Issue #110
Scanned and OCR'ed by Grant Youngman, NQ5T

AM in Amateur Radio: The Situation Returns to Normal

AM is increasingly heard throughout the lower HF bands, as it once again claims its rightful status as a legitimate facet of mainstream amateur radio. The ARRL and QST, once perceived by many as "anti-AM", have come round with recognition of AM in the magazine and willingness to petition the FCC to retain AM privileges. Most equipment manufacturers, including Ten-Tec, now produce HF transceivers that include the AM mode.

I hear old timers describe how much fun they have been having since getting back on AM, after finding themselves bored with SSB and "state of the art." AM has rekindled these amateurs' interest in ham radio, whether they are simply switching their transceivers to the AM mode or going all the way to equip their stations with vintage hollow-state transmitters and receivers like the ones they used in pre-SSB days.

I became licensed in 1959 and first got on AM phone in early 1960, so I remember the "AM days" very well. However, I cannot make any claim about being "back" on AM, because I was never- off AM in the first place. I can lay claim to being a purist, having worked AM continuously for over thirty years, except during short periods of inactivity when I could not set up an amateur radio station where I happened to be living. I have never owned a SSB rig or operated SSB from my station.

Around 1980 I ran the service department of a commercial two-way and amateur radio retail firm that did mail-order business nationwide. I became thoroughly familiar with the insides of many SSB transceivers of that era, having repaired and modified hundreds of them. But I found these rigs uninteresting and never felt any urge to acquire one for my own use.

I might have eventually operated SSB, had it not been for the high-pressure salesmanship campaign led by the amateur radio establishment during the 1960's to make everyone feel obliged to abandon AM. Frankly, I resented the propaganda in the amateur radio publications, and the attitudes of many "recent converts" who populated the amateur bands. With the advent of "affordable" low quality SSB transceivers circa 1963, I felt this was foremost a campaign to expand the market for a new product.

The campaign succeeded in hardening my enthusiasm for AM, and I felt determined to resist the pressure. Eventually, AM began to return to the bands as the situation slowly returned to normal, what it should have been all along: SSB taking its place, perhaps a dominant one, along beside AM and other modes, in the multi-faceted hobby of ham radio.

The near demise of AM began about 1963 with the advent of the cheap, low quality tube type transceivers, the early Swans, the Galaxies, the Heathkit "HW" series, and others. Until then, commercial SSB amateur equipment was limited to Collins and Central Electronics, which was too expensive for most hams' budgets. The only other way to get on SSB was to homebrew your own, but SSB was too complex for most hams, although the ham magazines and handbooks were filled with homebrew SSB projects. Many of the AM'ers of those days wanted to join the SSB crowd, but were "stuck" on AM for the above reasons. When the low-cost SSB transceivers came out, a real market demand was filled, and the manufacturers started a campaign to convince every amateur to change over to SSB. The campaign got vicious at times, and even today the legacy of deliberate QRM is still with us. After 1963 the decline of AM was rapid, and the shock of Incentive Licensing almost wiped out the remaining vestiges. Only a few of us "die hards" remained through the late sixties and early seventies.

About 1973, a group of amateurs in the Northeastern states became active on 75 metres, and the 1885 kHz "AM window" was established. From about 1974 onward, AM has made a steady comeback. To the FCC’s surprise, renewed interest was sufficient to defeat the infamous Docket 20777, which would have outlawed AM by restricting bandwidth. Since then, the AM community has fought off several other dockets and petitions to restrict or eliminate AM. Interest has continued to grow to the present, with AM firmly reestablished as a permanent facet of amateur radio.

The "come-back" of AM began about 1974, twenty years ago. This means that the period of steady growth is now twice as long as the decade of decline that lasted from about 1963 to 1973, when AM was written off as "dead" by the amateur radio establishment.



AM'ers Featured in Other Radio Publications

See the March, 1994 issue of Popular Communications, page 6, for a letter to the editor, written by the editor/publisher of The AM Press/Exchange. The subject is the FCC's internal "justice" system and the rulemaking process,

The cover of March, 1994 CQ features a full page photo of a well known AM'er from the Northeast, Chuck O’Neal, WA1EKV. Chuck's AM signal is well heard throughout North America and beyond on 75 metres, always with broadcast quality audio.


