PK232-3.... Recently, several new packeteers using PK-232s have appeared on our local duplex repeater which is dedicated to packet radio. This is one environment where the collision frequency should be very low since there are no hidden terminals. Almost immediately we noticed that the collision frequency had risen dramatically. After some investigation we discovered certain stations were almost guaranteed to be involved in stepping on in-progress packets. These stations were contacted and in all but 1 case they were new users of PK-232s. We were very puzzled as there have always been some stations on the repeater using PK-232s with no apparent problems. The new stations were asked how their station was configured and what method was used to get the DCD operation adjusted. We were very surprised at the answers we got. Every single one of the offending stations had set their station up exactly according to the instructions in the PK-232 manual. However, contrary to the advice given in the manual, none of these stations had configured their setup so that they could hear what was going on on the channel when the PK-232 was connected to the radio. None of these new operators knew what DCD meant, what it did, or why it was important that it should be working on a multiple access packet channel. We obtained a PK-232 manual to try and discover the exact nature of the problem. What we found was that although the manual is very complete and generally very well written, there are some areas where it leaves something to be desired. Specifically, in this case, the instructions given on page 2-16 (we had manual PK232UG Rev. B 9/86 for reference) for setting up the PK-232 and an FM radio for DCD operation are simply incorrect. If set up exactly as described, DCD will NEVER be asserted during a packet transmission by another station on the channel!! We had found the cause of our problem. If you have a PK-232 and haven't already discovered this problem for yourself, please disregard the instructions in the manual for settineration and use the method presented here. All your packet neighbors will appreciate it very much. The manual is quite corect in all respects EXCEPT where they discuss setting the DCD THRESHOLD control and receiver audio output level for proper demodulation and DCD circuit operation. The correct way to set these adjustments is as follows: 1. At least temporarily, arrange to be able to hear the receiver audio signal which is being sent to the PK-232. 2. Set the squelch circuit on the radio for normal squelched operation. The DCD circuit in the PK-232 is incapable of proper operation with unsquelched audio from the receiver. 3. While monitoring incoming packets adjust the receiver audio level so that the tuning indicator "spreads" fully when receiving a packet from the station on the channel which produces the LEAST amount of audio output level. There are several limiters in the PK-232 demodulator so louder stations will not be affected adversely by this. 4. Once the audio level is properly set, adjust the DCD THRESHOLD control on the PK-232 so that the DCD led lights when there is a packet being transmitted by the station on channel which produces the LEAST amount of audio output from the receiver. Make sure, however, that the DCD LED is extinguished when the circuit has cut off all audio from the receiver. If the above procedure is followed, the PK-232 will properly hold off transmitting during a packet transmission from another station and will not send acknowledgements to individual frames of a maxframe greater than 1 packet while it is still being transmitted. We hope PK-232 owners will find this information useful and take steps to assure that their DCD is operating properly. Multiple access packet channel throughput is severely degraded when DCD is not working. 