From owner-qrp-l@Lehigh.EDU Mon Oct 6 20:39:44 1997 Received: from fidoii.CC.lehigh.EDU (fidoii.CC.lehigh.EDU [128.180.1.4]) by oucsace.cs.ohiou.edu (8.8.5/8.8.5) with ESMTP id UAA29952 for ; Mon, 6 Oct 1997 20:39:43 -0400 (EDT) Received: from Lehigh.EDU ([127.0.0.1]) by fidoii.cc.Lehigh.EDU with SMTP id <34914-35114>; Mon, 6 Oct 1997 20:38:53 -0400 Received: from nss2.CC.Lehigh.EDU ([128.180.1.26]) by fidoii.cc.Lehigh.EDU with ESMTP id <35055-35114>; Mon, 6 Oct 1997 20:24:07 -0400 Received: from m7.boston.juno.com (m7.boston.juno.com [205.231.100.196]) by nss2.CC.Lehigh.EDU (8.8.5/8.8.5) with ESMTP id UAA176538 for ; Mon, 6 Oct 1997 20:22:01 -0400 Received: (from gsurrency@juno.com) by m7.boston.juno.com (queuemail) id U]N01180; Mon, 06 Oct 1997 20:20:07 EDT Message-Id: <19971006.171710.6566.6.gsurrency@juno.com> Date: Mon, 6 Oct 1997 17:16:54 -0700 Reply-To: gsurrency@juno.com Sender: owner-qrp-l@Lehigh.EDU Precedence: bulk From: gsurrency@juno.com (Gary L L Surrency) To: "Low Power Amateur Radio Discussion" Subject: HW-9: The (nearly) complete guide to the mods. Part 2 of 3 X-Mailer: Juno 1.38 X-Juno-Line-Breaks: 0-5,14-15,22-25,31-32,38-39,47-48,51-52,62-65,72-73, 82-83,89-90,98-99,106-107,115-116,120-121,125-126,134-137, 144-145,155-156,161-162,171-178 X-Listprocessor-Version: 8.1 beta -- ListProcessor(tm) by CREN Status: RO ******************************************************************* HIT DELETE NOW IF NOT INTERESTED IN HW-9 MODS ******************************************************************** AGC MODS The AGC is too fast for my tastes, and the S-meter tends to beat itself to death on CW signals as it jumps up and down on every CW character. There are a couple of ways to slow the AGC and S-meter operation. Either decrease the value of C317, a 3.3 uf electrolytic; or increase the value of R312. Or, do both! I decided to simply remove R312, a 47k 1/4 watt resistor that is the primarly discharge path for C317. This slows the AGC and S-meter quite a bit, and is easy to do. C317 then discharges through R311 and R309, as well as R316 and IF amplifier U301. Some discharge also occurs by way of D306 and the S-meter FET amp. While trying several caps for C317, I found a 33 uf cap produced about what I wanted for the AGC time constant / S-meter behavior. But I concluded such a large value of capacitance might load the AGC output level of U302, causing a loss of proper attack time and AGC peak voltage level. I therefore decided on removing R312, which produces about the same affect on AGC time constant with no additional loading of the AGC circuit. TRANSMIT / RECEIVE TRANSITION IMPROVEMENTS Much of the improvement of the T/R switching is achieved by the mods to the keying circuit mentioned earlier. However, some additional improvements can still be had by addressing the receiver muting. Transistor Q303 is a MPSA20 (Heath pn 417-801) general purpose device. It lacks the low ON saturation voltage and switching characteristics needed in muting the audio input to U306, the AF power output amplifier. Using Molex pins again, I tried a large number of NPN transistors for Q303. Most of the high gain, low VsatON devices performed better muting of the audio line and helped reduce the small T/R pops during fast QSK settings of the Mute Delay adjustment pot. Believe it or not, the best device turned out to be a spare Heath transistor I had for my SB-104A transceiver! It is a Heath part number, 417-233, or a 2N3643. If you can't find something similar, look for a high beta audio transistor with 600ma to 800ma maximum collector current rating. The NTE replacement guide shows a NTE128 as a possible part. Run the audio gain at near maximum while keying the rig at the shortest Mute Delay settings to find the quietest transistor. Note that you should *not* turn the AF gain to its maximum setting, as this will induce another T/R pop that is practically impossible to eliminate without further modifications. This was mentioned earlier in the section on KEYING mods. It is also *not* necessary to increase the CW Level so that the rig is transmitting, unless you think the additional current draw might introduce more T/R noise. I did not notice this with my rig. I should mention that a 2N7000 TMOS FET could be used as a muting transistor, but unless a discharge resistor is added from the Gate to ground, once the Mute signal is applied to the 2N7000's Gate, it will "latch" and not turn off during the return to receive! I tried a 2N7000 on mine, and although it worked well at killing the audio pops during Mute, I decided the 2N3643 worked just as well and did not require another resistor as the 2N7000 would. Several mods have appeared in various publications about putting a series Mute FET in the audio lead to the input of the AF power output IC, but this requires more effort and modifications than just finding the "right" muting transistor for Q303. HELP FOR THE BFO, VFO, ETC. If you still can't get enough output on your rig after doing the pre-driver mods for Q401 and Q402, you might have low output from the BFO circuit. Note on the schematic how the output level of the BFO is varied by the CW Level Control on the front panel. Diode D143 is used to shunt a portion of the BFO signal to ground before it is introduced into the SBL-1 bilateral diode ring mixer. This is a rather unconventional yet effective way to control the transmitted power level. Since Heath commonly used the MPSA20 (Heath pn. 417-801) device in many kits and circuits such as the BFO in the HW-9, I figured it was probably not the ideal device for RF applications. Once again, some improvement