Hand Soldering TSSOP's with Thermal Tab (Long)

I'm considering at using Linear Technology's LT1976 switching regulator in a new design. The package is a 16-pin, 0.65mm TSSOP with an exposed thermal tab on the bottom which needs to be soldered to the PCB for heat dissipation. Although the package was designed for reflow soldering, I would like to hand solder it since I don't have access to a reflow oven and don't really want to make a project out of converting a toaster oven right now. Also, even if it works in the design, the volumes won't support investing in a reflow oven.

I've gotten relatively proficient at hand soldering other SMT packages with similar pin spacing using a temperature controlled iron with a small tip and fine solder but the thermal pad on the bottom of the TSSOP presents some unique challenges. I also have access to a Hakko hot air station, but my success with using solder paste and hot air has not been as good as an iron. One of my real problems with hot air has been getting too much paste on the board. I always seem to spend way more time removing solder bridges and cleaning up boards I solder with hot air than those I solder with an iron and fine solder. That's particularly true when the pins get so tight that there isn't room for solder mask between them.

Based upon some earlier posts in this group, I've come up with two methods that seem workable for hand soldering TSSOPs with heat sink tabs and would like feedback from some of the experts that read this group. I'll be designing a

2-layer PCB with plated holes and solder mask to make assembly a bit easier. Descriptions of the methods and simple layout images for the TSSOP are attached for both of the proposed methods.

Method 1:

The PCB layout (Test3A.png) includes a bare pad directly under the package tab, one large via in the middle of the bare pad and four smaller vias near each corner. On the bottom of the board the solder mask will expose a large ring at the large via. Assembly will start by hand soldering the TSSOP on the top of the board. Then the board will be flipped over and each of the vias under the TSSOP will be filled with solder paste using a syringe and needle. The board will be preheated using an heat gun and then a large tip, temperature controlled soldering iron will be applied to the ring around the large center hole. If it works like I anticipate, the solder paste will initially conduct heat to the IC tab, then melt and reflow to fill the void between the exposed tab on the package and the exposed pad on the PCB. Heat transfer to the IC should be "OK" as solder paste is 85-90% solder by weight and will bridge the gap between the tab and PCB (.05 - .15mm) but preheating seems essential. By observing each of "corner" vias one can tell when IC bottom tab has been completely reflowed.

Some experimentation may be required to ensure that vias are completely filled but the viscosity of the solder paste should minimize the possibility of injecting too much paste in the space under the IC. Correctly sizing the center and corner vias to provide the volume of solder necessary to completely fill the void under the IC may be a bit of a challenge, but chances of creating a solder bridge between the heatsink tab and a pin seems minimal as I won't be adding solder as I heat the tab. The paste solder information I have indicates that paste solder is 50-60% solder by volume.

Method 2:

The PCB layout (Test3B.png) includes a bare tab directly under the package tab and two very large vias at the ends of the under package tab. On the bottom of the board the solder mask will expose large rings at each of the very large vias. Assembly will start by hand soldering the TSSOP on the top of the board. Then the board will be flipped over. The board will be preheated using an heat gun and then a small tip, temperature controlled soldering iron will be inserted in one of the very large holes so that it contacts the tab on the bottom of the IC and the via. Enough wire solder will be added at the first hole to ensure good heat transfer from the iron to both the board and IC. Wire solder will then be inserted into the second hole, allowed to melt and reflow under the IC. When solder is observed rising in the second via, the iron can be removed.

Some experimentation will be required to get the amount of solder added just right as the solder will tend to "equalize" between the two holes when the iron tip is removed and it might be possible to add too much solder and create a solder bridge between the tab and one of the pins.

So what do the experts think? Which method would you recommend? What other methods might work?

--
James T. White




begin 666 Test3B.png
 Click to see the full signature
Reply to
James T. White
Loading thread data ...

As bad of an idea as it sounds, I've had a lot of luck just using my heat gun at its hottest setting and paste to solder thermal tabs. I just design the pad with an array of the smallest vias that my fab can support, lay a little dab of paste at the middle of the part with a little more on a corner pin, and use the part to press it out across the thermal contact, and then use the gun at a distance to slowly heat the board, inching the gun closer until I see paste at the pin reflow. I give it a few seconds after the pin goes, then switch the gun over to cold and pray that I haven't roasted my chip. Hasn't failed yet.

Also, if you can provide heat from underneath to supplement the heat from above it helps heat the board more evenly. Be> I'm considering at using Linear Technology's LT1976 switching regulator in a

new

Although

Reply to
Rob Gaddi

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.