"Grey Sheets" Now Feature Collins Newsletter

The Classic Ham Radio Shopper, or "Grey Sheets" published by Dave, W3BJZ, has a new format. Because there is no monthly Collins magazine for the Collins collector, Dave has begun to publish a monthly Collins newsletter within his Grey Sheets. The new name for his publication will be the Classic Ham Radio Shopper and Collins Newsletter. Dave welcomes all his fellow Collins collectors to support his efforts to keep Collins people informed on a monthly basis. Technical articles, hints on restoring Collins equipment, including free ads will be featured in this new section of the Grey Sheets. Subscription price is still $12.00 per year for 12 monthly issues. For more details write to David Knepper, Box 34, Sidman, PA 15955.

FCC Drops Station Location Requirement

Since November 15, 1993, the FCC has no longer required that a station location be shown on amateur license applications. The Commission said that because portable and mobile equipment is now so often used by amateurs, a station's location often changes, sometimes daily. They also said that deleting the station location requirement would expedite the processing of license applications. Since this rule amendment is not likely to be controversial, the FCC declared it a "nonsubstantive" change in licensing procedures, therefore no notice and comment period was needed.

The amended rule "Section 97.21 Mailing Address" now reads that "Each application for an amateur service license . . . must show a mailing address an area where the amateur service is regulated by the FCC. The mailing address must be one where the licensee can receive mail delivery by the United States Postal Service."

- ARRL, Westlink Report

New "AM Window" on 160?

How often have you listened on the popular AM frequency of 1885 kHz with the intention of starting a QSO, only to discover that the frequency is already busy with an AM QSO in progress, and the roundtable is too large, or one or more of the stations is too weak to comfortably copy? Some of the East Coast AM gang, looking for an alternative to 75 metres, have suggested 1900.0 kHz This is close enough to 1885 to allow QSY without extensive tune-up, and will help maintain AM presence in this part of the band, which is increasingly becoming more congested with SSB activity just as the top end is becoming less available, cluttered with radiolocation beacons. How about it, gang?



Bacon, WA3WDR, 8 Paulding St., Huntington, NY 11143-1960


A typical amateur modulator uses a pair of push-pull beam tetrodes operating at a fixed screen voltage. Unfortunately, this does not make a very linear amplifier. The dynamic characteristic curves of plate current versus control grid voltage show an upward curvature, followed by a gradual flattening off as they approach saturation. The result is distortion.

Generally, screen grid tube linearity is best at lower fixed screen voltages and worse at higher fixed screen voltages, but in all cases the curve shows that general upward curvature, followed by saturation.


Just place a resistor in series with the screen grid. In a push-pull stage, place one resistor in series with each screen grid, as shown in Figure 1. These resistors should be film or composite types of suitable power rating. If large dissipation is necessary, a series combination of lower value units should be used. Don't use wire-wound resistors, because they may introduce enough inductance to cause parasitic oscillations.

Click here to see Figure 1 – Resistors in Series with Screen Grids

Full output power can still be obtained from the tubes if you raise the screen supply voltage enough to compensate for the voltage drop in the resistors at maximum screen current peaks.

I'm using push-pull 8417 beam power tetrodes in my modulator. This tube is similar to EL34 and 6550 and not very different from any of the 6L6 - 807 – 6146 class of tetrodes. I was running regulated 300 volts to the screens; I took two 820 ohm, 2 watt resistors, and placed one in series with the screen grid in each 8417, and raised the screen regulator output to 380 volts. Distortion went down!

On audio peaks, there is about a 60 volt drop in the screen resistors. By raising the screen supply voltage, the screen voltage on peaks is approximately the same as it was before the change. Otherwise, there would not have been as much output power from the 8417s as there was before the change.


Consider the screen current over the linear range of the tube. The screen current is generally low at control grid cutoff, and it increases until saturation.

Next, consider the effect of screen voltage on plate current. Generally speaking, the higher the screen voltage, the higher the plate current.

Now consider the effect of a resistor in series with the screen grid. The screen voltage will pull down below the screen supply voltage, by an amount which depends on the value of the screen resistor, and the current drawn by the screen grid.

When the control grid is near cutoff, the screen current is low, and so the screen voltage is close to its source voltage. As the control grid voltage increases, the screen current rises, and so its voltage drops further below the source voltage. The main effect of the resistor in series with the screen is to smoothly reduce the screen voltage at higher drive levels, which compensates for that troublesome upward curvature in plate current.


The screen current characteristic of a screen grid tube tends to match the plate current characteristic. This causes the compensation introduced by the series screen resistor to tend to match the plate characteristic, further improving the compensation from cutoff to saturation.


