The 20 February 2008 issue of Dataweek carried an article by Phil Zarrow, discussing the evaluation of solder pastes - specifically technical evaluation. However, there is more to it than that, as Phil elaborates.
As I mentioned before, I perform a number of solder paste evaluations for clients in the course of my business. Usually, either the client or myself obtain 250 gram samples of the candidate pastes, which the manufacturers send on a no-charge basis for evaluation purposes. Recently, I performed a very extensive study of no-clean pastes for a client.
Because we wanted 'statistically correct' sample rates, we determined that 250 grams of each paste was not going to hack it. We required something on the order of 2,5 kilograms of each. For this quantity, freebies being out of the question, we went about purchasing the paste. Because the client, a billion dollar firm, did not wish to be pestered with salesmen, it was decided that ITM would purchase the paste samples.
All of the pastes were ordered three weeks prior to the test date for arrival three days prior to commencement of testing. I should add that we were testing Sn63/Pb37, Type 3, very low residue, reflow in air solder paste - in other words nothing particularly special. We also obtained 250 gram samples of each of the seven candidate pastes.
Bear in mind we are talking about ITM ordering the material - a tiny company certainly not to be confused with IBM. In most cases this was not a problem. Some companies billed us while others required a credit card.
One manufacturer (and only one), whom we shall call 'Company X' directed us to their representatives stating I could not order directly from the manufacturer. Okay, that is cool - spread the wealth. I was a rep once, too. Upon placing the order with the rep, I was informed that we would have to order a minimum of 5 kilograms. My client reluctantly agreed - after all this was a leading paste manufacturer and we definitely wanted to include them in the evaluation.
It gets better.
Three days before the test date, when the pastes were due in, I had received materials from all of the companies except one. Yes, it was good ol' Company X. We had received their 250 gram sample but not the main shipment. A call to the rep (remember, we had to go through the rep?) yielded frustration on both of our parts. He got back to me saying there was a 'viscosity problem' and that the new batch is in good shape and it will ship overnight to me that afternoon.
Next day - no Brand X solder paste. A call to the rep ended up with the rep saying I should contact the factory myself (great - so glad I worked through a rep. Ah, the service industry). A call to the sales department at Company X went something like "ITwho?. Cannot seem to find your order…oh here it is...must have fallen through the cracks." Wow, talk about viscosity problems! "It will ship this afternoon. I will go down and take care of it myself right now.."
Imagine my surprise when the paste did not show up the next day - or the next, after repeated phone calls, etc. To make a long story short, the paste did not arrive until we were three days into the testing. Fortunately, we were able to perform the preliminary, low volume tests using the 250 gram sample we had (fortunately) received. Not that it mattered. After this fiasco, the paste could have technically come out as the best (it was not) and my client would not buy it due to their lousy service.
By the way, Company X is one of the larger solder paste suppliers, not some garage operation. I had previously been impressed with the fact that they had been beefing up their technical support staff. So what really happened? Did they really screw up my order? Were they having unfortunate quality problems? Was I bumped for a rush shipment from a more prominent client (like Motorola, Delco, Lucent or the one that sounds like ITM)? Was this a once-in-a-blue-moon occurrence or their regular mode of operation?
This brings up the question of how well a company you are buying from is going to support you after (and apparently during) the sale. How does one determine this?
A method we advocate is the 'counter-reference' and this works with materials, equipment, even contract assembly service. Here is how it works.
When you call up a supplier you ask him for references - people who are currently using his solder paste. The supplier, of course, is going to give you references that are going to, essentially, sing praises unto him. (You did not expect him to give you someone who is not happy with his service, did you?) What you do is, when you call these references, you ask them what other solder paste(s) they also considered that this brand beat out. Chances are, that list will also include your other candidates. Find out what their evaluation criteria were and why this brand was selected. Proceed to do the same cross-referencing with the references the other suppliers gave you. Of course, the depth and intelligence of the evaluation tests (if any) must be taken into consideration and factored in.
As mentioned, this method works for checking out equipment suppliers and contract assemblers as well. However, this is by no means a substitute for doing your own evaluation - these things are applications specific so no goofing off. But, references and counter references should be a part of your evaluation process. You are not just buying a material or a machine, you are buying the company behind it. Hey, we are all in this together.
Reel stuff
The reflow of through-hole process (also affectionately known as paste-in-hole, pin-in-paste, intrusive reflow, etc,), where through-hole components are soldered simultaneously with surface mount components in a reflow oven, continues to gain popularity, and for the right reasons. Most of us prefer to stencil the solder paste over the aperture prior to insertion. The solder can also be deposited via an auto-dispensing system or solder pre-forms can be used. The latter are traditionally little doughnuts of solder stampings that are placed around the lead to be soldered and which melt during the reflow process.
The problem is, they are usually hand-placed over the leads. This is not a lot of fun and when we start talking about 200 pin connectors and such, not very 'high speed'. Occasionally, there are applications where enough solder paste cannot be printed, particularly with very thick back panels.
One cool solution has been developed by Indium. They fabricate solder pre-forms in the shape and size of a passive component. These 'solder 1206s' are supplied on conventional 8 mm tape and reel. One can print a volume of solder paste and then supplement that with additional volume by placing these solder pre-forms with the pick and place machine - accurately and efficiently. The pre-forms can be placed on the top side of the board, into the printed solder paste tangent to the hole. In some extreme cases they might even serve as standoffs when the component has none.
Another interesting innovation is the Surftape. This is a tape and reel feeding methodology that does not use a cover tape. The carrier tape itself is dimensionally identical to conventional tape and reel tape except that each cavity contains a pressure-sensitive adhesive. The components adhere to this sticky tape. Dimensional formats conform to worldwide standards for conventional tape and reel carriers, meaning virtually any existing 8, 12, 16, 24 and 32 mm tape feeders.
Parts can be located in the tape cavity and stay there for accurate presentation to the pick up head. Components are accessible for visual inspection, testing, marking. This also appears to be a slick alternative to waffle-packs for carrying and dispensing bare die.
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version