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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

solder paste measurement machine

Views: 1910



solder paste measurement machine | 26 April, 2007

Hi all,

We plan to buy solder paste height measurement machine. However, not sure which is the better one..2D or 3D? As far as I know,3D is quite slow and but more accurate in term of measurement result. Pls share with me if anyone who has this experience about 2D and 3D solder paste height measurement machine. Really appreciate your help!

Rgds, JJ

reply »



solder paste measurement machine | 26 April, 2007

Hi, yes may be 3d is slow but it gives you great results as it shows the height while 2d picture shows error when the paste in your machine appears to end i.e. when it see that the pad starts to appear. I don't know which company brand you prefer but compare your choice with others because speed is not only the thing, there are other things too. bye

reply »


solder paste measurement machine | 26 April, 2007

3d is good for initial verification that your screen printer and stencil is producing the desired paste height. 2D verification is more than sufficient during regular production.

That being said, your best bet for a 3D machine is to go cheap - buy a tabletop measurement system - since you'll be doing this only on a sampling basis. It adds NO VALUE whatsoever to do 3D inspection as part of in-line production.

If your screen printer does not have 2D capability, or if it's an older DEK where the 2D capability doesn't work, then buy the in-line machine but only do 2D inspection during regular production, and do the 3D on a sampling basis.

reply »


solder paste measurement machine | 26 April, 2007

Hi Dub J,

I have worked for a few companies that did 3D sampling and found the data they produce to be useless. In the real world they are slow, so that means you won't sample every board. Now if you're not sampling every board, and you start to run low on paste volume starts to deviate, you lost the whole reason you would even try to keep SPC data on such a process. That being said and prolly stepping on a few toes, I would go 2D. 2D is fast enough to put in line and you can check 100% of your product. You will find real defects there. With 3D we always had to come up with excuses each time it deviated from sigma. The funny part was the solder joints after reflow met IPC Class 3! But the 3D machine said something was wrong.

reply »


solder paste measurement machine | 26 April, 2007

Question...when the machine said something was wrong, was it reading something wrong (false failure), or was there a true descrepancy but not one that resulted in a bad connection (bridge or insufficient)?

Also, did you perform a DOE on the process and find it inconsistent in production or did you use their recipe for pass/fail and find it useless? Not trying to find fault with your methods (I wouldn't question it myself), just trying to understand how you put it to use.

reply »



solder paste measurement machine | 26 April, 2007

Good Morning,

We have a 3D solder paste inspection machine. We purchased it on a recommendation from a consultant. We are a CM high mix low volume. The President of my company has attended some seminars and meetings where it was conveyed to her that 3D measurement will eliminate 70% of our defects. I do not believe that this is true. I find this type of measurement valuable at initial setup but accomplishes very little in production. Out system only measures preprogrammed points of interest. If a solder skip occurs and is not where the paste measurement takes place, it will get by the test. I feel that from a production standpoint, coverage measurement seems to be a more critical measurement. A solder paste deposition that is 1 mil less than optimal will still produce an IPC acceptable joint. A pad that is only half printed, probably will not.

My $.02



reply »


solder paste measurement machine | 26 April, 2007

Hello Firstly I�ll make a disclaimer that I do work for a paste inspection vendor. I will not mention which. I will leave the selection up to your due diligence. Paste inspection is the up and coming �fad� or required process depending on what your opinion is or precious experience is. A few of years ago 3D paste inspection was a slow and sometimes ineffectual process. This has changed some what now. There has been a lot of development from many manufacturers and improvement in the abilities of these machines.

Typically 3D paste inspection was difficult and involved long programming and debug. The current state of 3D inline paste inspection is that several vendors sell systems. Which will not be the gate in your line and for the most part will be faster than most system in your line. Currently 3D SPI is fast enough to keep up with the pace on most production lines. Which then leaves you with the decision on which machines fits your process better.

Sampling 3D paste height can give you some basic information on your process. But you must be very methodical on how you do this. You must take into account when you do the samples. Prime example are all you measurement done the same direction of you print head movement. Print head direction can and does give the most variation of paste topography.