The series screen resistor also produces negative feedback. For any given control grid voltage, any variation in plate voltage will affect the screen current in such a way that the resulting voltage drop in the screen resistor will increase the screen voltage if plate voltage increases, and decrease the screen voltage if plate voltage decreases. This produces negative feedback which further reduces distortion, and also reduces the plate resistance of the tube.


There will be some optimum value for the series screen resistor.

The characteristics of certain tube types may produce a distortion minimum with some specific combination of operating conditions and series screen resistor value.

A desired plate resistance may determine the optimum screen resistor value in some cases. Output transformers usually respond best with some specific value of plate resistance. The larger the screen dropping resistor, the lower the plate resistance will be.

Available supply voltages and output power requirements may place a limit on the maximum screen resistor value. If output power is critical, and the plate and screen supply voltages can not be increased, then the optimum value would be just less than the value which begins to reduce output power.

For tubes in the 6L6-807-6550 class, try a screen resistor value around 1000 ohms. If you experiment with this resistor value, remember that P=I2R. In this circuit, a higher value resistor will drop more voltage and therefore dissipate more power.


If a very high screen dropping resistor is used, output power will be greatly reduced unless the screen supply voltage is raised to a very high value. This can cause many problems: low gain, high input capacitance, excessive screen voltage around cutoff, screen secondary emission runaway, etc.

For tubes in the 6L6-807-6550 class, I would think the maximum value for the screen resistor would be around 10K.


I haven't tried the Williamson-Hafler circuit, so I'm not sure how different series screen dropping resistor values would affect its operation. However, I suspect that any number of balances could be struck between plate voltages, screen dropping resistor values and screen tap percentages.

Amateurs don't always have access to driver or modulation transformers with taps at exactly 40%, so a little experimentation with the resistor values and supply voltages may be in order.

Figure 2 shows screen dropping-resistors in the Williamson-Hafler circuit. Try using a screen tap percentage below 40% and adding screen dropping resistors; if the output is too low, try increasing the supply voltage. Be careful, because this will raise both plate AND screen voltage; note that the earlier warnings about excess voltages still apply.

Click here to see Fig. 2 – Screen Dropping Resistors in the Williamson/Hafler circuit


Grounded-cathode screen grid linears offer high gain, but always generated high intermod distortion as well. A screen dropping resistor can reduce distortion here, too.

Figure 3 shows how the screen dropping resistor is applied to a grounded-cathode screen grid tube RF linear amplifier. The resistor simply goes between the screen and the screen supply, The screen grid is bypassed just enough for RF, because it must be allowed to swing in voltage according to the instantaneous value of the RF envelope.

Click here to see Fig. 3 – Screen dropping resistor and bypass

Again, full output can be maintained by raising the regulated screen voltage somewhat to compensate for the drop in screen voltage caused by the resistor at maximum screen current.

In fact, a higher screen voltage can actually permit higher output in class AB1.


Pocketful of Power: The Gonset Commander

by Rick Ferranti WA6NCX; Arlington Heights, Massachusetts


Surprising as it may seem, it wasn't until 1948 that hams were allowed to operate mobile on frequencies below 28 MHz. The new ruling spurred manufacturers to respond to a whole new market for ham gear. In the later 50's, familiar equipment like the Johnson mobile, the Morrow twins, the Elmac PMR and AF series, and the Gonset G-77 transmitter, G-66 receiver, and the G-76 transceiver, appeared to answer the burgeoning need for compact hf equipment operating all the way down to 160 meters. But it was the very first of this gaggle of Gonsets that broke new ground for compactness and light weight, and it's still hard to find any hf plate-modulated AM transmitter smaller or easier to carry than the Gonset "Commander," manufactured between 1952 and 1955.

Click here to see graphic of the Gonset Commander

Faust Gonsett W6VR, was a frequent contributor to RADIO magazine (see Appendix) during the mid and late 1930's, often writing about a new ham transmitter design or a commercial adaptation of one, such as a diathermy machine. In the early 40's he joined his father's company1 and several years later designed a compact mobile/fixed station for six or two meters, the famous Gonset "Communicator." These were indeed hot performers for their time, with tunable superhet receivers, crystal-controlled plate-modulated AM transmitters running 5-10 watts output, and a modular design incorporating separate chassis for the AC/vibrator power supply, the receiver, and the transmitter. Known affectionately as "Gooney Boxes," and noted for their ability to hear VHF AM aircraft communications as well as the two meter band (the 5 MHz if and poor tunable front-end selectivity yielded a strong image response), Communicators are still seen at ham radio flea markets, sometimes painted garish Civil Defense yellow.