can be gained by selecting a better device for some of the BFO stages. There are a couple of choices here to obtain increase BFO drive - both for the transmitter stages and also for the receiver's Product Detector. You can either pull out several of the Heath 417-801 transistors and select the best ones for critical circuit locations, or use another device altogether if you don't mind replacing the original Heath parts. Since I had some Heath 417-801 spares from my SB-104A rig, I decided to pick a few of them for the BFO that gave me the most output without making any other changes. If you remove several 417-801 transistors from your HW-9, you can select the ones that perform best where they are needed. Otherwise, use MPS2222A, 2N2222A, or 2N3904 transistors that are widely available. Using Molex pin sockets for Q113, Q114, and Q115, try several different transistors for the greatest and cleanest output of the BFO. Monitor the BFO level using the Heath provided RF detector at test point TP104, and / or the S-meter during transmit as it monitors relative power ouput. You should not increase the BFO's output to the point that spurious output is developed from overdriving the SBL-1 mixer and succeeding transmitter stages. You wouldn't want to undo all the work just done in the pre-driver stages by applying too much BFO injection! I found a couple of 417-801 Heath transistors that were much better than the original devices in the BFO stages. The most critical device is Q113 for decent BFO injection to the receiver Product Detector, U303. For the transmitter BFO injection, Q114 seemed to be more a factor than Q115 (as you might expect!), and the increased BFO signal made available allows the CW Level Control to be adjusted to a lower setting for good output and smooth control of the PA output level. In fact, the additional BFO output actually seems to make the CW Level Control more broad in its adjustment. This is probably due to a greater available amount of BFO RF into D143 at lower settings of forward bias from the CW Level Control, rather than when it is almost biased completely off for greatest PA power output when only low BFO drive is present. D143 must have a more linear affect on the output level while it is still partially forward biased versus when it is nearly completely biased off. The MPF105 used for the BFO oscillator turned out to be more than adequate, and so it was not replaced. After selecting your favorite part for Q113 - Q115, be sure to re-adjust the BFO frequency, and BFO filter inductor L137 as outlined in the manual. Low VFO output can be corrected with a 2N3906 at Q106, or try several Heath 417-234 transistors (2N3638A) if you have them. My VFO does not exhibit much drift, and other info on tracking down excessive drift is covered in the HW-8 Handbook. I used a 5.6K resistor across Transmit Return control R131 to make it less touchy to adjust. Just tack it onto the outside leads of the trim pot from the top the the PCB. If 5.6K is too small, try a 6.2K resistor instead. The object is to make the adjusment of the Transmit Return voltage very broad, rather than the difficult adjustment it previously was. Another method would be to use a smaller control than the 50k pot Heath provided, and adjust the ratio of R127 / R131, but the parallel fixed resistor across R131's terminals is easy and neat. VFO DRIVE PROBLEMS The small vernier drive used to tune the VFO cap has low torque, and the VFO cap is often mis-aligned with the drive mechanism, causing binding and slippage of the dial. I checked the VFO cap shield for squareness, and it seemed OK. However, the VFO capacitor itself seemed to be non-perpendicular from the rear mounting surface to the shaft. Many HW-9s have one of the VFO shields removed, since this makes the shields more rigid to flexing, with the result being the dial and shaft will bind. Also, if the vernier is not properly mounted and the hex nuts at *each* end of the vernier tightened properly, there is not enough torque to overcome VFO drive friction. Be sure the vernier nut closest to the small pinion that is part of the VFO vairable cap is snug. If that is satisfactory, tighten *only* the front hex nut on the vernier to secure it to the VFO shield. It is wise to leave the three mounting screws on the rear of the VFO slightly loose and rotate the VFO through its entire range of rotation before final tightening of the vernier drive and the VFO capacitor mounting screws. Find the location of the VFO cap mounting that results in least binding. If it still binds, either place thin shims between the VFO capacitor and the rear of the VFO shield, or leave the VFO cap mounting screws just slightly loose, but not too loose. Try turning the VFO cap to the point of greatest binding, and then make small movements in the vernier drive and/or VFO cap mounting screws until it frees up a bit. I was able to retain the second VFO shield (which I had to fabricate, since it was missing from my HW-9) and find a workable adjustment of the shield and VFO mounting screws. The slide-on second VFO shield is useful in making the knob and tuning shaft "feel" more rigid, as without it, the single "U" - shaped VFO shield is inadequate for overall VFO rigidity. A "box" shape is many times more rigid than a "U" shape, so use both shields if you can without excessive dial slippage. The rotation of the dial should be light and free of binding if you can get it all to work out properly. It is a poor design, and *very* tempermental to assemble. END OF PART 2 AB7MY Gary Surrency Chandler, AZ (Near Phoenix), QRP-L #571, AZ ScQRPions, ARRL VE