TDI Programs: Reviews & Experiences

Hello,

It has been reported several times that TDIs suddenly exhibit increased power and/or reduced fuel consumption after approximately 10,000 km.
Since mechanical "break-in" effects tend to become noticeable gradually, abrupt changes are often explained by switching from a break-in program to the standard software, which should therefore occur within a range of approximately 10,000 kilometers.

I would like to gather some objective facts here, because the acceleration popometer seems to me to be an unreliable measuring device, and its daily performance fluctuates at least as much as the performance of TDIs .

The knowledge, observational skills, other competencies, and experience gathered here should surely be sufficient to answer the question of break-in programs?
Perhaps they are only available for certain models or engines...

Gremlin's observation of improved acceleration after the first WIV oil change, for example, suggests a significant reduction in friction due to the new oil, without requiring a software update.
Or should the first maintenance update have triggered a (further) switchover to the standard software?

Julian's car, as far as I know, now handles low-end torque better than it did before reaching the 10,000 km mark.

My wife and I have previously driven two former new TDIs past the 10,000 km (approximately 6,200 miles) mark (an AFN engine in a 1997 Golf 3 and an ALH engine in a 2001 Ibiza 6K, both without a water-cooled intercooler) – without any clear "aha" moments, but also without any visits to the workshop; I have personally taken care of all the scheduled maintenance.
However, back then, we didn't pay as much attention to the finer details of performance.

Regarding the acceleration test, I vaguely remember that initial attempts with the car in its stock configuration resulted in times around 6.6 seconds, while later attempts were closer to 6.2 seconds. However, I no longer recall the mileage at the time or the specific testing conditions. It's possible that I was initially timing based on the speedometer instead of the tachometer, or that I didn't account for wind conditions, or other factors.

With the ALH engine, we hadn't performed any dynamometer tests for the last 10,000 km, but the current measurements show approximately 97 horsepower.
Our existing fuel logs do not indicate any significant decrease in fuel consumption since approximately 10,000 kilometers.

Unfortunately, I can't contribute any more based on my own experience yet, as my electric vehicle is likely to exceed 10,000 km (approximately 6,200 miles) only in the fall.
By then, I want to log all performance-relevant data during DZR tests, so that I can potentially compare it with data collected "afterwards."
According to the DZR (German Technical Inspection Association), the engine currently produces approximately 134 horsepower (~2,500 km) without any modifications, just as it did at 500 km.

The only thing that seems a bit strange to me so far are the negative fuel injection start values according to VAGCOM (e.g., partial load = -3.7°, full load = -21°) and the STG coding 00005 (part number ???).
Should these be indicators for a break-in program, or just for OBD-compatible software, or for WIV, or for something else?

Okay:
Who can report objective / measurable changes in the area of approximately 10,000 km (or other mileages)?
-> DZR times
-> (reliable) dynamometer measurements.
-> Changes in data logs (that are not solely attributable to environmental conditions).
-> Fuel consumption trends (according to the fuel log).
-> Possibly, maximum speed tests under the most identical conditions.
-> additional notes (connections to oil changes / maintenance for VAG vehicles, etc.)
or potentially contribute valuable insider knowledge?

Please provide the following information for each case: vehicle type, engine code, year of manufacture, mileage at the time, and whether or not a warranty was active.

Even wagons where it appears nothing has changed are helpful in recognizing a pattern!

Once I have enough data that can be processed, I will create a table and make it available online.

@ Gremlin + Julian: Please post your oil change, mountain driving, and possibly fuel consumption experiences here again, including the details mentioned above.

Hi Ulf,

"I had my last DRZ tests done around 5000-6000 miles, and now I'm at 38000. So, this car is a suitable candidate. I'll try to finally solve the problem with the computer's power supply so I can start logging data while driving again."

I'll also check about the coding, specifically whether it's still 00005.

The data will be available soon... and there is a complete logbook in electronic format detailing all fuel fill-ups. However, I don't think it's very informative, as the city driving cycle is quite irregular, and long-distance trips distort the results even more.

The only thing that's certain is that nothing unusual was detected around the 10,000 km mark.

Best regards,

Jan.

@ Gremlin + Julian: bitte postet hier nochmal Eure Ölwechsel-, Berg- und evtl. Verbrauchs-Erfahrungen, möglichst mit den o.a. ZPlease provide the text you would like me to translate from German to English.


But sure, I will try!

MKB: AWX, GKB: GBQ, Audi A4 8E Sedan, Year of manufacture: 05/03.

- Inclination of approximately 30° (quite steep), approach with approximately 1100 RPM / 3rd gear. After ~10,000 km, it was no longer necessary to downshift (to 2nd gear) when going uphill (speed could be maintained). After ~13,000 km, any gear could be used. *at some point it might even be possible in 4th gear*.

