12.5kHz the easy way

Upgrading repeaters designed for 25kHz channel spacing to 12.5kHz channel spacing seemed impossible before. Now, the availability of new filter kits allows modifications to accomplish narrowbanding-without junking your repeater.

Changes in FCC rules are moving us from the present 25kHz channel spacing at 450MHz to 12.5kHz spacing, and to tighter spacing in the 150MHz band. This change will require more than just reducing the deviation to 62.5kHz.

Most repeaters in operation today were designed for 25kHz channel spacing. Converting them to the tighter 12.5KHz channel spacing seemed impossible because the manufacturers did not have a modification kit to do the job, and many would like you to buy new equipment anyway. Although the repeaters had the required frequency stability, the lack of narrow IF filters was a major stumbling block.

A few years ago, I wrote an article, “Converting to Very Narrow Band” (CQ VHF, July 1997), in which I discussed an experiment conducted on a GE UHF MVP mobile. I wanted to see if it was possible to convert the MVP from 25kHz spacing to 12.5kHz spacing. Checking the GE parts information, I only found a filter change for 20kHz spacing, far from what I needed for the experiment. Another problem was that the GE Master II series of radios (Mastr II, Executive II, and MVP) are single-conversion receiver designs with an IF frequency of either 11.2MHz or 9.4MHz. There are no 455kHz second IFs to play with where filters are available.

So we tried another angle on the conversion of the GE MVP to 12.5kHz channel spacing. Because the 11.2MHz IF frequency was only 0.5MHz away from the common 10.7MHz IF frequency used in many radios, I decided to try to change the IF frequency to 10.7MHz. This change in the IF frequency would require a change in the injection frequency to reflect the 0.5MHz change in the IF frequency.

Five 10.7MHz crystal filters, designed for 12.5kHz channel spacing, were acquired from Communications Specialists in Orange, CA. The GE MVP was disassembled, and its old filters were replaced with the new 10.7MHz filters. A new receiver crystal was purchased so the injection would be 0.5MHz lower than the original crystal.

The mechanics of replacing the filters is easy, but be careful to keep the crystal pairs together (the crystal filters are shipped in matching pairs). When installing them in the radio, make sure the “dots” on top of the filters face each other. The first filter board in the GE MVP has two filters (four poles) with input and output tuning coils and an adjustable capacitor between them to facilitate alignment. The second IF filter board comes in two versions, one with, and one without, tuning coils; neither has the variable capacitor for finetuning of the filter. Also, most boards have only two filters (four poles), but all of the boards have the holes to mount a third filter, for a total of three crystal filters (six poles) on the second IF board.

After checking the schematic, I installed the third filter (Mastr IIs have the extra filter), and an adjustable capacitor was added to help in tuning the new filter.

After installing the 10.7MHz filters and aligning the receiver, testing showed the receiver to be working just fine. No change in sensitivity could be detected. The squelch and CTCSS decoder worked fine. (And it’s good that GE put an audio level pot in the detector circuit to help compensate for the lower deviation.) So, the receiver was working fine at 12.5kHz spacing or 62.5kHz deviation.

The transmitter was changed in two ways. First, the deviation was turned down to a total of 62.5kHz-exactly half of the old level. Second, the GE 0.005% crystal was changed to the 0.0025% EC2 type.

This is not really a difficult change, and it’s a lot less than a new radio (less than $100), but it could have been easier-and now it is. I found that in trying to convert the current 25kHz spacing to 12.5kHz spacing, the major stumbling block was not the transmit crystal at 0.0025%, which most manufacturers have available, but the IF crystal filters. While in my article I pointed out that Murata makes many 455kHz filters (the “D” 610kHz [normal for 25kHz spacing], the “E” 67.5kHz, the “F” 66kHz, the “G” 64.5kHz [normal for 12.5kHz], the “H” 63.75kHz and the “I” 62.0kHz), finding them is not so easy, except for the standard ones used in normal radios.

Manufacturers’ charts show the various selective curves of the filters for comparison. However, remember that deviation and bandwidth are not the same; bandwidth equals about 2.4 times the deviation. The problem is where to find the filters; you cannot just order them from Murata, for example, on a one- or two-piece order.

The major stumbling block has been removed, and now it is possible to not only order the 10.7kHz and 12.5kHz filters, but 11.2kHz, 11.7kHz, 21.4kHz and others, along with the 455kHz “E,” “F,” “G,” “H,” and “I” filters from Communications Specialists. It is now easy to modify not only the GE Mastr II but also the Motorola Micor and Mitrek, along with many import radios, to the new 12.5kHz channel spacing.

You can get by with only changing the 455kHz filter to get 12.5kHz channel spacing, but why have your receiver process all those adjacent channel signals, when by replacing the high IF filters your receiver has to handle only a single signal?

As soon as the new 11.2kHz-12.5kHz filters became available, three sets were acquired and installed in a GE MVP. Five filters were installed. The second set of two needed a variable capacitor for tuning, but the others were simply installed in the normal circuit. All of the new filters were tuned normally using a 62.5KHz deviation signal modulated at 1,000Hz. At the conclusion, no difference in sensitivity could be detected. In fact, it might have been a little better. The newer generation of filters may be made better than the old ones-the simplicity of tuning made it easier Communications Specialists is putting together filter kits with instructions on how to modify many of the radios and repeaters in use today to the new 12.5kHz channel spacing. The cost of these kits is much less than replacing the repeater.

I can remember converting the old GE Progress Line radios in the 1960s from 50kHz channel spacing to 25kHz spacing. We removed the IF coil assembly and physically moved the coils apart and realigned the receiver. Then we added a few parts to the transmitter and reduced the deviation from 615kHz to 65kHz. The crystal filter changeout process of today is much easier and can bring many more years of service to our currently operational repeaters and radios.

Getting even narrower What about 6.25kHz spacing? It can be done. I have an operating repeater on the amateur 440MHz band using 61.25kHz deviation. The repeater uses two Kenwood TK810 mobiles (one for TX one for RX) modified with 0.0025% TCXOs, eight-pole 12.5kHz IF filters (6.25kHz is not yet available) and an “I” 62kHz 455MHz second IF filter. Kenwood TK810s are also used for the mobiles. The experiment is working well, and we are learning a lot.

You do not need that entire spectrum just to talk.

Acknowledgements Thanks to Spence Porter at Communications Specialists, Orange CA, and Ben Sarehnia at Metro Mobil Communications, Walnut, CA, for support with filters and surplus equipment that made the tests and continuing experimentation possible.