I was given a Drake R-4A Receiver! A local ex-ham was downsizing and wanted to make sure it went to someone (else he was going to throw it out). Went through the bandswitch which appeared to be the only thing to be cleaned. Superb receiver, so quiet. Now to see if I can acquire the accompanying Transmitter (T-4), Speaker (MS-4) and Power Supply (AC-3 or AC-4) can be found as cheaply.
TrueTone D3603A by Western Auto
This is a swell portable, sold around 1956. My mom had one of these, in Green (D3603A) — I’ve never seen an example of the Maroon model (D3604A). That old one had a melted speaker grill grid, probably where she’d set an iron too close to it. It found it’s way to a garage-sale table and was gone, sometime in the 1980s.
The Less Common Color
Updated: I finally found the Burgundy-colored model a few years later. The logo “Truetone” is very faint on the front cabinet, but otherwise it’s in grand shape.
Many thanks to Mike Stute for giving me this one a couple of years ago, when I casually mentioned that my mom had had one of these.
Another in the S-38 Series from 1957-1961 era
I purchased the old radio for $30 from Jim Heye (K5WLQ). I never plugged it in, respecting the fragility of possible old paper and electrolytic capacitors, but the first thing I noticed was that the main tuning dial was strung backwards! The cabinet was nice, with a few nicks and scratches from a normal life — not as nice as I would have liked, but cleanable. The S-38E is the grey hammertone finish; it was also available in S-38EB (beige) and S-38EM (mahogany). It’s a swell example of the era, with the short (pre-1970’s) AM dial and the CD — Civil Defense — marks at 640 khz and 1240 khz.
The S-38 Series was introduced in 1946 (S-38) and were produced through 1961 (S-38E). The S-38E
was produced from 1957 to 1961, making it the end of the line, priced at about $50. Internally,
the radio is a classic “All-American 5”, which makes it very easy to work on. The original S-38 has 6 tubes, but subsequent models have 5.
Tube Lineup for the S-38E, all 9-pin ‘miniature’ tubes — a departure from the older octals:
- V1 – 12BE6 Mixer / Oscillator
- V2 – 12BA6 IF / CW BFO
- V3 – 12AV6 Detector / Audio Amp
- V4 – 50C5 Audio Output
- V5 – 35W4 Rectifier
Left to right: Antenna lugs A1, A2, Ground lug (soldered directly to the chassis!), cw adjustment (Mark II model only), phone output.
Note the use of miniature tubes. Serial number looks to be: 252816, with an over stamped Q42301. Not shown, to the right, the power cord is permanently attached.
The S-38E is a transformer-less radio, and like the S-38B, presents quite dangerous shock hazards.
At some point, however, an attempt was made to isolate B- from the chassis a bit, and to provide for some safety by bridging the B- (ground) and the metal chassis with a 470k ohm resistor in parallel with a 0.06 micro-farad capacitor. However this doesn’t completely remove 120VAC from the chassis or the antenna ground-lug, which is soldered directly to the chassis.
For safety, the switch, which is designed to connect one wire of incoming power directly to the B-, is moved to the other leg, and the newly polarized plug’s ‘neutral’ is placed at the point where the last filament is joined to B-. This assures that (provided the electrical socket is wired properly) the chassis, antenna ground-lug and B- are close to ground.
The power cord is attached to the radio. After a safety rewire, polarizing the cord assures
that the hot leg always is sent to the switch and the chassis-ground is tied to neutral.
This little radio is now receiving on all bands. I’ve received 40m CW and
11m CB. AM Broadcast is strong and clear.
- Radio is tuneable from 540KHz to 32MHz across 4 switched bands:
- Band 1 (AM Broadcast) 0.540 – 1.65MHz
- Band 2 1.65 – 5.1 MHz
- Band 3 5.0 – 14.5 MHz
- Band 4 13 – 31.0 MHz
Completed Appearance Improvements
- Cleaned the cabinet, dials, and knobs.
