FreeEMS firmware feature wishlist! (out of date)

[AbeFM]

Skimming the last couple pages, a comment or two on stuff I didn't read:
WBO2 per cyl is FINE for NA, remember, most NA headers are pretty long, so you don't need to put the sensor right in the port blocking it!

Also, VVT can't be PWM vs RPM, well, to work well, you really want to have oil pressure in there. That should be pretty easy, though if someone lacks this, oil temp, or even water temp, should be enough to get you something predictable. Without it, I'm sure you'll have issues.

Abe, total tune switching should solve your issues. You can use table switching to do this. Your entire tune could be different, or just one little bit of it. etc. You could combine the lean one with a lower boost cut etc if you wanted to get flash Otherwise it should be a piece of cake to hack in a condition like that

I was thinking about that - honestly, I was turned off due to MS's poor handling of table switching (so I've heard). But that's not really the key - it's literally turning off EGO all together. I'm still a bit torn on this, but I think both would make a certain amount of sense.

Of course, a "points in space" style architecture would well support a virtual continuum of TPS influence.

[Fred]

Many things in this thread are outright wrong :-) Many of them posted by me a year ago. At some point I'll need to do a new one with implemented stuff and priorities for the unimplemented stuff.

People serious about getting their stuff done should do a mantis issue for it.

Fred.

[MotoFab]

Also, the probability of hitting the filled space drops as the code grows. right now 50% is used and therefore your hit rate is 50%, when we are using 90% (soon enough) it shall be a 10% hit rate and near useless.

I get that. The ruggedizing features work together to increase the success rate. The 'call counter' is another example. Due to a 'physical error' in the processor, the program counter gets corrupted and jumps to an unknown portion of the code, right. If it jumps to an 'filled' unused section, great. If it jumps to a used section, the call counter may catch it.

The call counter implementation doesn't have to be reserved for only calls. Incrementing a register at various places in the code, then comparing the register to a 'you-should-be-at this-value' value, also catches physical errors. Those types of code catches can be placed in sections of the code where the processor is operating a potentially noisy external peripheral.

Whoa. I had no idea the code is filling 256KB of program space.

- Jim

[MotoFab]

After recently enabling EGO correction on my MS, and picking newer, richer targets for "near idle" operation like 100 mph super-cruise, my mileage went up from a VERY consistent 17 mpg to something which seems to be bouncing between 25 and 29 mpg. This 70% reduction in vehicle operating costs has, um, made me very happy.

So my ultimate system, aside from having great, transparent EGO I could leave on all the time, would be to have four conditions which would turn it off:
Below ___ RPM
Above ___ MAP
Above ___ TPS
When a hardware switch is on.

I was thinking about that - honestly, I was turned off due to MS's poor handling of table switching (so I've heard). But that's not really the key - it's literally turning off EGO all together. I'm still a bit torn on this, but I think both would make a certain amount of sense.

Of course, a "points in space" style architecture would well support a virtual continuum of TPS influence.

Up to 29mpg, nice.

Lean of peak is a great way to operate the motor. The 25mpg is a 'best efficiency' afr, while the 17mpg is closer to or maybe even above the 'best power' afr.

Switching between the power and efficiency maps can be done using the various thresholds you mention, but anytime the motor is operated in a 'static' condition is a good time for a best efficiency afr. No reason why a high RPM, high MAP, and high TPS high speed cruise shouldn't be at best efficiency afr. The motor runs cooler, and particularly under those conditions, the exhaust valves are cooler. The exhaust valve temperature and heat dissipation being perhaps the most significant limiting factors in motor HP.

Whether operating at either best efficiency, or best power, the effect of transient throttle inputs is roughly the same. Because of the equivocal effect of transient throttle, table switching may not be a good solution, or even necessary. The implementation of acceleration or transient enrichment is either a source of roughness, causes a reduction of power by operating at above best power afr, or, it's unnoticeable.

- Jim

[AbeFM]

Switching between the power and efficiency maps can be done using the various thresholds you mention, but anytime the motor is operated in a 'static' condition is a good time for a best efficiency afr. No reason why a high RPM, high MAP, and high TPS high speed cruise shouldn't be at best efficiency afr. The motor runs cooler, and particularly under those conditions, the exhaust valves are cooler. The exhaust valve temperature and heat dissipation being perhaps the most significant limiting factors in motor HP.

Whether operating at either best efficiency, or best power, the effect of transient throttle inputs is roughly the same. Because of the equivocal effect of transient throttle, table switching may not be a good solution, or even necessary. The implementation of acceleration or transient enrichment is either a source of roughness, causes a reduction of power by operating at above best power afr, or, it's unnoticeable.