After World War II, Gonset also marketed a hf mobile receiver converter series, first the "10/11," then others like the "Triband" and the "Super Six." The latter was a fine tunable hf converter, and it even became the heart of a mobile transceiver designed by Don Johnson WA6AAQ, as described in his book 40 Years of HF Mobileering. Appearing shortly thereafter, and not much larger than the converters, was the Gonset Commander transmitter.

Anatomy of a Gonset Gem

The Commander is a 30-50 watt input, crystal controlled 160 to 6 meter plate-modulated AM/CW transmitter measuring 8" wide by 5" high by 7" deep, about the size of two modest-sized paperbacks stacked on top of each other. It weighs all of 7 pounds, an easy handful for the weakest couch-potato ham to heft. Inside, a 6AG7 is the crystal oscillator/multiplier, driving a 6146 final, which runs as a straight- through amplifier below 27 MHz, and acts as a doubler on 10 and 6 meters. The final is plate modulated by a pair of push-pull class B 6AQ5's (in earlier models) or 7C5's (later production), driven by a 12AT7 mike amp/driver. The rig will accept either a carbon or high-impedance (crystal or ceramic) microphone.

The rf section is a bit unusual. First, the 6AG7 crystal oscillator can be reconfigured for VFO operation by a front-panel slide switch. No, it does not become a buffer, but rather the 6AG7 accepts a series-tuned circuit at the front panel jack and becomes a free-running Clapp oscillator! The VFO that Gonset marketed for the Commander is thus a completely passive device. Though Gonset claims that their passive VFO was stable enough for operation even on 10 meters, I wonder if more FM resulted than AM. I've seen only one example of the Commander VFO at a ham flea market.

The 6AG7 oscillator/driver is tuned over a 1.7 - 27 MHz range by a bandswitched tapped coil and miniature tuning capacitor. Grid current is indicated by a switched front-panel meter. But the final amplifier is tuned by a plug-in coil, accessed by a little front-panel trap door. Loading is accomplished by moving the coil tap up and down the plug-in output tank until the correct plate current is indicated by the meter at the resonance dip. The output tank tuning capacitor has a large, color-coded tuning dial, calibrated roughly in wavelength, so that with the proper Gonset coil the operator has little chance of tuning the final to the wrong harmonic. Two coils were provided with the rig, one covering 80 and 40 meters, the other 20 to 10 meters. A 160 meter coil and one for 6 meters were available separately. The plug-in coils and the rig's ceramic feedthrough output terminals could be configured for either coaxial or balanced feedlines. Despite its harkening back to the plug-in coil transmitters of the 1930's, this final output design allowed for very compact construction, and it adds a bit of quirky nostalgia to the transmitter

Other front-panel controls include slide switches for tune-operate, hi-lo grid drive level (you need the higher drive capability when doubling in the final for 10 and 6), AM/CW selection, VFO/crystal operation, and power. Connectors include a FT-243 style crystal socket, the VFO input connector, and a 3/16" microphone jack. The rear-panel mounted mike gain control, power connector, and output terminals complete the control and connector complement. No antenna changeover relay is included. The rig can be wired for either 6 or 12 volt operation, and runs entirely on 300 volts, including the final amplifier. In fact, the manual cautions against higher voltage operation, giving elaborate details for connecting the then-popular 425 volt PE-103 dynamotor through dropping resistors so that the transmitter does not zorch over or burn up. With 300 volts running both low and high level stages, the Gonset Commander can be conveniently powered by a single vibrator-type power supply, or a simple AC operated source.

For a lightweight transmitter the construction is remarkably rugged, with shielding and layout cleverly designed to maximize rigidity without adding weight. High quality components are used throughout (my 3 Commanders have had no smoked or failed parts); the audio and modulation transformers are shielded as well. This was not a cheap radio to build, but it was built to last - no wonder it cost $125 in 1952.

Commanding the Airwaves

Operating the Commander is very simple. After you get an antenna relay and PTT control system connected, you key up in the "Tune" position and tweak the grid for maximum current. If your crystal is a bit inactive you may have to switch to the high drive position for the required 2.5 mA grid current on the 6146. With the proper final amplifier coil plugged in and the little door open, you switch to "Operate," key down, and quickly swing the plate tuning capacitor through its range, tuning for a dip. To load, simply change the antenna tap point on the output tank coil, shooting for 100 mA plate current at resonance. Adjust the microphone gain for 100% modulation and you're on the air - oh yes, don't forget to check that you're tuned to the right harmonic on the final tank capacitor dial, and do close that cute little coil door. You can't get a 300 volt shock from the exposed windings (the coil is isolated from the 6146 plate voltage with a coupling capacitor), but the high-z side of the coil could give you an unpleasant rf burn.

