soontobered84

My understanding is that Ken DID alter some of the logic, form and function and, while you are mostly correct about the 24/19ths fuel consumed, he did some other "magic" stuff that more than offsets the injector change.

wpgshark

Next tank I'll do a manual check and then compare it with the average and report back.

PorKen

I wasn't able to follow exactly. Were you able to get both S4.S300s chips working?

I think it may be possible to get the S4 chips to output a corrected consumption pulse (pin 31 on the LH plug) but I did not get to it.

FredR

Ken,

Exactly how is the "fuel remaining distance" signal generated? I have assumed the computer compares the number of pulses it is firing to the tank level signal, applies a calculation to some algorithm stored in memory and computes an output of some kind. How and where that output is converted to an analogue value I do not know but presumably there is a DAC of some sort in that section of the display unit that processes a digital signal and converts it to an analogue output for the digital display? Thus why it is necessary to calibrate the system with the 16 litres of fuel in the tank as accurately as possible [also helps the visual tank level analogue display to be more representative].

Maybe you can derive one of your "magic patches" like the one you programmed previously when I was trying to evaluate the impact of advance on individual cylinders in ST2. Type in the nominal poundage of the injector and "la voila" it corrects- got to be worth another large beer at least! Needless to say it would only work with ST2 or with your chips but I rather think it is a utility that is nice to have, especially if you have to drive significant distances between gas stations. Not too bad over here in that regard but in some parts of the States I would not be surprised if you can find yourself with some distance to go before you next see fuel, especially in the likes of New Mexico.

I keep meaning to give Sir John a prod to see if he can do anything within the ST2 programming. Since I converted to the 30lb injectors a few years ago I learnt not to look at that part of the display but it would be nice to get it working correctly irrespective of how it is achieved.

Rgds

Fred

John Speake

IIRC the calculation using tank fuel quantity and available range uses the injector pulse lengths.

If that query is correct then the range will be too optimistic when fitting larger injectors ?

FredR

John,

My thinking is that the LH is simply integrating the sum of the number of pulses times their duration and subtracting from the level signal derived fuel capacity algorithm to work out the logical range at the current rate of consumption. Thus just as ST2 is programmed to suit the poundage of the injector, logic suggests that an extra line or two of code can introduce the new variable that is already embedded into ST2 and thus modify the output to correct for the modified injector sizing specified. Easy to visualise but no idea whatsoever as to how difficult it would be to implement. If it was "so easy" presumably you would have done this already unless it was not considered to be of value.

If the range logic solver is located outside the LH then obviously some other means of correction for injector size would be needed. It is not a huge deal but nonetheless a nice to have feature for ST2 if possible.

Rgds

Fred

John Speake

The ST2 or otherwise modified code produces the required injector pulse widths, shorter for larger injectors.

These pulses are then used by the digi-dash processor. So there's nothing you can do in the LH software to correct the problem in the digi-dash.

PorKen

Although the 87- LH has a distinct consumption output, it looks like the digi-dash is hard wired like the earlier analog gauges with a simple input from the injector feed. So fuel level / injector duty. You'd have to change the digi-dash code to fix it.

Looking at Rob's LH connector you can see pos #31 is empty. Bummer for the consumption gauge but cool to have another programmable output from the LH. (Also not wired, pos #26 is for a shift light. Wired and used by the LH, but not remarked in the WSM, pos #15 is A/C on.)

I am pretty sure that the consumption output comes from the hybrid activated by processor port P1.1 - there is a separate counter for P1.1 related to injector duty. I can test this fairly simply.

FredR

Typical Porsche- why do it the easy way when you can do it a more difficult way!

Presumably some new code accessed like the last patch and a new output circuit gate from one of those not used terminal points could generate a conditioned output to send to the DAC in the display after a bit of re-wiring?

Rgds

Fred

PorKen

I think you may only need to reroute the wire from the EDU 'L-Jetronic' input to terminal #31, plus some code to speed up the timer (more pulses vs. injector output). I would bet it is 12V but but #31 might be 5V high. Not sure if that would be a problem. If the EDU doesn't like that you could use #26 which is 12V high as it was made to drive a dash bulb directly (switched ground = Shift).

The EDU 'L-Jetronic' input is a YEllow wire going to the CE at Q23 (89 J12), crossing over to W12 where it goes to the common injector trigger ground via a BRown/REd wire.

wpgshark

This is fun to watch.

I really like watching all of you when you're all pulling in the same direction to meet the challenge.

Thanks

FredR

Ken,

If I understand your note correctly, what you are saying is that at the moment the display unit is fed the pulsed signal via that yellow wire. It then takes this signal, processes it in some logic unit unit comparing it to whatever other signals it processes and then displays the remaining distance it thinks the car will run before it gives up the ghost.

Thus to solve the current dilemma, the output signal needs to be modified within the LH unit and then the conditioned signal outputted to whatever terminal can be made available and then send this signal to the yellow cable instead of the one that currently runs from the injector earth rail- correct?- presumably interdicted in the central electonics area.

Rgds

Fred

PorKen

Fred - yes. Probably can just remove the terminal at Q (J) or W and plug the new wire onto/into it.

(I have a big bag of the ECU plug terminals.)

FredR

Assuming it is possible, how difficult would it be for you to come up with a patch to modify the BIN file like we did during the individual cylinder retards I was experimenting with so that ST2 users like myself can self medicate the poundage rating thus reinstating the original capability?

Presumably you could apply such to future revisions of your chips as well for anyone so inclined? I still have the patch loaded but after the trials I just set the values to zero [for those interested Ken modified one of the maps that is not used so that we could enter differential corrections to the timing on individual cylinders to help compensate for the better air flow to the inner cylinder banks thus leaning out a bit and being more sensitive to knock].

Maybe Ken's next project should be to come up with a better noise filter than the one currently used which some of us think intimates knock when such may not be happening thus attenuating the motor output a tad.

Regards

Fred

PorKen

The patch would be simple. Hopefully the math would be simple as well, just a subtraction from the counter which regulates the frequency of #31. It would require a little bit of experimentation and logging. If it worked I could incorporate the patch in a future version of my S4 LH chip. This is winter time, off season stuff, though.