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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.



SDRSS | 20 July, 2000

Bob, Could you outline the processes used for this seemingly impossible assembly technique? Whenever mentioned, most folks shake their heads and look at me like I'm delusional (not true since I started my new medication, hehehe). Of course they're the same ones who still think that normal double sided reflow is physically impossible (and the earth is flat)....

reply »

Bob Willis


Re: SDRSS | 20 July, 2000

Yes no problem here it is in a nut shell. The sponsor of my CD ROM on this process was Loctite who specifically developed a glue for the process to cure in less than 10seconds. Have a word with these guys they may even have a few CDs.


The process of SDSRS includes the following stages:

Stencil Print Paste Side One Dispense Adhesive Side One Alternatively Screen Print Adhesive Side One (Using Recessed Stencils) Alternatively Dispense Paste & Adhesive Place Components Side One (Possible UV Curing Stage for Adhesive) Invert Board Screen Print Paste Side Two Place Components Side Two Possibly Insert Through Hole Components Reflow Solder Both Sides Simultaneously

The question has to be asked why consider SDSRS as it will undoubtedly cause problems with loss of components in the early stages of process development. If we examine the time taken for the reflow cycle of a product, typically this is 3-4 mins. By reflowing both sides of the assembly then the time equates to a total of 8 mins. By conducting simultaneous reflow then the process cycle time is reduced with time saved on the total process cycle.

It used to be the case that placement was the slowest part of the process but with increasing speeds this is not the case. Printing was the last part of the process to be attacked by the speed merchants focusing on the solder paste specification which was the limiting factor in high speed printing. The formulations of paste have now provided the speed to make other parts of the process the weak links in the chain with cycle speeds for printing being less than 10 seconds per board.

SDSRS Advantages

Eliminates one reflow process stage Only one heating process for double sided boards Reduction in capital equipment Reduced floor space requirements Reduced handling stages Increased PCB throughput speed Potential elimination of component weight issues Improved solderability of OSP circuit boards


New process introduction Possible loss of components Gluing stage required New glue materials required Glue curing process Possible limits on component positioning


Eliminates one reflow process stage

With simultaneous reflow soldering both sides of a surface mount board can be conducted in one operation. If through hole parts are also designed to be reflowed this can be done in a single operation.

Only one heating process for double sided boards

With only one heating process there is a potential benefit to the reliability of the joints with a thinner intermetalic formation being possible. The issues related to reduction in solderability of surface mount pads are also eliminated.

With only one heating process during reflow soldering, in which the board may be supported with a central wire, it is possible to reduce warpage of boards susceptible to this problem.

Reduction in capital equipment

Eliminating a reflow process means one less piece of equipment in a production line is required. The cost of power consumption and extraction systems for reflow equipment is higher than any other piece of equipment on the line and they are effectively halved.

Reduced floor space requirements

Eliminating a reflow oven reduces the floor space required in a manufacturing area. Even with the need for an additional printer or the use of a single dispenser for both adhesive and paste, floor space is still being saved.

Reduced handling stages

The number of handling stages is reduced from a conventional surface mount process, albeit only slightly reduced from a fully double sided reflow process.

Increased PCB throughput speed

The total process cycle time for each product is reduced by the elimination of one reflow soldering stage. The increased cycle time for one printing or dispensing stage is still far less than the period taken to reflow one side of a board.

Potential elimination of component weight issues

In a liquid state there is a limit to component weight during traditional double sided surface mount assembly. Large heavy components can fall off the board. If adhesive is used and cured it will increase the range of components that can be used on both sides of the board.

Improved solderability of OSP circuit boards

There has been great debate on the benefits of using OSP: improved printing and placement, cost reduction and improved yields. OSP coatings have been shown to deteriorate during two reflow cycles; with one reflow process stage the problem is eliminated.

Note: It is the opinion of the author that most of the problems seen on OSP coated boards stems back to the quality or preparation of the original coating. Potential users of OSP should understand some of the simple process evaluation techniques or goods receipt tests.


New process introduction

With any new process there is a learning curve, particularly with new materials and processes. There will also always be a degree of reluctance on the part of some company departments to invest time on a new process. Always ensure that a justification has been well thought out prior to starting tests.

Possible loss of components

Component loss may occur if the adhesive is not cured or if products are incorrectly supported during the different process stages. Trials have shown that if a curing process is used them loss of parts on a normal process are eliminated.

Gluing stage required

To allow this process to be conducted successfully a gluing process is required, which requires either a printer or a dispensing system. It is one additional stage at the start of the process.

New glue materials required

The use of UV curing materials has declined for surface mount application in favour of thermally cured products. Most development of materials has focused on either their high speed dispensing or printing characteristics. Examination of the material properties of UV curing products will require further engineering time.

Glue curing process

If the process is to be successful the curing cycle or part curing will need to be achieved in <10 seconds without heating the board surface. The UV lamp would need to be placed over a conveyor suitably screened to meet all health and safety requirements. If the time to fully cure was insufficient then it would also be possible to place a UV source under a conveyor section after board turnover to supplement the cure. This may also allow a lower intensity source to be used.

Possible limits on component positioning

When using glue to hold parts on the base of the board there will be some limitations just like with the use of double sided reflow. BGA, CSP and most four sided devices may not be useable. The access for positioning glue spots may be restricted. This is dependent on the capability of the adhesive application process.

reply »



Re: SDRSS | 20 July, 2000

Hi Bob,

From my memory about almost 4 yrs. ago, I remember we did something similar to this in Apple Computer Singapore.

Process sequence are:

Bottom Side Process:

Print Solder Paste(Dek 265GS) Dispence Adhesive (Cam/Alot 3800) Place Components (MV2F) Cure Adhesives (BTU Convective Oven) at Temp between 155 to 160 deg. C

Invert Board

Topside Process:

Print S. Paste (DEK265GS) Mount Components (MV2F, GSM1) Insert PTH Connector (DB9 PIN, DB25, keyboard socket etc) Reflow the board with peak temp of between 220 to 225 degrees C. (Using BTU Convective Oven)

Note: The above is a real time process which mean boards flow from bottom process to top process without any delay.

Results: hips of dry joints and missing chips mainly on the bottom side of the PCB. Most component chips were later on found inside the reflow oven. Dry joints were suspected to be caused by the curing of adhesives due to evaporation of flux ?

Is there a company testing this SDRSS or some data have been collected and posted somewhere ? Also please advise on what could be the possible cause of the above dry joints and missing chips.

reply »

Bob Willis


Re: SDRSS | 20 July, 2000

We set up and ran this process at Nepcon UK with some success but we used a UV adhesive less than 10sec cure so no heat. Push off force for the biggest parts was over 800 grams.

We have had to modify the design rules a little to make it work but it did as an inline process.

To my knowledge my CD ROM and the report that the SMTA Office have is the only detailed work that has been done so far.

I would have loved to have seen the defects you have seen I have hundreds of my own.

reply »

Essemtec Fox Compact Automatic Pick and Place

HeatShield Gel- thermal PCB shield during reflow