Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Reliable electronics design

Status
Not open for further replies.

DrBudz

Junior Member level 3
Joined
Dec 5, 2011
Messages
28
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,500
Hello guys,


I would like to start an open discussion about proffesional design of electronic devices that should work in harsh industrial enviroment.

What are the common protective measures for inputs and outputs, besides galvanic isolation with optocouplers and relays?

How to choose components and make filters for power supply and digital inputs?

How to make reliable NPN outputs? What kind of protection should be implemented on NPN output (beside flyback diode)?

Any contribution to the topic will be appreciated.
Thanks!
 

the best thing is that the devices must be placed in the circuit to receive the maximum ip but the effect of temperature and other potentially dangerous signals must be avoided

in npn trans op can be made proper by using heat sinks and thermal radiator to prevent thermal run away
 

In the 70's I worked for an Aerospace company. Each product had a different set of harsh environments and reliability requirements.
For Example, for airborne we were restricted from using any new parts < 2yr old for safety margin when no reliability data was avail.

There are many environments to consider;

- Climatic ( Temp & %RH)

- Mechanical (shake, drop , & acceleration )

- EMI ( ingress and egress // conducted and radiated from 50Hz to XRay if applicable. although conducted is usually 30MHz max and radiated 50KHz min)
- ESD/EOS test 15kV to 25kV, with various acceptance criteria. no operator intervention .. up to no failure but manual intervention allowed. You would be amazed how many early keyboards failed when the zap test when around the exposed LED to the board underneath and smoked it.

- Vacuum effects of mile high Denver and deep space

- Stress on components ( current use to max rating, temp use to max rating, bias stress to max rating. where Failures in Test (FIT) accelerate with % of margin or % ratio of rated spec., power dissipation to max derated spec)
- Cycle count on connectors vs rated.
- Contact current vs rated and exceed wetting current for non-gold plated contacts with signals
- Cable bend radius vs specified min.
- Flammability protection from spilled liquids on the lid.
- Locked rotor tests for UL on all motors.
- Ground leakage current for safety ( 0.5mA max for filter allowed)
- Ground impedance test for safety ground to case (10A test)
- Sledge hammer explosion proof and high voltage exposure ( for UL , in case something drops on your 1U rack panel ) with enclosed PSU
etc etc...

- Reliability power cycle test 10K cycles for fans to detect dead spot condition with no start. ( shifted hall sensor)

For rocket use in Black Brandt series we made ;
- Mechanical Test included 100G shock 10mS, 50G acceleration and 15g vibration for both operating and non-operating. Often ground use has three sets of specs for operating and non-operating and storage.


In the 80's as Test Eng Mgr for HDD production, we also tested OEM drives to corporate environmental standards for ground benign conditions.
We did all the above tests to ensure we Qualified the best drives for our system designers and customers in a process called DVT or Design Validation Testing. we could run several products concurrently and schedule 2~3 months to complete it.

You start with a good spec, then write a test plan around these specs. I went beyond this and employed HALT HASS methods to find the margin to failure, sometimes it was destructive testing, but usually if you can detect this before applying it, you can save the units but usually not destined for customers.

I hope this gives you a thumbnail sketch on what it takes to ensure you have a reliable design.
I found HALT HASS and margin to error detection the fastest most effective means of comparing product quality.
eg voltage margin test +/10% while thermal cycling. and measuring accuracy , power consumption and error rate or margin to error with injected jitter.


SO once you are aware of the Environmental Specs required to pass, you more easily determine if your design will pass when it is done.

How you do it is up to you.

I am retired now but have tons of stories on these topics.
 
A typical harsh environment is the industrial area and when you are interfacing microcontroller, you might find helpfull tips from this report : https://ichaus.biz/mcu_interface .

Enjoy your design work!
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top