Is Testing Difficult – comparing housings in the thermal and acoustic

Is Testing Difficult - comparing housings in the thermal and acoustic

Fundamentals Article: Why Case Tests Seem So Easy And Yet Almost All Are “Wrong” – This Is Really!

Even if the title may sound a bit martial and exaggerated – it is the bitter reality if you really want to objectively compare housings in the thermal and acoustic area.

And that’s what this article is all about. Of course, subjective factors such as appearance, production quality and usability are also part of the content of a good review, and both deserve a peaceful coexistence. You just have to call it that. Hands-on and usability on the one hand, the actual cooling performance, noise emissions and damping for an honest comparison on the other.

In today’s guest post I just want to describe the point of view of those who really want to make it resilient and objective, without diminishing the performance of those which present a housing from the user’s point of view in the normal reviews. But it is about legally secure assessments and certifications.

System2
System1
System2

Initial situation: how much effort do you have to make to determine what?

Some people really think that testing a case is a neat piece of cake, because all you have to do is install a system in it and mention in the text how easy or difficult it was, whether you like the internal and external design and what it is so as a bonus in the scope of delivery. And if you’re not being paid to look good, then you can say a few words about the quality of workmanship. You can (and should) do that, but then please not as a comparison. Let’s call it product presentation and user experience, because it has its justification and target group. But everything that continues is now getting tricky.

Some more savvy testers move on to the next step and also evaluate thermal performance by using the same system in all of their case tests and recording and comparing the temperatures of critical components such as CPU, GPU, drives, VRMs, etc. in a database. Even more experienced testers then use at least the delta temperature to compensate for the different ambient conditions. Assuming you don’t have a climatic chamber, you can’t guarantee a stable operating temperature in this case either. Thus, using the delta difference is the only way to have a reasonably useful reference point for all evaluations. But that is not physically exact either and does not do justice to some products. One cannot claim objectivity in this way.

After all, very few testers measure noise emissions in addition to thermal performance and try to normalize the results by applying the same test conditions to all cases. But you have to explain that a bit. Because you cannot directly compare an enclosure that emits 40 dBA at full fan speed with an enclosure that generates 35 dBA with its fans at maximum speed. The correct course of action would now be to reduce the fan speed of the loud enclosure and let its fans rotate at the appropriate speed that allows it to output 35 dBA. Only then could one also compare apples with apples. To do this, however, one would have to normalize the fan noise levels. To even be able to do that you would have to use a semi-anechoic chamber (hereinafter simply called a chamber) and a high-precision noise analyzer and then follow the relevant ISO guidelines, which specify exactly how to place the microphone and the test object in the chamber. Such equipment is expensive and it is difficult to build a suitable chamber. Well, what now?

Read This Now:   AeroCool Double Power, Thermaltake DuOrb, Zalman VF1000 LED and a few more good coolers for video cards

Another critical factor when it comes to honest housing reviews is the consideration of the noise attenuation (colloquially also insulation). How do you realize something like that? Very simple: you install a constant source of noise inside the housing and then take measurements. The noise source is then removed from the housing and noise measurements are taken again (or vice versa). In the end, it’s just a matter of subtracting to find the noise attenuation as the difference. But which noise pattern should you use? Pink or white noise? And yes, if you really want to be thorough, you could still use a long chirp pattern. Let’s assume that the tester wants to provide interested users with even more detailed information. In that case one could use many different frequencies from ultra low to super high and find the noise attenuation performance for each frequency (or different frequency bands). For example, this is what Cybenetics does in the case reviews. It is a demanding process that requires first class equipment and an advanced chamber. But it also goes far beyond the limits of a “normal” review.

