CB PLL data book/en/Section II: Background for modification methods

The CB PLL Data Book

Now we’re ready to get to the good part you’ve probably all been waiting for! Several ways to “trick” CB rigs into getting those high and low “funny” channels, as well as 10 Meter Ham conversions, will be explained in this section.

There are still two basic methods of changing frequencies in the majority of CBs. At this writing, the foolproof designs have not exactly flooded the market, especially for SSB use, and therefore most rigs can be modified using these tricks. These tricks are:

1. Change the Programming N-Code on the PLL chip’s pins;
2. Change the Loop Mixer signal.

It’s not the purpose of this book to teach the basics of aligning transmitters and receivers, so if you’re planning a large frequency conversion such as up into the 10 Meter Ham band, realignment of other circuits will also be required. A schematic circuit diagram of the rig is essential. However for adding one or two additional 40-channel segments to most rigs, the only alignment usually required is in the PLL’s tuning circuits themselves. This part is left up to you to figure out with the aid of the circuit schematic or a more experienced friend.

Before getting more specific, I think it’s important to describe an actual PLL circuit to make sure you understand its complete operation. Let’s walk through the complete circuit, step by step.

A typical synthesizer circuit

Refer to Figure 10, which is the PLL circuit of perhaps the most popular AM PLL rig ever made. It’s been sold under dozens of brand names and uses the ever-popular PLL02A PLL chip. The AM/SSB or AM/FM/SSB variations of this chassis are very similar when you consider the minor changes needed for SSB offsets and sliders and FMing the VCO circuit. You’ll also be referring to Chart 1, which is a breakdown of all the important operating conditions by channel number. Such a chart is normally included with the radio’s service manual but certain facts not related directly to 40-channel operation are often left out. I’ll be filling in the missing blanks for you.

A PLL circuit may be categorized very generally by the number of crystals it uses and by whether its VCO is running for low-side or high-side receiver IF injection. This example is actually the second generation PLL02A AM circuit; the first one used a 3-crystal loop and can be found in Section III. The newest chips use a single 10.240 MHz crystal and low-side VCO operation in the 16-17 MHz range where the VCO can be directly divided without a loop Down Mixer.

The key to synthesizing all the required frequencies is in the Programmable Divider, which is the only PLL section that you can control from the outside world at the Channel Selector switch. That switch is where the whole process begins.

Suppose you choose U.S. Channel 1, 26.965 MHz. (This description applies to all circuits and chips.) In the Channel 1 position, the Programmable Divider receives a very specific set of instructions at its programming pins, which are directly connected to the Channel Selector. This particular instruction set, called an “N-Code”, applies only to Channel 1 and is nothing more than a number which will divide down any signal appearing at the Programmable Divider input by that number.

Binary programming

Referring now to Chart 1, you see the N-Code for Channel 1 is the number „330” and the numbers progress down to „286” at Channel 40. The number 330 is the direct result of applying a +DC voltage of typically 4-8 volts to certain PLL program pins while grounding certain other pins at the same time. Recall that the PLL requires a digital or binary counting system rather than the common decimal system used by people.

In a binary number system, each successive programming pin or “bit” is worth exactly twice (or half) that of the pin next to it, such as 1, 2, 4,8, 16, etc. A series of „Is” and „Os” appears in the chart for each of the 40 channels. The „1” means +DC is applied to that pin, and the „0” means that particular pin is grounded. The greater the „Power-of-2” controlled by a pin, the greater its „significance”. As you’ll see next the greatest Power-of-2 for this example is 256 on Pin 7. Therefore Pin 7 is called the „Most Significant Bit” (MSB) and the „Least Significant Bit” (LSB) is Pin 15, which only has a weight of 1. A chart like Chart 1 that shows the logic states (“ 1” or “0”) of each PLL program pin for each channel is called a “Truth Chart”.

How exactly was the number „330” decided? Chart 2 shows the truth states for Channel 1 only. Above each PLL program pin are numbers I’ve labelled “POWERS OF 2”, such as 1, 2, 4, 8 up to 256, because this is exactly how a binary counter counts. By adding up the weight or significance of every pin where a „1” appears, the N-Code is determined. (The „0” pins are always ignored.) In this example, we have 256 + 64 + 8 + 2 = 330. Figure 11 shows the actual voltage switching. Try the math for a few other channels yourself; you’ll be using this knowledge eventually!

Notice from Chart 1 and Figure 11 that the logic states of pins 7 and 8 never change at all for any of the 40 channels. Instead they are permanently hard-wired to the chassis such that Pin 7 is always connected to +DC voltage (“1”) and Pin 8 is always grounded (“0”).