With the Commander running all of 30 watts input on AM and yielding about 12 watts out (10 watts on 10 meters), you're not going to blast through the QRM with shear power. But on a quiet band, the Commander really shines with fine, punchy audio, no doubt because of its integral high-level clipping which limits the modulation to near 100%. I've made contacts up and down the Northeast on 40 and 80 meters; on 10 meters the West Coast hears me fine (the antenna is a full-wave 80 meter horizontal loop about 30 feet off the ground, fed with a homebrew tuner). In fact, I was more readable with the Commander while talking with Walt KJ4KV 500 miles away on 75 meters, than was my neighbor Charlie KA1GON running 10 times the power with a much-audio modified Valiant and a dipole. I haven't tried to "improve" the audio with various mods; it doesn't require any work according to the reports I receive. In any case, running a stock rig lets others hear just how good the original design is.

If you're looking for a compact mobile or fixed-station transmitter that pours out a first-quality AM signal, you will find just the ticket in a Gonset Commander. Mated with a suitably tiny receiver,4 the Commander forms half of what must be the smallest and lightest AM ham station money could buy. Its high quality, convenient power requirements, ease of operation, and fine signal make it a, worthy avatar of hf ham radio mobile, in the early 50's or 40 years later.

Click here to see Commander schematic

Appendix: The Golden Age of RADIO Magazine

If QST experienced a golden age, it was certainly in the early 1930's, when they published a whole raft of technically advanced articles, written by several outstanding radio engineers of the time. These included articles on class B push-pull modulation, simple five-meter rigs using modulated oscillator transmitters and superregenerative receivers, articles on the National SW-3 and the HRO (by James Millen W1HRX), and articles on the single signal superhet and the if noise silencer (by James Lamb, one of the most talented engineers on the ARRL staff). The later 1930's, however, belonged to the lesser-known West Coast magazine called RADIO, a publication with roots reaching back to the teens, when it started as Pacific Radio News. Unlike its East Coast counterpart, RADIO contained many more technical articles and none of the endless traffic reports still plaguing QST.

Among the more famous articles found in the mid to late thirties RADIO were the flattop beam antenna by John Kraus W8JK, articles on "efficiency" and cathode modulation, and dozens of articles by the most prolific amateur radio writer of the time, Frank Jones W6AJF. Innumerable transmitters and receivers came from his workshop and pen, the more famous of which were the Jones exciter and the Jones "Super Gainer" series of receivers.5 Jones also wrote and published the first RADIO handbooks, starting in late 1935. These excellent books. Inspired by Don Wallace W6AM's 1933 Shortwave Handbook, became the standard "West Coast Handbook" still published by the Editors and Engineers and Bill Orr W6SAI.6

RADIO magazine reached its zenith in the late thirties, with hundred-page-plus January issues containing everything from commercial ham receiver and transmitter surveys to dozens of construction articles by prominent writers like Faust Gonsett and Grote Reber, the radio astronomy pioneer. Then the magazine began to shift focus, first geographically with a move of its editorial offices to the East Coast, and technically with more and more emphasis on commercial broadcast engineering. Having lost its unique West Coast flavor and its amateur audience, and having barely survived the Second World War, RADIO appeared for only a few issues in 1946. The new East Coast upstart CQ, The Radio Amateur's Journal, absorbed RADIO's ham readership a few months later, while other former staffers went on to found Audio Engineering magazine. Thus the amateur world lost RADIO, its finest magazine of the golden thirties.



1An excellent two-part article on the history of the Gonset Company appeared in Antique Radio Classified, December 1991 and January 1992 issues, by F.T. Marcellino, W3BYM.

2Novices may recall the time when they were limited to crystal control, and all sorts of mayhem resulted when a series tuned circuit was plugged into the crystal socket of their Heath AT-1 or Globe Scout. This gave VFO operation all right, usually drifting and chirping right out of the 40 meter band.

3George WIGAC, who knew Faust Gonsett, has shown me a Commander he modified with a full bandswitched pi-network built into the coil compartment behind the little door. I prefer the original plug-in arrangement; it's more fun to have a few coils on display around the rig so visitors can see what real radio is all about.