- Replaced some missing cabinet (back and bottom) screws.
- Finally bought some Bristol Keys to properly remove the knobs.
Always use Bristol Keys (McMaster-Carr #7048A55) to take the knobs off. Hex keys will occasionally ‘work’ to get the knobs off, but they will always destroy the little slug that holds the knob.
Completed Repair Items
- Cleaned the Band Switch, the front-panel switches and Volume Control with De-oxit
- Replaced the multi-stage Electrolytic with a terminal strip and 4 separate electrolytics
- Replaced all the tubular capacitors.
- Rewired the power connection and polarized the power plug for Safety.
Clean scans of schematics are not available via internet, as far as I can find. The usual sources list the user and service manuals, but in some cases only for the ‘non-Mark II’ version. The available Mark II version schematic is a poor scan, in some places illegible. I replaced a tiny mica cap, C16 and if it had been missing or in bad shape, I’d not been able to read its value: 82pf. I plan to take a trip to the Dallas Library Sams Photofacts section for this and a few other scans.
- S-38E Mk II Schematic
- Service and Owners Manual
- A must buy for this and other S-38 series radios is a 10-piece Bristol L-Key Set (McMaster-Carr #7048A55).
These can be found on eBay for around $21 (plus shipping), but McMaster-Carr has the same set for $15.50 (as of Jan 2009)
The S-38E was given to a local gentleman who had contacted the W5FC Club and wanted to try some Shortwave Listening (SWL). I donated the S-38E to the cause, although antenna issues in his location limits his reception. http://www.websdr.org/ fixed that and he is now happliy listening in on CW QSOs all over.
Fortunately an identical S-38E recently replaced it, an almost pristine MKII model, which only required fresh electrolytics and power cord rewiring.
Finished the <a href=”TO-Y-600_refurb.php”>TransOceanic Y-600</a>. This is a magnificent radio and turned out nicer than I could’ve expected. A fortunate turn on this model was that the 1L6 tube is in good shape. Although in fairly ready supply, the tube has reached ‘unobtainium’ prices, even though solid-state substitutes are available. In any case the 50A1 was replaced with a solid-state replacement, which provides very stable current and voltage regulation, which is a good thing given the expense of the 1L6 tube.
Over the Holidays, finished the recap the <a href=”http://mikeyancey.com/S-120_refurb.php”>Hallicrafters S-120</a> that had been in the office.
Finished the <a href=”http://mikeyancey.com/S-38B_refurb.php”>Hallicrafters S-38B</a>; it’s seen some daily use in my office at work. This is my first run-in with the AC-DC power issue and I found a way to route the power cord ‘HOT’ through the switch to eliminate (unless the house wiring at the socket is reversed) shocks from a faulty cabinet insulator.
I snagged three Astron power supplies recently from a listing on Craigslist. It was quite an inexpensive buy, but they were heavily abused and needed some repair, so overall, it won’t work out to a huge profit, but it gives me an opportunity to apply some skill and come out on top.
The previous user, I’m told, was a car-stereo seller and used the Astrons in a large display of car stereos.
From the group, there was a 35-amp RS-35A which, after testing, appeared to be in fine shape; only the fuse was out of order – the previous user had replaced the 8A ceramic fuse with a 10A glass, 3AG fuse – both incorrect and unsafe. The large value would allow more current to be drawn than what it was designed for; ceramic fuses are more resistant to breaking given the high current. Replaced the fuse, tested the supply for regulation under load and it’s now powering my FT-897D.
The two others, both RS-50A models were substantially abused and showed signs of having been run well over-current and for longer than the 50% duty-cycle.
But both still provided regulated voltage!
Both had incorrect fuses – replaced both with correct 10A, ceramic fuses. One had a 20A, 32V fuse (on the 120V primary!). Both had burned wires and the large filter caps (100,000uf in one and two 51,000uf in the other) were toasted.
