From netcomsv!decwrl!elroy.jpl.nasa.gov!swrinde!emory!kd4nc!ke4zv!gary Tue Sep 21 08:42:20 PDT 1993 Article: 349 of rec.radio.amateur.digital.misc Newsgroups: rec.radio.amateur.digital.misc Path: netcom.com!netcomsv!decwrl!elroy.jpl.nasa.gov!swrinde!emory!kd4nc!ke4zv!gary From: gary@ke4zv.atl.ga.us (Gary Coffman) Subject: Re: 2-m homebrew packet rig? Message-ID: <1993Sep16.032150.10590@ke4zv.atl.ga.us> Reply-To: gary@ke4zv.UUCP (Gary Coffman) Organization: Destructive Testing Systems References: <1993Sep12.134230.19253@ke4zv.atl.ga.us> <1993Sep14.160053.26686@newsgate.sps.mot.com> Date: Thu, 16 Sep 1993 03:21:50 GMT Lines: 99 In article sbrown@charon.dseg.ti.com (Steve Brown) writes: >In article <1993Sep14.160053.26686@newsgate.sps.mot.com> Chris Terwilliger writes: >> In article <1993Sep12.134230.19253@ke4zv.atl.ga.us> Gary Coffman, >> gary@ke4zv.atl.ga.us writes: >> >> >For 9600 baud, the Micor is fairly trivial to convert, >> >> please elaborate... > >Yes, please elaborate _on the net_. I tried to send you a message every >way I could think of and got kickbacks. Well (aaarrrgg) my Motorola manual is out on loan so I can't give you a part number by part number mod sheet. It's basically simple though. You just disconnect the audio amp from the modulator and connect your TNC output instead, and you disconnect the audio amp from the discriminator and connect your TNC instead. It can't get much simpler than that. Since I *do* have my GE manual handy, I'll give you the method for making a UHF Exec II or MVP work on 9600 baud. It isn't so simple. On the limiter detector board, U602 is the limiter detector IC. Pin 1 is the audio output, but we don't want to take our signal directly from there, we'd load it too much. Instead we take it at J601 pin 2. That's isolated by a 180k resistor from U602. Now we'll only have 0.2 volts of signal here, and that isn't enough to drive our TNC. We could try to take the signal from the emitter of Q603 but we'd have bandlimited de-emphasized audio, not good. So we outboard an OPamp and set it for a gain of 5, just use a 900k feedback resistor in conjunction with the 180k resistor already in the radio. This will give us a volt to the TNC which will make it happy. Note that the signal at J601-2 is offset +0.35 volt DC so we have to null that voltage with a bias pot in the non-inverting input to our OPamp. All is not well yet, however. The radio uses a 4 pole crystal filter in the IF which is too narrow for our purposes at +/- 7 kHz. We've got to widen this a bit. There are several ways to approach this problem, but the easiest is to solder a gimmick capacitor across pins 1 and 3 of FL502 and twist for the desired response. FL502 is the filter nearest J502, the +10 volt power connection. Solder two 1 inch insulated hookup wires, one to pin 1 and the other to pin 3. Now twist these tightly together for the desired response, about +/- 10 kHz response. This is easiest done with a sweeper. You may need to touch up C504 and L503 for flattest response. Cut off any excess lead length on the gimmick and put a touch of glypt on it to hold it in place. Now the easy part is over. We have to modulate this thing with our 9600 baud signal. The modulator is a classic variable reactance phase modulator. That won't do for our purposes so we're going to turn the radio into an FM modulated transmitter. First we're going to operate on our Icom (channel element) by removing C1. That's the square green bypass capacitor just below the piston trimmer. The varactor in the Icom is too small to give us the swing we need, so we rob the varactor from the phase modulator, CR180, and parallel it across the varactor in the Icom. Now we have to cut the trace going to XY101-4, the Icom socket on the exciter board. Be sure when you cut the trace that you cut *after* the connection to P902-3 or you'll lose your receiver temperature compensation. Ok, now we need about 2 volts of signal to drive the varactor in the Icom, and we want to keep our compensation voltage too which will be in the 4.0-6.0 volt range depending on temperature. To do that, we're going to use an OPamp as a voltage summer. Feed the TNC audio, 1 volt, and the compensation voltage to the inverting input of the OPamp. Take the audio through a 100k resistor and take the compensation voltage via a 400k resistor connected before the trace cut. Use a 50k feedback resistor. Use another section of the OPamp as an inverting unity gain amp with a 2.5 volt positive offset in the non-inverting input. This gets the compensation voltage going back in the right direction and biases the varactors properly. You should now have a 4.5-5.5 volt positive DC output with a 2 volt symetrical audio signal riding on it. Note that we've halved the compensation voltage swing because we've doubled the amount of varactor. Now connect the output of the OPamp to the XY101 side of the trace cut. You should be able to get 4 kHz deviation. There *is* another way, but it puts pre-emphasis on the signal. That's to just accept the phase modulator and put the audio in on the Channel Guard input, P902-9. It takes a *lot* of signal though to get 4 kHz deviation. You'll still need the OPamp. If you do this, you'll want to add a de-emphasis network in the receive section. You can do this by adding a capacitor in the feedback section of the receive OPamp to give a 6 db per octave roll off. Don't try to use the speaker output. It's bandlimited as well as de-emphasized. That won't work well at all. The modem operation will be compromised, but may work if others in your area are using pre-emphasis/de-emphasis in their radios. I don't recomend it, but it is easy to do. Gary -- Gary Coffman KE4ZV |"If 10% is good enough | gatech!wa4mei!ke4zv!gary Destructive Testing Systems | for Jesus, it's good | uunet!rsiatl!ke4zv!gary 534 Shannon Way | enough for Uncle Sam."| emory!kd4nc!ke4zv!gary Lawrenceville, GA 30244 | -Ray Stevens |