4The poor little transmitter would look ridiculous next to a 75A4; my Commanders operate with a Gonset G-66B or a Davco DR-30 receiver, both of comparable size.

5My favorite Super Gainer design was a two-tube shortwave superhet (ok, each tube is a dual-purpose type). wherein everything that can be made regenerative, is. Anyone who, like me, started his radio career with a Knight Kit "Star Roamer" 4-tube shortwave superhet (with regenerative if) owes the remarkably good performance of that 1961 receiver to the 1936 Jones Super Gainer.

6Faust Gonset was co-editor of the 1942 RADIO handbook. W.W. "Woody" Smith W6BCX, also a former editor of the handbook and of RADIO magazine, became Chief Engineer of the Gonset Company in the early 1950's. He may have designed the Commander transmitter.





(Part IX of a Series)

By George A. H. Bonadio, W2WLR


All these (approximately) 88 mH toroids are actually with a split winding so that we can use either. the two windings in series, for 88, or we can tie the two in parallel, with the correct polarity, to net only (near) 22 mH. We can tie the combination of two 22's, in parallel, to net only about 11 mH. All toroids are with cores which look like doughnuts overwound with wire, and are of powdered molybdenum permalloy. Detailed instructions on how to wire the inductor leads will be provided by Ed when he sends to you your inductor set.

Click here to see Photo 1 – Functional "6 pole ELPF"

bulletPHOTO 1. Functional "6 pole ELPF" of 500 ohm input & 500 ohm output. Notice that the C's across L2p are in series while most others are in parallel to reach the optimum values. The three potted L2's (s=series, p=parallel) while typically "22" mH, this batch measured 20.0 mH. Each is glued to the board. The minimum weight buss bar is shown, well soldered, by W3NQN.

In making up values of toroids we may have to unwrap old sticky tape. Either hold that tape up to a hot lamp, or dunk it in very hot water. (Do not put wet tape onto a hot bulb!) The wire pigtails are designed to be rosin core soldered, uncleaned, as their insulation is so composed, without special preparation.

To test to prove that you have either 22 or 88, as you have them connected, put your audio generator, to the ends of the coil and across them, a C of 0.1 u (100n) and a voltmeter. Putting them all in parallel and sweeping the range for a highest voltage reading, will give a very broad peak. If it is highest near 3,400 Hz you have 22 mH, and if near 1,700 Hz, you have 88 mH. With no very broad peak you have one winding backwards, or a bad C. Ed has a much more accurate way:

To confirm that you have either 22 or 88 mH, connect a 0.1-uF C across the L. See Figure 16.

Click here to see Figs 15 and 16

Then couple the output of your (audio) signal generator across the tuned circuit using a 100-pF C between the generator and the top of the tuned circuit. Connect the input of your a-c VTVM across the tuned circuit using another 100 pF C. The coupling C's are made about 1,000 times smaller than the 0.1-uF tuning C to minimize the effect of connecting the signal generator and VTVM across the tuned circuit. As you sweep the frequency of the generator back and forth you should observe an abrupt increase in the voltage level across the tuned circuit at its resonant frequency. If you use a tuning C of 0.1, the voltage peak should occur at around 3400 Hz with a 22-mH inductor and around 1700 Hz with an 88-mH inductor. If you do not see any peak you either have an inadequate signal input level or the inductor windings are connected incorrectly, or you have a bad C .

Test the ohmage of each toroid, in comparison with others. An 88 mH should test about 8.4 ohms. Most of the half windings will measure near 4.2 ohms. Test each C, before soldering it in place. It should read maximum, or well over 10,000,000 ohms.

I recommend that you measure and record the DC ohmage readings, end to end, and across each L, on a paper, and tape this data to the filter - just in case you might worry, at a later date, that you have blown a C therein.


In your filter in your PRE-box, mounting it upon the ceiling of the box is fine. It is going

to be only 500 ohm from G, which can stand much stray C (10n) to G. So, if we want more shielding, we may wrap the whole filter in layers of insulating material, then cover it with shielding kitchen foil, covering this with, perhaps, a plastic bag, or 3 layers of black tape. We do this grounding with several inches of bare twisted zip cord as grounding wire intermeshed to the foil, to reduce feedback problems.

