Personal statement on the test of the Alphacool Apex Stealth (Power) Metal Fan

For this necessary statement, I consciously took a few days, even though the community understandably demands urgency, as it turned out to be more complex than I had anticipated. Given my high value on transparency and respect for third parties, the backstory must play a role, as it explains quite a bit (though not everything). However, I’m not among those who casually return to normal daily life by just waiting things out.

We have thoroughly documented the development of our measurement setup, and there’s no need to repeat the details here. Everything was calibrated several times and cross-checked for plausibility with reference fans. If you’re wondering why I didn’t conduct and oversee this in my lab, understand that a day has a limited number of hours, and even spaces have their limits, despite being spread over two floors. Community involvement at this stage also requires gratitude, yet control as well.

Development of the Apex Stealth Metal Fan

First, I must clarify that I was not directly involved in the decision-making process for the fan’s development, certainly not contractually or for any compensation. However, I agreed that our employee, in his spare time, could use the existing technical possibilities to test various iterations and development stages of the fan. The actual development always took place at the OEM and Alphacool in the factory, not by my employee, who only had a supervisory role for the measurements. It’s normal that some insights were shared, but this was not engineering.

I only looked at the requested documents afterward, and it’s evident how things moved in a direction that should have been noticed sooner. I won’t question the employee’s work in detail, but there are certainly valid questions remaining, which we will now actively address.

As an enthusiast, one always hopes for a miracle, and I too was significantly influenced by the euphoria. In a way, I was even the first pre-orderer who wrote the corresponding article, only to be disappointed. Now, of course, I know better, as I only received the final model in December. In hindsight, I shouldn’t have written the article, as I only had a prototype with a 4-pin motor, which, however, performed quite well. My mistake.

This doesn’t deny the uniqueness of the overall concept developed for the fan, but in this version, the fan unnecessarily loses a lot of potential in close range. It’s definitely an extremely quiet and efficient case fan with good and very functional material selection and decoupling. But because of the “focus ring,” which concentrates the airflow too much in the center, it’s not optimal for radiators.

Necessary Consequences

Although this was not the first major deviation in the employee’s tests, I didn’t take yesterday’s decision lightly. The internal review I commissioned of all fans tested so far (over 50) showed a completely plausible picture of all measurement results. A lot of time and effort from the employee went into this project. Nevertheless, after considering all circumstances, I had to make a consistent decision.

Since I always place a high value on valid work and tests like those of the Alphacool fan should never have been passed on unvalidated, I relieved the employee of all further fan tests yesterday, as much as it pained me at that moment. I can’t fully assess what exactly happened and how, but I also feel somewhat deceived. Furthermore, I’ve taken another article temprary offline for review.

Certainly, my mistake was the lack of follow-up checks and the neglected four-eyes principle (not only in this case), but it wasn’t actually my job, as I wasn’t directly commissioned with the tests. Nevertheless, I acknowledge my negligence and blind trust in the results and marketing. That’s why I’ll soon take the test offline after a revision, as well as the prototype test values from the database. To avoid accusations of covering tracks, all content will remain online for a reasonable time, with a comment and preface, of course, and marked as a prototype.

The first step after completely relieving the employee of this duty was to hand over the complete testing equipment to an uninvolved third party yesterday. This person will conduct an inventory analysis, counter-tests, error determination, and a retest over the holidays. This will include modifications and work guidelines to prevent such errors in the future. I trust the competence and neutrality of the third party I appointed.

In this context, an expansion to practice-oriented temperature tests on a water circuit is planned, based on my existing waste heat testing device (thanks to Bernhard Baumgartner) with adjustable outputs up to 250 watts. Then, tests can be conducted to determine the influence of a fan on cooling performance for various performance and speed ranges.

For the fan tests, I will have to find a personnel solution after completing all scientific and practical investigations, capable of handling Failure Mode and Effects Analysis (FMEA) to avoid such deviations in the future and to be able to identify their causes. FMEA is an analytical method for identifying potential errors with the aim of recognizing and preventing them early. It’s described as a standardized method in the DIN EN 60812 standard, but requires the necessary training and software.

Conclusion

Blind trust in calibrated technology and lack of on-site control will eventually take its toll, especially when it leads to such results due to many circumstances. But I also need to protect my not insignificant investment, which is in the interest of the readers, so I must pursue a different personnel approach. Especially when something like this happens, it ends the possibilities of well-intentioned leisure time analyses. I advise anyone who wants to do more than just screw fans onto radiators and measure a delta to take this to heart. I’ve already paid my tuition fees here. But one should learn from mistakes.

And that’s exactly what I’m not excluding myself from. Yes, it’s convenient when you don’t have to take care of everything during a 12- to 14-hour day but let others do it, especially when it’s things that don’t directly concern you. But you can’t blindly trust everything and everyone, even if it’s the easiest way to end the workday without collapsing. I definitely have to own that. As for the employee, I wish him all the best, and he probably knows best that I went a huge step beyond what was necessary at the end of his tenure, although I didn’t have to. Let’s see it as recognition for his work so far, which, however, must end here if we are to make an honest cut and continue working at the demanded level.

The Apex Stealth Metal Fan is actually an excellent product for its price point, which should have been marketed differently. But I want to make it clear that I’m not responsible for Alphacool’s marketing and their internal processes up to pre-order. I can only apologize to our readers on behalf of all involved. Everyone should draw their own conclusion from the whole story.

What will happen in detail next?

Fritz Hunter will take over the inspection. First, new fans will be procured from different manufacturers already listed in the database. Not all, but some important ones. With one of these fans, the reference will be set. Not in terms of the fan, but the entire measurement system. In other words, he will conduct an MSA 1. MSA stands for Measurement System Analysis (MSA). This initially demonstrates the basic capability of the measurement system. Then, two additional fans of the same type will be measured 10 times each, creating an MSA 3. This must also turn out positive, as we would otherwise not have a valid measurement setup. Fritz assumes as of today that the measurement setup will be valid. However, this effort is necessary for safety. After demonstrating the measurement system capability, there will be a first article on the topic. As part of an FMEA, Fritz will not only theoretically record all possible errors but actually try them out. This means:

1. Fans will be deliberately installed improperly, skewed, or offset.
2. At various (even seemingly senseless) places, such as the chamber inlet/outlet.
3. In front of or behind the middle partition.
4. In front of the radiator (Push) and behind the radiator (Pull).
5. Measurement devices will also be positioned incorrectly, among other things.

These error scenarios will be measured, as far as possible, and compared with the reference. Everything will be documented visually to create a clean work instruction for the measurement setup. Currently, some equipment is still missing to even start. This includes fans, software, radiators, etc. We need to procure these again in the next few days. Therefore, it’s currently difficult to estimate when we can provide you with further information.

Fritz also secured one of the APEX pre-series fans from Alphacool at his home, which one exactly, we still have to find out. At least Fritz has already remounted and thus repaired it. Since we do not know under which circumstances the fan was measured, a correct retest is not immediately feasible. Therefore, I ask for understanding that we will only measure this again once the measurement setup is validated. Whether we will have to retest all previously tested fans will only become clear through the processing. Since we will use another measurement setup for temperature and plausibility tests in the future, this may also still be ahead of us. The measurement setup sketched above will also have to undergo an MSA 1 and MSA 3 at Fritz’s place. That’s for sure. Much safer!

Important addendum

In order not to escalate further and also to remove those involved (also with regard to the redistribution of personal views and comments of many posters), I have decided to remove the comment function. I also owe it to the emplyee and those who do not need to be drawn into a possible escalation. Thank you for your understanding, it should certainly have been done in the first place.