|Caution: This RF deck looks easy to construct, and it can be, but if you do not have extensive experience working with surface mount parts on high power SSPA's at 23cm, your results can vary widely. Even if you are an accomplished builder of other things, the assembled/tested RF deck may be your best choice instead of the kit; it could save you a lot of time, trouble and expense.|
The assembly instructions are next; please read the entire document, and carefully
This is what the properly assembled RF deck will look like before mounting to your heat sink; A high resolution photo is here to assist you with parts placement.
Note the two trimming pads nearest the output ground foil have been removed; these are not used, and were the cause of arcing at full output on some of the prototype RF decks.
I originally thought the cause was that green solder mask, and it did play a role in combination with those outer pads; later, I found eliminating those two pads was more than enough to solve the problem...and much easier to do than removing the solder mask around the output area.
The boards in your kit will be the ones without solder mask
(shown in the next photo), but for the purposes of placing components in these
kit instructions, I'll be using the green boards as examples because the board
markings have better visibility.
The RF deck assembled from the kit will look like this one. The solder mask and two of the outer trimming pads on the input and output have been eliminated, yet both PC board revisions perform the same, there is no operational difference at all. The boards are identical in thickness, material, trace design and copper weight.
OK, let's get started...
Locate the output board; referring to the schematic and the reference designators on the board, install the 6 high voltage microwave chip capacitors.
The output board is the shorter one.
Note: save the trimmed-off capacitor leads in the next step, they will be used later.
Trim the leads on the electrolytic capacitors to 6mm length and bend them as
pictured, about 2mm away from the body of the part. Note the position of the
negative lead on the body of the parts, and the direction of the wire bends on
each (they are not bent the same as one another).
Install them like this, positive to the bypassed end of the VDD trace, negative to the ground foil. Leave enough room for the two mounting screws and their flat washers to be placed at the top of the board; note the position of the negative polarity marking on the bodies of the parts, they face inward toward the center of the board.
Now, using a cut-off piece of one of the capacitor leads you saved in the previous step, jumper in one of the trimming pads at the output as pictured here.
How many pads you will need to jumper in to trim the amplifier output properly is going to depend on how it plays when tested; starting with one is a good place, I find them to require as little as none to as many as three. Most of them play best with one or two (at 1296).
The output board is finished.
Locate the input board (the longer one) and install the remaining parts. Checking the schematic and the high resolution photo will help you put everything in the correct place, and orient the Zener diode properly.
The input board is finished.
If you purchased the copper spreader and then mounted your own LDMOS using the flow-solder method (see the video here), make certain the area under the LDMOS tabs is free of flux or other material which may hinder good electrical contact.
Slip the input board under the transistor gate tabs. Use it as a lever to bend the tabs upward a bit as you see it done here. Do the same for the tabs on the drain side.
You probably already noted there is no room for mounting screws near the transistor tabs; with high power devices at microwave frequencies, it is very important to get good reliable contact between the PCB and the copper spreader, so we'll be arranging a very tight fit as the boards are finally set into place. You won't be soldering the PC board to the spreader using this procedure, as there are serviceability drawbacks to that method.
The copper spreaders I have available are machined with a .055 trench depth, and that positions the transistor tabs about .005 above the board surface when all is in place. There is some variation (+- .002 or so) in the machining of the trench, so we'll need to fill that gap with a small spacer placed between the pcb and the copper spreader (next step). The spreaders for this RF deck are shipped with spacer material the correct thickness for the spreader provided.
If you are making your own copper heat spreader from
the drawing available here, you can
eliminate the need for these spacers by changing the trench depth from .055 to
.060 as long as you can hold the tolerance to +-.002 (.058 to .062).
Slip the spacer in place partly under the LDMOS tabs, but not far enough forward to
climb up on the LDMOS footing.
You may need to insert the PCB at a slight downward angle to get it into place, it's going to be a tight fit. If you can't get it set into place properly, with the edge of the pcb in alignment with the edge of the trench, the spacer may have slipped onto the transistor footing.
Check to see if the mounting holes are centered where you see the two 3/16 screws and flat washers installed. If everything looks OK, install those 3/16 x 4-40 mounting screws with their flat washers.
Install the output board with it's shim in the same manner.
Observing proper ESD safety methods, bend the transistor tabs back into
place (level). Using just your clean fingers, push the tabs flat against the
Place a drop of liquid flux at the center edge of all 4 transistor tabs, so it flows a bit under the tab when it is applied. Not too much, just a drop will do. Solder the transistor tabs to the PCB as pictured here. You can clean up the flux residue with iso alcohol and a cotton swab if you like. Leaving it in place will do no harm, but I like to clean it afterwards myself.
Once the copper spreader is mounted to your heat sink (also available on the parts page), tightening the 8 outer mounting screws will apply a slightly (and beneficial) higher pressure yet to the board-to-spreader contact area.
You're all done and ready to test; here is the recommended test procedure.