You’ll discover that many service manuals won’t even indicate these pin states in their Truth Charts because they never change when programming the legal 40 channels only. This is a case of those missing blanks I’m filling in for you, and you can test this idea by checking the rig ’s circuit diagram . Compare the total programming pins available to the total number used in a 40-channel rig; you’ll find an obvious modification source!

The 18-Channel A ustralian CB service was recently expanded legally to match the standard 40-Channel FCC American service. Since many Australian rigs are simply U.S. rigs with a different (limited) Channel Selector switch, they can be easily modified to cover the extra channels. For example, Australian Channel 1 is 27.015 MHz, which is U.S. Channel 5. The N-Code here is 325. The N-Code for Australian Ch. 18 (27.225 MHz) is 304. Therefore by reprogramming N-Codes of an 18 channel Australian rig for numbers greater than 325 or less than 304, the rig can be expanded.

This particular chip, the PLL02A, has a total of 9 binary programming pins, which are pins 7 - 15. It therefore has what’s called a „9-bit” binary programmer. Some quick math will tell you that this chip actually has a potential channel capability of 2⁹-1, or 511 channels! (1+2+4+8+16+32+64+128+256 = 511). Only 40 channels are used for CB purposes but by proper connection and switching of unused pins, many more frequencies are possible.

VCO circuit

Refer back to Figure 10. This VCO runs in the 17 MHz range, going from 17.18 MHz on C hannel 1 to 17.62 MHz on Channel 40. The VCO is controlled by an error voltage it receives from the Phase Detector, which is always looking for a match between the Reference Divider and Programmable Divider outputs. The Reference Divider is very accurately controlled by a 10.240 MHz crystal oscillator whose signal is divided down digitally by 1,024 to produce the 10 KHz channel spacings. If the Programmable Divider should also happen to produce an exact 10 KHz output, the result would be perfect; there’d be no correction from the Phase Detector, and the loop would be locked.

What would it take to produce a perfect 10 KHz output from the Programmable Divider? We’ve already seen that the Programmable Divider is set to divide any signal it sees by the number 330. If it should see, for example, a signal of exactly 3.30 MHz appearing at its input, the resulting output would be 10 KHz. (3.30MHz-P 330 = 10 KHz.) If we can somehow produce an input signal of 3.30 MHz, everything will fall perfectly into place!

Loop mixing

It so happens there’s a very easy way to do this by cleverly borrowing some existing circuitry. If some 10.240 MHz energy from the Reference Divider is taken off and passed through a tuned doubler stage, the result will be 2 x 10.240 MHz = 20.480 MHz. Here’s where that very important loop mixing principle enters: By mixing the 20.480 MHz signal with the 17.18 MHz Channel 1 VCO signal, sum and difference frequencies are produced. The sum frequency is 20.480 MHz + 17.18 MHz ~ 37.660 MHz. The difference frequency is 20.480 MHz — 17.18 MHz = 3.30 MHz which is precisely w hat’s needed to lock the loop on Channel 1. And the 37.660 MHz signal isn’t wasted either; it’s used as the high-side injection signal to produce the first receiver IF when mixed with the incoming 26.965 MHz Channel 1 signal. (37.660 MHz — 26.965 MHz = 10.695 MHz first IF).

Phase detector correction

What happens if the mixing product entering the Programmable Divider isn’t exactly 3.30 MHz? Think about it. Since the N-Code is 330, any signal other than precisely 3.30 MHz will produce a slightly different output to the Phase Detector. For example, if a signal of only 3.10 MHz enters the Programmable Divider, the resulting output would be 3.10 MHz -P 330 = 9.39393 KHz. The Phase Detector will now sense this error and try to correct it by sending a DC voltage to the VCO. This correction voltage will drive the VCO up or down slightly in frequency, constantly being compared in the Phase Detector, until an exact match occurs once again. Although this appears to be a trial-and-error process, the whole thing happens in the time it takes you to switch from Channel 1 to Channel 2!

Receiver IFs

This completes the basic loop; everything else is icing on the cake. We’ve just seen how the Channel 1 PLL mixer signal of 37.660 MHz provides the receiver’s first IF injection signal. Now notice from Figure 10 that even a third use can be made for the 10.240 MHz Reference Oscillator. By mixing it with the 10.695 MHz first IF, the result is 10.695 MHz — 10.240 MHz = 455 KHz, the second IF. (The sum product is ignored.) Pretty smart these engineers! Almost all AM or FM CBs use this dual-conversion receiver process, and it’s commonly used in car radios, scanners, FM stereos, etc. where a lot of the circuit hardware already existed.

