TEC3r issue. EFI gurus; help!

[otto in norway]

I have now done some more troubleshooting.

I checked for vakuum leaks, none found. (Plugged the brake booster connection)
Checked the resistance for the wiring to the injectors, all OK.

Porsche 935:
I think you may be misreading the log windows. I can't see any significant difference in either voltage or idle pulse with. The battery voltage corrections to the pulsewith is only 65uS, so that's not going to affect this.

The enrichment is a TPS based setup, with a 2.5V/s activator treshold. It will start adding 5ms to the pulsewith from the MAP table for a duration of 0.12seconds.


I have now actually gotten in touch with Electromotive support. So far the guy that mailed me seems to point towards the cam sync operation. (he was wondering if I had any error indication, -wich I don't) So I'll have to wait for his next response.

[Uwon]

I can't see anything from the info you sent me. I was about to suggest plugging the hose to the brake booster ( which leak I've had) but I see that you just did that. The electromotive support guy may be on to something. I remember that I abandoned the crank trigger setup very early on for a flywheel sensor because I could not get a good idle and low speed smoothness (aka bucking). Maybe time to send in your unit for the hardware upgrade that they are offering. I understand that the tec4 software is pretty good and it's a natural migration from the tec3r. I'll be following your progress. Good luck.
Johan

[otto in norway]

After some more hours of testing, and finnaly getting confirmation on how the TPS should be set, I am now confident that the issue is related to the cam sync.

By the way, the ECU expects an idle voltage to be within 0.08V above the TPS idle voltage setting. I tweaked the TPS locator for the axle, and reduced the play so now it varies between 0.9-0.92. So, in my setup, the idle setting is now set at 0.84.

Anyway; The reason I am now confident in that the cam sync is not operational, is that it is first of all logical. Second, I can actually hear it fueling in the correct sequence or not. If it is 180 degrees out, I usually need to grind the starter a bit longer before it lumbers into life. But if it is at the right cycle, it tends to start more quickly, and it sounds crisper as well.

The weird thing is that I do not have any error indication for anything.

I am wondering if the setup of the .bin file has to be done in a specific order for it to really activate some of the features, such as cam sync operation. So I asked that to Electromotive support, and I'm waiting for a reply. I'll probably make a new file from scratch, making sure that I get those basic setting in from the start. (worth a shot)

[Walt Fricke]

All this makes me glad I use batch firing. It has been suggested that this is best for continuous high RPM operation (like a race track). But doubtless not so good for fuel mileage during ordinary highway or street driving.

[otto in norway]

Well, an update is in order;
I have now experienced that the engine has changed character while driving. And when I tried to adjust the PMO balance, I noticed that the cylinder no 3 and 4 is way off on the adjusting screws, and doesn't really do much at idle. So this is probably an air leak at the head flange. I'm gueassing that I haven't torqued the nuts enough, because there was a little more to go on when I re-checked it now. So the possibility for an air leak is very likely if you ask me.
Annoying that I have been "fooled" by a symptom for so long, and that amazed that it can have this behavior for so long. It feels like the gaskets are getting worse, because the "bad" runs are getting more frequent.
I'm now going to check this by dismantling it completely before it hits the road again. Hopefully before winter arrives for real.

Thanks again for your input!

[otto in norway]

Well, time for an update. -Again(!)

I have had the PMO's off and replaced all gaskets. Made sure there is no leaks anywhere, and also performed a really thourough balancing/setup of the throttle bodies. I even replaced the fuel pump and filter for good measure. All is well in that department for sure.
But the problem persists.

So I have got hold of an oscilloscope to check my cam pulse.
The result of my measurements leaves no doubt. -The problem is definately related to cam sync not functioning as it should.

The pulse from the cam sync goes from +5 to 0 V at apx 100 degrees BTDC, and back to +5V at about 40 degrees BTDC for cyl1. The pulse is strong and has no distortion or noise. This I measured at the connector to the TEC3r.
When comparing the cam sync pulse timing to the injector #1 firing, I can clearly see that it is firing the injector at the wrong time when the run is "bad". This apparently makes the engine go lean and pop when rapid pedal movement is applied. I even observed the injector pulse firing av various offsets on some runs, at say 90 degrees or 120 degrees. The timing for firing the injectors stays consistent right from the start. And to fix it I have to restart the engine. And if I'm "lucky" the timing is good, it's firing the injector at apx 360 degrees offset from the cam pulse, and all is well for that run.

To me this is really strange, because it looks like the TEC3r gets the perfect pulse it needs, and it has no error code. But it just don't use the pulse for anything. It randomly selects a time for firing the injectors. So right now, this looks like a bug in the Firmware. But it's strange that this topic isn't adressed more in here if it a known bug.

Anyway; I think I have to send the unit in for a service and upgrade. But I'm not convinced that this bug will be fixed by the upgraded Firmware. So maybe I'll switch to DTAfast. -seems like a good unit..(?)

Anybody got some input on this?