2d inspection is a stop gap measure at best. I have been on evaluations where the customer as been using a 2D paste inspection and 3D sampling. And 3D SPI still reduced the number of solder defects by up to 70%. (Please don�t start saying her comes the sales pitch I m the install engineer and I m not saying for who)

This statement is will agree with >�It adds NO VALUE whatsoever to do 3D inspection as part of in-line production.�< No inspection does add value there is nothing intrinsic, in any sort of inspection that adds value��.. But what inspection does do is give you the ability to increase yields. By giving real feed back to you. So you actually know how to use and setup your printer correctly and then use the inspection machine to catch the anomalies.

If you don�t wish to go the inline route there are other options as well. Some of the SPI manufactures also offer bench top versions of their machines which will inspect 100% of a board and allow easier SPC analysis of your process.


reply »

Upper Management


solder paste measurement machine | 26 April, 2007

Seminars and meetings are the best way to get a clear understanding of a machine's capabilities.

I have found that the manufacturing engineers will oftentimes skew the results of machine evaluations with facts and data. In practice, I've found it's best just to dictate which equipment the company purchases... assuming we can't build it in-house.

Your President sounds like she'll go far.

reply »


solder paste measurement machine | 26 April, 2007

In short this guys are not well updated on the new technology of SPI. True, 3D IN LINE SPI is no longer a slow machine. Try checking some brand like Cyberoptics,Synapse imaging,SEIMENS,KOHYOUNG and many others...they can match the speed os some mounters at the speed of 20cm per sec!!

The 3D in line type machine will provide better inspection rather than 2D.(I will not go into details)However, good SPI in line machine is not cheap, they cost around US$150 to US$180k. If you have a budget below than this then, buy a 3D table top which cost about US$40 to US$60K..

Hint on buying an IN LINE SPI: Try checking machine with 2 lightings- this will remove shadow effect on your paste. Shadow effect will give you an inaccurate reading.

reply »


solder paste measurement machine | 27 April, 2007

Jeremy, well said! Thanks for "validating" my theory... LOL

I agree that the sampling frequency, AND methodology are key things with 3D inspection. 3D will give you some insight onto how your paste (rheology, viscosity, thixotropy)AND process are holding up throughout production. The key thing need to be able to react to the data accordingly (ie knowing which process parameter to adjust, etc..).

Chunks was right in saying that at some companies, the personnel who take measurements just "make stuff up" if they use 3D inspection as part of SPC, and wind up adjusting the wrong things or not doing anything at all..this happened at 2 companies I worked for...

Bottom line...if I had a huge budget, I'd have a batch 3D SPI system, again, using it to establish baseline process parameters, and a "spot check" on how your process is behaving (or bouncin').

reply »


solder paste measurement machine | 27 April, 2007

Hi Steve-O,

Good question(s). This process was set-up by an engineer that had a PhD in statistics, took him a year to develop and then thru it over the wall at us. The biggest problem we saw was SPC to measure solder volume. SPC is good for measuring parts forming but not very good for solder volume. I seriously doubt you would find any different between the volume of paste on an aperture from the 1st print compared to the 1 millionth print. Stencil just don�t wear out that quickly. So running complicated SPC generating software programs is generally a waste of time.

The 2nd thing we found was alignment was very critical to get proper measurements. Having a machine that measures 3D that is not directly bolted to the conveyor was a problem. That and using the edge of the board to align to was not that great.

The 3rd thing was paste slump. This will change over time and does affect your measurements. Brick spikes for the poor release was another way to confuse the 3D of most machines.

Brick spikes and other such anomalies where a major contribution to miss readings.

Plus on larger pads, the fine line the laser uses can let partial printed bricks pass. If the 3D is used in conjunction with D this will be caught. 3D itself will not catch this unless the laser happens to be at the right spot. Most people teach this to the middle of the pad.

reply »

PCB equipment

PCB X-Ray Inspection