73 and happy packeting, Eric Gustafson N7CL -=-=-=-=-=-=-=-=- AEA HAS ANNOUNCED MODIFICATIONS TO THE PK232 TO IMPROVE PERFORMANCE. THESE MODIFICATIONS ARE AS FOLLOWS INSTALL A .047 MICROFARAD CAPACITOR IN SERIES WITH R79. THIS WILL IMPROVE PERFORMANCE ON VHF PACKET WITH SOME ICOM AND KENWOOD TRANSCEIVERS. CHANGE THE VALUE OF R35 TO 39K OHM. THIS MOD WILL IMPROVE PERFORMANCE BY 6DB ON CW MAKING THE SENSITIVITY TWICE AS SENSITIVE. THESE MODIFICATIONS WILL TURN AN ALREADY OUTSTANDING TNC INTO THE BEST THING SINCE SLICED BREAD. 73 AND GL DE STEVE, KF7K =-=-=-=-=-=-= I JUST BEEN DOING SOME WORK ON AEA'S PAKRATT 232. I NOTICED THAT THIS TNC HAS NO INTERNAL REGULATOR TO ISOLATE THE CHIPS FROM THE EXTERNAL POWER SUPPLY THAT THE USER HOOKS UP. I CONNECTED THIS UNIT DIRECTLY TO MY PYRAMID PHASE III RATED FROM 12 TO 15 VOLTS (ADJUSTABLE). THIS WAS CARELESS OF ME, BUT ON THE OTHER HAND I ASSUMED THAT THE PK 232 WAS REGULATED BEING A TNC. WELL NO IT ISN'T AND I WONDERED WHY I HAVEN'T BLOWN THE THING OUT BY ACCIDENT YET. ALSO YOU AS A USER OF THIS TNC MAY NOTICE A HIGH LEVEL OF RFI FROM THIS UNIT. WELL THIS MAYBE THE REASON, MAINLY IF YOU ALSO HAVE YOUR TNC AND RIG CONNECTED TO THE SAME POWER SUPPLY. I SUGGEST THAT YOU PLACE A 12V 1.5 AMP VOLTAGE REG INTERNALLY. AND FEED THE UNIT WITH 13 V. LOTS OF LUCK DAN... WB2ZTK.... -=-=-=-=-=-=-=-=- PK232 Batteries. The battery holder inside of my PK232 had a habit of falling from the top cover via detioration of the adhesieve foam used to secure it in place. I also had concerns about the 3x AA cells leaking with age and possibly damaging the P.C. board. I decided to replace the 3 AA cells and holder and replace them with a Panasonic 3V lithium battery. It appears to be working ok since I have noted no memory loss, even though I went from 4.5V with the AA pack, to 3V with the lithium. the lithium cell used was a panasonic type "BR-1/2A" - O.E.M. part no "P134-ND" this battery is available from Digi-Key. their part no is "P134". you can piggyback the cell (a cylinder .67 in dia.) on top of the gray electro-lytic cap, next to the solder points for the battery on the p.c. Board. I used ordinary scotch tape to secure on top of the cap. (no more battery pack drop-sies!) hope u pk232 owners can use this info if u had/have the same problem! Signed: George, WA2RCB @ N3FOA, Philadelphia, Pa. 19134. -=-=-=-=-=-=-=-=-=- From: G4UXD@GB7OAR Dear Reader! There was such a response to my note regarding "Cures for PK232 Lockup" that I've had to put out a general bulletin! Here is a synopsis of some of the replies I got to my Help! cry: Lenny (the sysop @GB7LIV) reported that he ended up putting 0.01 uF disc ceramic capacitors between ALL connections to the radio (232 also? Derek) and ground. He also fitted a heavy braided earth strap from the chassis of the TNC to TRUE earth (onto the earth chassis of the computer or to an earth point on a 13 amp socket outlet). The problem then stopped TOTALLY. Chris (G6FCI @ GB7FCI) reported a problem with his PK232 locking up when a fluorescent light (and other appliances sometimes) was turned to spikes getting onto the earths in the RS232 between the computer and TNC. Heto solve it and is not quite sure which effected a cure. 1) Stick a decoupling cap between the RTS and CTS lines and earth (he thinks that was where he put them). 2) Strap the signal and frame ground lines together at the TNC and computer end of the RS232 cable. He knows that technically the two are supposed to be separate, but suspects that the problem arises when the signal ground ends up at a different potential from the frame ground. He hasn't had the TNC lock up for 6 months or more. Alex (G1FBY @ GB7BMX) reported that he has had his 232 for two years. One of the first problems he encountered was lockup. All investigations of the unit proved OK. One Sunday his XYL was using the washing machine and dishwasher for long periods of time and the PK232 locked up 6 times. It was run on a fully regulated power supply. He fitted a Radio Spares spike protection socket to the mains plug of the power supply and all has been FB