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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

Measuring Board Deflection



Measuring Board Deflection | 11 July, 2000

Hi Guy's..

Been a long time since I've been on the forum so let's start by saying..dave..great recipe!!!..

Anyway's why am I on for...ha yes board deflection

Wave soldering is a fun past time at best, it's made even more fun when your board deflection over the wave is such taht you get your chip's knocked off the bottomside or things crash in the machine. So what I'm loooking for is a pointer in the direction of some device either mechanical or electrical that will measure and record the deflection of a board over time.

I'm sure there must be something but so far I can't find one and I'm dreading the thought of having to design something...

cheers guys


reply »


Re: Measuring Board Deflection | 12 July, 2000

John: Several thoughts:

1 What on your wave machine are these bowed boards catching on? It has to be something on either wave. Are you getting hung-up on the chimney? That doesn't seem correct!!! 2 Why do you want to measure the defection of you board? The CTE of most boards are fairly well defined. Is you supplier providing standard material? Or is this a weird substrate material? 3 Could you be loosing your components some place other than in the wave? We've talked about loosing glued SMT chip parts before. It's easy to blame it on the wave, when actually the components were lost way before the wave and no one noticed them missing until they check the quality of the soldered connections.

reply »



Re: Measuring Board Deflection | 13 July, 2000

Dave, the heck are you?..been a while....

Ok so I'm running an Electrovert Electra witha hot gas knife, & I've got large server boards going over the wave. I've seen us loose chip, mainly Tants on the hot gas knife...ain't good especially if they are tants..! I can run with the bath hieght lower and yes I don't get the problem but the design of teh wave nozzels means that I no longer get a good smoothwave know wetting time and all that jazz. I'm lothed to slow the belt speed to compensate (and hence redow all the preheats and the flux applications and so forth. If I can measure the deflection accuratly I can see if my boards are bowing by some amount outwith what you'd exepct. I can then go talk to teh PCB vendor and drag him into the equations to say right let's see why we're getting this...(material is std FR4 by the way) Board thicknesses run from 1.6 (that's a 400x300mm pcb!)to 2.4mm - OK so I know you guy's haven't done this metric thing but hey inches?? in the UK...well not at my young years! We do get missing components from other sources e.g. radial insertions adn poor hanling...but my job is to make the wave as robust as I can given the warping issues...thought it would be good to try and measure it...


reply »

Mike F


Re: Measuring Board Deflection | 14 July, 2000

John, The first thing to check is how tight the boards are being held in the conveyor (assuming you are using a finger conveyor). They should be a little loose at the input end so there is some room to expand. If the boards are held too tightly, when they start to expand in the preheat the only way they can move is to bow downward. Boards the size you mentioned will still sag, but you can offset some of that. Also, you can buy titanium stiffeners that would clip onto the leading and trailing edges of the board to keep it from sagging. They can be notched to clear components near the edge. The downside is the additional labor.

A third choice is to make a nested fixture that supports the board on all 4 sides. We have a 19" x 15" board that uses a fixture like that. This can be an expensive solution, but well constructed fixtures can last a long time.

Mike F

reply »



Re: Measuring Board Deflection | 20 July, 2000


we use carrier's as a matter of course on all wave products, it let's us keep the conveyor width the same all the time so speed's up change overs and eleimiates one more thing that can be set wrong.

The main aim of the exercise is to be able to measure the deflection of the boards and allow us to set up an optimum process for a known deflection range as it could vary batch to batch. However as a by product we should be able to demostrate to customers and board vendors excessive bowing and therefore drive improvements in the PCB manufacturing.

reply »



Re: Measuring Board Deflection | 20 July, 2000

It's called a strain gauge with data manipulation software. They are used for load cells and the like. Hope that helps.

reply »



Re: Measuring Board Deflection | 21 July, 2000

You're going in the direction of fab improvement, a great direction.

But in another direction; If you're already in a carrier, how about using a skyhook of some sort to support the middle of the fab from the top side? In some cases you can put fine wire thru a couple of vias and tie to a carrier support above the fab, just cut the wire flush with the fab later (both sides of the fab).


reply »

Capillary Underfill Dispensing

ICT Total SMT line Provider