ECU 021EF vs. 021DF: What's the Difference?

Hello,

Here's a brief overview of the setup: 96' Pasi engine, 1 turbo, 216µm injectors, 220 bar injection pressure, variable geometry turbo (VNT) 17, original 1Z injection pump, and now with a shut-off valve.
So, up until now, I had a control unit with the part number 028906021DF (from a 1997 Golf GTI TDI).

Recently, I took the boat out for a test run to check everything. Up until then, it had been running perfectly with plenty of power. However, during the test run, the engine suddenly experienced several instances of a loss of power. (--> Logfahrt Attachment)

The error code in the diagnostic log mentioned something about a "boost pressure sensor - short circuit to ground" (I don't remember the exact wording 100%). I then checked some values at idle and noticed that even at idle, there was a boost pressure of 1060 mbar, while the ambient pressure was 979 mbar (--> idle reading attached). However, the same thing was happening on the other turbocharger.

Okay, let's get started. I got a different engine control unit from a friend to try out. For simplicity, I'll just put both of them side by side: Both engine control units are for the AFN engine and are in pristine condition; 1998 Toledo, part number 021EF; 1997 Golf, part number 021DF.
Free adaptation channels: EF - 1, 2, 3, 4, 5, 12; DF - 1, 2, 3, 5.
Adaptation channel 1 to approximately 3mg/R, a 1Z pump is required, but this is feasible for both.
Das komische ist --> 1. Spritzbegin beim EF ~ 4°v. OT (ist aber Adaptierbar) ; beim DF ~ 1,3 v. Off-topic.
"The initial amount for EF is 20.6 mg/revolution, and for DF it's 23.8 mg/revolution, at maximum adaptation. And that's exactly where my problem lies now. Due to this small initial amount, combined with running solely on biodiesel, I now have to crank the engine for a very long time (10 seconds) in the morning, even at 0°C, before it starts."
3. The preheating time seems to be somewhat shorter on the EF model, as far as I can tell.

Well, in terms of pure riding feel, the DF is a bit more responsive. However, the EF seems to generate more power overall (subjectively) and feels more balanced throughout the entire pedal stroke. I would prefer to continue riding the EF, but I probably won't be able to do that in the winter.

How can I adjust the initial fuel mixture to make it start up better?

How do I get the starting amount right so that it starts up better?

By modifying the corresponding entries in the dataset .

Best regards,
Rudi

Hi Rudi, so, I can't access the dataset . I lack the necessary expertise. How about a power box? I already have the box, but I only activate it when needed. The problem with that is that it has a startup delay, so it's not online immediately after starting.

My opinion on Murkelboxes leans heavily towards considering them special waste .
The time delay is implemented to prevent the engine control unit from activating immediately upon being switched on.
goes into emergency mode because the self-test tolerances during 'booting' are stricter than those in the...
ongoing operations.

Best regards,
Rudi

Hello,

I have the exact same engine control units. I sometimes drive with an EF from an Ibiza and with a DF from a Golf. The problem, especially because of the more balanced subjective behavior of the EF, is that it only knows one coding: 00002.

That's probably why the EF-SG often has starting problems, especially when it's warm. This happens regardless of whether it's running on biodiesel or not. If the engine is really hot, I have to crank it for 10-15 seconds to get it started.

"I tested both versions on a dynamometer and had them both run through an EU emissions test (Euro 3). Although both failed (so far), the results are quite clear. The EF version has a significantly higher emissions output, but the torque and power curve is slightly more aggressive. I'm not entirely sure why this is the case, because, according to VW, the AFN should produce the same values in both vehicles. I suspect it might be related to the tuning for the automatic transmission."

But in your case... you can extend the preheating time significantly, at least I tried it this morning with the EF. Originally, it was 2.98 seconds at 4.5°C, but I increased it to 14.XX seconds, resulting in a value of 32468 instead of 32768.

Everything else is really a matter of tuning the engine maps. I would suggest contacting a chip tuner to see how much they would charge to copy the starting behavior from a DF engine to an EF engine.

I don't see any other way to manage the initial amount... at least, it didn't work for me with any of the possible variations.

Best regards, Max.

Hi atlantic,

Okay, both of my SG (sequential gas) injectors are coded to 0002. The engine starts immediately with a warm or hot start. I'll try adjusting the start quantity and preheating time again this weekend. Now that I think about it, the values didn't change last time. Could you possibly read out the start quantity for your EF SG injector? Or try to see how high you can adjust it.
I don't have access to a chip tuner.

Quote:

Originally 2.98 seconds at 4.5°, now 14.XX seconds with a value of 32468 instead of 32768.

Am I understanding correctly that lower values increase the preheating time? Is this analogous to the starting amount? Unfortunately, the temperatures are currently around 7 degrees. Therefore, there is no preheating, and I have to crank the engine for a long time.

Hi there,

Sure, I'll check it out tomorrow.

I tried the thing about the preheating, and it's actually true.

And you're not having any cold start issues with the SG? Hmm... that's strange. It's the opposite for me. It runs okay when it's cold, but when it's warm, it's awful.

We could actually compare all the values; that would immediately show us what might be the cause.

Best regards, Max.

So...

I have now been able to extend the preheating time through adaptation (channel 12). However, this not only extends the time, but also increases the temperature threshold at which preheating even begins.

--> Therefore, smaller values result in a higher temperature threshold and a longer preheating time.

Furthermore, I have increased the initial dose in channel 5 from a minimum of 19mg/R to approximately 26mg/R.

--> Similarly, smaller adaptation values increase the initial amount.

"The cold start performance of my tractor has improved significantly (although it's not as cold right now). However, it still makes quite a bit of noise immediately after starting. The warm idle speed is between 2.6 and 2.9 volts, with an oxygen sensor reading between 53% and 56%. I'll try adjusting the timing a bit earlier soon. The actual value in the timing checker is below the ideal line."

Perhaps we can incorporate these findings into the research article about adaptation channels.