The SW200 – A Budget AM Broadcast Processor for LPAM – and Hams Too?

Note that although the audio processor described in this article is intended primarily for Part 15 AM broadcasters, and perhaps also low power licensed AM broadcasters on a budget, it may also have applications for hams with their AM endeavors.  I’m thinking of the AM enthusiasts on 160, 80, and 40. Some of them are ex-broadcast industry professionals who like to use vintage broadcast equipment in their stations, including what one might call “legacy” audio processing gear, such as the CBS Audimax and Volumax. This little processor may not have the looks of classic vintage gear, but some hams are Part 15’ers and also do AM on the ham bands. This could be a useful box for them to have around.

It’s not news that I haven’t been doing much homebrewing for a while. Sorry about that – I really feel as if I’ve let you down somewhat! However, I haven’t been completely away from radio, and I have been transmitting, though not on the amateur bands. To be more specific, I have been broadcasting 24/7 since July 2017 in my immediate neighborhood, on the AM broadcast band.  Back in 2010, I posted about my first experiments with broadcasting under what, in the US, is referred to as Part 15. Part 15 refers to the section of FCC rules on low-powered devices that emit RF and are allowed to do so without a license. Baby monitors, garage door openers, many cheaper wireless microphones, FM iPod transmitters, and WiFi routers, are all examples of products that are allowed to emit RF without the need for a license. The allowable amount of radiation varies across the RF spectrum. Operation in the FM broadcast band (88 – 108 MHz) is covered by 15.239 and specifies that the field strength at a distance of 3 meters from the antenna shall be no more than 250µV/M. This is not a lot and, in practice, won’t get you more than about 200 feet of range. A receiver with a gain antenna might eke out a bit more range, but unless you’re living in a very densely populated area and use multiple transmitters, you’re not going to get many listeners if, of course, broadcasting is your mission. Most people utilizing Part 15 on the FM band are doing it to broadcast tunes, radio shows etc around the house and for this it works well.

To the folk who are purposely using the FCC Part 15 rules in order to try and reach a radio listening audience, 15.219 is the most interesting and useful rule. Unlike most, if not all, of the other rules under Part 15, this one, which covers emissions in the 510 – 1705 kHz band – the AM broadcast band, doesn’t use field strength as a limitation at all. Instead, it specifies a maximum allowed DC input power to the final stage of 100mW (0.1W) to a combined antenna + ground lead length of no more than 3 meters. Even at the very top of the AM BC band, the wavelength is about 176 meters, so a 3 meter antenna is going to be very inefficient. Nevertheless, with a good ground system, coverage of up to about a mile radius can be achieved with one transmitter, depending on the efficiency of the ground, obstructions, and local noise levels. People like me, who are in locations with obstructions and limited ability for extensive ground systems, achieve proportionally lower coverage. There are a number of folk who cover their entire small towns with one Part 15 AM transmitter. Others, like Radio Sausalito in Sausalito, CA use a network of linked and synchronized transmitters to provide continuous coverage across town. Other enthusiasts are content to broadcast over a smaller area, covering a few blocks, or perhaps just their house and a few neighbors’ houses.

I’ve thought about detailing my Part 15 operation in a post here, but don’t want this blog to stray too far from it’s main subject, for too long. If you want to inform yourself more about how to operate a legal and FCC compliant Part 15 broadcasting operation, the site and forums at Hobby Broadcaster.net are Part 15 central. Owned and run by long-time broadcast engineer Bill DeFelice, there is a wide range of technical resources, equipment reviews, and discussion forums available to registered members. Bill actively and carefully moderates the forums, keeping the discussions on point, and relevant to the topic of legal and compliant Part 15 operation. If you’re wanting to know how to build your own legal micropower broadcast station in the US or Canada then hobbybroadcaster.net, in my opinion, is the best place to do it on the internet. You do need to register in order to get full access to the forums and all resources, but it’s worth it.