Transmitter section

For the transmitter of this example, the on-channel frequency is produced very simply by mixing a separate 10.695 MHz crystal oscillator with the 37.660 MHz Channel 1 PLL mixer output. The difference is 37.660 MHz — 10.695 MHz = 26.965 MHz, which is then passed through tuned circuits and the normal transmitter RF amplifier chain.

You can use the preceeding explanation as the basis for any PLL circuit you find. We’ve already figured out some of the most common ones for you and their block diagrams appear in Section III.

Truth charts and progamming methods in detail

The Truth Chart is the most important first step in determining how a modification can be made, or even i/it can be made, so we’ll look at it in greater detail now.

The example just explained was a very easy PLL circuit using the binary type of programming code. It’s quite possible for the same chip to have different N-Codes depending upon how many crystals are used, whether it’s AM or AM/SSB, etc. The AM 2-crystal loop had N-Codes going from 330 down to 286, because those were the dividers needed for proper loop mixing. An earlier PLL02A scheme used a 3-crystal synthesizer with N-Codes going from 224 up to 268. And for the ever- popular PLL02A SSB chassis (American or European versions), the N-Codes are 255 down to 211.

Notice that these N-Codes may go up or down with increasing channel number; this depends purely on the VCO’s design. In Section III you can compare all the PLL02A block diagrams to see where and why these differences occur.

Meanwhile let’s return to a portion of Chart 1 to study some of its other features. Chart 3 shows only the channel number, channel frequency, and N-Codes from the original chart. Observe the progression of N- Codes from Channel 1 to Channel 40. Notice anything unusual? The N-Codes are not all consecutive and skip a few numbers any time there is no legal CB frequency. For example, Channel 3 is 26.985 MHz, and Channel 4 is 27.005 MHz. What happened to 26.995 MHz? It’s not a legally assigned channel. This is known to CB people as an “A” channel, in this case Channel 3A. There are also skips at channels 7,11, 15, and 19. In addition the American FCC Channels 23, 24, and 25 are assigned out of order. Therefore all N-Codes as well as VCO and mixer frequencies are also out of order in the chart. Many European countries having only 22 channels simply adopted the American scheme exactly for the first 22 channels. Australia uses 18 channels whose numbers didn’t correspond to American/EEC numbers but many of the actual frequencies are the same. And Britain originally used 40 consecutive channels having no skips at all. Remember this fact whenever you’re checking a PLL Truth Chart; otherwise you might think your math is wrong when it isn’t!

BCD programming

Another common programming method is called “BCD”, which means “Binary-Coded Decimal”. Think of it as a cross between the binary (Base 2) and human Decimal (Base 10) number systems. Chart 4 shows part of a BCD channel program used in the very popular uPD858 SSB rigs. (Eg, Cobra 138/139 XLR, Realistic TRC457/458, President Adams, etc.) This chassis is an older PLL circuit requiring a Down Mixer into the Programmable Divider. If you check the block diagram for this chip in Section III, you’ll see that the downmix frequencies are .910 MHz to 1.35 MHz. Therefore the N-Codes are 91 to 135 for standard 10 KHz spacings. Note that the N-Code between channels 3 and 4 skips in exactly the same way as in the PLL02A circuit, since Channel 3A is not a legal CB channel. What’s the big difference? Above each PLL program pin number is now something called “BCD POWERS” rather than the previous “POWERS OF 2”.

In this system , the pins have been assigned such th a t each successive group of pins has a w eight or significance 10 times greater than the proceeding group. Within each decimal group, w eights still double in the usual binary progression, except that the highest possible number in any group c a n ’t exceed 9 or its decimal multiple, such as 90, 900, etc. (Assuming there were that many pins on the chip.) Each decimal group can only have a maxim um of 4 bits; in this chip, there are only 10 rather than 12 programming pins so the Hundreds Group can only add up to a maximum of (1 + 2) x 100 300. Figure the total binary value in each group, multiply it by 1, 10, or 100 as appropriate, and add the groups together: Ones Group + Tens Group + Hundreds Group, etc.

