I just replaced one of my GNSS units that failed to get gps fix as that particular device was a couple of months out of the production window for this specific failure. When I first installed the new unit it obtained a fix although with significantly weaker performance compared to my other two. After a few connects and bench testing it started to oscillate between fix and no fix and at this point in time it simply refuses to obtain a gps fix. I have been using Zubax GNSS since its first year of release and been through all the growing pains but at this point I am very disappointed in the quality control system. These have always been a cut above all other gps units i have used and I want to continue with them but others are catching up and without the same functional issues. These have been purchased through Titan.
As I just wrote elsewhere on this forum, we have identified that the LNA SAW chip we use in Zubax GNSS 2 (manufactured by Qorvo, Inc.) is defective. The problem cannot be detected by the output quality control because it is often latent in its nature: a unit may be fully functional during the QA test, and fail later once reached the customer (which happened in your case). We are currently preparing a minor hardware revision that will be free from any components manufactured by Qorvo, Inc. Will report on the forum when done.
I understand how frustrating this issue with Qorvo may be to you. Could you please provide an official statement on this latent issue? Items I would be interested in hearing are:
What is the characteristic of the latent failure? Could the failure happen mid-air? Does the failure manifest itself only during boot-up?
You mention it is a problem with the LNA SAW chip. Could you elaborate? Is there a drop-in replacement chip that customers can install by themselves to bypass that chip?
You’re still selling the Zubax GNSS. If I bought more today, May 23, 2019, would you expect those items to be potentially defective? If so, what percentage of them would be impacted?
You mention the fix on your end is a minor hardware revision. Would you have a timeline for the fix to be found, how you’re ensuring you detect latent issues in the future, and how long it will take for the upgraded products to start being shipped?
It is supposed to be installed on top of the existing LNA SAW footprint. Please contact firstname.lastname@example.org if you need more information (please reference this post in the email).
We are not. We stopped doing so immediately once we discovered the defect. We also made sure to inform Titan Elite and are keeping them up to date on our progress.
We are expecting the new revision to be available by July 2019. As to the detection of latent issues in the future: by definition, it is theoretically impossible to do, just like you can’t detect unknown bugs in your software because by definition they are not yet known. Our recently updated internal policies dictate that all new designs are to incorporate only parts certified by AEC-Q100 grade 2 or equivalent, which we believe is at least sufficient to prevent similar problems from occurring again.
There is no robust way of diagnosing the problem unless you have GNSS emulation equipment. I recommend that you put your units on a windowsill and leave them there for 30 minutes. If there is no fix, they are likely to be defective.
Heads up: we are slowly rolling out fixed GNSS boards. Please email email@example.com for replacement. We have extracted generalizations from this experience and taken preventive measures to ensure that similar issues will not occur again.