Part 15 operations run the gamut from radio stations at educational institutions with a full cadre of volunteers running the programming, to church groups broadcasting services and announcements, to enthusiasts operating stations that cover their small towns or neighborhoods in larger areas, to……. well, you name it really. One gentleman I know of switches his station on at specific times during the day and broadcasts the school lunch menus for the week to his neighborhood. You can get started very cheaply if your budget is limited. New transmitters begin at around $100, if you want to get your feet wet without a large outlay. At the other end of the scale, the Chez Procaster, at about $700, and Hamilton Rangemaster at close to $1000, depending on options, are generally considered to be the best performing AM Part 15 transmitters, putting out the most power for the allowed 100mW DC input, and having the best modulation quality. They are also the most expensive.

The Hamilton Rangemaster and ChezRadio Procaster are good transmitters, but they’re not exactly cheap.  I did think about building my own Part 15 AM transmitter from scratch, and did some initial research to that effect. However, I was a little burned out with scratch-building, from previous projects, and wanted this project to be more about the programming of the station and engineering of the overall system. In addition, I didn’t know how much I could expect from a Part 15 transmitter in terms of performance, so I figured that by purchasing one which is considered just about the best, it would act as a benchmark for any future installations and experiments.

On deciding to have another stab at running my own little micro-broadcast station back in the early part of 2017, I ordered a Hamilton Rangemaster transmitter, and put it on the air with programming provided by a piece of automation software called Zara Radio. I had been working on the programming on and off, for several years, and it was ready to go. Here’s a simple diagram of the  airchain at the time –

The airchain consisted of a cheap refurbished laptop fitted with a low-priced outboard sound interface (which gave better audio quality than the laptop on-board sound). Thanks to hobbybroadcaster.net member Carmine5 for the suggestion of using the Behringer UCA202 to improve the sound quality. The very original form of audio processing I was using was a simple musician’s single band compressor, the Alesis NanoCompressor. The main issue I experienced with this, was that on tracks with heavy bass, the bass thumps caused the compressor to turn down the gain on the entire track (being a single-band compressor), causing the audio to “pump” with the bassline. Moving to the Behringer SPL3220, with it’s dual-band compressor, was a big improvement. I have hobbybroadcaster.net member Carmine5 to thank for that too. His review of the SPL3220 processor is here. I purchased mine brand new for $99 inc. shipping, and it was a heck of a deal at that price.

At that point, I was fairly happy with the sound of my station. The dual-band compressor in the SPL3220 was doing a much better job with the audio than the single-band NanoCompressor did. To be fair to the NanoCompressor, broadcast audio processing is not what is designed for. In addition, the SPL3220 has an AGC/leveler on the input, and limiting on the output. It’s a good budget processor, and the only major omission, for a broadcast processor, was pre-emphasis. Carmine5 uses outboard EQ to provide pre-emphasis for his transmitter, and it works out well for him. I don’t have an EQ unit, and was starting to feel the need for a “proper” broadcast processor. As I haven’t done this sort of thing before, I was looking for benchmarks to be able to compare future equipment acquisitions with. The Hamilton Rangemaster is a good reference to compare future transmitters to, whether home-brewed or purchased. In addition, and for the same reasons, I wanted to know what a complete broadcast audio chain sounded like with my station.

This was the point at which things got exciting.  Jim Wood, the founder of Inovonics Broadcast, decided that he wanted to design and manufacture a budget AM broadcast processor with Part 15, TIS*, and low power licensed broadcasters on a budget, in mind. This was not going to be an Inovonics product – it would be solely the work of Jim, and would use commonly available through-hole components, a 2-layer circuit board, and analog circuitry. The brand name of Schlockwood, a fun play on Jim’s name, serves to draw a clear distinction between this product and those developed by the company he co-founded over 45 years ago, and is still closely involved with. An excellent user manual is included with the processor and extra information, including a full engineering brief with block diagrams, full schematic, and notes on circuit operation, are available on request. The whole idea is to make a processor available to broadcasters on a budget who, with a little electronics experience, can keep the unit running for many years into the future. While on that subject, unless you are mistreating the SW200, you are most likely to be able to let it sit and run 24/7 for many years with no issues. Perhaps after a couple of decades of service, you might want to install new electrolytics. If you’re treating it nicely, that may well be all you’ll need to do. Oh – you might need a new wall-wart as well.