Since each group h as a value, the sum of the groups produces the N-Code. For Channel 1, we therefore get 1 + (10 + 80) = 91. Try the math yourself for the other pins. Notice also that pin 22 is permanently grounded, since its w eight is “200” and we never need an N-Code bigger than 135. (100 + 30 + 5 135). By using all 10 programming pins (pins 13 to 22) there’s a potential channel capacity of 9 + 90 + 300 = 399 channels if N-Codes could be programmed from 1 to 399. This fact has been put to much use in frequency modifications! Once again, the 858 chip has this excess capability for possible use in other synthesizer circuits besides CBs.

Before you get too excited about all the potential channels hidden inside some PLL chips, I must point out that most rigs can ’t possibly cover as wide a range as these chips without a lot of retuning. Modern rigs are capable of about 1.2 MHz to 1.8 MHz total bandwidth, which means 120 to 180 10 KHz AM /FM CB channels.

The BCD method was originally used in about 15% of the older generation circuits. The reason was because certain support hardware such as BCD switches, keyboard controllers, and 7-segment LED displays required BCD inputs. The current generation almost always uses BCD inputs. Some examples are the LC7120, LC7130/31, LC7135, LC7136/37, uPD2814, uPD2816, and uPD2824. These chips also only have 6 programming pins.

Pressetable dividers

An interesting variation of the programming expansion scheme is used in the new Cobra 148GTL-DX, which is a very popular rig sold only in Europe and the U.K. In order to get 120 channels, they start off with the same very flexible chip, the MC145106, as used in the typical PLL02A chassis. Only this time, the N-Codes can be preset to a new set of 40 channels each time you change the L,M,H band switch. This is done by using two special digital counter chips, the MC14008s, wired such that each band selection also changes the set of N-Codes. The net result is that a single Loop Mixer crystal (15.00 MHz) can be used to provide 120 channels. In previous 80 or 120 channel schemes, additional loop mixing crystals are switched in while maintaining a single set of N-Codes all the time.

The reason for doing this is purely economic: The Cobra 148GTL-DX can offer 120 AM/FM/SSB channels, and a dual-conversion AM receiver, for a total of only 3 crystals in the whole radio. Compare this to the typical 120-channel Cybernet (Ham Infl, Colt, Major, etc.) or the 120-channel Uniden Superstar 360, which require 5 and 6 crystals respectively. The cost of a crystal to the manufacturer is about $3 each, while the cost of two MC 14008s is about $1 total. Since there is roughly a 5:1 mark-up from manufacturer’s cost to actual retail price of a rig, this means a savings to you of $30 to $45 on the total retail price!

The idea of presettable dividers is also found in several other PLL chips. The most common example is the MB8719. While the chip at first appears to have 7 binary programming pins, closer study shows that Pin 10 is actually used to preset different N-Codes for use with different loop mixer crystals in the American rig versions. (11.1125 MHz vs. 11.3258 MHz crystal in an otherwise identical chassis.) In the newest Uniden European rigs (Eg, Stalker ST9F-DX, Superstar 360FM) the MB8719 or MC145106 chip is used along with the MC14008 presettable dividers to provide 80 or 120 channels. In the Stalker, they even provide an additional Loop Mixer oscillator on its own small PC board which can be switched in to give the 40 U.K. channels as well. (15.4825 MHz for the first 80 “FCC” channels, and 15.55625 MHz for the 40 U.K. channels.) Of course the N-Codes are different for each band and the Band Selector switches both the crystals and the proper IC programming at the same time.

Multimode programming

There’s one chip that deserves special mention, even though it’s not being used much anymore: the uPD861. NEC really outsmarted themselves with this one!

The 861 has some special control pins so that the designer can choose either binary or BCD programming. There are 8 binary programming lines, which means a possible 28—1 or 255 channels when used in the binary mode. In the BCD mode, a special “ROM” Code Converter is connected to allow only the legal 40 channels. Thus the 861 could be used in other synthesizer applications. In Section III you’ll find the exact specs; note that simply changing the voltages on a couple of control pins will allow you to convert a rig which when used in the BCD/ROM mode is non-modifiable. You can then program it directly in binary with switches.

A few chips such as the uPD2810, uPD2814, and uPD2816 allow multiple choices of N-Code sets such that several possible downmixer circuits can be used. This feature is intended only for design flexibility; it won’t help you in your modification attempts. (Actually the feature was intended to make the chip usable in both AM/FM and SSB circuits, but to date only the AM/FM design has been found in CB rigs.)