Astron supplies replacement parts, so I’ve already repaired and tested the newer model with the single 100,000uf cap. Now waiting for two 64,000uf caps to repair the other. I had to make replacement 13.8-volt ‘common’ cables with black, 10ga wire. That ought-a hold ’em.
How a Club Loaner Rig Becomes New Again.
This seemed like an extensive redo, but was not, because many of the boards went mostly untouched. It began with a redo of the HP-23 power supply with an HP-23RL board from The Heathkit Shop. And it ended with replacement of two of the Carrier Oscillator crystals, which with age had changed too much to adjust with serial or parallel capacitance.
For kits anyway, this was the pinnacle – the tip-top. No one would ever make a 20-vacuum tube transceiver again.
I purchased the rig and power supply for $75 from a W5FC
club sale of some old gear. The radio worked, having been gone through by the venerable OM, Don (W9VE).
However, a few issues cropped up, including an un-nulled carrier and a significant difference in
USB and LSB – USB is very muddy sounding and power output was low; LSB was completely normal and contacts
on 40m were made every time the rig was powered up!
In addition, CW output was almost nil with the CW filter in. Both the USB and CW issues pointed to the
center frequency of those two modes being outside the filter passband. Still, it was quite a buy, since
the HP-23B power supply can be found for well over $75 all the time on eBay.
The rig has stickers from the Heathkit factory where apparently it had been shipped to correct some
problems. These stickers date the rig’s assembly at before 1972. The rig works well with a Turner mic
(also from the same W5FC sale) and a later, fortunate fined was a brand new Electro-Voice 719 microphone
with the box, instructions, and (blank) registration card for $9.99 on eBay.
I chose to only do invisible or ‘functional’ mods which didn’t significantly alter the radio. But here are a list of Service Bulletins and the most popular (and necessary) Mods.
Subsequently found an SB-600 speaker, with an additional HP-23A inside for cheap. This had been
originally built by WB8LOL – now K5LOL, Thomas, who’d built it originally built the unit in Detroit. He,
and the rig, found their way to Texas and via K5BJI (Mike Goidl), I obtained the supply.
As a result of all the research, I’ve found some superb resources for part. Here are a few.
- McMaster-Carr – superb online catalog and search tool
- 9540K33 7/8″ x 5/8″ w/washers – feet for HD-10
- 9540K56 25/32″ x 9/16″ w/washers – feet for HW-101 – fits #6 machine screw
- Elliots Hardware – great stock of fasters (McMaster-Carr Numbers)
- 90054A148 – #6 1/2″ Hex Washer sheet metal screw
- 90054A146 – #6 3/8″ Hex Washer sheet metal screw
- 90053A144 – #6 1/4″ Sheet Metal – for RF Cage
- Ralphs Electronics
- Amphenol 2-pin, Mic Plug – http://www.ralphselectronics.com/ProductDetails.aspx?itemnumber=AMPH-80MC2M
- Amphenol 2-pin, Panel Jack – http://www.ralphselectronics.com/ProductDetails.aspx?itemnumber=AMPH-80PC2F
- Type 86-3-24 – strain relief cover –
- Type 86-CP11- 11-pin Plug –
- Type 78-S11 – 11-pin Socket –
- Panel Mount of 78-S11 requires –
- Plain Cover –
- Type 86-3-24 – strain relief cover –
- Leeds Radio in New York.
- http://www.leedsradio.com/parts-sockets.html (78-S11)
- http://www.leedsradio.com/parts-connectors.html (86-CP11)
There appear to always be a few items remaining to do, but the HW-101 operates properly now and
I’ve had two contacts so far: first on 20m (WA7ND) and the USB appears to work, but the Electro-Voice
mic connector shorted out temporarily ending that QSO. Secondly, on 80m with KC9MOS and the
ElectroVoice mic cord appeared to be working again for the duration. I’ll continue to be looking
for bad out-of-spec parts that might show up in performance, but the rig is working nicely!
Completed Appearance Improvements
- Replaced the front panel with a fresh, clean one.