For several months during 2017, a small focus group of Part 15’ers offered input to Jim as he designed and breadboarded the processor, then transferred the design to printed circuit board, and produced the first prototypes. One of these went to Bill DeFelice, and the other found it’s way to me. The term “pre-production version” is a more apt description, as the unit I received was very, very similar to the final version. The only differences were –

 

1) 3 extra components, a molded choke (L1), a resistor R23), and a capacitor (C12) were added to the board to form a low pass filter at the input. It filters out everything above audible frequencies, as there is a possibility that if RF appeared on this line, it could heterodyne with harmonics of the 100KHz PWM oscillator, producing birdies that would appear at the output of the SW200. It is for this reason that it’s a good idea to keep the SW200, which is not housed in a shielded enclosure, away from strong sources of RF. These 3 LPF parts, incidentally, were present in the pre-production version, but were soldered in place underneath the board. Pulse-width modulation (PWM) is used throughout the SW200 for signal level control. As in a light dimmer, the signal is switched on and off at a very fast rate, with the ratio of on to off determining the degree of attenuation. In the Engineering Brief, Jim says that PWM is straightforward to implement, and colorless. FET’s can be tricky and drift-prone, while VCA chips can be costly. All the chips that Jim ended up using are widely available – no proprietary devices here that are likely to become unavailable anytime soon. It’s a little like the IC version of building something using all 2N2222’s – even if society collapses, you can be fairly certain you’ll be able to find parts to perform repairs!

2) The front and rear panels in the final production version are aluminum, and connected to the ground plane, whereas in the pre-production version, they were plastic. In both cases, they are covered with a printed vinyl overlay. A professionally-produced polycarbonate overlay would have increased the cost unnecessarily, contrary to the main mission of this project.

3) My pre-production version had sockets for the IC’s; in the final version, they are soldered in.

4) This one is very, very picky, but I have an eye for detail, so here goes. The round holes in the printed vinyl overlays on the front and back, for the DC power connector and the LED’s are very slightly ragged in the prototype. You can just about see it in the photos in my User Review on hobbybroadcaster.net linked below. In the final version (shown in this post), they are nice, clean, round holes, making for a very agreeable finish.

 

My personal favorite Part 15 site, hobbybroadcaster.net, has already published a lab review, and video review, both by site owner Bill DeFelice, and a user review by yours truly. All the reviews contain slightly differing information, so if you’re interested in this processor, please check them all out. They are here –

SW200 LPAM Broadcast Audio Processor Lab Review By Bill DeFelice of hobbybroadcaster.net

SW200 LPAM Broadcast Audio Processor Video Review by Bill DeFelice of hobbybroadcaster.net

SW200 LPAM Processor User Review by Dave Richards AA7EE

I was very lucky to be able to have a pre-production unit running in my home broadcast chain for a few months. It was in constant service during that period and, when the time came to return it to Jim, I knew that I needed one in operation permanently here, and immediately ordered a final production version. The first 3 photos, of the unboxing and unpacking process, are of the pre-production version. All subsequent images are of the final production unit. However, the differences (described above) were very minor, as Jim had already finalized the design before shipping out the evaluation units to Bill and myself.

This was the intial pre-production version hidden in it’s packing because, you know, when you’re very excited about a new acquisition, you need to document the whole process –

A picture of a Medium Flat Rate US Priority Mail Box filled with packing material. There is an SW200 LPAM Processor in there somewhere.

