Select a technical image to view it in detail.
In Part 1, I introduced a seller who “manufactures” thousands of refurbished injectors. Along with the injectors, the deceived customer received test protocols that supposedly confirmed their excellent performance. Here I look closer at what those protocols actually show.
Here are all the injectors:

And these are the test results:






Identification problems in the reports
The tests were run on Carbon Zapp equipment, which is already familiar from many other entries.
What I dislike most is that nothing ties a test result to a specific injector except the hand-written #1, #2 and so on. There is no other unique identification, and the serial numbers in the S/N fields do not match these injectors. The person who ran the tests typed those numbers in by hand, so the excuse that nobody bothers to fill in those fields does not apply. Yet the numbers do not correspond to the injectors. This reduces my trust in the test report to zero.
The reports also carry no company data. That is more than strange, because the company advertises actively on eBay and sells thousands of refurbished injectors. I wonder why.
If those points are not enough, let’s go further.
First test: R2LC electrical measurement
Electrically, an injector behaves like a capacitor. I measured it with several different instruments: a regular multimeter, a specialized LC meter and a 5.5-digit reference-grade meter. For more or less recent releases (07 to 11), the injector capacitance lies in the range of 2.2 to 2.6 uF. Exact figures appear in the other Carbon Zapp tests (see the reports in other entries).
The reports here show 4.1, 3.8, 2.8, 4.1, 3.5 and 3.5 uF. Interestingly, the value closest to the correct one belongs to injector No. 3, and that is also the injector in a brownish body, so it looks like a genuine *048 injector.
I see two fundamental problems in this data:
- The electrical parameters are very far from the expected average of 2.4 uF, by as much as +70%.
- The spread is wide: the difference between the minimum and maximum is 46%.
For me this means:
- Apart from No. 3, these tests were definitely run on injectors that are not from the *048 releases of the last ten years, and certainly not from releases 07 to 11.
- The five tested injectors in black bodies are either poor-quality “alternative” injectors, or injectors of different types, or some of them are damaged. Because of the mixed-up parameters, it is hard to say which one might be damaged.
Why capacitance matters
Here is what the injector control signal looks like:

This is the control signal at idle, with an opening time of around 200 us as shown by INPA. Two things stand out:
- The voltage supplied to the injector reaches its maximum (130 to 150 V) only after around 100 us.
- The signal rises for around 100 us and falls for the same time.
So the injector spends most of its time opening or closing. The reason is the nature of the injector itself. The DME high-side switch uses PWM to charge and discharge the capacitance of the injector’s piezo element with a large current of around 10 A. That is the maximum current the MSD80/81/87 can deliver and the injector can endure.
If the equivalent capacitance is 70% larger than 2.4 uF, as with these tested injectors, the rise and fall times also grow by 70%. That is the time during which the injector opens and closes.
The minimum opening time (confirmed by INPA/ISTA) is only 100 us. With “normal” injectors, this means there is no voltage plateau at all: the injector opens for 100 us and closes straight away. Injectors with increased capacitance simply cannot work in this mode. My conclusion is that injectors with such large capacitance cannot be used in N43/N53 engines, nor in N54/N63.
I have no doubt that these injectors are not even worth installing. But since we have started examining the test protocols, let’s look at one more oddity.
TP4 test: injected fuel quantity
The TP4 test measures the fuel injected during a short opening (240 us). The test pressure is also close to that of the N53 engine (180 bar versus 200 bar). In the test reports, the measured parameters of all injectors are similar, yet the coding data differ by as much as 3.2 times (224 versus 715). How is that possible? I see two scenarios:
- The results are completely fake.
- The test conditions (for example, the charge and discharge of the injector) were radically different.
On a test stand, charge and discharge currents of 50 to 100 A are not possible. They would be needed to make the data correct given the large difference in capacitance, and the injectors themselves would not withstand such currents. I have no slightest doubt that these test reports, except No. 3, do not belong to the injectors that were sold.
Conclusion
The seller may have been very cunning:
- During the electrical test, the real (sold) injectors were connected to the Carbon Zapp equipment. This can be done in-house, and it lets the seller check that they match.
- For the other measurements, different, correctly working injectors were used.
That yields a set of tests that is hard to challenge, and hard to use as grounds for demanding a replacement.
If my version is true, then with all due respect, this is fraud at a professional level. The image of the honest seller has broken down.