- Replaced the rubber feet – McMaster-Carr 9540K56 is a perfect fit for the HW-101
- Replaced some missing cabinet screws.
Completed Functional Mods and Improvements
- Improved the power supply with a re-cap via the HP-23RL board, cleaning up some poor assembly and soldering.
- Converted to handle Low-Z headphones – external speaker now mutes properly with “modern” 32-ohm headphones.
- Improved the CW operation by increasing drive to the VOX relay
- Killed most of the CW side-tone audio on key-up by dumping sidetone to ground.
- Some mods had already been done, including the meter zeroing issue and some TX/RX improvements.
- Replaced the poorly soldered Amphenol MIC jack.
- Rebuilt the old power cord to supply 120V AC to the Power Switch on the HW-101
Final Completed Items – December 2008
- Replace the USB and CW carrier oscillator crystals – bringing the CW and USB right back into IF passband, probably within 100hz or so.
- Replace the old RCA RF Out jack with a BNC connector. The BNC is better than either the old RCA or a ‘UHF’ connector, plus the single-hole, bulkhead mount BNC didn’t require enlarging the hole.
- Actually found a nearly broken output connection while replacing the RCA antenna connector – fixed.
- Decided to not add a volume control to the side-tone. Maybe at a later date.
- Replaced the grotty old 1/4 inch headphone jack.
- Replace the Carrier Null pot with a new 200 ohm trimpot
Continuing Updates – May 2009
- Swapped the 6EA8 Speech Amplifier (V1) with the 6GH8A which is a higher output version.
My 25-year-old Weller WTCP gave out.
I was repairing one of the two Astron RS-50A power supplies I picked up recently and when I swapped the tips to get more heat out to the massive transformer center-tap… nothing. No clicking. The neon light was on, but no heat was home. It was a terrible discovery.
I’d had that old iron since Mostek. I think I bought it at some employee discount. Lots of projects from the old days and from recently were completed with that good ol’ tool.
So, my choices were to buy a replacement soldering pencil from Fry’s for $69.99, or just get a new one; the WES51 is not much more and is ESD and has a variable heat control. So I sprung for the new iron. But now: what to do – part out my old friend? Can’t let that 2A transformer go to waste. But I can’t hack apart an old friend. Aw, heck, went right back to Fry’s and snagged a TC-201A pencil to go with it. Now it’s the garage soldering iron. I think we have soldering covered here at the house.
Part 15, Low Power, FM Stereo Transmitter on a Budget
I have two good AM transmitters – one I’d built using a single 6888 Tube plus an old KnightKit Broadcaster that I’d refurbished, as well as a high-quality solid state transmitter from SSTRAN that I use to play music over the several antique AM radios I’ve repaired or refurbished.
I wanted a high-quality FM Stereo transmitter to stream iPod / iTunes output around the house and to my FM-band radios.
FM Stereo, however, is a bit more difficult to home-brew. I wanted to avoid the poor frequency control of the Ramsey FM-10C (with the BA1404 chip), and the low modulation of the little iPod FM transmitters you find for use in the car – although frequency control is quite good on these, the audio on these is just terrible. I’ve had about 3 of these iPod transmitters and they were all completely unusable.
You can get really GOOD FM transmitter kits but you have to go on up to $140+ to find a kit with suitable audio quality and frequency stability (think: Ramsey FM-25B).
To home-brew, first you have to build a stable exciter, preferably PLL synthesized, but the ICs for doing so are simply no longer readily available (Motorola MC145170, Plessey NJ88C30). Secondly, you’ll need to encode the left and right channels into Left+Right, Left-Right and tack on the 19 khz pilot tone, the 38 khz sub-carrier (See: Wikipedia, FM Broadcasting, FM Stereo).
The NS73M FM Transmitter module from Niigata Seimitsu Co. is ideal for this task. Unfortunately,
it needs a controller to setup the pre-emphasis, modulation level, frequency and power level.