Removing the top layer of packing revealed a modestly-sized box of – excitement, with a manual on top! If this sounds as if I’m going a little heavy on the hyperbole bear in mind that, at this point, I had assembled an almost complete broadcast chain – with the exception of comprehensive processing. This box represented the final link –

One final view of the pre-production version in it’s packing –

 

This is what it looks like, with the included wall-wart, and a steel rule for a sense of scale (it’s 8″ wide). The regulated power supply, Triad brand,  comes in it’s own small yellow and black box (not pictured). This processor is built into a readily available Hammond project enclosure (Hammond part #1598DSGY), a move designed by Jim to keep the cost down. At $11.96 plus shipping from Mouser, it’s a lot cheaper than a custom-made and printed steel enclosure. I have already purchased an extra, “just in case”. In truth, I’ll probably never need it and at some point, it will end up being used to house a home-brew project. Perhaps a nifty “On-Air” light to sit above the processor?  The front and back panel printed overlays are vinyl stickers, which help to keep the cost down, and give the unit a very presentable appearance –

The SW200 accepts either balanced or unbalanced inputs and outputs, via 1/4″ TRS or XLR connectors, and gets it’s power from a wall-wart power supply that outputs 18V regulated and floating. The floating supply is used internally to provide ±9V rails for the op-amps –

SW200 Rear panel. Both the XLR and 1/4″ TRS jacks can accept either balanced or unbalanced signals. The User Manual has details on these connections.

 

SW200 Front panel controls and indicators. The larger LED’s, for AGC Gain, and Triband Limiting, are bicolors, for effective real-time indication of processor operation.

 

The SW200 contains an entire broadcast audio processing chain in one small, light package. It consists of the following –

 

-An initial gated AGC/leveler stage. It’s an AGC with a slow time constant, so as not to interfere with any of the subsequent processing in later stages. Basically, it ensures that the input audio is roughly at the level that the rest of the processor can work with. Think of it as a guy with his hand on the master volume control, slowly turning it up during the quiet passages, and gradually turning it down when things get a bit too raucous. I understand that in the early days of broadcasting, there was a fellow whose job it was to do just this. Imagine having your job replaced by an op-amp, a CMOS switch, and a handful of passive components! The gating impedes the AGC during pauses in the program audio. This guards against the noise floor rising too much during prolonged quiet periods. I’m not sure if you’ve ever worked as an announcer at a station that had no gating on the AGC, but I can assure you that it keeps you on your toes. For me, the station in question was a Top 40, and the lack of gating, combined with the heavy compression, kept my voice breaks snappy and, until I gained some experience, ensured that I sounded nervous when on the air. I was too scared to stop talking!

-Pre-emphasis. Modern AM BC band receivers tend to have limited bandwidth, which can de-emphasize the higher frequencies. As an attempt to compensate for this, pre-emphasis of the higher audio frequencies is added at the transmission end. The SW200 has a “Peaking Pre-emphasis” control which, when rotated fully clockwise, has the standard NRSC pre-emphasis curve of a straight-line boost from 1KHz to about a 10dB boost at 10KHz.  Many modern receivers, though, are so limited in bandwidth, that this pre-emphasis doesn’t make a lot of difference. When the control is rotated anti-clockwise, the 10dB treble peak moves to the left, providing an approximately constant 10dB lift at any frequency down to 3.5KHz. In this way, you can add presence to your signal as heard on receivers of more limited bandwidth, bearing in mind that it will sound somewhat more shrill on older receivers. It’s a balancing act.

