TKR-820 Repeater - BART - Border Area Repeater Team

Checkout the BART on Yahoo !! THE 448.35 PROJECT Pictures taken on Mt. Lemmon are at the bottom of the page. <click here>
Note: The modifications mentioned below have been posted on the http://www.mods.dk pages under Kenwood. DPL Mod KEY Mod Line-Out Mod
September through November 2006
Sometime in September 2006, Gary - N1DHS suggested that he would like to replace his Mt. Lemmon 448.350 General Electric antique repeater with a little more modern one. He had a Kenwood TKR-820 repeater in service on 449.80 for several years, so it was chosen to be the candidate for the job. He presented the machine to Craig - KD7TXO and myself and we proceeded to figure out what needed to be done. You'd think we should know what we're doing, right? Right ! Read on.
<Click Pictures to Enlarge>
This is what it looked like after I put a label or two on it to identify the machine and show what position the switches needed to be in. It was later decided that a shield would be placed over the switches on the right with access holes to prevent the inadvertent changing when it's on the mountain. Dave - K7IOU made a bracket, however, I didn't get a good shot if it. Just use your imagination. Here's the manual. Next, it was apparent that the CAT- 300 Controller interface was not quite up to what we were planning. First priority was to suppress the transmitted PL when there was no input signal to the receiver. This would prevent the IDs, hang times and beeps from coming over the link. We had been using an ICOM-2720 for the 224.74 to 448.35 link and it had Digital Private Line capability, so we decided to control the DPL encoding of the repeater. My first crack at controlling the DPL was to install a relay in the TONE output of the repeater Signaling Unit (X57-3140-010).

This proved the concept would work. The relay power was connected to the 12.8 VDC input power to the Signaling Unit available on CN1-1. The other side of the relay coil was connected to the COR signal on CN1-7. When COR is active, it energized and opened the path on the TONE output on CN2-1. <SCHEMATIC> One concern was that when using this technique to cut off the DPL encoding it would remove the short PL tone burst out of the Signaling Unit after the DPL dropped. Gary was concerned that it might cause the Digital Controlled Squelch in his Motorola rigs to act up. Turns out it didn't make any difference. That was a GOOD thing !!
It turned out that the reed relay was not so reliable. I swear that it would miss energizing sometimes. So, I moved on to a different relay. I used a Radio Shack SPST relay that was suppose to have a 100 K life cycle under full load and a 10 M life cycle under no load. Here's the installation of that one. The schematic is the same, just the names have been change to protect the innocent.