In your 500 ohm high level audio line filter, you do not need to do any shielding. However, there may be accidental high voltages on this filter, so it should rest in an insulating box, and be labeled as containing HIGH VOLTAGE. In your audio tube plate fed low pass filter (LPF) you have up to 1,000 V to protect from G. This means special fastenings or thick insulation, before you foil cover it. Take a grounding wire out of that foil by passing it through a C of at least 1,000 V and not less than 0.001u (1n) up to 0.1u (100n) to G.

New zipp cord can connect this filter output to the two terminals of the primary of that first transformer. The hot input connects to the plate of the tube. The filter hot side does replace the lead between the plate and the transformer. Don't connect the output end to the tube.

Do insulate over all the foil, for safety. Most of the DC of the plate supply is blocked off from the filter C's by connecting the common "colder side" to the B+ side of the transformer primary. The C's of the filter can, then, be of 600 V rating. A choice of values for all the filter C's, in a 1,000 V rating, is hard to find. However, two equal value 600 V C's, in series, will rate for 1,200 V, but at half the marked capacity.

See Figure 15 for how two three or four C's of 400 V or 600 V may be assembled to be safe to 1,200 V. Note the effects on the resultant ratios of the C values, too.

Click here to see Fig 15


Be sure that you take and record all your ORIGINAL response curves before you start any changes.

What I do not want you to do is to jump to a wrong conclusion about responses. The roll off of the bass which you see on your original curve was needed because receivers were getting sharper IF's and chopping off more of the treble. This made a nearly flat bass response sound very bassy. You will be able to demonstrate that to your contacts with the turn of your Preemphasis Control from optimum to off.

Without enough treble the bass must be cut. Those transmitter engineers compromised on top of the receiver design compromises.


The "500,000" formula, for balancing tones, divided by, say, 2,000 Hz = 250 Hz as being the lower limit on balancing bass. Between receivers and muffled transmitters, a top figure of 1,500 Hz was sometimes effected. That left about 330Hz for the low roll off, not very bassy, if balanced. We are using up to 3,333 Hz, so, 150 X 3,333 @ "500,000," an excellent balance. That is a factor in causing the good reports received. Listeners hadn't been hearing balanced signals. Now it surprises them. They initiate comments.

Thus, you see, with the receiver selectivity effectively deemphasizing, if you put on a flat response, you are severely deemphasized. You will sound severely bassy, as 150 X 2,000 = "300,000," compared with "500,000." Even worst would be 100 X 2,000 = "200,000." We find a few hams with deeper bass, 80 X 2,000 = "160,000" @ very unpleasant, and difficult to copy against any noise.

Try to stay between a bassy "400,000" and a thin "600,000." The CB range of 300-3,000 nets a very poor, thin "900,000."


We do need real functions in our PREbox. The mike socket is identical to the socket on the transmitter. Then, if you wish, or need to, you can unplug the mike cable and put it into the transmitter quickly. Yes, then, you will have "high fidelity" going out but you will be muffled being received. This is because you are using normal bass instead of the reduced bass, as in Figure 1. (See issue # 102, p.12)


What we need in values is "near enough" accuracy. What is adventuresome is looking for values of 1%, and of the voltage you need and of the capacitance. In the PREbox, and on the loudspeaker circuit, small is nice, and we can use 100 V, even 50 V values in the PRE-box, on that 500 ohm design. In the 500 ohm line, even with a few watts of power, we use 600 V. In the high Z filters we do need 600 V. Please do not use any old C's. Buy new C's and test them. We are looking to preserve AM, not to crash it.


An ohmmeter test should show 22's as about 2.1 ohms, 11's near 1 ohm and 88's near 8.4 ohms. 6 X 88 @ 50 ohms, 4 X 88 @ 33 ohms. Don't 'smoke test" these. Measure the DC from end to end of the entire filter to see if it agrees with what the series toroid ohmage should be. Test audio signals through them to show you the roll off. There should be a flat response up to 3,000 Hz, then about -3dB near 3,300 Hz, then at least -45 dB by 4,000 Hz, and continuing to stay beyond -45 dB of losses for all higher frequencies.


If local operators do hear your signals when they tune beyond 8 kHz (counting their IF bandwidth which they may forget to allow for) they will not hear splatter. They may hear a more normal audio bleeding through their own receiver's selectivity limits. This undistorted audio is their problem, not yours. Prove to yourself, in your own receiver, how much neater your signal is. If your signal swamps your receiver, disconnect your receiver's coax. Even stuff a little shorting foil into the receiver's coax socket. Use a dummy load on the rig, so that you can test as a modest strength incoming signal.

