rt320lsb.pdf

The VMARS Newsletter Issue 28

Adding Lower Sideband to the RT 320

Dr Andrew Smith. G4OEP

The RT230 (UK/PRC 320) is a super rig for amateur use, particularly on 7MHz, where its limitation to

30WPEP is not a problem. But it is precisely on this band that the absence of LSB is particularly felt.

Adding LSB is quite an easy modification, but anyone who has one of these rigs is aware that they

are very much sought after by collectors, and any modifications must be done with great care if the

interest value of the rig is not to be destroyed. Fortunately it is possible to do the modification in such

a way that no-one who is unaware of it could ever discover that it is there without opening the case.

General.

The final IF of the RT320 is at a nominal frequency of

1.750MHz. Due to a subtractive frequency conversion

earlier in the system, an incoming USB signal appears as

LSB in the 1.750MHz system. So to avoid confusion,

USB and LSB will henceforth be used to refer to the

incoming signal (or transmitter output) unless specifically

stated otherwise.

The existing carrier insertion oscillator (CIO) for SSB

generation is a 1.750MHz signal derived from the

reference oscillator for the synthesiser. This signal is fed

to the IF board (SD-D 218834) via pin F on the board.

The modification takes the form of an additional crystal

CIO running at 1.74680MHz (that is the crystal marking)

whose output is switched to the IF board by a relay when

LSB is selected. The relay also switches power to the

LSB CIO so that it is only active when LSB is in use. The

relay is activated by grounding pin G of either of the

handset connector plugs. [See note at end of article].

expected. Start with 4.7k W and adjust to suit the activity

of the crystal.

It is important to get the LSB CIO frequency exactly right

(within 100Hz or less). If you have no instruments you

can do this (approximately enough with care) by

disconnecting the antenna and listening to the broad-

band receiver noise. There should be no noticeable

change of pitch when you change from LSB to USB. If

you have a reasonably accurate frequency counter you

can get a very exact setting - see notes below on RT320

frequencies. Alternatively if you have an oscilloscope

and an audio sig-gen proceed as follows. - With the '320

connected to a dummy load, select SSB, USB and low

power. Using an appropriate audio attenuator feed a

1KHz audio signal into the mic socket, and connect the

'scope to the dummy load. Press the PTT and observe

the rf output. Raise the audio frequency until the rf

output voltage falls to half its original value (or any

measured amount). Note the AF frequency. Then find

the lower cut-off frequency by lowering the AF frequency.

Switch to LSB and do the same. Adjust the CIO

frequency until the upper and lower cut-off frequencies

are the same for USB and LSB. On the fourth hand - I

have found that the panoramic frequency display which is

part of the Digipan PSK32 software (search for Digipan

on the net) gives a very clear impression of the ssb filter

pass band if you disconnect the antenna and adjust the

volume so as to view the receiver noise. I expect you

could use this to view the effect of shifting the CIO

frequency- the passband would move up or down in the

audio band, and should be the same (about 300Hz to

2800Hz) on both sideband settings.

c) Reservations. The system as described here works

well and produces very acceptable LSB SSB which

attracts good reports on-air. However, there are one or

two theoretical points of the nature of a disclaimer. - i) the

sideband filter is an LSB filter specifically intended for the

1.750MHz LSB system. This means it asymmetrical and

is non-optimal for the new USB role. In practice this

seems to make no difference at all - carrier suppression

is excellent on transmitted LSB. ii) It is not possible to kill

the existing USB CIO when using LSB. The resulting 2.4

kHz beat frequency due to leakage of the USB CIO into

the mixer is just perceptible during receive when no other

signal is present. Use screened connections to keep this

under control. iii) In use, if you select CW make certain

that you have also selected USB. Your CW note will be

wrong (lower than normal) if you select LSB with CW

Wide, and will not get through the narrow IF filter at all if

you select LSB and CW Narrow. Also - always return to

USB when selecting AM if you want to avoid some very

strange effects. Ideally the USB/LSB selection system

should be interlocked with the mode selection - but that is

a refinement for another day. None of these errors of

settings will harm the rig.

d) Opening the case. To open the case first undo the

screws fastening the back panel and then remove it

Modification details.

Pins G of both handset connectors (on the back panel)

were originally connected to pins D (audio out) behind the

connector. These connections have been cut. Pins G

have been joined together, and to pin 9 (previously un-

used) of the 9-way D plug marked 2PL3 on the back

panel of the rig. Pin 9 of the mating socket on the main

chassis is connected to the coil of a relay which is located

(in this arrangement) under the IF board, and the relay

coil circuit is completed to the incoming 24v via the front-

panel switch. The correct connection to the switch can

be found by testing with a continuity tester for a

switchable through-connection to the external 24v

battery-connection lug, with the battery removed.

The track connected to pin F on the IF board has been

cut at the pin, and the pin, the track, and the new LSB

CIO are joined to the relay change-over contacts as

shown in Fig 1. Additional relay contacts are used to

supply 24v power to the new oscillator, which is wrapped

in insulating material, and tucked into a corner near the

front-panel mode control switch. If the track on the pcb is

cut carefully, it can subsequently be repaired (if required)

with a small blob of solder. Fig 2 is the circuit diagram of

the new oscillator.