Controlling program pins

You know that to control a program pin, a voltage or ground must be connected to that pin. Most chips have resistors built into the chip structure which are connected internally to the main +DC supply or ground pins of the chip. These resistors are called “pull-up” and “pull­ down” as they automatically force the logic state to a “1” or “0” respectively, unless controlled externally. The external control takes the form of the Channel Selector switch if the pin is needed, or a direct connection to the rig’s circuit board ground or +DC if not needed for only a 40-channel set of N-Codes. When you need to control a pin for modifications, cut the circuit board trace leading to that pin and bridge the cut with a small (*4 watt) resistor of about 1K to 4.7K ohms. This will isolate that pin until it’s ready to be switched by you. In addition it can help protect the chip from possible damage due to static electricity; a pin should never be left “floating” and should always be connected to something externally.

Returning to our first example, the PLL02A h as in tern al pull-down resistors, which means that each program pin is always in the “0” state until +DC is applied externally. So if your modification calls for control of say, Pin 7, cut the foil trace going to Pin 7 and bridge it with a resistor. Figure 12 show s the principle of external pin control; it’s com m only used with the PLL02A, MB8719, and uPD858 chips. You can also use this idea to get the automatic Ch. 9/19 recall feature in the LC7131, LC7135, or LC7136/37 chassis if the rig doesn’t already have it installed.

ROM program converters

The reason the new est chips use BCD programming is purely a legal one: By using BCD combined with a sneaky additional circuit inside the chip called “ROM”, any illegal frequency modifications by changing programming voltages are now impossible. The current FCC rules require that the PLL chip can only contain a total of 6 programming pins. So even by using the straight binary system , this alone limits possible channels to 2H—1, or 63 total channels. (1 + 2 + 4 + 8 + 16 + 32 - 63.) By combining BCD and ROM, even this number is reduced to exactly 40, 22, or 18 as the case may be for various countries.

To date, the British government has given their “CB 27/81” approval stam p to a few rigs using the simple binary chips (Eg, DNT 2740 and M4OFM) but most legal U.K. rigs are using the LC7136/37 Cybernet chassis or the TC9119 Uniden chassis. In the U.S., the only rigs still being sold with modifiable PLLs are the SSB rigs using the PLL02A or MB8719 chips, because the manufacturers have never changed their p articular model number from the time it w as given the original FCC approval. If they ever changed their model numbers, they'd have to go through the entire re-approval process and end up with the newer PLL circuits. For this reason, rigs like the Cobra 148/2000 GTL will probably be sold forever because as soon as a non-modifiable model appears, sales will drop drastically when the word gets out. I f s no accident th a t some models are so popular! Be aw are of this and if you’re in the market for a good rig, see if the model is listed in Section III under the “good” PLL chips.

Since governments finally got wise to all the bootleg CB frequencies being used, ROM was the answer. A “ROM Code Converter” inside a PLL chip is the key to preventing modifications. The term “ROM” means “Read Only Memory” and is commonly used in digital computer system s. Inside the chip, safely out of reach by you, is a ROM Code Converter. The required N-Codes for only the legally-authorized number of channels were permanently written into the IC chip during manufacture.

The programming pins that connect the chip to the outside world at the Channel Selector switch are used only to command the ROM to release its stored N-Code information to the Programmable Divider circuit. In other words, there’s now a “middleman” to interfere with your modification plans, and he’s untouchable! The Channel Selector instructs the ROM, the ROM releases the correct N-Code, and the Programmable Divider then performs its usual division of the input signal. In addition, by using BCD it’s a very simple matter to say to the chip (in BCD language), “Give me Channel 6”. The BCD code adding up to the number “6” is then applied to the correct pins by the standard use of voltages and grounds. This code is among those the ROM will recognize, and the correct N-Code divisor is set up in the Programmable Divider. If you should try to force an illegal program code with other voltages and grounds, the chip either ignores you completely, kills the transmitter, or in some cases calls up Channel 9 or 19 instead. Also, the chip uses a T /R shift with different ROM N-Code sets for each mode to provide the 455 KHz IF offset for AM or FM dual-conversion use. A nasty trick!

Chart 5 is a section of a ROM Truth Chart showing how this idea works. This is a typical example of a 6-bit BCD programming method. It’s obvious however that every single channel is generated by calling out its number directly in BCD. This is received by the ROM, which then converts the program code to whatever the real N-Code happens to be.

