Author Topic: Continuous improvement - increasing the reliability of the HEX ezCAN GEN-II  (Read 803 times)

StephanT

  • Administrator
  • Sr. Member
  • *****
  • Posts: 449
  • Karma: +21/-2
    • HEXcode
Over January we noticed a slight but noticeable increase in specific ezCAN failures. Albeit a small portion of the overall units, it was an uptick in failures highlighted by our trending.  Although we have seen a small portion of similar failures the numbers were so low that previous investigations had failed to pinpoint the origins of the failures. Armed with more information at our disposal we launched a detailed investigation into the cause of the failures.  As the initial investigation hinted at a possible hardware issue, we temporarily halted production, until such time as we had a definitive reason pointing to the origin of the failures. 

The symptoms associated with this particular failure mode are ALL of the following:
  • The ezCAN is completely dead - no blinking LED, no permanent LED, irrespective whether ignition is ON and/or powered by USB
  • The physical 30A fuse has continuity and is in working order
  • No USB communication with the ezCAN from the PC/MAC software
  • No physical damage
We assembled as many of the failed units as possible. Every unit was put through our rigorous manufacturing test jigs. Additionally, many units were X-rayed. Internal physical damage was found in almost every single case. This was mostly contained to one component near the edge of the board where cracks could be seen between the pins of the component and the solder pads.

After thorough analysis of the data collected during the investigation, it was determined that the potting seemed to be the accelerating factor.  Although not all potting or epoxies are suited for electronic components, the one that was in use was carefully chosen and complied to all the requirements, especially the shrinkage factor that so often leads to similar failures if the incorrect compound is used.  It is our opinion that the shrinkage due to the colder than usual winter was a contributing factor that highlighted the interaction between several individually otherwise well-suited parts. We are of the opinion that it is not the potting compound per se, but a combination of the different shrinkage constants between the aluminium housing, its form and the potting epoxy that cause the forces on the component level to increase around the rounded edges of the housing where the shrinkage of the aluminium translates to a maximum physical deviation. This conforms to the physical failures we see on the component level on the edges of the PCB, nearest to the rounding of the housing.

We could replicate the failure symptoms on some ezCAN devices by super-cycling a sample of devices in our environmental chamber. This included high stress, accelerated tests (multiple hot to sub-freezing cycles per day), increasing the expansion and contraction forces on the devices. Naturally these fast cycles do not occur in the real world, but their extreme nature helped us to highlight the issues and verify the proposed fixes.

A clear fix was to replace the potting epoxy with a softer polyurethane type. Various products were tested, and a suitable replacement was selected.

Devices manufactured with the new potting compound were put through the same accelerated super-cycle tests in our environmental chamber and we are highly satisfied with the results. Not a single unit failed the extreme cycle test.

We have asked our distributors to replace any GEN-II ezCAN failures that have failed with the above-mentioned failure symptoms, that were manufactured prior to the manufacturing update, irrespective of whether they are within their 2 year warranty period or not. These include only BMW, KTM and HD GEN-II devices sold prior to March 2021.

If you are an ezCAN customer affected such a failure, please contact your distributor.

Finally, there have been some rumours circulating that the GEN-II ezCAN has been discontinued and is being replaced with a GEN-3.  Although a GEN-3 will most likely replace a GEN-II at some point in time, it will most certainly have to be conceived in concept first, then designed, refined and finally industrialised… and as we have just put incredible effort into increasing the reliability of the GEN-II devices, GEN-3 will not be anytime soon… if however you are waiting on something new, wait for some interesting accessories for your GEN-II ezCAN!


Whether you are traversing the Outback, crossing the Savannah, getting lost in Mongolia or finding your reflection in Lake Titicaca, your HEX ezCAN should serve you well!  Be safe and enjoy the ride!

Stephan
« Last Edit: May 12, 2021, 04:13:55 PM by StephanT »
consider supporting my fundraiser for the Heart Foundation:
https://www.justgiving.com/fundraising/stephan-hex
currently riding: BMW F850GS Adv - slightly modified

NH1200c

  • Newbie
  • *
  • Posts: 4
  • Karma: +3/-0
I'm in my first year of ezCAN ownership and haven't had any issues. However my entire working career was in high reliability electronics for both military and aerospace applications.

Just wanted to say thank you for taking the time to understand the failure mode and undertake the investigation including DPA to truly understand what is going on. Very nice to hear you have modified your process to enhance reliability and will stand behind the products which failed.

Very well done to you and your team -

NH1200c
Larry
BMW R1250RT
GS-911 & ezCAN

CaptainBob

  • Newbie
  • *
  • Posts: 2
  • Karma: +0/-0
Thanks for the write up, I got the same issue with mine now and support is working on getting it replaced.  I use mine for two heated gear ports because it is still cold in Colorado.  Only lasted a few weeks, I'm happy you found the root cause.

Thanks,
ME
______________
2016 K1600 GTL

StephanT

  • Administrator
  • Sr. Member
  • *****
  • Posts: 449
  • Karma: +21/-2
    • HEXcode
thanks Gents - appreciate the feedback!
consider supporting my fundraiser for the Heart Foundation:
https://www.justgiving.com/fundraising/stephan-hex
currently riding: BMW F850GS Adv - slightly modified

fschuurman

  • Newbie
  • *
  • Posts: 1
  • Karma: +0/-0
After reading some dreadfull posts the last months on amazon and some biker forums, i have renewed confidence from this post. Appearently this is not broadly communicated but i will purchase my ezcan within a week. It should be in de square boxes and not the bags i believe.

Cheers.

motoUX

  • Newbie
  • *
  • Posts: 9
  • Karma: +0/-0
I have received my new GEN-II version Friday and plan to install on Wednesday. I'll report progress.

FYI, my (2nd) unit currently on the bike works when the temperature gets above 80º F. so that tracks with the idea of contraction on the existing unit's epoxy issue.