A 150 watt 902 903 linear amplifier5

Some time ago I made available a number of boards for 903 amplifiers. Most of them were 150 watt boards but some were in the 30 watt range while other were much lower power. They all came out of surplus Motorola cellular linear amplifiers. The boards were very high quality commercially made products. This page details my use of one of the 150 watt boards in a 903 station. This article was also printed in Cheese Bits, the publication of the Mt. Airy VHF Club of Philadelphia.

At this point I have not personally done anything with the lower power boards. In all likelihood I will not either since my transverters has 10 watts output and I have a 50 watt linear already. No support beyond the scope of this web page is offered or available. No additional boards are available at this time either.

The following is the original test from Cheese Bits:

Using those 903 MHz 150 watt amplifiers

I finally got around to packaging a 150 watt 903 amp. When I started my objective was to go to the junk box and find the needed parts avoiding spending more money. That goal was met without difficulty. Most of the members can accomplish the same thing.

Dave, W3KM suggested making an PC board interface to the amps since they are on a fragile alumina substrate. This was a great idea; one that I supported. But I got looking around and thinking about making a board and decided that would take too much time. I then looked for an easier (faster) way to get this project done.

What I did was use 2 female chassis mount SMA connectors with a short piece of .100" Teflon coax that I had acquired and mount them very close to the RF in and out tabs on the board. (See Cheesebits 2-02 or my written description for pin layout) I cut most of the Teflon coax off leaving only about 12 mm of cable. I then cut off the shield and Teflon insulation and bent the center conductor over to mate with the tabs and soldered them in place. Next I soldered the ground tabs of the boards directly to the SMA connectors. Use lots of heat here; once the solder flows onto the SMA connector push the ground tab over with the iron and solder it in place. DO NOT keep the heat on the ground tabs while waiting for solder to flow on the connector body, they may break!

Next I brought out the bias tabs to a 1000 pf feedthrough. Following that I brought out the extra tabs that do not seem to be required for our purposes, just in case I ever wanted to do anything with them. I am not clear on how Motorola used these terminals. Finally I brought both 26 VDC tabs to a common point through RF chokes. Use a VK-200 here using wire sufficient to handle about 5 amperes. Since I was soldering directly onto the tabs of the board I then used a large glob of silicon adhesive to hold the RF chokes in place avoiding stress on the tabs themselves. From there I ran a #14 wire to a terminal with a .47 UF tantalum capacitor to ground and then used a third choke to another terminal with an additional .47 uf tantalum. Finally a #14 to a 1000 pf feedthrough.

Looking at the pictures you will get an idea how I laid this project out. I had to do some creative cutting of the heatsink around the SMA connectors. Notice that I used metal tape over the wires running next to the board. I did this for two reasons; to hold them in place avoiding stress on the tabs, and for RF shielding as they pass the board. The box I used is a little large but I had it on hand.

A possible modification would be to include the voltage regulator within the amplifier box for the 15.5 VDC bias voltage. This must be a well regulated voltage to maintain the linearity of the amplifier. Of course a much smaller box would also work. Make sure you use enough heatsink. BNC or N connectors would be alternatives though they would require long leads to and from the amplifier RF tabs potentially causing problems.

Good luck,

Randy Bynum, NR6CA