Mainboard 1
Mainboard1a
mainboard2
mainboard3
Mainboard 1
Mainboard1a
mainboard2
mainboard3

Why a real system must lead to errors

However, the biggest mistake with most of the reviews is that they rely on a real world system that has the following drawbacks:

  • It needs to be replaced frequently to keep up with technology trends and to provide the “wow factor” you need. Nobody will be pleased to see that a 3-5 year old test system is in use, although thermal performance is what counts most in these tests, not performance. But that doesn’t interest a superficial observer.
  • You cannot apply a uniform, precise and, above all, adjustable load with it if you are using a real system. You rely on how well the benchmark software can hold the load stable. However, even small load deviations can affect the recorded temperatures by several degrees, which in turn is sufficient to distinguish one housing from another (unjustifiably). In addition, you have to set all options that can influence the wattage of the CPU first manually in the BIOS and of course the same applies in the end to all corresponding options of the GPU. Measurement tolerances and errors in reproducibility add up very quickly here, unfortunately.
  • It is not easy to test different scenarios, i.e. different hardware, because you have to use a fixed system to keep all measured values ​​compatible between the evaluations.

With a simulated system, however, it would be possible to adjust several parameters quickly and easily, so that one could test different scenarios comfortably (and reproducibly). For example, if you had a 250W CPU and 350W GPU installed in a real system, you would not be able to see the performance differences between a system using a 100W CPU and a 200W GPU . If, on the other hand, you can set the required load precisely, this is absolutely simple and you can simulate all CPUs and GPUs from low-end to high-end without any problems. Sounds good? It is too!

Sensors and fans – other imponderables

Incidentally, many testers rely on the temperature sensors of the mainboard, the CPU and the GPU and you can never know whether these sensors are adequately accurate and really report correct temperatures over the entire range. Well, one could argue that none of this would be a big problem from the moment you used the same system in all of your tests, but every measurement system needs to be calibrated from time to time and that is exactly something you built in with Sensors (aging, firmware updates, etc.) can not do. Certainly there are a few testers who use correct and calibrated temperature loggers, but you have to be able to ensure that these temperature sensors are placed in exactly the same place for each test. Otherwise, the results provided will not match those of previous tests. The best way to ensure that the temperature sensors are providing accurate information with each test is to drill small holes in the appropriate heat sinks or metal parts and then reinstall them in place over and over.

Read This Now:   Colorful and quiet budget: Arctic P12 PWM PST A-RGB 0dB in the test - that's how it has to be!

Controlling the fan speed is another challenge. One would have to have total control over all system fans to normalize fan noise and also get a detailed log. However, if you want to carry out automated tests, a programmable fan control is essential. And already we have come to the actual core topic, which must be clarified for efficient processing of large numbers of test objects.

1
2
3
4th
1
2
3
4th
5
1
2
3
4th
1
2
3
4th
5

The automation of the processes minimizes errors – the test system takes shape!

Correct implementation is therefore the most important thing of all. We are all people who are prone to mistakes that can be made while running the test or when submitting the results to an Excel spreadsheet. In order to avoid all these mostly unconscious errors and to enable long test sessions in which all data is logged and reported, you need an automation, a control and monitoring program that controls all instruments and various test scenarios with variation of the applied load and the fan speed.

Finally, when everything is ready, the same application should be able to deliver the data in the required format for further analysis, from CSV to Excel files. This part of the code can get very complicated and expensive and requires long beta tests. Still, once it works well it will give the most accurate results and the whole testing procedure can take many hours without anyone having to oversee the whole process. In addition, with such a program one can also determine average values ​​at certain points in time, which are much more precise than if one just carried out a test and recorded random spikes.

Jalapeno 1
Jalapeno2
Jalapeno 3
Jalapeno 4
Jalapeno 1
Jalapeno2
Jalapeno 3
Jalapeno 4

This is precisely why we decided to develop a fully custom system that will only be used in our case tests to take into account all of the above. We have already developed something similar for our heat sink tests and the accuracy achieved is impressive. We now had two options: we could either design something from scratch or we could use a real system and adapt it to our needs. Since we don’t have a CNC machine in our workshop and 3D printing was out of the question, we therefore decided to find some suitable PC parts and modify them for our needs. This has two advantages, because you can use a real system and the costs are significantly lower in the end.

