Daily WTF

[Fred]

Between your lack of sleep and my D minus reading comprehension, I lost you :-) Having re read it, I see it now though.

Fred.

[Fred]

About the "lenient wheel decoder" :

This only applies to the generic trigger wheel decoder though. We haven't added this feature to all of the other wheel modes. Some would be easy, some would be quite tricky.

James

It would have been nice if Ken had said this in his release notes. I was wondering about how it functioned with strange wheels. I guess that answers it quite well.

Abe will be interested to see this.

Fred.

[AbeFM]

I am interested to see it. Where did he write it? I start to feel I'm terrible at reading these posts.


Speaking of WTF. So, I had sticky valve guides leading to improper valve adjustment, to the point that one cyl wasn't even functioning at idle. So I get all my new shims (shim-over-bucket, thanks mazda, for building a motor that can rev to 7,000 with technology my last 14,000 rpm vehicle didn't resort to), put them in. They are all perfect but one, so I weasle it out with the help of a friend and some trickery. Now the valve is even more wrong. So I put the one I just took out back in and it's perfect.

Start the car, and it runs perfect for the first time in weeks. Awesome! But there's this ticking it's always had, which has gotten slowly louder. Well, this time, it's so freaking loud I can hardly hear the motor over it. Sounds like a wrist pin. Or a water pump.

So I get sick of it, go to my room to mess with my computer, rip some DVDs. Well, the program crashes. While I wait for it to relaod, I check my email. My web browser locks up. I figure I've overstayed my computer welcome, so I reboot.

Then the computer won't boot. Gets to one of the boot screens, and sits there, black. While I try for the third time to reboot it... I decided to change the name of someone in my cellphone, since I had him as Aaron when it's really Adam. I go in, edit it, and when I hit save, it says "Opperation Not Allowed".


GGGRRPRRPRHAAAGHTTHTHTPPTTTHTHT FUCK!

Damnit.

[Fred]

Start the car, and it runs perfect for the first time in weeks. Awesome! But there's this ticking it's always had, which has gotten slowly louder. Well, this time, it's so freaking loud I can hardly hear the motor over it. Sounds like a wrist pin. Or a water pump.

I doubt your water pump is making that noise... long before they make noise, they leak all your water out. Wrist pin failure? Highly unlikely. Improper valve shim adjustment is my guess :-p

LOL at the rest of your misfortune though :-)

[AbeFM]

Gee, thanks.

And no, the valves seem to make a pretty obvious noise, with a corresponding sensible volume.

After several restarts yesterday, unable to fnd the noise, I knocked off the belt for the WP and alt, and no noise. Put it back on, still no noise. Now the motor runs great, but down on power - I guess I need a retune.

Who knows if the noise will come back.

[Fred]

Depends where that set of feeler gauges you left inside the engine ends up I guess :-p

;-) lol

[Fred]

I really feel like I'm being mean to poor old Ken :-( I don't mean to be mean, but he keeps saying things that just aren't right! So Ken, if you are reading this, I'm sorry :-(

Today he has again said that "multiply map = off" is a good thing for resolution based on the range of values you end up with in your table. This is false. resolution and accuracy of tune comes down to granularity of the adjustment.

Multmap on = higher min value

if min is 50 and 1 is the unit of change then a single unit of change is 2%
if min is 100 and 1 is the unit of change then a single unit of change is 1%
if min is 200 and 1 is the unit of change then a single unit of change is 0.5%

mult map off = bad idea IMO

There *could* be other accuracy issues with the math involved that means mult map = off means less loss of precision, but I doubt it. And, that aside, its totally wrong. The only down side to adjusting your "ve" high like that is that it becomes totally non representative of the real engine ve. I don't see that as a problem however.

Secondly :

Yeah, this is one of the problems with injectors that are too large for the engine, which is why I added lag compensation.

That reminds me of something an msefi member called "Jim" might have said...

I don't buy the "injectors too big" BS, I buy the "ECU too incapable for such large injectors" thing. Reality is that if you put in a single shot of fuel independently from each injector for each cylinder firing event, and calculate as close before that as possible, then by default you have the best possible transient response before trying to play with fancy algorithms for "AE" which I have dubbed "Delta Enrichment" which seems more apt to me.

For reference and if you don't know what it is, multiply map is discussed here : http://www.msextra.com/viewtopic.php?p=185702#185702

Fred.

[Fred]

Another dubious piece of code (Sorry Ken) :

when MAP goes up, even if the VE lookup is a little behind, the MAP being multiplied in will make the PW increase faster.

Now, I can see the potential advantage here, and 9/10 or even 10/10 this probably helps, but I still think it is wrong to use one parameter and not use the rest that go with it as a set!

I mean, if the map VE figure decreases as map goes up (possible with restrictive intakes etc) then you could have the situation where you get more fuel instead of the same, or more fuel instead of less! Hmmmmmmmmm.

Fred.

[AbeFM]

Ha. A friend took my schematic, admitedly, only so-so:


And made this out of it:


I guess it'll work.

[Fred]

http://www.diyefi.org/forum/viewtopic.php?f=6&t=224