Uniden CB Radio "Dynamic Squelch Control"

Background	Discussion of problem	Key new DSC functionality
How to Fix Problem

Summary

The Uniden Dynamic Squelch Control feature in the Uniden PC78 and PC68 series CB radios is fairly useless as it comes from the factory, as seen in numerous online reviews and comments. A simple alignment procedure is discussed to make the Dynamic Squelch Control a very useful feature, in that by a simple on/off switch, the squelch threshold is determined by a preset SNR threshold--which is what a squelch circuit truly should be based on - SNR.

References

This document is based mostly on the Uniden PC78. Only Uniden PC68 and PC78 radios have the dynamic squelch circuit. However, the Cobra 29 is very similar to the Uniden PC78, and the schematics (property of CBTricks.com) may be obtained at: CBTricks.com.

Background

While the Cobra 25 and Cobra 29 CB radios have the SoundTracker feature, which is a compandoring feature (best used when both ends have SoundTracker), the Uniden PC68 and Uniden PC78 CB Radios have the Dynamic Squelch Control feature.

The Cobra 25 is roughly analogous to the Uniden PC68, while the Cobra 29 is roughly analogous to the Uniden PC78. Again, this document is based on my measurements of a PC78.

The Cobra 29/Uniden PC78 use a sophisticated squelch circuit that prima facie appears to modify the attack/decay time using two transistors (Q11, Q12) plus a third muting switch transistor (Q13). All three of these transistors are NPN, which implies that as Vbe exceeds about 0.6V, the transistor impedance from the collector to emitter becomes low, i.e. the "switch" is "closed" from the collector to emitter. The user adjusts the bias voltage Vbe through the front panel squelch control. As the FP squelch is turned fully anti-clockwise, the base of Q12 is shorted to ground, making the collector to emitter path of Q12 look like an "open circuit." When Q12 is not conducting, the Q11 Vbe goes over 650mV (through the power supplied via R42 that comes from the microphone PTT switch 8V RX, 0V TX), which makes Q11 conduct, thereby reducing Q13 Vbe to near 0V, stopping Q13 from conducting. The Q13 bias power comes through R41, which also bias powers Q6 through the FP RF gain control. We note that Q11, Q12, and Q13 are 2SC945 transistors. Measuring the base of Q13, it is observed that as the Vbe increases past 500mV, the audio output drops dramatically, until by 600mV the audio is completely muted. The Q11/Q12 pair acts to quickly transition from about 50mV Q13Vbe to about 630mV Q13Vbe. The C36 33µF capacitor is used to delay the opening & closing of the squelch, to avoid a "chop-chop-chop" effect on a rapidly fading signal. Q13 "shorts to ground" the audio input (from the volume control) to IC4, the TA7222AP high power audio amplifier. IC4 is the flat chip attached to the right side frame of the radio, under the weather board. The basic functionality of the Cobra 29/Uniden PC78 AM squelch circuit may be summarized by that the user set the FP squelch control as desired, say to just mute the audio when no one is transmitting. Then, when a carrier signal is received, D6 injects a reverse voltage that reduces Vbe, causing Q12 to stop conducting under the discipline of the time constant set by C30 (0.01µF) and R125 (3.9K ohm). D6 feeds the AGC (automatic gain control) of the radio as well. Per preliminary PSPICE simulations, D4 has no effect on the operation of the squelch. It's just connected to the same +8V supply line as R42 through R127. D6 alone appears to govern the operation of the legacy (non-DSC) squelch. VR3 (100K ohm pot)adjusts the FP squelch range of adjustment--increase VR3 and the FP maximum squelch level is reduced, and vice versa.

The Cobra 25/Uniden PC68 use a less sophisticated with only two transistors. The FP squelch sets a bias reference point for Q6, and then the reverse voltage from D6 pushes the Vbe down such that Q6 stops conducting, which thereby stops Q10 from conducting. The attack/decay timing is less sophisticated due to the use of one less transistor vs. the Cobra 29. I would have to do further analysis to better characterize this circuit.

Discussion

Now that we know how the basic squelch circuit of the Uniden PC78 works, we can examine how the DSC interacts with the legacy squelch circuit. The DSC opens the connection to the FP squelch control, and in its place connects the DSC control output. While the FP squelch control gave a constant bias voltage, the DSC automatically increases the squelch bias voltage to Q12 as the ambient noise increases. A voice transmission of adequate strength will "quiet" the receiver, lowering the DSC output voltage such that it opens the squelch.

