Here are my thoughts on a systematic approach to optimizing D-Jetronic for a specific stock or near-stock engine - just putting this down and looking for comments:

1. First step - make sure the engine has no other faults. Mechanicals perfect, all systems present and working. correct part numbers, timing/dwell set to spec, valves adjusted to spec, fuel pressure to spec. etc., etc.....
2. Set the MPS using a dyno
   • Set the basic mixture (inner screw) - goal here is fuel economy and low emissions with reasonable performance (suggested AFM targets?)
   • Set the transition vacuum to full load (outer screw) - goal here is to eliminate flat spots
   • Set the full load mixture (full-load stop screw) - goal here is to set for maximum performance at WOT
3. Get car fully warmed-up and set idle mixture to spec CO - goal here is to reduce emissions and fuel consumption, and to minimize idle droop with electrical loading (the leaner the better).
4. Set idle speed with bleed screw to 1000 rpm
5. Let sit overnight
6. Insert 300 ohm potentiometer into the CHT circuit (make cables long enough to adjust from the cockpit
7. Start car and drive for 15 to 30 minutes, with frequent stops - you're looking for:
   • Proper aux air regulator operation (at least 5 minutes of idle higher than 1500 rpm after starting)
   • Idle stability without drooping, oscillations, or stalling
   • No popping on overrun
   • Minimum idle droop when electrical loads are on (lights, heater, blower)
8. If the conditions above are noted, increase the pot setting slowly until the best compromise performance is achieved.
9. Remove pot from circuit, read value - fabricate 1/4W ballast resistor to insert in CHT circuit of same value.

Comments?

If the intake system has been significantly modified (e.g. different cam, big valves, etc.), then to do the dyno procedure correctly, an assessment of the VE changes would need to be made, and the ECU daughter-card SC circuit would need mods - but this is beyond a reasonable approach for most people.

You know what would be nice here.

I detailed drive and tune section to replace step number 2.

A Dyno is not always within reach, and can make it cost prohibitive to do.

Rich

Thats basically what I did when I had my idle hunt problem. I never touched my MPS, but I basically refurbed the rest of the system. The only parts I didn't touch were the MPS and the ECU. Otherwise, everything else was new or refurbed.

I had the idle hunt, so I put in a variable resistor and dialed it up to about 380 ohms (where the idle stabilized). Now it runs great, idles stable when hot, and passed AZ Emissions no problem. (other than the monkey that worked there couldn't shift my car!)

My FI is 73 and obviously requires at least the factory spec'd 280 ohm resistor. Oh and my car has MSD, Pertronix, and Euro-race header.

I would be real intrested in what size of resistance did other 914 owners place in series with thier TS2. I know I added a 270 ohm resistor in series and it made a huge difference in stabalizing my system (76 2.0, early exhaust, petronix). This weekend I plan on playing with a pot to find the ideal size.

I'd try a 1000 ohm pot. I've heard about people taking old pots from radios and using them, but I think the resistance is too high. After a half degree or so of turn, it would probably do a rich shutdown.

I picked-up a 2000 ohm variable pot (0-2,000) at Radio Shack for $2...I should be able to dial in pretty good.

I think my pot is 1k and I got it at the Shack too.

If adding a lot of ballast resistance makes your 914 run better, then do it! Be aware, however, that adding too much may make the mixture vary once the engine is fully warmed up, possibly causing drivability problems.

This chart is from my ECU web page:



It shows the relationship between the Cylinder Temperature Compensation (CTC) circuit output voltage and the resistance of the TS2 (also called the CHT sensor) plus any ballast resistance. Higher values of CTC Vout translate to richer mixtures. The blue trace is for an 037 ECU with an 012 TS2 and no ballast, the other traces show the effect of increasing ballast, up to 500 ohms. The temperature ranges on the chart were deduced from calibrating the resistance values of the TS2 to actual temperatures.

What you see is that if you add between 50 and 270 ohms of ballast, you richen the mixture only through cold start and warm-up. If you add 500 ohms, now the mixture varies once the engine is fully warmed up. The ECU circuit is designed to linearize the TS2 characteristic and to cut off its effect once the engine is fully warmed up.

The chart was very helpful understanding what the resistance does...especially after warm-up. Since I currently have 270 ohms...I'm only seeing the effects during warm-up...it will be interesting when I crank up the pot between 270-500+ how that will effect warm running.

Do you suspect the 043 MPS responds simarily to the 037 MPS?
Thanks

So, from what I see there, if I want the motor to be a tad richer over the entire heat range, I have to start at 500 ohms.

Anything less seems to only affect the warm up time period, which may only last a few minutes.

Now, back to the MPS.

What is a simple and effective way to set up the MPS on a bench, like it was done at the factory, BUT...big BUT...with normal tools, normal guages...I will consider normal a hand vacuum pump.

