D-Jet & #73 Web cam - Am I expecting tooo much!

[3D914]

Well, its been over two months now since my 2.0L engine had complete rebuild with Euro pistons, and #73 Web cam/lifters/push-rods, and I still can't get the stock FI running right. See links - Manifold Pressure Sensor (MPS) for 2.0L and ECU Quirk for details.

Has anyone out there - "been there, done that." I just need to know. Tired of banging my head against the wall.

TIA,

[evill]

Gerard, the Web 73 "works" with D-Jet. That means D-jet can run that cam, however, it tends to run rich at lower rpms like idle. Emissions can be a problem with the 73 grind. You can adjust the MPS to dial it in better, however this is time consuming. I adjusted mine on a 1.7L with a 3-wire O2 sensor and trial and error, it can be done.

Good luck,
Ed

[seventythree]

I used the #73 cam on my otherwise stock rebuid of a 73-2.0L. It did not run well on a rebuilt MPS, but has been running flawlessly with a factory MPS for over 25K miles. So, I know that is a workable combination.

[chrisreale]

The rebuilt MPS's are being calibrated closer to 1.7L specs. All of them. This means if yo have a 2.0L, you are kinda screwed.

[3D914]

Ed,
What type of O2 sensor did you use, and how was it wired up?

seventythree,
Did you get to test both MPS's to determine the differences? And what were the final parameters for the factory MPS?

I have a working MPS that passes all tests including aneroid cells, but I can't get it dialed in right. And I'm reluctant to go into it an adjust the inner or outer screws. Especially without first knowing whether its range of adjustment will succeed in meeting the needs of the cam. Also, as Chrisreale mentioned about the rebuilds, it could have been rebuilt and I have know idea.

If you can offer any solid direction, I'm all ears.

Thanks,

[Bleyseng]

Gerard, What is the number on the MPS you are using? For a 74 it should be ending in 043. Does it have the yellow epoxy over the end plug? Does it have screws instead of rivets holding it together? Is it painted silver?
To adjust the MPS is no big deal IF you have the right tools (Wavetek meter) and go to a dyno shop.
I have reworked about 6 MPS's so far to excellent working spec for 2.0l and my 2056cc.
Geoff

[3D914]

Bleyseng,

I agree with you that's its no big deal, I just don't want to do it if its not going produce the results I need. So I'm doing my homework first to find out what the adjustment range is, and what affects that will have.

Then I'll need to get an O2 sensor so that I can measure, dynamically, the various conditions I'm running with now.

Once I can determine what I'm going to need the MPS to do, then I'll be in a better position to judge whether its up to the job. If not, then hello CIS, SDS, or MegaSquirt.

Thanks,

[3D914]

Well, I have some better news.

As I continued my research on the MPS adjustment range, a couple of things occurred to me as I was reading other posts. One was the need to try increasing fuel pressure across the entire range to compensate for the increased air flow into the combustion chamber.
It also occurred to me to advance the timing slightly.

So I tried both of these tonight with some success. I was able to get the idle high enough that it would adjust properly with the air screw, leaving the ECU in middle position. This was with the engine warm. Now I'll see tomorrow morning how cold idle goes, and will take the interstate to work and see how the hot idle turns out.

Wish me luck!

[3D914]

Just to update with my results.

The slight advance and increase fuel pressure (up to 34spi) seemed to help cold and warm idle nicely. It also seems to have improved overall power (to be expected). The hot idle still drops down very low, and almost dies though.

I'm going to start from scratch this weekend with a valve adjustment, reset timing minus a degree or so from where I'm at, and try my good MPS. Then Monday, I'll get it down on the analyzer to check O2 and Hydrocarbons.

[DuckRyder]

so, whats the verdict?

[3D914]

DuckRyder,

Thanks for asking.

For the most part I would not suggest the #73 cam with the stock FI, unless one is willing to get into the major components, like ECU and MPS and make changes to them. I cannot use my stock MPS as it produces a very rich mixture that cannot be adjusted out with the ECU. I'm relying on a used one that drops about 5ingh in a couple minutes time. Its affect on manifold pressure produces a leaner mixture within the range that I can adjust.

I have mine running decent, but not at all as reliably as with the stock cam. I don't get startup and idle with no intervention. I have to keep the peddle down for a minute or two, then it idles fine.