Below Chart 5 is a drawing showing the equivalent circuit for Channel 6 in such a chip. This particular chip can directly divide a VCO signal in the 16 MHz range. I already know from the service manual that the input to the Programmable Divider from the VCO for Channel 6 (RX mode) is 16.330 MHz. The BCD equivalent of “6” is presented to the ROM input, which then converts it to the real N-Code. Since this chip uses 5 KHz steps from the Reference Divider and Programmable Divider, some simple math will tell you that the real N-Code for Channel 6 is 3,266. (16.330 MHz ~ 3,266 - 5 KHz). You can’t fill in the “A” channels, go below Channel 1 or above Channel 40, and even the FCC/EEC skips are already pre-programmed into the ROM. Examples of ROM chips are the LC7120, LC7130/31, LC7135, LC7136/37, PLL03A, PLL08A, TC9106, TC9109, TC9119, uPD2814, uPD2816, and uPD2824. (NOTE: A new U.S. chip has just appeared, the MB8733.)

Other ROM variations

There are several newer Toshiba chips (TC9106, TC9109, TC9119) that first appear to use very odd-looking program codes in their Truth Charts. You won't be able to figure out any kind of binary or BCD progression w hen studying the sequence of ‘"Is” and “Os”. That’s because these chips contain two sets of ROM, and are designed to work with standard rotary or LED Channel Selector switches. The code you see in the Truth Chart actually does two different things:

1. Signals the second ROM set to release its stored N- Codes into the Programmable Divider. It does this only when a legal program code is presented to the first ROM set.
2. Applies the correct set of +DC voltages and grounds to light up all the proper segments of the 2-digit, 7-segment LED channel numbers. This is another bit of digital magic that we won’t get into here!

These chips use 8 programming pins to control the LED channel display, where the BCD chips only need 6 pins. You can think of the first ROM set as nothing more than a converter which translates an 8-bit rotary switch code into a language that can be understood by the channel display and the second ROM set. Figure 13 shows the general idea. The chips by the way are nearly impossible to modify by any easy method.

About the only good thing to be said in defense of all these newer ROM chips is that they’ve helped to keep radio prices affordable by greatly simplifying the PLL circuits. And they’re more reliable because there are fewer parts to go bad. Compare all the block diagrams in Section III and you’ll get some appreciation of just how far the PLL has evolved.

Loop mixer modifications

Now let’s look a t the second possible conversion m ethod, that of changing the Loop Mixer frequency itself. This is one of the easiest ways to modify a PLL circuit that contains a downmix signal. A few chips such as the PLL02A, MB8719, and uPD858 can be modified by either the programming pin change or the downmix change methods. No wonder rigs using these chips are so popular even today!

Changing the mixer crystal is commonly done in CB-to-Ham 10 Meter conversions. Since the rig will never be used on the lower frequency CB band, it can be permanently retuned up in the higher Ham band. However most of you are probably expanding the CB band to add another 40 or 80 channels. The very popular European rigs from Ham International, Major, and Superstar are basically an American PLL chassis with the extra mixer crystals already installed.

Never forget there’s always a trade-off for any modification involving a large desired bandwidth. You can’t stray too far from the rig’s design limits without retuning the entire radio. Modern solid-state rigs can generally cover about 120 to 180 channels without much work, and sometimes even more when certain broadbanding tricks are used by a qualified technician. One major problem that results from broad­ banding is that receiver selectivity and “bleedover” rejection suffer. It’s no accident that American CB rigs which are designed for only 40 channels have much better Adjacent-Channel Rejection specs than their European counterparts which allow 80 or 120 AM/FM channels. The difference is typically 60 dB for American vs. only 40 dB for European rigs; the American rigs thus have 100 times better rejection. (20 dB difference = power factor of 100.) The new line of high-frequency CHANNEL GUARD crystal filters available from CB CITY INTERNATIONAL can add razor-sharp selectivity and bleedover rejection for the popular 7.8 MHz, 10.695 MHz, and 11.275 MHz SSB rigs. Write for details.

CB-to-Ham conversion problems

When permanently planning a CB-to-10 Meter change, the PLL’s Lock Detector may need to be disabled before you can retune the rig. Substituting a new loop mixer crystal that throws the rig 2 MHz or more from its original design limits may place the VCO outside of its normal lock-up or capture range. You can easily defeat any Lock Detector by either cutting the foil trace to that PLL pin or lifting one end of the switching diode normally connected to it. Just unsolder one diode end from the PC board until the VCO is retuned to its new range; then put it back.

Sample modification

Once again I’m using the PLL02A chassis, this time the SSB version. Refer to its Block Diagram in Section III as we proceed.

