A New Method for Accurate OTA Production Testing of Wireless Products
TESTxpert is the fastest, most repeatable and most flexible family of customizable production line over-the-air test systems.
Difficulties of a traditional conducted measurement:
- Conducted measurement is slower. It relies on a person to connect/disconnect the RF connector.
- It is expensive because the RF connector cable will wear out fast in a production setting.
- It doesn’t take the antenna into account. For wireless device, antenna is the critical aspect of the entire system performance. So, this test is missing a large source of problems.
- Many new designs don’t even have an RF connector. So this test has to be completely removed and replaced by the over-the-air test.
- Difficulties of a traditional over-the-air production testing
The path loss which is driven by the coupling factor is very inaccurate. Small variations in the positioning of the AUT can create large uncertainties in that path loss value.
It has to be tuned for every device model that will be tested. If the production has to switch to a different model, the coupling has to be changed and every time a new model is introduced, the coupling has to be redesigned and retuned. This often takes weeks to months to perform.
Current coupling accuracy of +/-2dB can rarely be met for the production line testing.
OTA testing is antenna specific and there are multiple antennas in a device. Each of these antennas require a separate coupling element. Some devices have 7 antennas or more which means you need to have 7 coupling elements or 7 offsets factors for the coupling element. Each element or factor has to be individually designed and tuned, making the process even longer.
Difficulties of a traditional full quality control testing
This very often done by sending daily samples in an anechoic chamber. This is a slow and expensive approach.
Such test takes several minutes per device. If you want to test a substantial amount of the devices being produced, this will be a very long process that will delay identification of production issues.
A trained operator is required which means this can’t be done in the production environment.
Novel Approach to over-the-air production testing
The RFxpert is our product development tool. It is a mature technology. It’s well understood and been very well researched. The RFxpert is actively being used in R&D and design facilities around the world (scroll down to see some RFxpert customers).
TESTxpert: The Technology
The coupling plate does not need to be redesigned for each specific model; only a probe has to be selected for the best coupling performance. This means that once the probe effect is understood for each antenna under test, it’s the same for every device of the same model that is being tested.
In this technology, the path-loss is very well researched and controllable and the coupling effects are well understood.
TESTxpert: The Concept
How does it work?
Instead of a test box with coupling elements underneath the AUT and the jig to position it very accurately, we have this very simple setup where the AUT is simply placed on the surface.
TESTxpert is composed of an array of probes. Each grid corresponds to a unique probe. Probes underneath the AUT will be used as coupling elements for OTA testing.
The AUT is placed on top of the array of probes. The different antenna elements in the AUT will lie on top of several probes. Each one of these probes can then be selected as the optimum coupling path for every antenna on the phone.
TESTxpert is flexible: Test different models, different frequencies without changing the test setup
TESTxpert is the generic test fixture for all product models.
As you switch to different technologies, different frequencies, and the device changes between antennas, the array of sensors can simply switch to a new near-field probe and provide a new coupling path.
So regardless of the model you are testing, regardless of the position of the antenna elements on the AUT, the same array can be used and re-programmed on-the-fly.
As you switch between different models, you simply make a change in the programming of the array. If you’re moving to a different technology or a different frequency, it doesn’t matter; these probes are already very broad band from 300 MHz to 6 GHz and they don’t need to be re-tuned.
So this gives you incredible flexibility in adapting to different models and different antennas in a production line.
TESTxpert is reliable and repeatable: Future proof your production line
One of the biggest problems right now for the current OTA testing is the fact that the coupling path has a very uncertain path loss. The fact that the TESTxpert coupling path is very repeatable with a +/- 0.5 dB tolerance is a critical advantage. The impact from positional errors is almost negligible. This means even a rough positioning of the device is not going to affect the measurement accuracy.
High measurement accuracy helps reduce the threshold for your PASS/FAIL test, decreasing substantially the false positives and the false negatives. Immediate benefits are less sub-standard products in the hands of the end-user and less rework in the production line.
TESTxpert transparent to your current parametric tests
Once the probe is selected on TESTxpert, it’s operating as a passive connection like your current OTA coupling plate. You have nothing to change in your test equipment and test plan.
TESTxpert deals with complex MIMO-based devices
You can integrate parallel OTA paths inside the same TESTxpert array which will allow you test multiple antennas at the same time and do other more complex measurements as required.
TESTxpert results: comparable to traditional conducted measurement
The variation in OTA measurements on the TESTxpert is almost as small as the variation in conducted mode. This is far better than the variation that a typical OTA setup would be able to produce.
Add new capabilities to your production line with TESTxpert
TRP: absolute radiated power measurement with a 2-sigma accuracy of +/- 1.6 dB or relative PASS/FAIL radiated power comparison with a +/- 0.5 dB accuracy.
Pattern: rapid PASS/FAIL pattern comparison; important for devices designed with beam forming or other adaptive capabilities.
Dynamic effects: detuning due to body parts can be evaluated in a production line with a phantom kit equipped TESTxpert.
Tuneable antennas: the very fast frequency by frequency measurement capabilities helps you verify the antennas are tuning to the frequency as expected.