I would love to find a way to do this without buying the two (one time use) tools to set up the car.
Without an O2 sensor, and without the special meter (cant remember the name)
These two tools seem to be one time use only..once the car is set, its set...no need to futz with it iny more, at least the FI settings..they are not dynamic, so no need to watch them with a meter/guage...who cares.

Rich

of course as they are the same unit other than different calibrations.

For bench set up the Wavetek works with setting a MPS to known working specs ie Stock...

For a engine with different then stock parts you must tune it yourself or find a motor build thats very close to yours .

You can try it, but again, the issue is you'll never have a stable mixture when warmed-up.



It only affects warm-up if it's 200 ohms or so. Warm up lasts a lot longer than a few minutes, even in warm areas (I live in Phoenix) it takes a minimum of 10 minutes in the summer and up to 20 minutes from what I've seen.



There is no simple way to properly set up the MPS using "normal" (i.e. found in nearly every garage) tools ang guages. See my web pages for details. The only way I've found that you can accurately set an MPS to factory spec is to have a setup that simulates the FI system running at a set engine speed, and to measure the injection pulse duration over a full range of load conditions. This requires a specialized D-Jetronic tester or a homebuilt setup that simulates the contact trigger points operation at a set engine speed. You also need a reference characteristic for the MPS you're calibrating to. I've gone and done this for the 043, 037, and 049 MPS's.

My web page describes a procedure using a Wavetek LCR meter, but I do not consider this to be an accurate procedure. It's OK for getting an MPS to a workable state, but has too much variation to be a calibration method.

It'll be richer, and as the head temps vary within the normal range of operation, the mixture will vary instead of remaining constant.

Why do you want a richer mixture once the motor is warmed up? From what I've seen, there are more problems with 914's due to rich mixture than due to lean mixture. The only systemic lean condition problem I've seen is on warm-up, where Porsche had to meet tough standards from the EPA for the technology of the time.

Does the adjustment on the ecu only affect idle mixture or is it across rpm range, or what? I added a 500 ohm on my '76 and then dialed down the ecu as low as possible to maintain stable idle after aar closes, but not really warm. But if the ecu knob is idle only I may be running lean at speed. I never monkey with mps's.



lean running will lead to higher temps and possibly valve seat problems (as in falling out).

I have a desmogged 76 motor with stainless exhaust--which leans the mixture as compared to the restrictive stock 75-76 exhaust. Now removing the air injection and the recyclyed gasses????who knows....I really need to get this thing on analyzer or wideband meter to truly see what's happening...other than just driving. I not going to monkey with the mps...period.

The knob on the ECU affects the mixture only at idle and depending on the ECU, during overrun (throttle closed, clutch released, coasting).

Rich running also has problems. High fuel consumption, engine wear, oil contamination, high emissions. Too many 914's I've seen are running extremely rich. 5% is common, I've even seen one that was at 8%!

Well, I added 470 ohms today, and the car runs pretty darn good. The idle is a little high (1200 at full warm-up, and the warm-up works well with the AAR functioning well) rpm but I should be able to fix that with the ECU knob (lean it out some)..if not I'll drop 100 ohms and run with 370--I built the resistors with quick connects so its easy to add and subtrack a resistor ( 270 +100 +100=470). The idle screw on the throttle body is screwed all the way in.

For what it is worth. Here are the measured resistances and temps for sensor 0 280 130 017

32 degress = 3455 ohms
66 degress = 1627 ohms
212 degrees = 120 ohms.

Unfortunately this is not a linear curve. The points on the curve can be calculated by using the Steinhart-Hart equation. The math is too much work for me but if someone wants the values these represent I'll post them. Or I'll wait for one of the kids to wakeup and solve it on their graphing calculators.

Dave

I would suspect that you have an air leak.

At one time, I could have use my trusty graphing calculator to plot the progression...been tooooo long, I would be lucky to figure how to turn the thing on now. I would like to follow this one up...a lot of good info here to be found...if anything for my own understanding of what's going on. One of the problems that complicates my car, is the TS2 is on the wrong head (PO switched heads when motor was gone through and I didn't catch it until the motor was in).

The car is running really good now...just got to address the higher idle...perhaps I'll play with the ECU knob today and then open up the throttle bypass screw some.

Any info you come across would be appreciated.
Mark

I'll wait for my 13 year old daughter. She can figure it out for sure.

Dave

Back on topic...

I think for the "best economy" you should use the 052 ECU with its fuel over run shutoff feature. I have installed one just to see how they run compared to the 044 and it runs nicely and I bet I am getting better MPG. I have data logged with the LM1 and AFR is pretty much the same as with the 044, maybe a hair leaner. I have another 052 to install and test to see if its the same.

Here you go . . . If my reasoning is right it should be an exponential curve. The r^2 value supports my thoughts.

This is a great thread. I think we're all learning something.

Jim