The VCO for this chassis runs in the 17 MHz range, and is mixed with a 20 MHz signal to produce the downmix signal into the Programmable Divider. This downmix signal is 2.55 MHz on Channel 1, down to 2.11 MHz on Channel 40. The 20 MHz mixing signal can be generated in two different ways, and you’ll find both methods used. Either a crystal in the 10 MHz range is doubled, or a crystal in the 20 MHz range is used directly. The American 40-channel version uses a 10.0525 MHz loop mixer crystal oscillator. You can add complete new 40-channel segments by switching in new crystals according to the formula:

New Crystal = 10.0525 MHz ± (N x .1125),

where N is the number of 40-channel segments above or below the “legal” 40 channels where you want to begin.

As an example, using a crystal of 10.165 MHz (10.0525 MHz + .1125 MHz) will give you a 40-channel band segment starting at 27.415 MHz in the Channel 1 position. If you do this, you will still have the same skips in the “A” positions and Channel 23-25 positions. In some European versions of this chassis, a small PC board containing the 10 MHz or 20 MHz crystals is installed; a front panel switch is already there to choose among the Low, Medium, and High (L,M,H) bands. The same idea is used in the Superstar 360, which is a European version of the basic American MB8719 chassis. It contains an extra PC board with additional 11 Mhz tripler crystals that are switched in from the front panel.

In this type of modification, there will always be the exact same skips at the “A” positions and Channel 23-25 positions as there are for the normal 40 FCC channels. That’s because the Truth Chart and programming N-Codes are still the same; they are already pre-determined by the Channel Selector switch. In other words, the N-Codes are identical. If N = 255 for Channel 1 with the 10.0525 MHz crystal, it will still be 255 with any other loop mixing crystal. The new loop crystal simply drives the VCO higher or lower as required to maintain the identical downmix input to the Programmable Divider.

Crystal switching methods

Figure 14 shows three ways to switch in extra mixing crystals. For diode or transistor switching, the actual switch can be a little-used switch already on the rig, such as the C B /PA . Otherwise you can drill a small hole in the side or rear of the metal frame and in stall a miniature SPDT toggle switch. For the CD4066 IC switch, you can drill a fram e hole for a miniature rotary switch. The important point is that the crystals and electronic p arts them selves must be physically very close to the existing crystal; long wire leads are out! There’s enough capacitance in 6” of switch wire to pull the oscillator off frequency or kill it completely. With the new crystal and its associated p arts rig h t by the original crystal, it’s perfectly safe to use long wires to the switch itself. Any of these three switching circuits can be built on a small piece of perf board or PC board and mounted near the original oscillator circuit.

External crystal oscillators

As PLLs developed more on-chip functions, the process of loop mixing w as simplified. In stead of needing a separate transistor oscillator, the chips began providing a suitable mixing signal directly off one of their pins. T his signal is typically 5.12 MHz, w hich is half the Reference Oscillator frequency. (10.240 MHz -r 2 - 5.12 MHz.) Since the most common VCO frequencies are in the 16-17 MHz or 34-37 MHz ranges, it's an easy matter to multiply the 5.12 MHz signal up by the proper amount to mix with the VCO and produce a downmix signal into the Program m able Divider.

The most common circuit uses a 16 MHz VCO and triples the 5.12 MHz up to 15.360 MHz. (5.12 MHz x 3 - 15.360 MHz.) This is usually done by passing the 5.12 MHz chip output through a tuned coil and then mixing it with the VCO. In addition most chips having th is feature also have the T /R shift feature needed to produce the 455 KHz IF difference for the receiver. To make matters even worse, they also use ROM! Examples are the LC7120, TC9102, uPD2814, and uPD2816.

Since the 15.360 MHz signal never changes between Transmit and Receive modes, all you need is to replace this signal with a slightly different one to get a new 40-channel band segment, To do this, you m ust build a very sim ple crystal oscillator circuit with the proper new crystal. Figure 15 show s the general idea for both the A M /SSB rigs with the uPD2824 chip and the uPD2816 chip used in the more common AM or AM/FM rigs. Inject the new signal at the points marked “X”. NOTE: You can’t use a crystal oscillator exactly as shown for the SSB circuits, because both the Clarifier and the mode offsets are generated from the 10.240 MHz crystal. These would be disconnected and would no longer work. Our EXPANDER 160 includes special jumpers which allow the correct modification in these chassis types.

The new signal will be in the 14-16 MHz range. To calculate the new crystal, the formula is

New Crystal = 15.360 MHz ± (N x .450),

where N is once again the number of 40-channel band segments above or below the “legal” 40 channels. (The standard CB band is 450 KHz wide.).