[Walt Fricke]

Otto - just for grins, why not reset the program to use the crank fire system? Waste injection isn't "wasted" the way waste spark is, but works equally well. Though I guess you only have the cam signal equipment, and no toothed wheel for the flywheel pulley. And maybe your goal is the most fuel mileage you can get with an engine usually running at modest RPMs and little WOT on regular roads?

Is the cam signal supposed to be just a trigger? If so, is it rising edge or falling? And whichever it is, are those signals - which you say are consistent - falling where they are supposed to? Not just regular, but regular and in the right place on he engine rotation?

How does the single injection system know when to do things? The cam signal can resolve the ambiguity of the crank fire system. What does it use for its second signal if it produces only one pulse per engine revolution?

Is the spark firing when it should? Advancing some with RPM increases?

But I can see the dilemma - you have a demonstrated good cam sensor signal, but the ECU isn't using it properly.

[otto in norway]

Walt:

The cam sensor is just there to let the ecu know which cycle it is in. I have a speed sensor at the flywheel as well. (60-2 trigger wheel)
The injectors is the only thing that needs to be sequencial. So the sparks are "Waste" configuration.

I'm told the ecu needs to see a rising edge in the Space of 180 - 6 degrees BTDC for cyl#1. (Reflected in the manual as well) I'm also told that the injector should fire at 0 degrees TDC for #1 compression stroke. (Which happens sometimes, but not always)

My goal is to get a crispy throttle response, and stable running car. And I'm soooo close. If it would fire the injectors when it is supposed to every time, the car would be a true joy to drive. -Every time!
The support at Electromotive doesn't reply to my emails anymore. This is just frustrating.

I have seen the spark for cyl# 1 fire well after the cam sync is done, (at say 50 degrees after the pulse) indicating that the cam sync position is correct.

[Uwon]

otto, at this point you may want to consider having Richard at Clewett Engineering look at your issue live through TeamViewer. With you in car and on cel. phone with him, you tether your laptop to your phone and then call up TeamViewer wherein Richard can then view, analyses and change your settings on the spot while you are driving. At about $50 per 30 minutes, that is pretty cheap expert second opinion at the very least. Call him and you'll get a sense of his expertise.
Good luck.
Johan

[Walt Fricke]

How well does intercontinental cell phone service work? But $50/hr for Mr. Clewett sounds like a huge bargain.

Is the 60-2 wheel on the flywheel, 3.2 style sort of? Or on the crank pulley (where it is for batch fire and waste spark on my race motor using the Tec3R)? In any case, it sounds like you could try to configure for batch firing of the injectors or whatever the alternative to the sequential is. If that works, it might tell you something. And if it does not, that might tell you even more.

[otto in norway]

Thanks for your input guys. I have already tried running it in Batch fire mode. That makes it behave consistantly. So that points to a problem with syncronization when in sequential mode. I have made some more measurments now, comparing the syncronizing with the 60-2 flywheel signal. All this makes me now feel overwhelmingly certain that my unit has either a bug, or a fault. Either way, I've now decided to replace the unit for another kind. Perhaps DTA?

Just for those who might be interested;
I'm using an old type of cathode ray oscilloscope. (Fully analog) I tried to take pics/film it, but the quality is not very good because of screen update inreference. Anyway, I posted some vids on youtube.

Bad run: https://youtu.be/TgiAT8AXVPA
Good run: https://youtu.be/oXBTElID_eQ

Apologize for bad quality, but when looking at it with my eyes it is pretty clear what I'm looking at. I even had the opurtunity to compare these measurments to a 3.2 SS engine with the next gen. TEC3r running it. On his car, the signals looked identical, apart from being in sync every time. But he also struggles a little bit with hesitant response when fast pedal motion is applied. (allthough it is identical every run) My theory is that the TEC3r is not ideal for ITB setup like ours. (Because the injector timing is fixed at a non-ideal angle)

[otto in norway]

Well, I ended up putting in a DTA S80 instead. And I am very happy I did!
The DTA is a much more comprehensive product, but extremely flexible unit. And it works, all the time! The throttle response and driveability is now very crispy and consistant. The DTA has the possibility to change the injector angle. This function confirmed to me that my measurments and suspiscions were correct. When setting the injector angle to end the pulse at 350degrees after TDC, it has a really crispy response. But when changing it to say 0, or 120 degres BTDC, I get the same symptoms as I did with the TEC3r. So my theory of that this engine is sensitive to injector angle was correct. An now I have full control over that subject. NICE!

I have no problems reccomending the DTA for anyone. The TEC3r is just a pale overpriced comparison to the DTA.

[rkeele]

Originally Posted by Uwon

otto, at this point you may want to consider having Richard at Clewett Engineering look at your issue live through TeamViewer. With you in car and on cel. phone with him, you tether your laptop to your phone and then call up TeamViewer wherein Richard can then view, analyses and change your settings on the spot while you are driving. At about $50 per 30 minutes, that is pretty cheap expert second opinion at the very least. Call him and you'll get a sense of his expertise.
Good luck.
Johan

+1 Richard is the Guy