Key point on DSC functionality

The key new functionality given by the DSC circuit is that on a legacy AM receiver, any strong signal, including 100% noise, will open the squelch with a threshold set just above ambient noise.
The DSC circuit puts a new condition on the squelch logic--specifically, the signal must be strong AND not noisy. If the signal is strong and noisy, the DSC amplifier amplifies the noise and effectively "turns up" the squelch to keep the noise out. If the signal is strong but not noisy (i.e., the typical voice signal), the squelch will open, since the DSC circuit's 6.8kHz notch filter will pass little energy from the normal voice transmission.

The DSC uses two op-amps, namely the NJM2904 and NJM084 op-amps. It appears that a 6.8kHz notch filter is setup. I have not yet measured the parameters of this filter. The DSC detects energy outside the normal voice band via D852 and D853 on the DSC board, which are MA728 Schottky diodes. If no one is transmitting, there is energy from low to high audio frequencies, and the receiver mutes. If someone is transmitting, their energy is focused on lower (<3kHz) frequencies, and so at the DSC amplifier at 6.8kHz, there will be little (unless your modulation limiter is clipped) energy above say 3kHz. Thus the DSC would unmute.

However, many CB radios have been modified to eliminate the Automatic Modulation Limiting circuit. This increases the postive peak modulation of the radio, within the limits of the circuitry. Unfortunately, without the benefit of negative peak limiting, and as the positive peaks are also clipped, the bandwidth greatly increases (adjacent channel splatter), without any more benefit to the user. Increased positive peak modulation can be beneficial (e.g. most all AM broadcast/shortwave stations use over 100% positive modulation), but while the broadcasters negative peak limit and watch that their positive peaks don't go beyond the capability of their transmitter, Billy Big Rigger is not going to. He just turns his mike gain wide open and runs.

Thus, the REDUCED signal quality that comes from an excessively clipped (as many CB transmitters are) transmit audio causes the DSC to accidentally close.

The DSC circuit has a little problem with the PC78 in that when it's set to require a high SNR (Signal-to-Noise Ratio), the DSC circuit will accidentally close the squelch when someone is talking loudly.

View of NOAA Weather board inside PC78LTW.
The yellow wire coming out of J701 at the left middle of the picture is carrying the 162MHz from the CB antenna to the WX board. The black wire going to J701 is "ground". It seems Uniden is using a 2 wire transmission line! I noticed that touching the wire didn't seem to affect the receive strength of a weak NOAA transmitter, so I guess it's OK.

Almost immediately upon first use of the Uniden PC78, I noticed the same complaints I'd seen on the internet about Dynamic Squelch Control ("DSC"). Specifically, it is noted that only extremely strong signals (S9) could break squelch. Secondly, no matter what signal strength, a person talking in a normal voice would clip in and out due to the squelch closing. The DSC feature seemed useless.
I decided to take a quick look using nothing more than a voltmeter and a second CB radio with a dummy load.

On the underside of the radio, underneath the channel selector board (in the PC78), is a small SMT (surface mount component) board that's held in place with sticky foam above and below.

This board has four connections to the rest of the radio, as seen in the table below.
NOTE: The wire colors in your radio could differ!
NOTE: This table is from the Uniden PC78
I speculate that in the Cobra 25/Uniden PC68 that the DSC output connects to the base of Q6, the squelch amplifier, but this is only a guess!

Uniden PC-78 DSC connections

Wire Color	Function
Black	Power Common/"ground"
Red	+8V Power
White	Signal (AC) Input from AM Detector diode (D4)
Brown	DC control voltage to squelch amplifier Q12 (via R45)

---

Fixing the Problem

Currently, my suggested procedure is as follows. What you're doing is isolating the two squelch systems from each other, so that the standard squelch system works as it has for the past 20+ years, and when you switch to the new Dynamic Squelch Control, it works as well as possible without redesigning the circuit.

Part 1:

Setup the dual-mode squlech by switching R43 out of the circuit when DSC on, and back into the circuit when DSC off

1. Remove top and bottom covers from the radio. Remove the 4 screws that hold the faceplate on and remove the faceplate.
2. Take the 'yellow' wire that goes from the front panel squelch control to the bottom of the DSC switch and resolder it to the center pin of the DSC switch (same pin as the 'pink' wire)
3. Desolder the leg of R43 (39Kohm resistor) that is toward the left (microphone jack side) of the radio.
4. Solder a wire from the leg of R43 now hanging free to the UNUSED middle pin of the DSC switch (be sure you're using the side of the switch that no wires were soldered to).
5. Solder a wire from the bottom of the formerly UNUSED side of the DSC switch to the hole on the PCB where the leg of R43 used to be. This makes it so R43 is connected when DSC is off, and R43 is disconnected when DSC is on.

Part 2:

Calibrate the DSC squelch board pot (on the little surface mount board).

