Tech. Support
(request for new user)
  • This email address is being protected from spambots. You need JavaScript enabled to view it.

Switch on and forget

The basic idea for industrial production of devices, machines or appliances, is to produce thousand or more of devices working 24/7..

Our entire engineering and production process has been designed to accomplish this target.

First of all, safety means reliability of the board during your system’s lifetime, many years, with an appropriate mission profile that considers the real usage of the board and it’s accessories, starting from the power supply. A 24/7 mission profile, is different than 8/5, and not just for the global working time that changes so much, but also from the number of bootstraps of the system that, not in a Novasom board for what I’m going to tell you, can compromise the functionality of the system itself.

Then there are the environmental stress factors, from the temperature to the EMC noise and spikes, that can easily break a power stage or GPIO not correctly protected.

A big problem with temperature range, is that most people don’t think , as we always do, to propose the conformal coating on the “outside temperature range “ board , and fail to consider on the datasheet the minus something indicated there.

The main problem is the humidity condensation that make short-circuits between test pad or components, especially 0201 often used for the stabilization of the DDR and USB3-Pci express bus.

Then, this condensation may freeze under at 0 (yes it is not necessary to go to -40, water becomes ice @ simply 0 Celsius), this increases the volume and cracks the BGA from inside the ball.

Otherwise, instead of thousands of fruitful installations, somebody will gain thousands and thousands of problems, angry customer’s calls, loss of reputation and other, not minor, problems.
The production phases, are complex but manageable. You can do extensive tests, with good testing machines, several runs of prototype to be sure that your products work like you planned them to work.

Yes, it’s complex and expensive I agree, but to do without it is not feasible.

The main hidden problems that appear when it’s too late, are the architectural- engineering problems.
A complex product line, an SBC for example, has not only the functional need you should accomplish to develop your products, but many other hidden, or not so visible, needs, that will be a problem if it’s improperly used or...when it’s too late.

How many times have you said “oh if I had only known” front of a problem.

This is our principal target. Our idea is to give you an SBC solution that’s not only fast delivery or customized, but one that comes from the experience we’ve had, so a lot of problems have been solved. And you’ll get the solution, not the problems to solve...much better.

Think of EMC protection from pulse and surge. All our boards are designed to survive in this kind of real ambient noise, with big margins certified also from third party labs that take care of our certification (see www.kksdk/ certification), but also thinking to well protect the power line of the board from any accidental short circuit you may do, or polarity inversion.

And any output, is protected to avoid damage to the board if something strange happens.

Or I should say when, because problems happen.

Consider now the mission profile of a semiconductor: is the mean time between failure measured in hours and on-off cycle. Industrial components we use are designed to be used 10Years on a 100% duty. Commercial components are for 2-3 Y with a 20% duty. This is not the same...

Last but not least, for the high end level computing up to our windows x86 proposal, we have studied how to get those kinds of systems to work for a long time, and have determined that the main failure mode is the power stage.

So we add a dual power stage input. Because two, is better than one. Sure. And when one fails, the unit can continue to work and notify you to change the broken one.

Isn’t it amazing? We think -- not enough

M-Line Novasom Industries

But we’ve studied how we can solve an unpredictable black out for a windows machine, where as everybody knows the OS doesn’t like to be switched off without notice.

So we‘ve added an internal UPS, managed @ bios-kernel level where you just add a battery, and let the system survive or switch off correctly. (See M11 line)

This is amazing.
This is our way of thinking switch on and forget

Just call us

We have the solution for you