a much more slippery fish. Simply 'Opening up' the intake and exhaust sides does NOT necessarily increase volumetric efficiency. Sure the idea that you can have less resistance to the air coming in and the spent gasses going out is attractive in simplistic terms but it is just that, simplistic. It does not take into account different head designs, porting, combustion chamber and piston shape and probably most critically cam timing.

To get the best out of an engine at any rpm requires the best cylinder fill possible at that point and the correct Air/Fuel ratio. Because all engines are an enormous bag of compromises there can be no hard and fast rule on how to achieve this. Where the AT falls down is that it depends on ONE input to make all the decisions about fueling and because of the charge transition issues at certain engine speeds and loads it seems to often get this wrong. A problem greatly exacerbated by open pipes and, worst of all, open airboxes.

Others may disagree. That's fine. I'm just reporting what I have seen not once but over and over again.

Pete

+1 on what Pete said.

You also need to consider what the exhaust system is doing as well. Two stroke engines have discovered the benefits/problems of an expansion chamber and this also applies to 4 strokes, to a lesser extent. An exhaust system will rarely work efficiently at a wide range of revs, normally just a narrow rev range - this is what 2 strokes take advantage of when they hit their power band: nothing, nothing, nothing, EVERYTHING - front wheel is in the air and the rider is mouthing the mating call of a crow - "FAAAAAARRRRRRK".

4 strokes can use the same principal but it is a further compromise to the volumetric efficiency as stated by Pete above.

"Opens can of worms"...

pauldaytona wrote:

TalkingGriso wrote:

I am starting to come to the conclusion that the PC/Autotune/open air box/open pipe setup is the "full race" setup for our bikes, where ultimate power is the goal and mileage and longevity are no longer a concern.  Because a race bike/track bike doesn't have to worry about stranding its rider miles from the nearest fill up, and will get rebuilt regularly.  Am I off base with this observation?

well, don't see what pc5/autotune has to do with a full racebike?

Seems that a lot of race bikes run Power Commanders. A PC5/Autotune Kawasaki won Pikes Peak this year.

@Pete: I was thinking that anyone building a race engine would take those elements you mentioned (cylinder head, cam timing, etc) into consideration when trying to get the most power out of a GRiSO without regard for fuel efficiency or accelerated wear. I thought those last two items were the main objections against the use of PC/Autotune, and they were caused by over fueling. I've been trying to keep up!

This would all be fantastic, if only the GRiSO were a race bike...

Some looney guy on apriliaforum rides his GRiSO on track often. I think he is amateur racing in some dinosaur class. But he's hanging it out there.

DM

Hi guys,

I'm fairly new on here but used to mess around with this stuff on a supercharged Miata. I have a bit of a problem with keeping my hands off engines...

Been doing a bit of reading about the guzidiag tool and am amazing the standard ecu is so flexible. Surprised that there's not an easy way to integrate a wideband AFR though, although i didn't previously see it as a necessity on NA engines.

Anywho, the favoured solution to these issues on the Mazda was to install a stand alone after market ecu (see megasquirt, MAP2ECU). This gave a fully programmable and expandable solution (e.g. add oil temp sensor, AFR, induction air temp, etc). Have any of you guys gone the standalone route? Does such a solution exist for motorcycles? I'm guessing one of the of the shelf standalones would work, assuming the guzzi is a wasted spark engine.



FB

Wasted spark works on in-line 4's but it still uses 2 ignition circuits. Nothing wasted on a Guzzi, except money, maybe.
Have seen a Link Atom fitted to a Stelvio, due to crash damaged instruments etc.
But Beetle can re-write ignition maps as well as fuel.