SMT相关知识国外人的看法 SMT, 看法, 知识, 国外

luoman

All the Questions and Answers below have appeared in one of Bob Willis editorial columns in EMP AMT Ireland Electronic Manufacturing Products or EM&T Electronic Manufacture and Test magazines in the United Kingdom or in Asian Electronic Engineer. Further editorial on electronic manufacture is available for inclusion in magazines throughout the world by contacting EPS.

Q Dear Bob - What type of thermocouple wire should I use for profiling surface mount boards ?
The wire most often used for this application is "K" type which is a twisted pair and used on most profilers. The "K" wire is used for temperature ranges between 150 C and is approximately -0.010". For occasional use an insulated wire may be used; for regular continuous use a higher temperature insulation wire.

Q Dear Bob - We use one profile for all our surface mount products we have a convection oven is this OK ?
Well it is fair to say that a convection oven can provide single profile for all products large and small. But it is poor practice as inevitably some products will be in the liquid state for an unnecessary period of time increasing intermetallic thickness affecting solderability for second side reflow and possibly component damage. Be a SMART engineer not a lazy one!

Q Dear Bobby - How should thermocouple junctions be fixed to the board surface ?
There are three methods: adhesive Kapton tape and high temperature solder. The best is quite simply the last but do not bury the probe in a pile of solder as this will affect temperature. You may need to use an activated flux to make the wire wet.

Q Dear Bob - When profiling a reflow process should you use a fully populated board ?
Yes always. You may find that soldering is successful with a profile developed with a sample board but have a look at the voiding in the joints. It is difficult sometimes to get a populated board if you are a sub-contractor customers are loath to let you have one. Insist this will help you provide a better quality and potentially more reliable product to your customer.

Q Dear Mr Willis - I keep hearing that nitrogen is beneficial particularly with low residue solder paste how can I justify the cost ?
Most people justify the cost on changing to no clean or on quality grounds but that is difficult to show a cost saving benefit. The best way to justify the cost is by the use of copper based finished boards which will cost less than tin/lead. Check out the guide to nitrogen video available from the SMART Group it includes interviews with Colin Lea and Chris Tanner.

Q Dear Bob - I would like to purchase a Mole to profile my reflow oven what should I look for in the equipment ?
First of all Mole Kick and Datapaq are all trade names for the mid to top end profile units on the market. Generally the price increases with the number of thermocouple inputs to the data logger. The high end units also have excellent software for manipulating data. The cheaper units between 6-800 pounds sterling only have very basic software. It depends on what temperature measurements you want to do in production. I personally use an ECD Mole for running all my profiles on ovens wave soldering systems and even the oven at home!!.

luoman

Q Why reflow solder through hole components?
In many electronic assemblies there are large multi leaded components still being used in combination with surface mount. If wave soldering is to be eliminated then either hand soldering single point automatic soldering or reflow must be used. The major driving force is manual cost reduction and a simplified process.

Q Do I need two stencils for through hole printing one for through hole and one for traditional SMT parts?
Depends if you are feeling charitable to your stencil supplier. Some people have used the technique to increase paste volume. (Bob's note) Ensure you tell your stencil supplier that the round apertures are required in your stencil for through hole printing. They are so often removing them for customers who have not supplied a solder paste file for the stencil those wonderful people may get carried away. I forgot to tell my stencil manufacturer last week; do as I say not as I do!!!

Q How many components can be soldered in this way?
I don't know the answer to the question. Each component needs to be considered for this process and needs to be assessed and discussed with the component manufacturer just like immersion cleaning wave soldering etc. Generally speaking it is the high point count devices like connectors pin grid arrays post headers sockets and dual in line parts that have been specifically produced for reflow applications.

Q What is the best use of this technique?
Back plane or junction boards where you have loads of connectors and functionality with lots of surface mount components. It can also replace some press fit designs. Its a godsend to manual assembly lines.

Q What will the solder joint reliability be like for reflow joints?
There should be no difference in the solder joint. There may be a difference in the solder volume due to the limitations of the printing process. Just try ripping a through hole lead out of an existing soldered plated through hole if you are strong enough. The microsections I have done look good.

Q What happens to the paste on the pin tips during reflow?
As reflow takes place the solder does remain on the pin evening out the thickness on the pin. However there will always be some slight build-up of solder on the pin tip. There is also a difference depending on which the up the pin is during reflow.