And, if you’re going to use a controller, you might as well include an LCD so you can know what
frequency you’re on. I named this the “FM Stereo Broadcaster” since it reminded me ofthe old
Knight-Kit Wireless Broadcaster of the 1950’s (I have one of those too!).
I selected a Bare-Bones Board (BBB) from Modern Device Company (that I had on hand) to provide an Arduino controller.
The Arduino is an open platform, the development tools are free, and can be programmed in a variant of “C”
language. The LCD is a 16 x 1 device from AllElectronics.com
made by Varitronix. Finally the NS73M is provided on a convenient breakout board from
I found some initial code built by Cai Maver (Arduino + NS73M = ARRRduino!)on the SparkFun forum. The original (ur-code?) sample BASIC code from Sparkfun / ZAPNSPARK (Jim G.) gave the original ‘protocol’ for interfacing with the NS73M.
The code was first built with 3-wire mechanism using 3 digital pins (after the sample code)..
After some back-and-forth collaboration, he changed the Arduino to NS73 communication it to use
the I2C protocol (Arduino Wire.h library).
I added the 4-bit LCD interface and did some fancy-schmancy handling of the up/down/set buttons so you can
take the transmitter offline, change frequencies, and put it back on the air, and I added some code to save
and restore the frequency in EEPROM so the last frequency is restored on power up. The Feature-List
- Power-up and recall the last-known frequency
- Provide access to the entire FM-broadcast band (USA; code is easily modified for other markets)
- Allow the FM Carrier to be taken ‘off-air’ or ‘on-air’ as needed
- Show the current frequency and carrier state on an LCD Display
The project becomes a matter of not assembling discrete components so much as putting together 3 highly integrated modules.
The LCD4bit library was altered in only two spots:
1. Disable the RW Pin – the LCD RW pin is tied to ground (LOW). We’re only ‘writing’.
2. Change the Enable Pin from ‘2’ to ’11’ (use the unused RW pin).
The Arduino pins are budgeted this way: Digital Pins - D12 = RS (from LCD) D11 = RW (NOT USED - The RW pin on the LCD is tied LOW) D11 = Enable (from LCD) D10, 9, 8, 7 = 4 data bits for LCD D6, 5, 4 = UP, DOWN, SET buttons Analog Pins: A4 = SDA, A5 = SCL There are a few pins remaining for future expansion.
Final code is in this
Arduino Sketch for An FM Stereo Broadcaster.
As currently configured, the NS73M transmits at 2 mw power output, with a 75 us pre-emphasis, and 100% modulation to occur at 200mV of input audio. The first time it powers up, it will start at 97.3 mhz.
Afterward, the start-up frequency is remembered from the last time.
Everything is reconfigurable for other countries, including the FM Broadcast band edges
(87.5 mhz to 107.9 mhz USA), and the channel spacing (200khz USA). The 4-Bit LCD interface is as follows:
LCD is being used as Write-only, so we can save a pin by tieing RW LOW and disabling RW in the LCD4bit library.
Also the LCD4bit library was slightly modified to move the ENABLE pin from Arduino Pin 2 to the (now unused)
The Two LCD4bit library changes are two lines:
int USING_RW = false; // make sure the USING_RW value is set to 'false'... ... and Change THIS Line: int Enable = 2; TO: int Enable = 11; // making use of the now unused RW pin...
- Frequency stability is tip-top – I connected a frequency counter and it NEVER drifted.
- Transmitted Audio quality is superb – I don’t hear much hiss at all and the audio has great dynamic range, so FM modulation is quite good.
- Range – I didn’t expect much, but with proper input volume (iPod nano, about 60% volume), and a short (read: legal!) antenna it reaches my living room about 50 feet away!
- Frequency Agility – I’ve tested it down to 87.5 and up to 107.9 and other than some very small ’rounding’ inaccuracies, it reaches all of the channels on the US FM broadcast band.