-Topspeed was first tested at 10,000 km, and since then (at 27,000 km), it has remained within the same range (approximately 240 km/h according to the speedometer). However, I rarely use it.

- Fuel consumption decreased; initially, it was around 7.5 liters per 100 km according to the car's computer, then from 10,000 km it dropped to around 6.5 liters per 100 km, and now it's at 6.1 liters per 100 km. The average speed was consistently between 110-120 km/h, as indicated by the car's computer.

-The fuel consumption decreased by approximately 1.0 liters. To determine the fuel consumption under constant driving conditions, the following data is needed, ideally obtained under similar conditions. (Output: FIS).

@ 100 km/h: 3.3 liters.
@ 120 km/h: 4.5 liters.
@ 140 km/h: 5.2 liters.
@ 160 km/h: 6.1 liters.
@ 180 km/h: 8.0 liters.

- Air filter for the engine replaced with a new one at 16,000 km.

- Oil consumption: 50601, approximately 0.75L per 27,000 km. Always maintained at the maximum level.

By the way, something that occurred to me recently is that the AGR (presumably a company or service) didn't 'survive' my car pickup at the dealership, and I've been driving 'without' it ever since.
Despite everything, the engine emitted a noticeable amount of black smoke during acceleration (@darkness) for the first 5,000 km, but this gradually decreased until it became completely invisible. At 16,000 km, the black smoke started to reappear slightly, which is why the air filter was then replaced. This completely eliminated the black smoke.

Hello everyone,

There is definitely no introductory program.
So far, the programmers have always complained about a lack of memory in the ECUs.

There is no room for additional features beyond the standard driving functions. Not even for a kind of 'correction factor' for the first 10,000 kilometers.
It's probably due to the normal adjustment process and the moving parts.

A test BMW M3 with SMGII accelerated a few tenths of a second faster from 0 to 100 km/h after 100,000 km compared to when it had only 20,000 km.

cu smith2

smith2 wrote:

There is definitely no easy entry program. "So far, the programmers have always complained about a lack of memory in the ECUs." There is no room for additional features, besides the standard driving functions. Not even for a kind of "correction factor" for the first 10,000 km.

Hi,

An interesting aspect...

Quote:

A test car, a BMW M3 with SMGII, accelerated from 0 to 100 km/h a few tenths of a second faster after 100,000 km than it did at 20,000 km.

I don't really care about the 0-100 km/h acceleration time, especially because of the imprecision of the starting and shifting processes – unless the first gear goes up to 100 km/h, or the car has an automatic transmission.
For a more "precise" comparison, we have "invented" the DZR test here...

Regarding 0-100 sprints, I don't put much weight on them due to the imprecision of the acceleration and shifting processes - unless the first gear goes up to 100, or the car has an automatic transmission.
For a more 'precise' comparison, we have 'invented' the DZR test here.

The vehicle is equipped with SMGII!
In 'S6' mode, the clutch engages abruptly, and the gears are shifted almost without any position control (which makes a lot of noise). Since this process is electronically controlled and not subject to human error, any blurriness in this regard can be ruled out, of course, excluding external climatic conditions.

cu smith2

An interesting aspect...


Could the minimum length of the zero-crossing point, relative to the driving performance, also be responsible for this?

There are likely several factors at play here that, while individually small, have a cumulative and ultimately decisive impact on the overall system.

cu smith2

Hello Ulf!

Unfortunately, I can only vaguely remember the exact mileage figures, but I'm quite certain that with my A3, a steady decrease in fuel consumption was noticeable until around 15,000 km.
Regarding performance, I actually felt quite good from the very first kilometer, and I don't recall any significant increase in performance.

The steady decrease in consumption, however, indicates a reduction in engine friction.

Allein diese Tatsache straft alljene Lügen,die immer wieder behaupten, dass Motoren bei der heutigen Fertgungstechnik nicht mehr einlaufen müssen. VUtter nonsense!

Just today, I had to experience this (bitter) lesson again at work. We installed and tested a new prototype of a hydraulic axial piston pump. We took the measurements much too early, after too short a break-in period, as we later discovered. The mechanical efficiency changed so significantly throughout the measurement series that all the measurements were useless. Afterwards, we ran the pump again for several hours under a moderate load and then repeated the measurements. The subsequent measurements then showed constant values, which was a sign that the break-in process was largely complete.

Regards,
Alex.

smith2 wrote:

The vehicle is equipped with SMGII! In "S6" mode, the clutch is engaged aggressively, and the gears are shifted almost without position regulation (which makes a terrible noise). Since this process is electronically controlled, far from being subject to human error, any blurriness in this regard can be ruled out.

Aha, sorry: BMW's high-tech is largely a foreign technical territory for me . No offense intended!

