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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Hunter

#10553

testing after SMt assembly | 20 July, 1999

Dear all Please help me to know what kind of tests usually done after SMT Pick & place and reflow oven. Our appliocation is Mainly fabrication of mother boards. Thanks

reply »

Peter Brant

#10554

Re: testing after SMt assembly | 20 July, 1999

| Dear all | Please help me to know what kind of tests usually done after SMT Pick & place and reflow oven. | Our appliocation is Mainly fabrication of mother boards. | Thanks | Hi,

The usual method is ATE, also known as a "Bed of nails" Basically, your design should incorparate "test pads", eg. 50 thou round pads, probes then press down on the pads and give a voltage reading. The designs owner can caculate what these voltages should be and therefore when comparing what "should be" and what "is" you can highlight any problems. Here, we tend to put test points on every circuit node, but it changes from company to company.

Hope this helps

Pete Brant

reply »

ScottM

#10555

Re: testing after SMt assembly | 21 July, 1999

| Dear all | Please help me to know what kind of tests usually done after SMT Pick & place and reflow oven. | Our appliocation is Mainly fabrication of mother boards. | Thanks | In theory, ICT is the classical answer (In-Circuit Test). In practicality (more often referred to as reality), NONE. The reason is: 1. SMT only components lack the PTH components to complete the network for proper ICT to work. This requires a seperate program and some modifications to the fixture, all at a cost ($$$). 2. Getting the ICT on line for the first article test often takes more time that what most production managers can stand ($$$). Just getting a ICT program stable often takes several test runs and several boards to "wake up" the fixture and get knowledge of the quirks. Time is money. 3. It often takes a Test Engineer to interpret the results for a "pass". This is because no program is perfect unless the parameters are so loose as to pass failures.

The above reasons are becoming less of a problem as testers improve but still exist none-the-less.

Although everyone would love to say they do it; I would bet that 80% of the companys that have ICT do not do test until after final assembly simply due to the cost and time. That's reality.

A good setup process and a good visual first article inspection is what I've seen as the usual "test" (though some use the test features in the pick and place).

I would certainly love to hear from those who are actually using ICT to verify SMT assembly right out of the oven. Just one?

Not a pleasant answer, Scott

reply »

Mark Quealy

#10556

Re: testing after SMt assembly | 21 July, 1999

| | Dear all | | Please help me to know what kind of tests usually done after SMT Pick & place and reflow oven. | | Our appliocation is Mainly fabrication of mother boards. | | Thanks | | | In theory, ICT is the classical answer (In-Circuit Test). In practicality (more often referred to as reality), NONE. The reason is: | 1. SMT only components lack the PTH components to complete the network for proper ICT to work. This requires a seperate program and some modifications to the fixture, all at a cost ($$$). | 2. Getting the ICT on line for the first article test often takes more time that what most production managers can stand ($$$). Just getting a ICT program stable often takes several test runs and several boards to "wake up" the fixture and get knowledge of the quirks. Time is money. | 3. It often takes a Test Engineer to interpret the results for a "pass". This is because no program is perfect unless the parameters are so loose as to pass failures. | | The above reasons are becoming less of a problem as testers improve but still exist none-the-less. | | Although everyone would love to say they do it; I would bet that 80% of the companys that have ICT do not do test until after final assembly simply due to the cost and time. That's reality. | | A good setup process and a good visual first article inspection is what I've seen as the usual "test" (though some use the test features in the pick and place). | | I would certainly love to hear from those who are actually using ICT to verify SMT assembly right out of the oven. Just one? | | Not a pleasant answer, | Scott | We were attempting to set up a test merely for opens and shorts on SMT after reflow. We had identified a small bed of nails tester which could use programs originating from our HP3070. But as Scott said it was cost. $75,000 for the tester, plus fixture design and fabrication as well as test engineering time to fine tune the program for the tester. We do all testing after the board is fully loaded.

Mark

reply »

Boca

#10557

Re: testing after SMt assembly | 23 July, 1999

| | | Dear all | | | Please help me to know what kind of tests usually done after SMT Pick & place and reflow oven. | | | Our appliocation is Mainly fabrication of mother boards. | | | Thanks | | | | | In theory, ICT is the classical answer (In-Circuit Test). In practicality (more often referred to as reality), NONE. The reason is: | | 1. SMT only components lack the PTH components to complete the network for proper ICT to work. This requires a seperate program and some modifications to the fixture, all at a cost ($$$). | | 2. Getting the ICT on line for the first article test often takes more time that what most production managers can stand ($$$). Just getting a ICT program stable often takes several test runs and several boards to "wake up" the fixture and get knowledge of the quirks. Time is money. | | 3. It often takes a Test Engineer to interpret the results for a "pass". This is because no program is perfect unless the parameters are so loose as to pass failures. | | | | The above reasons are becoming less of a problem as testers improve but still exist none-the-less. | | | | Although everyone would love to say they do it; I would bet that 80% of the companys that have ICT do not do test until after final assembly simply due to the cost and time. That's reality. | | | | A good setup process and a good visual first article inspection is what I've seen as the usual "test" (though some use the test features in the pick and place). | | | | I would certainly love to hear from those who are actually using ICT to verify SMT assembly right out of the oven. Just one? | | | | Not a pleasant answer, | | Scott | | | We were attempting to set up a test merely for opens and shorts on SMT after reflow. We had identified a small bed of nails tester which could use programs originating from our HP3070. But as Scott said it was cost. $75,000 for the tester, plus fixture design and fabrication as well as test engineering time to fine tune the program for the tester. We do all testing after the board is fully loaded. | | Mark | We seriously considered ICT after SM but found the same concerns ($$$). We chose to look into AOI (yes, pun intended). We found one which is easy and quick to program, and works very well. We only use it to check for right components, in the right places, with the right polarity. Of course, comoponent verification only works with marked components. It has been working well for us. We let end of line inspection people to focus on solder joint inspection (no, pun not intended). If the stencil and printing process are qualified and good then solder joints should not need to be a great concern.

Its not an exhaustive test but it gets us the most 'bang for the buck'.

Boca

reply »


DL

#10558

Re: testing after SMt assembly | 29 July, 1999

| Dear all | Please help me to know what kind of tests usually done after SMT Pick & place and reflow oven. | Our appliocation is Mainly fabrication of mother boards. | Thanks |

Hunter, YES ICT after SMT reflow is expensive, we currently onlt test our HIGH VOLUME BOARDS and BOARDS That have some type of Driver IC or QFP on them that will eventually have a LCD on top. For us this worked because having to remove LCD's to replace components is a job for the best rework person.

Lets look at this from another point of view. How accurate are your placements after the SMT process? How many SMT errors are caught in the ICT after the product has moved out of the SMT area? If you can anwser Real Good and Almost None then you dont need an ICT in the SMT area, of course I dont know your layout but you get the idea. I posted a link to a Power Point Presentation I participated in and gave to the Plant Manager and his Staff. Check it out if you have problems with excesive rework and sufficiently trained personnel.Ppoint97 or better to view.

reply »

ii-feed SMD Intelligent Feeder

 Reflow System