In fact this type of testing should be recorded for duplicatable testing before and after this whole event. Be aware that some of monitoring receiver signals can be detection of RF by cathodes in your audio stages, especially with your receiver gain very low. These may sound like there is too much treble because they do not have IF selectivity deemphasising. You are not a satisfactory tester of your own signal. Listen for spontaneous reports.


For your new dummy load use a PL-259 to RG58U of over 5 feet long, to get the RF away from you. End the coax into a ceramic socket, like available for screwing up to a ceiling. It can be heavy enough to hold your bulb upright. Test it on your normal frequency on your longest wavelength. The bulb wattage, for low power, should very roughly approximate that of your final watts input. For high power, use the wattage output sizes. Notice that your SWR into a bulb changes as that lamp warms up. To be continued


open forum

Update On Heathkit Equipment List

Dear Marty;

This letter is an addition of HEATHKITS to the list you published in AM press/exchange # 107. These date back to the late 1950's and the 1960's and are still in use. I assembled 29 kits and they are in working order except for xmtr DC 35. I am re-winding the power transformer but I am not optimistic how successful the project will be. Many thanks for all your work and effort making the list of kits.




Howard K6TER

Copy to: Don Chester, K4KYV




WANTED: Tubes unused and made in USA or Western Europe. Please state the type of tube box in your reply: 6060, A2900, B309, E81CC, ECC801S. B329, ECC82, E82CC, ECC802S, 6057, B339, CV492, CV4004, ECC83, ECC803S, M8137, ECC88. GOLD PINS CCA, E88CC, E188CC and E288CC. ECC32, ECC33, ECC35, 5691 and 5692. Made in USA 6L6G, 6L6GC, 6550, 7027 and 7591. Made in Great Britain or Holland 6L6G/CV1947, EL34/CV1741 and EL37/CV586. Made in Great Britain KT66/CV1075, KT77 and KT88/CV5220. Vintage tubes 45, 50, 211, 801, 845, 2A3, 6A3, 6B4G, 6A5G, VT-52, VT-62, WE274, WE300 and WE350. U52 and U77 MOV. Also wanted test equipment, Sprague Vitamin Q and Black Beauty capacitors, oil-filled capacitors and mounting straps/clamps. "Ex-valve equipment" choke and transformers. I would like to hear from you soon. Thank you for your kindly help, Wai Kei Leung, electron tube buyer. (past issue)

WANTED TO BUY SURPLUS TUBE INVENTORIES, Collins KWM1, KWM2, Johnson 350 watt Matchbox, for cash. (Past issue)

WANTED: RCA Chanalyst, Philco 7008, National FB7. FOR SALE OR TRADE

Jennings type "UCS" vacuum variable condenser, HP200 audio oscillator.

Kxxx (Past issue)

FOR SALE: Collins 75A4 with matching speaker in excellent condition, needs work on AGC, Asking $350, pick-up only. Kxxx (Past issue)

WANTED: AM/FM board for Yaesu FT-102, SP-102/SP/767 speakers, FC-102 ant tuner; Drake MS-4/MS-8 speakers; Kxxx (Past issue)

WANTED: For personal restoration projects. Old-style pushbutton 110 v.a.c. light switches and cover plates. Jewell meter: "Pattern No. 54", 3 1/2 inch diameter with silver scale, 0-300 ma DC. Brown glazed E.F. Johnson type 210" bayonet type tube sockets (for 866-A, 811-A, etc). HF-300 tubes. Will pay cash, or trade. Have many vintage components on hand. What do you need? Kxxx (past issue)

WANTED: E.F. Johnson Matchbox and general coverage receiver. Kxxx (Past issue)



This is the AM PRESS: An amateur radio publication dedicated to Amplitude Modulation.

This is the AM EXCHANGE: Offering FREE ADVERTISING to enhance the availability of AM equipment and parts.

SUBSCRIPTIONS: $10/year (12 issues) in North America. Foreign rates upon request.

CLASSIFIED ADS: (AM-related) free of charge to subscribers and nonsubscribers.

DISPLAY ADVERTISING: Rates upon request.

Edited and published by Donald Chester, K4KYV

NOTICE; The purpose of this publication is the advancement of Amplitude Modulation In the Amateur Radio Service, and there is no pecuniary Interest. Therefore, permission is hereby expressed for the use of material contained herein without permission of the publisher, with the exception of specifically copyrighted articles, provided that The AM Press/Exchange is properly credited.