There is s small temp range (don't have specific values) where the idle will want to drop to 200-300rpm. With the lights on it will die unless I keep my foot on the gas. Either side of that temp range it will idle fine at around 1150rpm.

Power is good, but gas mileage has dropped. Since the #73 cam has longer duration and higher lift for intake, you're gonna use more gas.

Another note is the noise. This cam does produce more noise than the mild stock cam. I now remember what it was like with my first 914 which had a similar cam and carbs - noisy!

I'm planning on getting the MegaSquirt FI controller and program it for this cam. The flexibilty of this controller looks very promising. I'm hopeful that with it I can get all fuel delivery conditions addressed so the car runs properly.

Regards,

[pbanders]

Originally posted by 3D914
For the most part I would not suggest the #73 cam with the stock FI, unless one is willing to get into the major components, like ECU and MPS and make changes to them.

FWIW, this is the same conclusion that John Larson (30+ years experience as a Porsche mechanic specializing in 914's and expert on D-Jetronic) on Rennlist came to years ago.

I'd also suggest that you don't get your hopes up too high for using any type of aftermarket FI to completely "fix" your problems. The WebCam 73 will likely never idle as as smoothly as the stock cam, as it's designed to promote VE at high engine speeds, at the sacrifice in idle and low speed characteristics. All static cam systems are subject to these kind of compromises, which is why modern engines use variable cam timing, lift, and duration systems on both intake and exhaust in order to have fully optimized performance (torque, fuel economy, and emissions) over all engine speed and load conditions.

[seventythree]

Brad:
My 73-2.0L was rebuilt, 30K miles ago, using US spec pistons and a #73 Webcam. It ran poorly with a rebuilt MPS at all engine speeds. Some D-jet guru (you know who that is) recalibrated a used stock MPS to the 043 factory specs, and the motor has been running flawlessly ever since. It idles better than my 85 Carrera ever did! Most likely, John Larson has a valid concern with higher emissions of a #73 Webcam, but I know from personal experience that it can be setup with stock D-jet to produce excellent drivability.

3D914:
No components were tweaked or recalibrated other than the MPS, which was recalibrated against a stock 2.0L 043. But, as Brad will surely concur, the inductance measurement technique is highly inadequate for proper calibration of an MPS. The only sure way to calibrate an MPS is by measuring the injection pulse width.

Regards,

Hamid

[Bleyseng]

The only sure way is too calibrate the MPS on a chassis dyno doing multiple runs to sync in the Part load and WOT settings to the proper Air/ Fuel ratios for that engine.
It is amazing how well the engine will run when this is done. Brad Anders settings are for totally stock engines, using stock parts to get you back to the Factory settings without doing the Dyno work.

Using a MPS that leaks will cause some problems find one that is good and calibrate it to your engine. You would rejet the carbs if you were running em right?
Geoff

[pbanders]

Hamid, Geoff, good comments.

As far as doing it "right", the big problem with changing a camshaft in the system is that this changes the VE curve. No amount of adjustment of the MPS will properly correct for the change in the VE curve and will be a compromise at best.

The "right" way to accomodate a different cam is to change the hardwired VE curve on the daughter card of the ECU. If you look at how Bosch designed D-Jetronic, they tried to minimize the number of different subassemblies they used. For example, for the most part, all MPS's are calibrated exactly the same for part load conditions. That's why you can use an 049 MPS in a 2.0L with pretty good results. MPS's differ generally only in the setting of the full-load enrichment. By taking this appoach, Bosch could build a full run of MPS's for all applications, adjust them to the same part-load spec, and customize them only at the last step (the setting of the full load mixture), then stamp them for shipping.

Similarly, the main loop circuit in the ECU that sets the basic mixture is exactly the same in all ECU's. The main difference in ECU's is the daughter card that has the VE correction. This again enabled Bosch to make a generic run of ECU main boards, then simply customize by attaching the specific daughter card. Also note that the same air intake sensor is used on all D-Jet cars, and that other than the odd '73 2.0L CHT, all the CHT's have the same resistance characteristic as a function of temperature.

The way Bosch accomodated differences in cylinder displacement was to use different injector flow rates.