- Cost – compares favorably to the Ramsey FM-10C ($45): the Bare-Bones Arduino($15), the FM module ($15, Sparkfun.com),
an LCD module ($5, Allelectronics.com), and some parts on hand (buttons, a 3.3v regulator,
resistors, trimpot for LCD contrast), but has the features of a Ramsey FM-25B ($139.95).
- I still need to package it in a suitable enclosure.
Finishing: since this is an RF project, an enclosure should be metal. I’ve settled on a Hammond 1455N1201 extruded aluminum enclosure – they’re easy to work with and I like the style. The datasheet indicates the RF Output is 50 ohms impedance, so a BNC Connector would be suitable. Each of the separate ‘modules’ (LCD, Arduino, FM Transmitter) can be mounted to a perf board and interconnected. Breadboard power is from a 5-volt lab supply, so a 5-volt regulator (and filtering) will be added to power the Arduino and the LCD; the NS73M uses a separate 3.3-volt regulator.
I initially forgot to add a level-shifter between the Arduino and the NS73M. The Arduino will produce 5-volt swings and this needs to be buffered to 3.3-volt swings in the Clock and Data. A great guide for this is SparkFun’s Tutorial on 3.3v Sensor Interfacing.
However, I avoided the more complicated solution (BS170’s on each side) and simply used a pull-up resistor to +3.3V on both I2C
pins of the NS73M. Connecting these to the +5V Analog pins (4 and 5) now restricts the up-side voltage to +3.3V, but allows enough
of a swing to assure a good I2C signal.
Finally, connecting the buttons (or the rotary control) requires debouncing of the mechanical switches.
A good look at this great tutorial on debouncing: http://www.ganssle.com/debouncing.pdf
will turn up the RC method (on or about page 12-14). I chose this for it’s simplicity and it’s quite good at
cleaning up either pushbuttons or a mechanical encoder.
- Notes on the Schematic
- 1/18/2011 Discovered a MISSING connection on the Schematic: Note that for I2C operation, “LA” must be pulled ‘high’ or 3.3V. So: Change U1, Pin 7 (LA) to go to +3.3V
- This is by no means a pre-bottled solution; you’ll probably want to tinker with it a bit, depending on what components or
enclosures you have handy (switchs, power choices, LCD types).
- I used the simple, cheater-way of handling the I2C – just two 10k pullups … to 3.3V! This limits the up-swing since the
Arduino is at +5V.
- S1, S2 and S3 – can be switches (version 1 of the code), or a Rotary Encoder (with a set switch) (Version 2).
- LCD code is written with a 16×1 in mind; 16×2 LCDs are now more readily available and cheaper. This just gives you
more display real-estate to use; tinker with the LCD output lines
- Not shown on the schematic – I used the ‘TEB’ output on the NS73M to provide an ‘On-Air’ LED. Just use a current limiting
resistor (maybe 1000 ohms to start). The TEB goes HIGH (+3.3V) to signal a LOCK (Low for Unlock), so this is a good visible check for ‘On Air’.
- Some builders put a small electrolytic (10 or 22uf) between the LCD Contrast and Ground.
- Remember this is an RF application; shield audio inputs, bypass early / bypass often.
- This is my first EagleCad schematic and I’m too new at it to know any ‘best practices’; please email any suggestions or errors. Thanks.
Source Sketch Files
- Version 1 – works with Arduino-0011, uses buttons for UP, DOWN and SET.
- Version 2 – works with Arduino-0012, replaces add-in LCD4bit library with the new built-in LiquidCrystal library, uses a single Rotary Encoder for Frequency change and Set.
- Version 3 – works with Arduino-0017, works with new, improved LiquidCrystal library.
I used a Bourns rotary encoder with a built-in push-button to replace the 3 buttons with a single control.
Final arrangement with an LCD, ‘On Air’ or PLL-Lock LED (in blue), and a rotary encoder with push-to-set switch for tuning.