Julian wrote:

Could the minimum elongation of the Z-axis relative to the driving performance also be responsible for this?

Hmm, any elongation would manifest itself within the engine control unit on the "tension side" between the crankshaft (KW) and the camshaft (NW).
That means the ignition timing would always be lagging behind the optimal point – but precisely that (being 3° behind the optimal timing) cost my engine about 2% of its power at the factory setting in the middle of the ignition advance range.

To gain power from the ZR extension, in my opinion, the engine control unit would have to be set much too early from the factory - so that the engine initially lacks power, and the extension would gradually compensate for this deficiency... but I can't quite believe that .

Okay, I don't have much to say about it, as I only received my car with 16,000 km on the odometer, and the dyno runs were performed on a dynamometer that wasn't properly calibrated.

Has anyone ever considered that this might be a "10-cent tuning" issue related to the break-in period? The fuel temperature sensor could provide different readings over time, which could cause the engine control unit (ECU) to inject more fuel.

Bertil wrote:

Has anyone ever considered that it might be a "10-cent tuning" issue related to the break-in period? The fuel temperature sensor could provide different values over time, which could cause the engine control unit (ECU) to inject more fuel.

Hi Bertil,

Not a bad idea at first.
According to the VAGCOM temperature comparison on Christel's 01 ALH, the diesel injector reports approximately 2°C more than water and air when the engine is cold. In my (new) car, it's exactly the opposite.

But even if my donor were to "increase" its power by 4° over time, I don't think it would result in a significant increase in performance: based on my previous attempts, I estimate the effect to be about 1-2 horsepower .

The question could probably be answered more reliably if we knew the "thermal growth" (% per °C) of diesel fuel -> does anyone know this value?

For further clarification, drivers of "older" engines could check measurement block 7.

smith2 wrote:

Hello @all, There is definitely no introductory program. So far, the programmers have always complained about a lack of memory in the ECUs. There is no room for additional features beyond the standard driving functions. Not even for a kind of "correction factor" for the first 10,000 kilometers. ... cu smith2

Well, if I look at my older TDI chip, there's still plenty of space available... Nevertheless, there should be other maps to be found there, which would then be switched to. And there's nothing like that to be seen – and to achieve all of this with just one or a few correction factors that would cover all the maps without the car "acting up" in any area, I consider highly unlikely.

Regards,
Ronny.

The steady decrease in consumption, however, indicates a reduction in engine friction.


@Ulf:

In a way, I completely agree with Alex. It's a bit of a stretch, but when I think about all my engines (two-stroke engines with glow plugs) in model airplanes and cars, they always had to be run in for quite a long time before they behaved in a way that could be called 'operable.' And that was really just due to the mechanical parts.

Okay, while the TDIs have fewer variations in component dimensions, overall, there are still a significant number of parts that need to be broken in.

Don't you think it could simply be due to that?

P.S.: The only argument that truly contradicts my theory is that a 220 CDI engine disassembled at 112,000 km showed no signs of wear or excessive abrasion. Nothing was measurable. All values were like those of new parts, as if the engine had just been assembled.

Julian wrote:

Okay, the TDIs have fewer variations in component dimensions, but overall, there are still a significant number of parts that need to be broken in. Don't you think it could simply be due to that? P.S.: The only argument that really goes against my thesis is that a 220 CDI engine disassembled at 112,000 km showed no wear or excessive abrasion. Nothing measurable. All values were like those of new parts, as if the engine had just been assembled...

Hi Julian,

"Of course," I believe that break-in effects, with the resulting lower internal friction losses, influence the performance characteristics of an engine over its lifespan.

But I've often read (in essence) that TDIs sometimes start running "suddenly" better, as if something has been switched.
And something like that (if it's true) cannot, in my opinion, be caused by initial setup processes, because these changes tend to develop relatively slowly.

Regarding the disassembled CDI: It's possible that only the friction surfaces were smoothed, without any further material being removed from the original stock.
Could the tools used also be used to measure the minimal reduction in surface roughness?

Unfortunately, my colleague who works at the ASZ (presumably a specific institution) didn't pay much attention to the details. The stories only refer to the PD-TDI (likely a specific product or technology).

From a software perspective, there's no problem. It would also be possible to integrate characteristic maps, but that's not necessary. I can simply reduce the entire characteristic map using a correction factor, which would result in a lower injection quantity. It works similarly for over-temperature situations, etc.
The EDC offers the possibility to do so.
Unfortunately, Bosch has no information available on this matter, as all the relevant data and specifications are customer-specific and therefore provided by VAG.

Regarding the oil change: The viscosity of the old oil had changed drastically, which made it difficult for the turbocharger to spin up. After the oil change, the turbocharger spun up much more easily, which also resulted in noticeably better acceleration.
The effect is slowly fading again.

CU Gremlin.