since. ********************** Hope this is useful, 73 de Derek [G4UXD @ GB7OAR] =-=-=-=-=-=-=-= Remove the six screws that hold the cover. It won't be necessary to remove the circuit board to install the state machine. Remove the jumper from JP6. If you wish, drill a hole in the front panel just to the left of the word "THRESHOLD" and above the DCD led. The hole size should be correct for the led that you wish to install. The led anode can be tied to any +5 volt source, and a 1000 ohm 1/4 watt resistor is connected in series with the cathode . The other end of this resistor will be connected to the orange wire. This hookup is optional. Connect the wires as follows: BROWN - (+5 volts) - Connect to U8, pin 14. RED - (Ground) - Connect to U17 pin 7. ORANGE - (LED) - Connect to cathode of installed led via a 1000 ohm resistor (described above). YELLOW - (DCD) - Not connected GREEN - ( -DCD) - Connect to center pin of JP6. BLUE - (Clock) - Connect to U8, pin 13 (74LS393). VIOLET - (CD IN) - Connect to the pin on JP6 that is closest to the front panel. GRAY - (NRZI) - Connect to U15, pin 6. Important: If you have the TAPR Modem Disconnect board installed, follow the directions given in the kit's instructions. Insulate the state machine board with some electrical tape, replace cover and put the screws back in . This completes the installation of the state machine board. Operation: Adjust the audio level from the VHF receiver such that the tuning indicator fully spreads on the station with the weakest audio (lowest deviation) on the channel. Set the THRESHOLD control so that ANY station's audio will light the factory DCD LED, as well as noise from unsquelched audio. The DCD LED should go out when there is no signal (squelched audio or full quieting from a carrier). Note: The DCD modification may affect Morse reception at certain speeds. It seems to not affect AMTOR operation or BAUDOT. It is recommended that an additional LED be mounted on the PK-232 front panel per the instructions given. Proper tuning of signals in all modes will be easier if this is done. It is important that the new LED as well as the old DCD LED be illuminated during MORSE and BAUDOT operation for proper decoding by the PK-232. There it is. These are corrections and rewording to the TAPR instructions that came with the DCD state machine. The last paragraph was taken directly from these instructions. 73 Dave, N2DS @ KC2AZ.NY -=-=-=-=-=-=- It always bothered me that the PK232 draws about 700mA while the KAM draws about half that. Besides the need for mA there's the heat issue. I solved both problems by replacing the CMOS Z80 with a CMOS part. Toshiba makes a very nice version that saves 100mA and is a drop in replacement. I used them with great success in a CP/M system up to 12MHz. I used to buy them from a commercial source but have seen them advertised in Computer Shopper. Too bad AEA doesn't use them. The PK88 could benefit from them too but unfortunately it's is soldered in. -Jim WA2ZKD@WB2PSI -=-=-=-=-==-==- These modifications apply to PK-232s with serial numbers under 1000. Four simple hardware modifications improve the PK-232's performance. 1. Change C54 from: 0.1 mfd 10% mylar (increase low-frequency rolloff) to: 0.01 mfd 10% mylar 2. Change U16 from: 74LS04 (do this only if bbRAM does NOT hold memory) to: 74HC14 3. Change R35 from: 20K 5% (improve VHF/Morse sensitivity)to: 39K 5% 4. Add: 0.047mfd mylar 10% in series with R79 (same as above) -=-=-=-=-=-=-=-=-=- Subject: PK-232 MOD (RF LOCK-UP) Are you having trouble with your PK-232 Locking up (not returning to receive mode) when a high RF signal is present? If so AEA says to add a .01 MFD Capacitor across D30. D30 is located in lower right hand corner of Main Board, parts pictorial page I-1 of operating manual, just to the left of JP2 and JP3 beside SW-2. I hope this helps, it stopped my problems ...... 73 Rick N7KQB @KE7PV -+-+-+ PORTED AND REALIGNED FROM PC TO AMIGA FORMAT BY BOB G0LBQ - N6XDI.