OK, now that I'm done ranting, a person could conceivably modify the VE circuit on the ECU daughter card to correct the mixture for a different cam. In reality, it would take about the same amount of effort to install a programmable EFI system, like the Megasquirt, and do the required dyno testing to spec it out.

[seventythree]

Geoff:

Your comment about using a dyno is valid if the intent is to tweak any FI component for a specific engine. However, what I meant was that bench CALIBRATING an MPS to a reference specification should be done by measuring injector duty cycle as opposed to primary or secondary coil inductance. I believe, the latter RECALIBRATION is what 3D914 has been attemting to do in order to confirm a good factory baseline setting; however, the inductance measurement technique might be the short fall in his effort.

[3D914]

Hamid, Geoff, Brad,

Thanks for the feedback.

Hamid & Geoff:
You're correct in your conclusions about my bench test method for the MPS. I would be interested in the "injection pulse width" method you're suggesting, as I'm not familiar with how to go about that at the moment.

Brad,
I would also be interested in hearing some more about how to "change the hardwired VE curve on the daughter card of the ECU." Is this something that is addressed on your web site on the ECU page?

I have two spare ECU's for the '74, so I'd be willing to give it a try.

Geoff mentioned something about the MPS that I was curious about also, and that's regarding the slight leak-down. On your MPS page you mention that for testing it should not loose more that 5inHg in a minutes time. Is this by design, or should it not loose any?

My good MPS (which runs way too rich) does not leak down at all. It holds 15inHg until I release the pressure. The one I'm currently using looses 5inHg between 90-105 seconds.

Thanks again for the feedback.
PS: I'll also call around to find out what it costs to get some dyno time.

[Bleyseng]

If a MPS is leaking down 5hg in 90secs it is cracked and not reading vacuum correctly. It will also produce varied running conditions as the crack in the diaphram will seal itself as it flexes. It is not as sensitive to the vacuum changes as the engine moving thru the rpms.
I would go to a dyno w/a Wavetek meter and adjust your good MPS that doesn't leak. (Before going prepare the MPS by removing the stop screw using Brads web page. Also check to make sure you have no vacuum leaks on the engine)
Set the Air/Fuel mix at partload to 13.7 to 1 with the inner screw using the dyno runs to check your adjustments. Then set the WOT to 12.5 to 1 using the outer screw again check your adjustments using the dyno runs.
I really like using the dyno as it checks the engine under load conditions thoughout the whole rpm range.
Set your idle.
Finally, set the machined stop screw to limit/support the diaphram at WOT using the Wavetek. Glue in w/ hotglue.
Take readings w/Wavetek so you can make another MPS if this one fails!!
It has taken me about an hour of dyno time to properly adjust a MPS with this method. Cost about $100. Why do I have to do this? My engine is far from stock and a stock MPS is way to lean. Sure it runs but gets way too hot and pings like crazy!
I think that if you adjust the MPS to be spot on for your motor you will be suprised how well it runs!!!
Geoff

[3D914]

Thanks for the insight Geoff.

Is there any way using the Wavetek meter, to preset the MPS near the values you indicated on the bench?

I've checked into Dyno time, and it shold cost me a little less than $100 for an hour. Now I just have to get the money for that and the Wavetek meter.

Thanks again.

[Bleyseng]

That is what you use the wavetek for, to set the MPS on the bench. The "stock" or known values are on Brads site and work for stock cars. Set your good MPS to the 043 settings and try it.
To establish the MPS settings Porsche/VW/Bosch ran the engines on a dyno then they bench adjusted the MPS's in batches to those settings. All stock cars use about the same partload setting but that is too lean for my car. You are doing you own tuning by going to a dyno . You really only need the Wavetek for the WOT/machine stop screw and to record the settings for the future. You will be using the dyno's A/f meter to adjust the inner screw and outer screw setting the partload and WOT.
This will be the optimum settings for your engine, so it will run very smoothly and have the most power.
Before doing dyno work, do a general tune up, adjust the valves, new plugs, points, cap, rotor, oil, blah blah so the engine doesn't have that to also deal with. Set the fuel pressure to 29-31 lbs.
Take the MPS apart and remove the stop screw per Brad site so the MPS is ready to adjust when you are there. Bring some tools too as they won't have any idea what you are trying to do, but will be interested in learning how.
Geoff