Initially used a 16 x 1 character LCD from AllElectronics.com — A Varitronix #MDL16166.
If I were making another unit, I’d use a 16 x 2 display.
Ground seems an optional item, as it works very well with just a short whip.
Here the FM Stereo Broadcaster is connected to a TM-100 Clone
for an antenna.
Screws are standard-head, socket-cap screws with a black-oxide finish (McMaster-Carr: 91251A148)
This is the essance of a ‘modular’ project: The FM Transmitter IC, the LCD, the Arduino that controls them all, along with a bit of inter-module glue: a 3-to-5 volt interface from the FM IC to Arcuino, 3.3 volt and 5 volt regulator ICs, and a RC-debounce circuit between the mechanical rotary encoder and the Arduino.
The 16 x 1 character LCD from AllElectronics appears very faint when looking directly on, or from above. Investigating other LCD options turned up some similarly sized LCDs that will probably work better; the final version (at left) uses an LED backlight and had the exact same measurements as the Varitronix MDL16166 (to fit the opening!).
Screws are flat-head, socket-cap screws with a black-oxide finish (McMaster-Carr: 91253A148). They work well and look great with these
extruded aluminum enclosures.
What’s it take to get the nice, square, straight LCD hole in the front panel? About 30 minutes with a Nipper tool.
- Mark the exact opening, making sure it’s square. I use a red Sharpie, but on the back side.
- Drill a hole large enough to admit the end of the nipper tool.
- Nip, Nip, Nip. Nip around the edges, always – ALWAYS – noting the position of your nipper against your red line.
This is for consistency more than accuracy in size.
- Nip slightly small; you can always use a file to ’embiggen’ the opening a bit. It’s less desirable to have
the exact-sized opening than an even, square one.
Any mistakes are probably my own. Thanks very much to Cai Maver for his original work on the Arrrduino FM – I’ve only extended his code a bit to include EEPROM store and some of the LCD behaviors. Many thanks to all the superb folks on the Arduino.cc website and their libraries, code samples and tips.
This project is not guaranteed for any specific outcome or purpose. I cannot be held responsible for illegal uses. This project is solely for educational, hobbyist, and experimental purposes.
Please check your local (and national) regulations regarding unlicensed transmissions. Do not interfere with other, licensed transmitted signals. FCC’s Part 15 Regulations are recommended reading, particularly
Bulletin 63 (October 1993)
“Understanding the FCC Part 15 Regulations for Low Power, Non-Licensed Transmitters”. This project describes an “Intentional Radiator” and as such (operating in the band 88 – 108 Mhz), according to Section 15.239 (b): “The field strength of any emissions within the permitted 200 kHz band shall not exceed 250 microvolts/meter at 3 meters.” Using their calculations, this works out to about P = 0.3 E2 watts, or 0.3 x (250 x 10-6V)2,
or 0.3 x 0.0002502, or: 0.00000001875 watt (.01875 μ-watt).
Updated May 13, 2009
Don’t know how I missed this one, but there’s a great rule-of-thumb allowance for AM and FM, unlicnesed, low power broadcast transmitters. The Document is from July 24, 1991:
Page one has the technical power rule: AM – .05 watts (or 100mw to the final RF), and FM – 0.01 microwatts. But since these are difficult to measure since calibrated RF meters are expensive, the general intent of the rule is defined as an ‘Approx. Maximum Coverage Radius’ or 200 feet (radius) for both AM and FM low power transmitting.
Updated May 13, 2009
Found a swell link that produces a list of unused FM Frequencies in your area. Searchable by City / State, or by Zipcode, it will return a graphic showing the signal strength of the stations in your area, plust a list of “best”, “better” and “good” frequencies. For example, in Dallas / Ft. Worth, Texas, there AREN’T any unused frequencies (!), but it gives a list clear of all but the weakest stations.
This made it to HackaDay.com. The project has an enclosure now, although I wish I’d used a back-lit and 2-line LCD. Too bad – the hole is already cut.