Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

Production Testing - HASS

Views: 4112


Production Testing - HASS | 27 March, 2007

Just wondering whether anyone is currently conducting Highyl Accelerated Stress Screening (HASS) testing on production samples.

I've read an old post which mentioned that stress testing production samples is not a good thing to do. I'd like to argue that stress testing is an excellent way to test your products provided:

1> You know the products operating limits,and these have been pushed out during development using Highly Accelerated Life Testing (HALT)

2> Proof of Screen has been conducted to ensure that wear out failures are not induced by the screen

If the above two conditions are met then I believe temperature and vibration stresses applied to production units will not cause a serious depreciation in product life. It will instead:

a> Get the products that have been tested closer to the first "knee" in the bath tub curb.

b> Provide advanced warning of shifts in production processes, and or component supplier issues.

c> Ensure your suppliers keep you informed of any changes.

reply »


Production Testing - HASS | 27 March, 2007

Hi Donovan,

I guess it really depends on the product, how it's built and how you'd stress test it. Many arguments can be made about how far to test, then when one fails in the field was it due to manufacturing or test? But the bottom line is most manufacturers do not want to product test. It costs money to product test. If initial testing is performed on a proto basis, with known manufacturing parameters and they pass, then no further testing is needed unless something changes. If the parts and the process parameters are controlled then known good units should be built.

Now this whole theory is based in a rather large glass house and the landscaping is nothing but rocks. Depending on the size of the rock and velocity it's thrown will determine the damage made. Kind of like the tolerance stack up of various parts bought from various vendors can do to your product. Not to mention the testing itself. I've seen some stress tests that do not represent real life situations. Why would someone want the control unit for your windshield wipers to withstand a one hour open flame test? Maybe it's me, but if I'm driving a car and it bursts into flames, the last thing I'm gonna think about is the wipers. Granted, if I were still in college, it's raining, I'm driving to a kegger, and my car bursts into flames - I'll probably keep driving as long as the wipers keep moving.

Back to real life. Product testing cost money. Manufacturers want to build and then test to assure how they built it is good. They do not want to test to make sure the customers knows what they are doing or 2nd guess at what they really want.

reply »


Production Testing - HASS | 27 March, 2007


I agree that it does depend on the product; and typically medium to high cost equipment, or extremely high volume production will benefit from this testing.

For my previous company we used HASS - which is a combination of vibration and temperature stress testing on a sample basis to monitor our production. Sure the costs were high, but these were spread over the entire build, making the per unit cost low.

We found an alarming amount of problems in the first few months that were attributed to component batch problems. Don't get me wrong, we had our share of production fallout issues as well, but the majority of problems were introduced by our suppliers. These suppliers were all known for their reliability, and were big name brands.

Another thing that people don't realise is the stress levels encountered during transport can be extremely high. Shock levels of > 60G are common, as are vibration events of >1.5Grms. And drop heights are just outrageous, old billy throwing your new server off the truck doesn't care too much about the glass house theory, but loves to throw anything that he can lift... say he cracks a poorly formed solder joint and this leads to failures at the customer which reflect badly upon your brand.

Stress testing will help to alleviate this by minimizing the weak points in a design during development, and maintaining the process/supply chain quality during manufacturing. Tuned screens will then break only "bad" units.

Also - once you have a production screen you can use this to qualify alternate components, or processes - such as lead free solder.

Lastly, in recent years, HASS has been picked up by some of the largest companies in the world. To sell to these guys you now have to perform both HALT & HASS.

I completely understand the example you've provided being somewhat overkill or unrelated to end use. Classic call on the kegger though, now thats the kind of human stress testing i'd be into.


reply »



Production Testing - HASS | 28 March, 2007

I'd like to side with Chunks on this one mainly as I come from a component manufacturing background, but also because your arguement comes from the perspective of someone who is an engineering consultant for a company that specialises in HASS & HALT.

reply »


Production Testing - HASS | 28 March, 2007

Wow, I fell for that one hook line and sinker. Sorry guys for carrying this thread/commercial on for so long. My bad.

reply »


Production Testing - HASS | 28 March, 2007

If I was trying to sell you something I'd be more direct, i'm more interested in peoples views of this testing then I am in convincing you to use it. In addition, you're both in America and i'm in New Zealand, so what am I going to sell you?

I didn't realise that we're not supposed to talk about our specialties here.

Its all about sharing knowledge, now that I see you've taken my views as being tainted due to the area in which I operate i'll stop posting on subjects that I know about.

reply »


Production Testing - HASS | 28 March, 2007

Shame on you Chunks for not nowing who these people were! Next time check source before opening your mouth. Now we are stuck reading advertisments which everyone hates. Stupid!

reply »


Production Testing - HASS | 29 March, 2007

This does not seem to be a sales pitch. It seems like Mr Johnson was trolling the web and came across a SMTnet Archival posting on the pointlessness of burning product life doing HASS and wanted to engage a conversation on his issue with that comment.

Maybe someone learned something about his points. He didn't even leave a phone number or web link. No harm, no foul.

For definitive comments on reliability testing of boards: IPC-SM-785 - Guidelines For Accelerated Reliability Testing Of SM Solder Attachments

reply »


Production Testing - HASS | 29 March, 2007

Thanks Dave...

To all that took offence I do apologise, its a fine line that i'm walking here. Accelerated life testing is an integral part of todays development and manufacturing environment.

I've spent 3 years living with 50% of my time in a design centre and the remainder at manufacturing, I completely understand both sides and see a reason for both sides objections - whether it be designers complaining about quality, or manufacturers complaining about cost and resource.

I really was trying to stimulate a debate on this topic, without penultimate views and with respect to this community.

Once again, I apologise for any misunderstanding.

reply »


Production Testing - HASS | 29 March, 2007

While I don't have a dog in this hunt (my only comment is that I wouldn't want to buy a candy bar that has been taste tested), I do very much appreciated the validation that I DID in fact retain something from college.

"Granted, if I were still in college, it's raining, I'm driving to a kegger, and my car bursts into flames - I'll probably keep driving as long as the wipers keep moving."

In answer to your question in another thread, Chunks, I spent my weekend watching horses go around in circles. It's not for everyone.

reply »

Ecolloy Soldering Alloy

SMT Reflow Soldering Equipment Company of the Year 2020!