-Tri-band compression. There are two main benefits to multi-band compression over using a single-band compressor. Firstly, it will help to make the program signal sound more dense, by boosting the parts of the audio spectrum that typically contain less energy. These are usually the low and high end, so your signal ends up sounding brighter as well as more bassy. Basically, it gives your signal that “dense, busy, bright, and beefy” sound that is commonly associated with AM broadcast stations. Secondly, if you have ever put music through a single channel compressor, you may have noticed that on bass heavy tracks, the bass thumps cause the volume of the whole track to “pump”. It can be quite disconcerting. Multi-band compression avoids this by compressing the mids and highs independently of the low frequencies. There is a Processor Drive control, that adjusts the amount of drive to this stage, and thus the depth of compression (and, as a result of course, the dynamic range). Also associated with this stage are separate Low EQ and High EQ controls, which, beyond the influence of the Processor Drive control,  allow the user to exercise a little more dominion over the bass and treble frequencies.

-Final Peak Control via the functions of limiting and clipping. The SW200 provides both of these, and the SW200 User Manual describes it in greater detail – as it does with all the functions of this processor. While compression increases the program density, the Final Peak Control directly influences the perceived loudness of the program signal. Being able to control the ratio of limiting to clipping allows you to decide whether you value smoothness in your station sound, honest-to-goodness loudness, or a combination of the two. Regardless of what the ratio of limiting to clipping is, the absolute maximum output level is set by the Output Level control; the limiting and clipping are the processes via which the output audio to the transmitter is guaranteed not to go above a certain level.

-For transmitters (like the Rangemaster and Procaster) that will support asymmetrical modulation, the SW200 will drive the transmitter up to 135% on positive peaks. Licensed broadcasters will need to limit their +ve excursions to +125%, but us Part 15’ers have no such limitations imposed on us. We are free to turn it up to 11, baby – provided we can live with any slight audio distortion that might result. The way this is achieved is by applying a bias voltage to the peak limiter/clipper transistors via the multi-turn pot that is accessible from the front panel. Firstly, with the positive peaks pot turned to minimum, the output level pot is adjusted for close to 100% modulation, ideally with the use of either an oscilloscope or a modulation monitor. Then, as the positive peaks pot is turned up, the increasing bias on the output stage shifts the positive excursions of the audio waveform further into positive territory, beyond the 100% modulation mark, while keeping the negative excursions at the same sub-100% level. We want to avoid >100% on the negative peaks like the plague, as this causes sideband splatter, as well as sounding terrible.

If you have any interest at all in the inner workings of the SW200, I’d recommend that you drop Jim Wood a line, using the contact info on his website at the bottom of this post, and request further information. The Engineering Brief is very well written and illustrated, and contains a detailed description of circuit operation.

Block diagram of the SW200 LPAM Broadcast Audio Processor, from the Engineering Brief.

A few more features offered by the SW200 –

 

-A mix-to-mono utility. This sums 2 stereo channels to mono, and can be set by a jumper on the board.

-A 5KHz filter cut-off option, set by on-board jumpers. The default 10KHz filter is what most Part 15’ers are going to want to use. Why limit your higher frequencies if you don’t have to? This default option doesn’t meet the FCC requirements for licensed broadcasters, but the 5KHz filter does. Licensed broadcasters will want to set the 5KHz filter. See the diagram below, taken from the Engineering Brief –

 

-The ability to disable the Pre-Emphasis and/or the AGC/Leveler sections completely. Normally, there would be no reason a user would want to do this, but it might be useful for experimentation and comparison purposes.

 

The circuit boards are traditional 2-layer types. The upper board holds the LEDS and driver circuitry, and is connected to the main board by a ribbon cable. The front and back panels are aluminum, and are connected to the groundplane on the boards. The upper and lower halves of the enclosure are plastic, so the unit is not screened. It would be advisable to keep it away from high RF environments. In the radio room here, it is located just a few feet from my low power ham gear, and the 300 ohm balanced feeder that leads to my antenna. That feeder carries a maximum of 10W on the HF amateur bands, and I didn’t experience any issues with interference. Probably best not to situate it next to the antenna feeder from a high power transmitter though. Some common sense is advised 😀

The boards, and front and rear panels form a cohesive unit. The slots in the upper and lower clam shell halves simply slide over the panels. The halves are completely stock – they have no holes or cutouts drilled in them at all. If you were ever to damage the case, or modify it, and later want to return it to stock, it would be a simple task.