Practical Notes .

a) Sources - I got my crystal from QSL Marketing (PO

Box 19, Erith, Kent, DA8 1HL, 01322 330830.

www.quartslab.com ) who charged me £10 total for one-

off. They could only supply an HC33 (wire-ended HC6U)

package which is VERY BIG considering the amount of

available room in the '320. If you can find a smaller one it

is worth trying.

b) Setting it up. Check the amplitude of the existing USB

CIO on pin F before you start, and then adjust the series

resistor in the supply line of the new CIO to give a

comparable output. About 200mV peak is what should be

9 April 2003

The VMARS Newsletter Issue 28

when switching from USB to LSB. No output at all is

produced if LSB and CW (N) are selected. The following

notes attempt to explain these effects.

It is known (from the User Handbook) that the CW(N)

filter has a bandwidth of 250Hz, and is used with a CW

audio tone of 2kHz. It is also known that the SSB filter

has a bandwidth of 2.7kHz.

Using all of the above information, the following

frequency scale can be constructed (note, however, that

the sense of the frequency differences is in reality

reversed, due to the USB/LSB reversal) -

f1) reference,

0Hz USB CIO

f2) +250Hz

Lower cut-off of SSB filter

before undoing the front-panel screws. If your rig is a

virgin you will have to use gentle force the first time, since

the gasket will have stuck the two parts together - be

careful how you separate them. The back panel

connects to the inner chassis by means of several multi-

way connectors which should be separated (and re-

assembled when the time comes) very carefully. Just do

it slowly and gently, and make certain that the connectors

are aligned properly when putting it back together. When

re-assembling, DO NOT tighten the screws unless you

are certain that the connectors are mating properly.

When the back is off you can undo the front screws and

remove the front panel from the case - the main chassis

will come with it. You can now put the case aside and re-

plug the back panel to the chassis for functional tests,

being careful that it does not all fall apart unexpectedly.

2) Break-Out Box. If you use pin G as mentioned above

you will need a spare plug to give access to the LSB/USB

control. These are available at rallies, or the Military

Radio Company (on the net). I do not use the Clansman

handset, preferring a pair of headphones and a fist

microphone. So I have made up a little box with a

Clansman-type plug attached. The box contains a socket

for the fist microphone, a jack socket for the 'phones and

a switch for sideband selection. The box is fixed to the

rig using the existing tapped holes on the left hand side of

the case. I can plug the standard handset into the other

socket if I want to use it.

f3) +1200Hz

LSB CW output (see below)

f4) +1875Hz

Lower cut-off of CW(N) filter (f5 -125Hz)

f5) +2000Hz

Indicated frequency - mid-point of CW(N)

filter.

f6) +2125Hz

Upper cut-off of CW(N) filter (f5 +125Hz)

f7) +2950Hz

Upper cut-off of SSB filter (f2 + 2700Hz)

LSB CIO. (f7 +250Hz = f5 +1200Hz)

f8) +3200Hz

Now, if it is assumed that the CW signal is produced by

inserting a 2kHz audio tone into the SSB mixer the

following behaviour will be expected -

Setting Behaviour

a) USB, CW(W) and CW(N) Output at indicated frequency as

specification.

b) LSB, CW(W)

Output at f8-2000Hz: i.e. f3 or

+1200Hz, This is 800Hz below the

indicated frequency.

c) LSB CW(N)

Same as (b), i.e. +1200Hz, but

this is below the CW(N) lower cut-

off frequency (f4, or +1850Hz) and

is therefore suppressed. No

output expected.

These predictions agree with observed behaviour except

in the details of the precise frequencies (see below -

Calibration), and so the interpretation given here is

confirmed.

Calibration.

The frequency table allows the correct setting of the LSB

CIO to be determined. It is precisely 3,200Hz below the

existing USB CIO. Further, a very convenient way of

setting the new CIO can be found. Set the RT320 to

CW(W), USB. Key the ptt, and measure the output

frequency (on any frequency setting) with a digital

counter. Then switch to CW(W), LSB, and adjust the new

CIO until the output frequency is exactly 800Hz different

from before. Extreme precision in the calibration of the

counter is not required, since it is only the difference in

frequency which is important. As a matter of interest, the

original 'by-ear' technique produced a setting which was

about 20Hz off. The specification of the RT320 states

that the frequency accuracy is 1ppm, so this is a good

way of checking your counter !

Editor’s Note: I have carried out this conversion, and it

works very well! However, pin D and G of the audio plug are

strapped for a good reason – some audio gear has the

earpieces wired separately, one to D and one to G,

apparently so separate comms could be fed to each ear. To

avoid separating the connection, I looked for a simpler way

of providing the LSB switching: Pin 16 of the mainboard,

which is easily accessible, goes to +24 volts when the meter

switch is in the ’batt check’ position and I used this to feed

the relay (other side earthed) to provide LSB switching:

switch to ‘batt chk’ and you have LSB.

RT 320 Frequency Readings

The User Handbook for the RT320 states that the

frequency indication on the decade switches is 2kHz

above the suppressed carrier frequency on USB, is the

same as the carrier frequency on AM, and is equal to the

emitted frequency on CW. The reason for the 2kHz offset

on USB is not known. When the RT320 is modified by

adding a LSB CIO, a different offset (~1.2kHz) becomes

apparent on the LSB setting, and moreover, in CW (W)

mode, the emitted output changes frequency (lower)

10 April 2003

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: