Jaws Mk2 CB - Conversion to 10M
You may ask why.. Well, it looked an interesting little project and a chance to get the very radio that I first used in 1982 back on the air. Poacher turned game keeper..
After reading the excellent T4TT articles in PW I made various enquires to Spectrum communications re conversion ideas. Basically a new 17.055Mhz signal is injected to move the top 10 channels up to 10M. This gives 10 10Khz spaced AM channels from 29.00Mhz to 29.100Mhz - I bet that's a busy place. I also ordered a copy of the circuit diagram as I expect it will be more involved than it first appears.
Following more Ebay madness I now have 3 Jaws MK 2's to convert. Well at least there will be two other local stations.. Now all we need are some FireStiks..
From Spectrum Communications
"The Jaws II uses the circuit of figure 1 of Practical Wireless
T4TT August issue. Only that it probably doesn't have the
Crystal Oscillator board for L & H.
Output from the divide-by-two is on pin 13 of the LC7120. This is coupled to a parallel tuned circuit C229 and L202 via C207, 5pF. There is also a 1K resistor to ground R329 across the tuned circuit.
Remove C207, C229 & L202.
Make up a Colpitts oscillator with a 17.055MHz crystal and inject it at the joint of C207 and C229.
Retune the rig for channel 30 = 29.00MHz, and channel 40 = 29.10MHz."
A Colpitts oscillator PCB and component kit, type SC29F, with low level output is available (without crystal) at £5.00. A3 size circuit of the Jaws II will cost £2.00.
The FM rigs listed can all be modified using an SC29 to give
29.30 - 29.70MHz. Additionally an RS10L repeater shift board can
be added for the required offset.
SC29 £23.00. RS10L £3.50. P&P inclusive.
To start off I ordered -
For the Jaws 2
A Colpitts oscillator PCB and component kit, type SC29F
A3 size circuit of the Jaws II £2.00.
For the Cybernet 1000 FM
FM rig SC29 £23.00
RS10L repeater shift board £3.50
10M AM – I tested both Jaws 2's on RX and TX. One has a scratchy volume control, but that should be easy to fix. Next step is to get the circuit diagram and oscillator board from Spectrum Communications.
CB – Found a very useful CB PLL Simulator that helped me understand the JAWS MK2 PLL system.
Ordered the crystals from QuartzLab.
10M – The crystals finally arrive! The Elves must have been really busy or perhaps they had just been drinking too much Guinness?
Soldered in the crystal and tested the board. Tunes across a wide range, but does not quite go down to 17.0550 The cap in series with the trimmer may need to be changed.
Checked the junk box and I have almost all the components required for the other two oscillator boards. I intend to use plain perf board and point to point wiring for their construction.
The kit oscillator works with the 17.055 crystal but is
slightly high in frequency.
I used perfboard and point to point wiring to build another version of the circuit for the other rig. It uses a 2n3904 instead of the BF199. This caused a little confusion with the pin out as B and E are swapped. The frequency was slightly high, probably because the circuit was designed for 17.065. The addition of some capacitance bought it down, but the tops of the sine waves were still a bit jagged. We want a nice smooth sine wave!
The 2n3904 transistor has a lot more gain than the original BF199. I ended up using a 2nf capacitor. The output is now a nice sine wave - 540mv PTP. I also increased the padding cap for the tuning so it sits nicely on 17.055Mhz.
Tuning Notes from Spectrum Communications
"There is no tuning procedure written up for all these rigs. Simply find the main synth varicap and the resistor feeding it with control voltage. Connect your multimeter probe to the decoupled end of that resistor. Set the channel change to CH1, adjust the VCO core and watch it achieve lock, and swing the control voltage to max and min. Temporarily set it about mid range, and then switch to CH40, see where it goes. Go to transmit and see where it goes. Then adjust the VCO core so that it doesn't hit the max or min level between CH1 and CH40 on TX and RX. The synth is then set up.
Now peak the transmit chain by rotating the TOKO coils clockwise. Large coils with big cores at the driver and output stage rotate anticlockwise.
Receive TOKO coils tune clockwise for maximum receive signal."
Jaws MK2A - Found pin 13 on the LC7131 PLL chip reference freq / 2. Scoped it and found no signal, it is grounded! The Jaws MK2A is certainly different from the MK2. It is not yet clear if this will make conversion of this rig easier, harder or impossible.
Back to the Jaws MK2 - Removed the trippler cap and inductor. Connected the new 17.055Mhz oscillator but nothing seemed to make much sense.
Reviewed the PLL datasheet and spent a while just probing
about. Reinstalled the old trippler and got better earth
connections - not all the chasis is "earthed". Pin 2 on TA 7310P
then showed sensible steps through the channel selector again.
Reinstalled the 17.055Mhz oscillator. Pin 2 on TA7310P now fixed at 11ish Mhz. Found the VCO and measured the control voltage on resistor feeding the varicap. Winding the VCO core in and out altered the control voltage between .5 and 9v, changing channels also altered the voltage - as expected. But the lock pin on the PLL would not go high and the frequency on Pin 2 of TA7310P remained around 11Mhz. Some progress, but further reading is required. Getting late.
Studied the documentation. Revisited the CB simulator to further aid my understanding. Scoped the downmix into the PLL but could not make much sense of it, infact was I even looking at the right pin.
The CBSIM diagram could do with some arrows to indicate directions. But it does show the basic ideas. Here it is on receive using the new offset frequency of 17.055Mhz and set to channel 30, the first 10m channel.
Now, all I had to do was to get the radio to follow this.. Went back to trying to scope the down mix, but still no progress. Decided to go back a step and reinstall the old trippler and take a look at the output from the VCO. With some winding about of the VCO core the channels settled back to their old frequency and spacing. Scoping pin 1 of the 73010P mixer seemed to give the most reliable readings of the VCO output. Still no lock indication on the PLL.
Encouraged I removed the old trippler components and reinstalled the 17.055Mhz oscillator, after more winding of the VCO core with it "locking" on various intermediate frequencys it alighted on 18.3050 - Some sucess at last. Still no PLL lock and the core is wound almost all the way out. But it steps through the top 10 channels.
Wound the VCO core almost fully in setting channel 30 to 18.3050Mhz. Steps through the top 10 channels correctly, but still no PLL lock indicator. Further investigation shows pin 15 the PLL lock indicator is held high by pin 7 and 9. Pin 16 PLL (Lock high) is used for transmit inhibit. The VCO voltage measurement still remains around 6.7 volts regardless of channel or VCO core position, which does not make sense.
The VCO control voltage on pin 19 of the LC7120 changes as the channels are stepped. Checked the spec for pin 16 PLL lock high. High is only .9 volts, which explains things. Wound the VCO coil and watched it go in and out of lock. Excellent.
For next time - Check the VCO voltage range on RX and TX, consider changing the oscillators capacitors to avoid needing a 2nf and thus improve stablity, tidy up and install the oscillator circuit, start the RX and TX tuneup.
Connected up the SWR meter and dummy load. Rechecked the VCO and ensured the PLL pin 16 went high to 800mv on RX. Keyed the TX, the PLL lock pin goes to 4.89 volts? No RF output detected, which could be down to the TX chain tunning? But it seems strange there is no output at all. No modulation lights on the rig might indicate that TX is still inhibited. Drawing about 250 milli amps. Clearly the PA is not doing much..
You can almost tell what channel is set by measuring the current drawn.. Scoped the connections on the mixer again and found very variable results, reflowed the solder on all the mixer pins and got better connections..
Pin 1 - VCO
Pin 2 - PLL input
Pin 3 - RX?
Pin 4 - RX
Pin 5 - Grounded
Pin 6 - TX
Pin 7 - TX
Pin 8 - Power
Pin 9 - TX
Leaving the RX to one side for the moment, pins 6,7,9 all
basically go to the start of the TX chain L301 and L302. When
keyed no sensible signal can be seen before the first transistor
amp Q301. After Q301 there is a low level rough 25ish MHZ
signal, this can be modulated slightly by speaking into the
mike. This increases the current draw as the audio amp operates.
Next questions - Is the carrier output getting sunk by L301 and L302 before reaching the first amp? What is happening with the outputs from the balanced mixer?
Used the scopes frequency counter to adjust the signal generator.. Then for the first time the rig received 29.00 Mhz - Channel 30. Tried TX - Pin 9 on mixer IC203 (TA7310P) should be the 29.00Mhz output signal, but nothing to be seen. The power rail for IC203 pin 8 during TX was 4.8volts, which seemed low?
Retuned the VCO coil again for the exercise. Watched PLL pin 16 (high level lock) going sharply to 800mv for lock on RX and 4.95v for lock on TX. Mixer IC203 (TA7310P) pin 1 gave 18.3050Mhz on RX and 18.7600Mhz on TX for Channel 30. Which is correct. No output on pin 9 of the mixer. Heard the sign gen again on 29.00Mhz Channel 30.
Pins 3 and 4 of the mixer could not be usefully checked, as at this point the core of the VCO broke. But it was receiving prior to that..
Got some very helpful responses from the CB2HAM yahoo group, including scans of the circuit diagram and service notes. Thanks!
VCO - Followed the service notes to get 5v on channel 40 in TX and about 1.4 volts on channel 1 in RX. Channel one is a bit high, the notes suggest up to 1 volt. But it is close?
IC 203 - The RX and TX frequencies measured on pin 1 are correct, but on the scope they certainly show some jitter compared to the reference oscillator. Is it worth looking at the loop filter r203 and c206 (I think)?
TX - 1.5v dc on Base of Q301, 800mv on Emitter and 2.5 on Collector. DC volts are about right but no sine wave on IC203 Pin 9.
I have not yet revisited the TX inhibit circuit.
Removed R203 and injected a variable voltage into the Varicap diode. Tweaking the voltage to around 5 volts gave 18.30500 on RX Channel 30 and reducing the voltage to about 1.5 volts gave 18.7600 which is correct for TX.
Interesting as this could not be achieved when the PLL was connected up.
Selecting RX or TX made no difference to the frequency - only the voltage altered it. The PLL would not be indicating LOCK so TX would certainly be inhibited.
As expected the voltage required very careful tweaking and the frequency did vary slightly..
Interesting to see the frequency changing as the voltage was varied.
As Greg from the CB2TEN Yahoo group had suggested I turned my attention back to the VCO
Disconnecting the VCO from the PLL and injecting 0-6.5v at
TP1, I can swing Pin 2 of the
7310 mixer across a part of the required range, but not all of it.
What is the function of C211? I have reduced its value and it
does increase the
max frequency. But seems to have a greater effect on the frequency "spread"?
What is the cleanest way to increase the VCO operating frequency?
Too many questions and it is getting late.
27/03/09 - A Happy VCO
From all this experimentation with the VCO it was clear that it did not really like working at the new frequencies, it was always a bit of a tweek and push, never quite enough range and with control voltages always on the limits. The basic frequency of the VCO needed to be increased.
I decided to take the VCO coil apart and either reduce the number of windings or the capacitance. The coil has seven connectors soldered to the main board. Using a solder sucker these were soon cleared of solder and the coil removed. A little prising with a small screwdriver to remove the coil's can and the mystery of the seven connectors was revealed. Two tabs are for the can's earth, two pins are unconnected, one pin for each end of the coil and a pin for the centre tap - which is not used. The tiny capacitor to one side of the coil broke while I was attempting to remove it. I was hoping to measure it, but never mind.
After cleaning everything up I fitted the coil back onto the board, soldered in the two coil pins and one of the can connections. Then added a small 1-28pf trimmer across the coil pins, setting it to a few pf and finally centred the coil slug.
Powered everything up and measured Pin 2 output of the mixer while clicking through all the channels - perfect lock on RX and TX across the whole range and the signal looked cleaner. Excellent!
30/03/09 - 29Mhz TX
Took a closer look at the PLL Lock and Inverted Lock signals The inverted appears to be strapped high by R205. Removed R205 and Inverted Lock goes low on lock. Replaced R205. The PLL was certainly working correctly. Had a closer look at the T/R and TX inhibit but it ran everywhere. Pin 14 the down mix input of the PLL still did not show on the scope which was very strange as the PLL was working. Still no output on pin 9 the TX output of the mixer.
Revisited the RX tune up procedure and carefully tweaked up L204, a single turn went from almost nothing on RX to S9. As previously suspected (16/12/08), perhaps the TX was getting squashed too? Carefully tweaked up L301 and L302 - 29Mhz finally appeared on the scope.
As I scoped pin 9 of the mixer the TX sine wave collapsed - the scope was loading it too much. The act of measuring destroyed what was being measured, which certainly caused confusion.
01/04/09 - Where's the Mod good buddy?
Tuned up the TX stage then tried the mic on 29Mhz but only picked up hum on the 817. The TX light also failed to light - no modulation. Further investigation showed the microphone PTT switch to be unreliable, swapped it for another Jaws 2 mic and it worked perfectly.
Next tasks are to finally tweak the TX/RX and fit everything back in the case, instead of it all hanging off the board.
Then... on to the JAWS MK2A with the unmodifiable 7131 PLL.
Revisted the VCO control voltages, over 6 volts on channel 40 TX and 3.5 volts on channel 1 RX, plus the mixer TX output is now looking very ragged. Ideally the control voltage should be < 1 volt on channel 1 RX and 5 volts on channel 40 TX. The VCO "spread" capacitor C211 may need to be changed.
C211 = 62pf 1 RX 2.73 volts 40 TX 4.81 volts
C211 = 39pf 1 RX 2.37 volts 40 TX 5.08 volts
C211 = 22pf 1 RX .8 volt 40 TX 4.41 volts
40 TX still very ragged. Removed the link to the PLL and fed in 5 volts.
40 TX VCO 18.860 - 5 volts
1 TX VCO 18.420 - .3 to 1 volt
I think RF was getting fed back into the VCO, perhaps through the scope leads.. as it only goes ragged when it locks.
With a 22pf cap for C211 the VCO voltages were broadly on track.
RF Driver stage tune up
Q301 - 640mv Peak to Peak - clean sine wave.
Q302 - 3.7v Peak to Peak - twin topped sine wave.
Q302 - 2.0v Peak to Peak - clean sine wave.
Peaking of the final stages was done by observing a CB SWR meter, as no suitable power meter was available. As expected without drive the VCO produced a clean TX sine wave.
Reworked the area around the PLL chip to provide some space for the oscillator board and connected up the power. Refitted the caps and replaced the temporary variable cap with an 8pf one. Checked everything before refitting the case. A quick transmitting test into a dummy load sounded fine.
Still to be done -
Properly measure the output power, as the SWR meter does not
give much of a guide.
Ensure the RX chain is fully peaked.
Check the modulation level.
But it is time for a test QSO with M1ECY - Sean.
11/04/09 - Testing
Power Output - For a basic power check I added a diode and capacitor between the "Hot" end of the dummy load and earth then measured the voltage across the capacitor.
Peak voltage measured across the capacitor (excluding diode
RMS voltage equals the peak voltage divided by the square root
19.5 / 1.414 = 13.79 Volts.
Power equals the square of the RMS voltage divided by the load
(13.79 * 13.79) / 50 = 3.8 Watts.
Testing Time - In the best traditions of CB I checked the "SWR" of the 10m Hamstick - as no Firestik was to hand.. then wired up the Jaws 2 ready for it's first QSO. The LED display glowed brightly, perhaps with expectation? A moments pause, then I rang Sean to ask for a QSO...
I could hear him calling but the signal was quite weak, although we could just maintain contact. We were quickly reminded that our antennas were cross polarised, but we did some useful tests.
TX and RX frequencies were spot on.
RX was working at least reasonable well.
Audio was very slightly muffled, or at least missing some of the high frequencies - the microphone is 30 years old?
Perhaps the modulation level was low, but hard to tell without a stronger carrier and a way to measure the modulation.
A very successful test. Thanks Sean.
12/04/09 - Further Testing
Tried another mic which seemed to have slightly better on air audio, although still lacked drive.
Once back on the bench the mic socket started playing up again, producing funny noises and cutting out as it was touched. I cleaned it as an interim measure and things seemed to settle down. This will need to be revisited.
Mic element dimensions - 30mm diameter across the top, 27mm diameter across the middle, 10.5mm deep
Had another go at retuning the RX stages. Without a reliable signal generator it was all rather hit and miss, but it seemed to be picking up the signal generator at broadly similar levels to the 817. It was certainly not completely deaf.
19/04/09 - Dummy Load
Modified my Dummy Load so it can easily be connected to the scope - at QRP power.
Connected up the scope and played some tones into the mic. On the scope at any level above quiet speech the frequency changed. The unmodulated carrier appeared clean and at the correct frequency.
Last week I built a USB Synth kit. I plan to use it as a sig gen and to investigate its suitablity for a future, darker purpose...
25/05/09 - 2nd Jaws Mk2
I did a quick test of my 2nd Mk2 and it seems to be working OK
– tricky to tell though, as I don't think these rigs work well
with a 30 year old mic element. Built another 17.055 oscillator,
copying my original one was so much easier than doing
it all from scratch. It worked fine.
Did some basic tests on the rig and went straight ahead with the first part of the conversion - removed the tripler, temporarily installed the new reference oscillator and reworked the VCO coil. Selecting channel 30 now gives 18.3050Mhz on pin 1 of the mixer. Copying the first converted Jaws Mk 2 certainly made the process easier. Next tuning up of the RX and TX stages and then revisit the modulation and power measurements.
30/05/09 - PSU issues
Started testing and tuning up the RX of the second Jaws MK2. The RX was very noisy. I puzzled over this for a while, until I realised the oscillator was running on a separate PSU. As soon as I connected it to the rigs supply the noise disappeared.
1. Understand, test and measure the donor rig throughly before
taking it apart.
2. Rig up a PLL lock LED to indicate if the PLL has "lost grip".
3. Try and avoid getting bogged down in why one detail does not work. It will still be there later.
4. Even a 10x probe can squash things.
5. 4 watts of RF will get everywhere.
6. Highlighted the need for a proper power meter and signal generator, although you can succesfully improvise.
Jaws 2 Information
Scratchy Volume Controls - Greg OH2FFY
Explanation of Jaws PLL operation - Greg OH2FFY
LC7120 - Jaws 2 PLL
TA 7310P - Mixer for PLL
TA 7205AP - Audio Power Amp
LC7131 - Jaws 2A PLL
CB2HAM Yahoo Group - Lots of helpful advice, schematics for the MK 2 / MK 2a and a host of other conversion information. Thanks Greg.