N5FC QRP-L Posting 1 Re: SMK-1 (4-30-2000)
Hi gang! I just night-before-last finished the assembly of my SMK-1, and wanted to report, since mine did not go so smoothly as many of the others. Also, I'll report some performance data taken that will be of interest to many of you. First, the assembly went quite well, and I had all of it done (except connectoring) in 2-1/4 hours after opening the box. I've had a little experience with SMT, so I wasn't put off by it. Assembling it on my kitchen table, using only a temperature controlled (650 deg) 25 W soldering iron with a 3/32" tip, a small pair of needle nose pliers, a jeweler's screwdriver, and some cheap head-mounted magnifiers. I must commend the SMK-1 crew, who did a bang-up job of kitting and instructions. Very innovative, and their technique really made it easy to be successful, and hard to fail. [Red soldermask... not what I see everyday, very unusual... subliminal advertising? hmmmm...] After checking the transmitter and keying, which seemed fine, the receiver was noticed to be rather grossly UN-sensitive, though it was working. I did a little signal tracing with an o-scope and a strong input signal and discovered that the series-mute transistor (Q1 circuit) was the culprit. I placed a short across the transistor, and BOOM, there's my receiver. But I needed to trace back further, to find out whether the mute transistor was being activated by the incoming control signal, or maybe it was itself bad. The level at Q5's base was seen to be high, turning on Q5, and causing the mute circuit to mute. A short clip lead from ground to Q5 base caused the circuit to un-mute, giving me receive audio. Q5 was working. I checked the collector of Q4 (which drives Q5), and it was high (6.1V). Apparently, Q5 was not turned on as expected. Grounding the collector also un-muted the receiver. I checked the base of Q5, which read 0.56 Volts, which seemed like a reasonable value to me to turn ON Q5. Thinking I'd found the problem, I replaced Q5 with a TO-92 package 2N2222, which I had on hand. Hey!, it STILL didn't work. A further examination of the biasing circuit shows that the voltage provided to the base is marginal, based on the voltage divider R15-R16-R17-R18. R18, a 1K resistor, was swamping the base circuit. I removed it, and all worked fine; Because I wasn't sure why R18 was in the circuit in the first place (just what is it's function, anyway?), I replaced it with a 2K chip resistor, which worked just fine, and allowed keying and muting circuits to perform as expected. Here's some performance data I collected : V-sup = 9.0V 12.0V 13.8V _______ ____ _____ _____ I-rx(mA) 14.9 17.5 18.4 I-tx(mA) 64.5 84.8 96.9 P-out(mW) 161 217 345 into 50 ohm F-tx(KHz) 7038.6 - 7039.4 (all) F-rx(KHz) 7035.6 - 7039.7 (all) MDS-rx(uV) 15uV Frequency Stability (transmit, versus Supply voltage): V-sup (V) Freq (KHz) ________ __________ 9 7038.795 10 7038.791 11 7038.784 12 7038.786 13 7038.792 14 7038.800 Frequency Stability (transmit, 12V, versus Time-since-cold-start): T=0 plus Freq (KHz) ________ __________ 2 sec 7038.828 15 7038.818 30 7038.812 60 7038.800 120 7038.784 240 7038.770 480 7038.752 All transmit signals looked sinusoidal on a high-end oscilloscope. Keying looked hard, but not square, and sounded clean, with no noticeable chirp, and no click-y keying. Sidetone sounded fantastic, and at reasonable levels. I had thought the receiver would have a little more sensitivity (15 uV MDS, as measured), but in practice it was more than adequate for the noisy 40 meter band in summer... The front-end attenuator is fantastic (wish I had one on my TenTec). I'm waiting for the cabinet and better band conditions before I put it on the air. My hat's off to NORCAL and the SMK-1 gang, who have done a truly great service for the hobby. Surface mount is here to stay, and we'll all be less fearful of it now... I hope the club will make it available for a long time to come. 73, monty N5FC