- Jim

Thanks, I'm pretty happy with the improvement myself. I know the car is capable, having seen 32 on the 100% OEM set up before. And cruising at 80+ verses 60 should take some hit. I agree table switching as a way for dealing with transients seems a poor solution - it always seemed to me a perfect map wouldn't need accel enrichment of any sort - I still have it in the back of my head that someday I will put a flow meter in each runner and stop messing around trying to guess what's going on. Another project, right?

The point I take issue with is "high TPS, high MAP" cruise. That's not a cruise. I used to say that on my kawasaki - I don't know how the engine behaves under steady state load because I can't get a steady state on a 500 lb, 250 hp vehicle.

Similarly, with the car, putting anywhere near 300 hp to the ground (high map, high tps) means I won't be cruising. period. In fact, no time where the throttle is above SOME number am I cruising. That number might vary vehicle to vehicle, but aside from scooters or other 1cc/10lb vehicles... I want power above some particular TPS value.

[MotoFab]

Thanks, I'm pretty happy with the improvement myself. I know the car is capable, having seen 32 on the 100% OEM set up before. And cruising at 80+ verses 60 should take some hit. I agree table switching as a way for dealing with transients seems a poor solution - it always seemed to me a perfect map wouldn't need accel enrichment of any sort - I still have it in the back of my head that someday I will put a flow meter in each runner and stop messing around trying to guess what's going on. Another project, right?

About having a perfect fuel map without using transient enrichment, I know, it seems intuitive. Like it should work. I've read the technical documents as I'm sure you have too. And also like you, there's this small background thought that won't go away that believes it can work without acceleration enrichment. It's a nice hope and I wish it were possible too, Abe.

Direct cylinder injection will not use acceleration enrichment. But then direct injection doesn't buffer and evaporate fuel from the port wall. A condition that requires active control in all but one condition, steady state cruise.

A possible solution is to go for creating a best power afr during acceleration, best efficiency afr during cruise, and lean-limit-of-combustion during decel.

The point I take issue with is "high TPS, high MAP" cruise. That's not a cruise. I used to say that on my kawasaki - I don't know how the engine behaves under steady state load because I can't get a steady state on a 500 lb, 250 hp vehicle.

Similarly, with the car, putting anywhere near 300 hp to the ground (high map, high tps) means I won't be cruising. period. In fact, no time where the throttle is above SOME number am I cruising.

Maybe this amounts to a difference in the use of the word. Cruise traditionally means steady state, as in no acceleration. A motor moving the car at a steady speed, at terminal velocity, at WOT, no acceleration, is in a cruise condition.

That steady terminal velocity cruise can be at a best power afr, or at a best efficiency afr. The only difference being, on average, some few percentage points of power. At medium triple digit top speeds, a few percent power increase doesn't increase terminal velocity very much.

It's like the 1001hp 250mph Bugatti. It takes 10hp to do 62.5mph, 100hp to do 125, and 1000hp to do 250mph. The new model has 350 more hp, 1350 total! But travels only 11mph or so faster with those 350hp.

- Jim

[Fred]

I think you are being quite liberal with the term cruise. You can not escape the strong connotations that the word has of being slow and leisurely. IE, not WOT and not top speed ;-)

About having a perfect fuel map without using transient enrichment, I know, it seems intuitive. Like it should work. I've read the technical documents as I'm sure you have too. And also like you, there's this small background thought that won't go away that believes it can work without acceleration enrichment. It's a nice hope and I wish it were possible too, Abe.

Although what you infer (but don't say) is true to some extent, it is minimally true on modern engines with a good design. For example, a modern 4v/cyl engine with V injectors deposits VERY LITTLE fuel on anything except the hot valve from which it evaporates rapidly. A large part of the transient enrichment issues with megasquirt are due to large latency between measurements and action. This can NOT be avoided with batch injection, at least, not without sacrificing all injection duration accuracy.

In summary, fast updates and sequential injection combined will give a very reasonable driving experience with minimal or no transient correction on a well built and setup engine.

Fred.

[WTDeuce]

My latest feature request is a 60-2 decoder early on... And an EMS that doesnt have sync lost issues

Must stop looking at Motecs....

[Fred]

Those two are, of course, the hardest two wishlist items to achieve :-)

My latest feature request is a 60-2 decoder early on...

Toyota is number one. Missing tooth is number two.

And an EMS that doesnt have sync lost issues

I'll do my best, but many of the complaints about that are noise related. Hopefully our hardware fundamentals will cause this to much less of an issue.

Fred.

[WTDeuce]

Those two are, of course, the hardest two wishlist items to achieve

My latest feature request is a 60-2 decoder early on...

Toyota is number one. Missing tooth is number two.

And an EMS that doesnt have sync lost issues

I'll do my best, but many of the complaints about that are noise related. Hopefully our hardware fundamentals will cause this to much less of an issue.

Fred.

I wish mine were noise related.... Im not sure they are though as im not using VR sensors :p