For example, a new crystal of 15.810 MHz (15.360 MHz + .450 MHz) will give you a higher 40-channel band starting at 27.415 MHz in the Channel 1 position. This is exactly how it’s done in the European versions of the common American LC7120 PLL chassis sold by Midland, Colt, Commtron, and others. Again, there will be the usual skips because the Channel Selector was designed that way, and these chips also use ROM.

or 10-Meter Novice ham conversions, a new crystal might be 16.455 MHz, which gives you 28.060 MHz (Ch.l) to 28.500 MHz (Ch.40).

Now you’re probably wondering, “Well great, but where do I get a new signal to replace the 15.360 MHz that comes from the PLL?” The answer of course is that you must build one, but it’s very easy. Figure 16 shows a proven crystal oscillator circuit that you can build on a small piece of perf board or PC board and place close to the original injection point. You can even combine this oscillator with one of the crystal switching methods shown earlier on the same piece of board. This could then be remotely switched from the front panel again, giving you 80 or 120 channels total. (For the lazy, you can order our EXPANDER 160 which is a combination oscillator & switch, or our EXPANDER 240 which is just a 6-position crystal switch. Write for details.

Crystal sources

For those of you having trouble locating a special-cut crystal, the following companies are very good if you have no other local source: CRYSTEKCorp. P.O. Box 06135 OR Ft. Myers, FL 33906 U.S.A. (800) 237-3061

JA N Crystals P.O.Box 06017 Ft. Myers FL 33906 U.S.A. (800) 237-3063

I’ve dealt with both these companies and they’re both good. The cost is about $6 these days plus shipping. Be very specific when ordering. You must state the exact frequency desired, holder type, accuracy, and load capacitance. Holder types are normally HC18/U for solder leads and HC25/U for plug-ins. Accuracy should be at least .005% or better. Load capacitance is typically 32 pF which is fine for AM-only rigs; however for SSB rigs you should get the 20 pF crystals because they require less external capacitance to trim and when part of the Clarifier circuit they will slide much further.

The reference oscillator crystal

So much for the modifiable ROM chips. It’s important to emphasize now that you can never modify any AM/FM PLL circuit using the single 10.240 MHz design by changing this crystal. Many people wrongly believe it can be done, but it can’t. Too many internal chip functions depend upon this exact frequency. For example, the 455 KHz T /R shift is the direct result of digitally dividing down this signal. If the signal were changed by changing the crystal, the T/R shift would change because the output from the Reference Divider would also change. This of course would change the VCO and mixer frequencies, there would be no 455 KHz receiver IF injection and therefore no operating receiver. The guys who designed these things are way ahead of you! (This does not apply to the LC7131 SSB chassis on Page 71.)

The impossible chips

As if to pour salt into your wounds, governments and engineers have now created a generation of PLL ICs that are almost totally foolproof. In addition to using a single 10.240 MHz crystal, T /R shift, and ROM, there is no loop mixing either. The Programmable Dividers are now so fast that they can directly divide down a VCO frequency as high as 20 MHz. Since there’s nothing to be mixed, you can ’t change the ingredients! These chips use a VCO running in the 16-17 MHz range and include:

• LC7130,LC7131,MB8733,TC9106,TC9109(40-channelU.S.)
• LC7135 (22-channel EEC)
• LC7136, LC7137, TC9119 (40-channel U.K.)
• LC7132 (40-channel U.S. and U.K.)
• SM5123A, SM5124A (40-channel U.S.)
• C5121 (40-channel U.S.).

The best way around this problem, if you can’t get one of the older rigs, is to buy a rig having SSB in addition to AM or AM/FM. The SSB circuits either use a loop mixer or don’t use the T /R shift, at least not yet. They’re a bit more expensive but th a t’s part of the price you must pay if you ever expect to go “upstairs”.

Another possible solution is to use an EPROM modification board. These allow you to customize the channel programming to your needs, and will work in many of the newer ROM PLL circuits. You can program out the skips, include 5 KHz spacing, or even program in a 100 KHz T/R shift for 10-Meter repeater use. Our new publication, THE CB EPROM DATA BOOK by Martin T. Pickering, explains not only how to make these boards, but also includes schematics and PC artwork for the most popular SSB chassis types. Highly recommended! Further­ more we’ll be offering these conversion boards soon. You can get full details by writing to us and enclosing a stamped SAE:

CB CITY INTERNATIONAL P.O. Box 31500 Phoenix, Arizona 85046 U.S.A.

THE BASIC MODIFICATION RULE IS: The simpler the PLL circuit appears to be, the harder it will be to modify. There are fewer and fewer places where you can jump in with your own program codes or loop mixing signals.

Good luck and Happy DXing!