Guess what ? This one would skip a beat once in a while too. I was pulling out the remaining hair on my head by this time. Not only that, we discovered that the ICOM-2720, when using the DPL Encoding/Decoding, worked just great when going into the repeater, but when the DPL dropped on the output of the repeater, the COR signal from the ICOM wouldn't drop until the carrier was gone from the repeater. What a bummer !! Dave kept telling me it was the Kenwood that was causing the problem, but I was sure it was the ICOM. Turns out I was right...ha ha ! I proceeded by investigating the CAT- 300 controller and tried to figure out a couple of serious deficiencies. First, there was no way to turn off the repeater via DTMF. You could turn off the controller, but the Kenwood would revert to it's own simple repeater action. That wouldn't do at all ! Also, a method to keep the repeater cool would probably be necessary so what kind of fan could we use and how could it be controller. It wouldn't be good to leave the fan running all the time, would it?
January - February 2007
After a couple dozen tries at various methods of repeater ON/OFF control, fan control, DPL encoding control from the controller plus fighting the thought of using relays for any of this, I came up with the following: First the fan. I ended up using a computer ball-bearing 12 VDC fan used in most of the CPU cases. You can see it setting on the back of the rack mount and pointing at the heat sink of the transmitter. We discovered that when you use DC power as opposed to 115 VAC, there is very little heat generated. That power supply in the Kenwood generated enough heat that when you run it on AC, the fan would come on often. Oh yeah, I forgot to tell you....click on the picture and you can see a group of wires on the black heat sink. Craig - KD7TXO and I installed a bi-metal strip thermal switch on the heat sink that closes at 100 degrees F. Also, you can turn the fan on and off using DTMF commands because one of the user outputs of the controller was harnessed to provide a ground to the fan when you want. You can see the AC power cord coiled up, unused. Also, the red and black DC power lines are there. There is battery backup for the entire system in that Forest Service shack. Pretty neat ! And, I had to modify the Kenwood to accept DC power. Here's a picture of that: If you look close, you can see the bridge rectifier installed in the bottom of the case. Had to add a power jack on the rear of the repeater case, but it was the easiest part of the whole project. You can see the power jack just below the blue adjustment pots. There is a Kenwood kit for that mod, but I just followed the schematic and did the modification. It's much cheaper that way....even if you can still buy the kit.
I had several sleepless nights trying to figure a better method of controlling the DPL without using a relay. After a few internet searches, I found what I figured would do the trick. A "solid state relay", or more correctly, a MOS/FET analog switch. I found it in the Omron product line sold by Mouser Electronics for about $3. I ordered two of 'em and came up with this: Click on it to see the thing of beauty. I used a Radio Shack small project board and mounted it as shown with a single screw. The 8 pin DIP analog switch is in a PC board IC socket and there are three resistors and a diode. Click <here> to see the schematic of the whole operation. It will control the DPL by closing the path of the TONE out of the Signaling Unit when the receiver is busy. It turns on in less than 0.3 msec and turns off in less than 0.1 msec. Fast enough. It doesn't need power because the thing has a LED activated MOS/FET switch that acts just like relay contacts. AND, it will allow the repeater ON/OFF in the other half of the analog switch that's in the same DIP. It interrupts the KEY out of the Signaling Unit similar to the DPL control. You can even DTMF and let the DPL go back to normal operation if you want. Pretty cool, if I do say so myself. I'm happy !!
March 2007
By this time, we had a date to replace the repeater on 12 March. I can't tell you how many times that Craig, Dave & I fiddled with the audio trying to make it sound half way decent. Man, it SUCKED ! The standard method is to use discriminator output from the received of a repeater and then, in the controller, use a simple capacitor feedback de-emphasis circuit as part of the receiver audio amplifier. Craig and I must have tried ten different capacitors to try and get the audio to roll off at the -6dB per octave curve. No way...it just SUCKED. We found a big chunk of the problem was caused by a squelch module that either was put in incorrectly or there was a broken circuit trace. We found that both the discriminator audio input and the transmitter output had been conditioned with RC networks. Also, it appeared that it was not operating properly because a little red LED that was suppose to be OFF when there was no COR shined dimly all the time. So, Craig and I removed the whole kit-n-caboodle. Then we started trying the capacitors in the controller. It was much better, but it still SUCKED ! And here it is ONE WEEK TO GO ! HOLY SMOKES !!!!
Monday Morning, 7 March 2007, 4:00 in the A.M.
I says, "That's it !! I am getting up and I'm going to go get that damned repeater audio straightened out or it ain't going up on the mountain !", says I ! I retrieved the repeater from Craig's QTH and tore it to pieces. I had noticed a LINE-OUT audio on the accessory jack that probably had proper de-emphasis. Heck, it had several stages of amplification and filtering and I figured that it must be a better method than a single feedback capacitor. I patched it in and it sounded pretty good. I installed the squelch module in and played around with it with no success. Dave - K7IOU came over and we played around with the squelch module some more using the line-out and the discriminator audio and ripped it out again. The line-out was sounding pretty good, only thing was.... IT WAS SWITCHED ON & OFF BY COR !!! DAMN !!!! YOU SHOULD HAVE HEARD THAT CHIRP ON THE SQUELCH !
Tuesday, 6 March 2007
Dave and I spent a couple hours Tuesday afternoon trying to figure out how to let that line-out pass through all the time, the same as discriminator audio. Dave says he can't see too good, but it was his eyes that helped me find the R45 and other components on the display panel and even remove those tiny surface mount resistors and a transistor, however, removing the components didn't seem to matter. He reinstalled the parts and had to go home. I took a ground wire and started prodding around in there and found how to turn it on continuously. VOILLA !!
Wednesday, 7 March
From there it was just a matter of time until I had the thing put back together and over at Craig's by Wednesday afternoon.
Thursday, 8 March
W8GRI - Jim told me it was still pretty sharp, I put in a 0.0047 uFD capacitor in the receiver amp in the controller. It' better.
Friday, 9 March
Gary - N1DHS says, "The Voice ID sure seems sharp !". I thought about it a while and slapped myself in the forehead and went over to Craig's and moved the capacitor to the transmitter audio output amplifier. Why didn't I think of that before?
Saturday, 10 March
Saturday morning after the breakfast, Dave and I hooked up a GM-300 donated by N1DHS to replace the delayed drop-out ICOM-2720. It works pretty darned good and the COR drops immediately. No more hang time on the 224.74 and the 146.94. There's still about a half second delay for the 448.35 to come up. We'll work on that.
Sunday, 11 March
After listening to it Saturday, I went over on Sunday and changed the capacitor to 0.002 uFD and said to myself "FINALLY ! That's the best I have ever head the audio on this repeater !" Talk about waiting until the last minute !
Monday, 12 March That's me smiling on the left ! I am a happy camper ! <click the photos to enlarge>
Below are the photos taken up on Mt. Lemmon. You can see the patch of ice that defeated Gary's Tundra (sans 4-wheel drive) and the damage resulting from the big fire. Also, there are a few taken on another site on Lemmon where the Cactus guys have their shack and equipment. All in all, it was a very easy install !