Q What about flux residues on the pin tips will it cause problems during in circuit test?
If you use a high solids paste or you don't tell your test engineers your preferred process the answer will be yes. Conventional joints that are to be hand soldered or reflowed should not be used for test access. With a little planning at the start of a project you eliminate the problems before they hit the shop floor.
Q Dear Bob - What are the main problems with reflow soldering copper finish boards ?
Despite the literature from the protective finish manufacturers heat does affect the copper circuit. It increases the time to obtain full wetting hence increasing the possibility of poor yield. Double sided reflow again increases the surface oxides and poor yield. You need to confirm the quality of the basic coating with your supplier and consider the other process stages which affect solderability. Cleaning the board washing off poor paste prints poor board handlin and the original surface preparation of the copper before treatment can all affect reflow yields.

Q Dear Mr Willis - What are the benefit of using 62% tin/2 % silver alloy paste during reflow ?
Today the benefit is a slightly lower temperature for reflow. The eutectic allooy 63 tin/37% lead has a reflow temperature of 183-184oC. The silver alloy reflows at 179C. The lower temperature is worthwhile for any process. The silver alloy in the past has been beneficial for soldering silver palladium parts to reduce the leaching of the termination. As people purchase chip parts with a nickel and tin/lead plating this should not be a problem today

Q Dear Mr Willis - Can I use the same temperature profile for second side reflow on a double sided product in a convection oven ?
Yes you can but when does the reflow take place on your existing process? You must fully understand your process and know when it reaches reflow temperature. In the case of doubled sided reflow the first side components can absorb head and slow down reflow. This may still allow reflow using the same profile but the joints may only just reflow causing a lot of voiding.

Q Dear Bob - I want to profile my reflow oven to try and reduce reflow temperatures where should I place my thermocouples ?
Positioning of thermocouples comes down to experience. They should always be soldered in place using high temperature solder. First position a probe at the centre of the board and one at the corner edge of the board. This should theoretically give yiou the lowest and highest temperatures. Next consider where the largest mass of components are located on the board and place a probe at one termination point. In a convection oven you should see a maximum temperature differential of 15-20oC. Experiment again with the same board placing the probes in different positions noting the temperature changes.

Q Dear Bob - If I change my extraction system or my process settings do I need to perfom further profiles ?
Yes depending on the type and position of the extraction points it can have an effect. If the ducting is poorly designed other equipment in the factory can affect the extraction rates and your board temperatuure during reflow.

Q Dear Bob - What information is available on the assembly process for BGA ?
Although a lot of parts are being processed only limited information is available. The SMART Group has produced a BGA report with all proceeds from the sale of the report going to charity. For a copy of this report contact the SMART Group tel: 01494 465217 Two videos on BGA are also available from EPS see video listing.

Q Dear Bob - There may be advantages with BGA but isn't the cost of parts higher ?
Yes the cost of the actual parts is higher but the price will come down as the technology matures. Even now you can make savings if you work out the total cost of manufacture not just the material cost of the components.

Q Dear Bob - As the use of BGA is relatively new what reliability data is available ?
There is a wide range of articles written on the subject for applications in the commercial sector and in telecommunications. Tests have been conducted on temperature cycling vibration and flexture testing. A bibliography is available from the SMART Group Tel: (+44) 01494 465217

Q Dear Mr Willis - I work in quality engineering how can I inspect BGA solder joints ?
Because the solder connection is under the device you cannot inspect the part using any traditional means. If joints are to be inspected then some form of X-ray inspection equipment will be required. As any form of X-ray will still only show a solid solder ball special design rules are required for the assembly.

Q Dear Bob - I understand that BGA devices are very prone to cracking during reflow and rework. Is this like QFP popcorn ?
Yes it is generally known as the "popcorn" effect. If moisture is in the package during reflow or rework it may cause pressure to form in the device. This then causes delamination between the moulding and fibreglass substrate. As the pressure is relieved you can hear a pop! Control in the manufacturing process can eliminate the problem.

Q Dear Bob - What are the advantages of Ball Grid Array technology to the design and assembly engineer ?
Generally speaking components are using finer and finer pitches well below 0.020". This increases the difficulty in tracking the boad design and increases the need for circuitry below .006 Both have an effect on board price. The assembly engineer is faced with high proability of poor lead quality and the need for higher levels of optical inspection and correction on assembly equipment. With BGA the pitch actually increases to between 0.050-0.060" but still providing over 300 termination points on the package. Would you rather handle 0.020" or 0.006" pitch? The process yield reported by BGA users is far higher than with fine pitch. BGA technology also allows users with lower specification for assembly equipment to use the technology. EPS have a whole range of BGA training products just call for further information.