1. Make sure you're always using the negative (black) power wire of the radio for the "ground" of your voltmeter. The chassis of the radio is isolated from ground!
2. Turn the front panel squelch fully clockwise. Turn RF gain fully anti-clockwise.
3. Turn the DSC on. Be sure to leave the RF gain at full anti-clockwise until the end of the procedure.
4. Measure with your voltmeter from the negative power supply to the brown wire of the DSC squelch.
5. Turn RT851 (the surface mount potentiometer on the DSC board) clockwise until the voltage on the brown DSC output is at about 875-900mV
6. Complete!

Side note: To tell if the squelch is open, put the voltmeter on the base of Q13 (this is under the weather board near the TA7222 IC mounted to the side of the chassis). When Q13 base is above about 570mV, the squelch is closed. If the Q13 base is below about 570mV, the squelch is open. You will see this voltage changes extremely fast when passing the squelch threshold, thanks to the amplification of Q12 and Q11.

Now, when the DSC is off, your PC78 will act just like a PC78 has for the past several years. With the front panel squelch set to about 9 o'clock, your radio should unmute when a signal is about 1 S-unit stronger than background noise --e.g. S3 background noise, should hear signals about S4 or stronger. E.g., S7 background noise, squelch opens for S8 or stronger. At full squelch, it takes about 4 S-units over the background noise for the squelch to open (i.e. the signal is about 24dB stronger than the background noise; a fairly clear signal).

The true beauty of this setup is that now you can set the squelch for as weak or strong signals as you like, but noise burts will very rarely come through. So I really love this--say I'm driving through the city, the background noise can go from S1 to S9 and anywhere in-between and the squelch never opens unless someone is more than about 1 S-unit stronger than the noise. And if I shut the vehicle off, maybe I have so little noise the S-meter doesn't even move. Then, if someone is even just S1, the DSC squelch opens.

---

The SMT board has a single potentiometer on it. Turning this potentiometer clockwise increases the threshold (makes the squelch harder to open). Turning this potentiometer anti-clockwise lowers the squelch threshold. Without yet having researched the board's internal functionality, a few empirical facts have been preliminarily established.



Preliminary Observations of factory default (unmodified) Dynamic Squelch Board functionality

1. Squelch transitions on/off at 805mV (at brown DSC wire).
   (as measured at the pink wire in the center of the DSC switch).
   >780mV squelch closes. <805mV squelch opens.
2. The DSC switch in the "off" (down) position connects the front panel squelch control, disabling DSC.
3. The DSC switch in the "on" (up) position disconnects the front panel squelch control, instead connecting the DSC output in place of the front panel squelch control.
4. The DSC output saturates at 3.35V. The DSC board potentiometer seems to control the gain of an amplifier (possibly amplifying a peak detector).
5. The factory setting for the DSC seems awful and useless, muting on "good" transmissions.

Having observed these facts, I noted that the factory has set the DSC brown wire output to about 1016mV. The following table details collected preliminary DSC data.

Notes

1. "DSC measured voltage" is measured at the brown wire output of the DSC board, WITH THE RF GAIN SET FULLY ANTI-CLOCKWISE.
2. The "muting" data refers to the squelch closing for a signal at constant maximum modulation (like "ahhhhhhh" or a long whistle).
3. You may reduce the muting data by about 1 S-unit for normal maximum modulation conversation. That is, for 925mV DSC output, a normal conversation will experience muting at S2, and going "ahhhhh" will mute at S3.
4. NOTE: The muting level is referenced to zero background noise (i.e., antenna disconnected). Think of it as, for the squelch to open, the signal must be at least that many S-units above the background noise.
   Thus, if I have S5 background noise, and my DSC output is 916mV, (set as in note #1), it will take about S5+2=S7 signal strength to open squelch, or about 10dB SNR.
5. My estimated SNR is pretty crude. For reference, 10dB SNR is considered to be "usable". A fair/good SNR is 20+dB (well for AM CB at least).

This table applies ONLY for the default setting of VR3! The brown wire values change when you change VR3. I kept this now obsolete table just for reference.

DSC measured voltage (brown wire) with RF gain fully anti-clockwise [mV]	Estimated SNR [dB]	Muting level (squelch closes with full modulation) [S unit]
825	6	<1
830	8	closes at 0.5, open at 0.75 (with a very quiet night, virtually zero background noise in a rural location)
841		<1
876		<1
886		1
902.5	10	1.5
916		2
925	13	3
963		4
975		6
1006	20	7
1016 (factory setting)		9+

I noticed something else--ignoring (turning off) the DSC, I noted that my PC78 clips voices on strong signals when I have the front panel squlech turned up high. Examing further, it's not all the DSC's fault. The AGC circuit voltage appears to fall in the time domain between modulation peaks.

Future Work

Block diagrams explaining functionality of common squelch circuit and new DSC circuit. Perhaps even explaining how the classic CB radio dual-conversion AM receiver works. Someday.