We quickly discovered that the relay switching was inducing lots of noise throughout the system. Most notably, the backplane power was showing 6V-10V high-frequency noise on the 5V and 24V supplies, GND and frame GND, that was synchronized with the relay opening. The supplies (also made by our partner) were specified to limit transients to 10 percent over the supplied voltage. The noise was going right through our supplies and filters, and appearing on the outputs. With this much noise in the system it was a minor miracle that our module was operating at all.
We looked at the voltage across the relay and found that they were arcing badly every time they opened. The designer came up with a voltage snubber to suppress the arcing, and the problem went away. The supply noise was still there, but with smaller amplitude and shorter duration so that our supplies could handle it. We also discovered that if we put our module on an extended backplane (adding about 8 inches to the length), we could no longer duplicate the failure.
The designer read though the relay module installation instructions and found recommendations that “sensitive analog modules” should not be located near the relay module and that all inductive loads have snubbers installed as close to the load as possible to limit contact arcing and RF noise production. We decided that in a real-world installation, the noise was tolerable after installation, but as the relay contacts aged the transients got worse and eventually caused a failure in our modules.
We had observed the relays degrading during our investigation, and had to replace them twice when the contacts got welded together. We asked our partner to find out whether the customer had snubbers installed, and asked them to move the analog module away from the relay module. At this point, they told us that the customer had already moved it away and the failure was gone. We are now waiting for our partner to let us know whether the snubbers were there or whether they helped. I suspect that we will not hear any more about this.
This entry was submitted by John Elliott and edited by Rob Spiegel.
John Elliott graduated from UC Santa Cruz with a BA in Physics, where he fell in love with electronics and software. He spent 17 years as a test engineer working with computers, disk drives, optics, and instrumentation. After an eight-year detour as a software project manager, he is now managing a new product verification team for a small industrial automation company.
Tell us your experience in solving a knotty engineering problem. Send stories to Rob Spiegel for Sherlock Ohms.
Your story demonstrates one of the keys to trouble shooting many problems: Being able to duplicate the failure. So many times there is a problem that appears randomly and until I can find out why, I am never totally satisfied with the fix. Even if it is my own error, just correcting it without knowing the thought process that caused the error to begin with drives me crazy.
It's called surge impedance Zo, which is defined as √(L/C), where L is the inductance in Henrys, and C is the capacitance, which for a coil is the interwinding stray capacitance.
The back EMF V= Zo (δI/δT), and it will have an oscillation frequency 1/(2π√LC). When you open up contacts in an inductive load, δI/δT goes to ∞: You see this as an arc when you unplug an iron; and also when relay & motor starter contacts are switching off an inductive load.
This is also why contactors have serious current deratings when switching off DC: Once the arc is established and current flows through the ionized channel, there is no zero crossing to extinguish the arc, as occurs with AC,
When dealing with AC, you design using the peak (not RMS) value of the load current when calculating, because you don't know where in the AC cycle the contacts will open.
Note: Those of you who are RF jocks will quickly recognize Z(o) = √(L/C) as the equation for the characteristic impedance of a transmission line: Yes, it's the same thing.
We had changed the output driver from a completely analog design, to one with a digitally controlled output driver. The new driver would reset occasionally with the noise or the communications to it would get interrupted.
On the old module the noise would pass through to the output, but the frequencies were well above the system response time.
John, did you guys compare the old and new analog modules that you had designed? It would be interesting to know what the design engineer changed between the two.
From Dell / Intel® New Paradigms in Design Work Scott Hamilton, vertical market strategist for Dell Precision workstations, 5/2/2013 5
Early in my career, I worked as a draftsman and remember the days of drawing on vellum with numbered pencils and Mylar with plastic lead. This was a fun experience in the sense that I ...
I've been using workstations for more than 10 years and love finding ways to get more performance from my system. With demanding professional applications that require more power each ...
A lasting memory from my first job as an engineer in an auto assembly plant is standing on hard concrete at six in the morning, vending-machine coffee clutched in hand, listening to ...
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This radio show will show what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.
To save this item to your list of favorite Design News content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Tweet This
7 
