NefMoto

So I'm trying to figure out how to calculate TEMIN and the forums seem to be extremely light with data on whether or not this can be calculated or if it just has to be fiddled with.  Can TEMIN in fact be calculated? I'm using Bosch EV14 52lb/hr injectors, and the calibration sheet for them lists the minimum pulse width, in ms, for the injector.  Is this TEMIN?  Also, it lists the minimum pulse width when measured at 39.15psid... is this indicative that if minimum pulse width was in fact TEMIN that I would have to scale it for my given fuel pressure if I wasn't running a standard 3bar FPR?

Thanks. :D

TEMIN has nothing to do with your injector spec sheet. TEMIN is the minimum pulse width ecu WILL ALWAYS use regardless of actual MAF flow (unless the ecu is in EV state which means no injection whatsoever when engine is in deep vacuum on deceleration). I posted enough on this in the other thread. You usually don't touch this unless you have gigantic injectors and you run rich at idle.

Gotcha, thanks. :)

And by gigantic I mean at least 1000cc at 4bar or over :)

TEMIN & TEMINVA

Case: block 32 shows at bank 1 idle that O2 is correcting by 6%.
Then you add 6% at TEMIN & TEMINVA values to lower it as closer you can to 0.

Correct or not?

I believe Tony described this in previous posts, and that most tuners just take old injector flow divided by new flow and multiply by original TEMIN.

Let's say you have 210cc and your adding 630cc, so;

210/630=.33
.33*(your original TEMIN value)= new TEMIN.

Again, as Julex stated, this is probably way overkill as the ECU hardly comes close to original TEMIN value to begin with.

I guess it's also a moot point because TEMIN isn't defined in this map pack even, heh.  Thought it was for some reason, too.  ;D  I'll stick to worrying about the other injector basics before I worry about things like TEMIN.

Thanks for the input guys. :D

No... If you're off by 6% for low range (idle) LTFT, you will also be off for mid range LTFTs. There is also high range LTFTs which don't show in block 032 unless you actually drive the car and exceed certain load/rpms threshold (defined in bunch of maps for that) which might also be off.

If your mid range LTFT is smaller than idle and high range even smaller, it means your TVUB is off and needs more time added (or subtracted if values are negative for LTFTs) to all points. TVUB weights the most at idle and decreases close to nothing % wise as the load raises as it becomes smaller and smaller portion of injector pulse width.

How much to add to TVUB? as much as needed to bridge the 6% gap albeit you must know that LTFTs start moving if instantaneous lambda correction exceeds 10% either way... so if your LTFTs move by 6%, you're really off by much more than that, To zero out corrections in such case would imply correcting TVUB or KRTKE or MAF maps (depending on scenerio) 16% towards opposite afr point.

Of course you can't just add 16% to TVUB as that's not the way to do it. You need to add as many ms (percentage wise) as is missing from actual injection time, not TVUB value. Let me explain.

If your total pulse width when @ 14v is 1.5ms with TVUB @ 14v being 0.8ms (this is just an example), then your real injector open time is:

1.5ms [Injector On Time] - 0.8 [TVUB] = 0.7ms   <-- real amount of time the injector is fully open and squirting

Now you need to figure out how many more fractions of ms it needs to stay open to supply 16% more fuel which is:

0.7 * 0.16 = 0.112

You need to add 0.112 to TVUB @ 14v since that's how much it is missing to be perfect... How about other points? They need to be increased by some multiplier to much how much the 14v point got increased...let's see:

0.8 + 0.112 = 0.9112    <-  new TVUB at 14v
0.9112 / 0.8 = 1.139   <- how much the 14v point was increased, this is now your multiplier for old TVUB values

Solution: Multiply all values by 1.1139 (you already did 14v!) and you should have what you were looking for.

Dropping knowledge. Thanks for the example.

julex your post is very instructive.
applaud added  ;)

Where does one calculate/derive/log the total pulse width [injector on time] that you refer to here?

This may be basic for everyone, but bear with me.
So, from the FR, here's basically the equation you gave above (for bank 1):
ti b1 = rk_w *frkte + tvu_w
total injection time = relative fuel mass*KRKTE + TVUB offset

So how do we know exactly what the relative fuel mass is, associated with a specific voltage? (Like your example, at 14V)

Incorrect. This is not only to prevent rich running at idle.
Idle is not the lowest load situation where the engine control unit is injecting.

You should always scale minimum injection time with the injector size.
If you don't, then you will get rich spots in very low load conditions.

This is a perfect example of thorough vs lazy tuning.

True, on theoretical level. But in practice it is not.

Let me explain. Most if not all of the injectors we are using in aftermarket application are much larger than stock ones because that's the whole point of using different injector. Larger injectors come with the same issues as the small ones only that these drawbacks are amplified due to larger orifice, valve and general mass of assembly that has to open/close to let fuel flow. This means that their minimum open time is longer, are usually less accurate in delivery and moreover has gigantic errors in amount of fuel injected below certain pulse width.

The TEMIN value of 0.5ms is already far below recommended minimum width for any of the injectors we are using here (well, maybe 52lb EV14 can do it, I give you that). Adjusting it even lower will cause the injector to deliver very inaccurate amount of fuel.

I've seen how that fares on my own car. There is a point where while decelerating you can press the gas pedal ever so slightly that the engine resumes normal operation and starts injecting fuel. The load is so low in such situation that closed loop is still disabled (no o2 corrections) and ECU works purely off the MAF reading... which is very inaccurate in this region as there is almost no air flow. In this situation the required pulse width will most likely be the TEMIN of 0.5 or less if you lowered it.

On my 1000cc injectors with TEMIN of 0.5ms I usually drop to around 11-12 AFR here but still fire the cylinders just fine. When lowered TEMIN is in place, you'd calculate this value as something liek 0.5ms/3.x = 0.15ms or so, I start misfiring very badly and car starts jerking like crazy.

So... is TEMIN of 0.15 right for 1000cc injector if going purely by math? YES. Is it practical due to how injectors work in real life? HELL NO.

To be honest I would probably benefit the most by lowering TEMIN to 0.4 to get that 11-12AFR leaner, I will give you that.

I know, that this was a while ago - but this is exactly what I meant by scaling TEMIN with injector size. Your big post will be helpful to people who read it, but writing all that while implying I don't know what I am doing and then saying at the end the same as I did looks a bit funny :)
Sorry, I am not used to spelling out bare basics all the time - but yes, obviously you would not scale these 1:1, you tune this by letting off and seeing how AFR and car reacts on fuel resume.

Just to give my .02

Leaving TEMIN @ the default 0.5ms would cause rich conditions under ultra light load for me (ie just holding the throttle pedal just enough so that the car doesn't go into overrun).

Lowering TEMIN to 0.12 solved that issue.

That was running ID1000's @ 4 BAR (around 1170cc)

Yup, that's exactly what I was talking about.

Hi all,

One basic doubt:
If i know flow rate of injector and engine volume, then how to calculate TEMIN?

Thanks

Hi
What about TEMIN VA?
Is this needed for deceleration?

teminva is when engine cold...

EV14 60# here. TEMIN = 0.299 MS . Works great.

TEMINVA is for the intake manifold wall film model, it is the minimum injection time during deceleration enrichment.