We started with a broken mainboard with sufficiently strong heat sinks. We replaced all FETs and chips under these heat sinks with special, powerful resistors – our heating elements. Simulating the CPU is not easy because you have to remove the socket and squeeze several resistors together. With a lot of effort we paved the way and installed six 140 W resistors in the area of ​​the CPU socket. We will use a standard heat sink to cool the emulated CPU. The areas where appropriately sized resistors are installed next to the CPU socket are the VRMs, RAM slots, chipset, and NVMe slot. We will also modify an HDD. The loads are adjustable in all parts and the CPU load can be up to 600 W if required!

Read This Now:   Fitting of fans in one housing including the overpressure concept - How many fans do you actually need where? | Short test

In addition to the mainboard, we also “butchered” a broken graphics card, from which we removed the GPU and installed several power resistors in its place, which can also deliver up to 600 W load if required. Is that enough? Yes it should.

GPU1
GPU2
GPU3
GPU4
GPU5
GPU6
GPU7
GPU8
GPU9
GPU1
GPU2
GPU3
GPU4
GPU5
GPU6
GPU7
GPU8
GPU9

The CPU and GPU loads are controlled by a powerful and high-precision laboratory power supply (AimTTi QPX600DP), which is able to deliver up to 1200 W together. It’s an incredibly heavy burden, and we will probably never use it to the full in the end. We plan to use different load combinations for both the CPU and the GPU. For example, we could use three different loads for the CPU, 100W, 200W, 250W, and three or more other loads for the GPU, 150W, 250W, and 350W. The possible load combinations are therefore nine, and if you consider three of four different fan speed scenarios (for the case fans) (25 dBA, 30 dBA, 35 dBA and full fan speed), then all test combinations can go up to 36 variants!

AIM-QPX1200-1k_0
AIM-QPX1200SP-bk-1k
AIM-QPX1200-1k_0
AIM-QPX1200SP-bk-1k

If you now assume that each test takes at least 15 minutes, and if you allow for 5 minutes of idle time between each test so that all parts can cool down, the total time for all test sessions is 720 minutes or a full 12 hours! For this reason, you simply need a fully automated test suite that controls and monitors everything. A major problem we had was with the automatic control and monitoring of the fan speed, but luckily the Corsair Commander Pro turned out to be easy to hack and very capable of the task. Please forgive us this sacrilege, it was just too tempting.

System_Complete1
System_Complete2
System_Complete3
System_Complete4
System_Complete1
System_Complete2
System_Complete3
System_Complete4

What will come what will

This was a brief presentation of what’s to come. Follow us for more on Hardware Busters, and Igor, keep rocking it! Together with igorslab.de we will enter into a kind of strategic partnership and thus make the measured data available to the readership. I am sure that the readers of this site in particular will appreciate scientific methodology and a professional approach. This will not be able to replace “normal” case tests and it should not be used at all. But it will add reliable facts to the testers’ point of view, which otherwise could only have been determined with extreme effort.

We will certainly expand this partnership to include the measurement and evaluation of power supplies, because Cybenetics has been certifying and measuring power supplies for years. However, it is still necessary to implement a suitable interface, then it is a fine win-win situation for everyone involved, including the readers, of course. And Hardware Busters’ YouTube channel is always worth a visit!

https://www.youtube.com/watch?v=GbsPHDcALbA


Notice: ob_end_flush(): failed to send buffer of zlib output compression (1) in /home/gamefeve/bitcoinminershashrate.com/wp-includes/functions.php on line 5373

Notice: ob_end_flush(): failed to send buffer of zlib output compression (1) in /home/gamefeve/bitcoinminershashrate.com/wp-includes/functions.php on line 5373