The front and rear panels are fairly thin, and light, aluminum – this is by no means a heavy rack-mount steel unit that can be bumped and knocked around. It is, however, very repairable. In fact, the only part of this processor that could be a little tricky to replace would be the vinyl overlays, were you to damage the front or rear panels. If this did happen, I wouldn’t be at all surprised if Jim were to provide a file for printing a new vinyl overlay. Alternately, you could always drill a new front or rear panel from the panels that come with the Hammond enclosure, and either design your own overlay to print, or label them with Dymo. At that point, you could consider this approach as giving your SW200 a look that is unique to you – a feather in it’s cap, and a symbol of it’s long and trusty service!

In the next view, in the bottom right-hand corner, the 2 blue rectangular components are the Bournes 15-turn pots that are used to set the Output Level and Positive Peaks –

The SW200 was easy to set up – a fact that makes it appealing to even the first-time user. Some audio processing gear requires extensive experience and knowledge in order to get maximum benefit, and often to avoid producing sub-par results. This is definitely not the case with the SW200. In fact, I’d go even further, and say that it’s actually pretty difficult to produce an unlistenable sound with this processor. Basic setup, once the input and output leads, and power cord are connected, consists of –

1)  Adjusting the Input Gain control so that both AGC Gain LED’s are shining green while representative program material is playing. This is not a critical adjustment – we are only looking for a ballpark here,

2) Turning the Pre-Emphasis control fully clockwise, for the NRSC standard pre-emphasis boost,

3) Turning all other knobs to the 12 o’clock position. These are good default settings for a starting point,

4) Turning the Positive Peaks trimpot to zero (it doesn’t have an endstop, so if you turn it anti-clockwise 15 turns, this will ensure it is at minimum), and adjusting the Output Level trimpot so that you are close to 100% modulation on your transmitter. If you don’t have a modulation monitor, you can approximate this by increasing the control until you hear distortion, then backing it off a little. The Output Level control is actually the most critical, as if you go over 100% modulation on the negative peaks, your signal can sound pretty awful. This, of course, is not a fault with the processor – it’s the way that amplitude modulation works!

The photos of  my SW200 in this post show the controls in the positions that I ended up at for the settings on my station. If you look at the close-up of the front panel earlier, you’ll see that I have the Processor Drive set to the maximum (+15dB) position. I was not expecting to find that the unit would produce such listenable sound at this setting, but it does. Heavy compression often becomes fatiguing to listen to after a while, but I’ve been listening to my own little station for a month or more with maximum processor drive, and haven’t noticed any such issues. It sounds good. If your program material is heavy on music that requires high dynamic range, such as classical, or some forms of jazz, you won’t want to drive the triband processor as hard. Similarly, if you are broadcasting mainly talk, you’ll want to go a little easy on the drive level.

One more thought before I wrap up this post and hit the “Publish” button. For many, the natural companion to the Rangemaster has been the Inovonics 222. The Rangemaster documentation has a description and diagram detailing how to connect them. The 222’s main functions were as an NRSC compliance processor, providing audio filtering to keep licensed stations strictly within the 10KHz FCC limit, as well as support for asymmetrical modulation, and peak limiting. There was a certain amount of dynamics compression inherent in the design, such that some users found it was all they needed in front of their transmitter. Others added the AGC action of an Aphex Compellor and were also happy. On using the SW200, the one thought that keeps popping into my head is that if the 222 was a good companion to the Rangemaster, then the SW200 is the perfect electronic pal, representing as it does, a complete broadcast audio processing chain. The 222 is no longer manufactured, but the price of the SW200 is significantly less than that of the 222, when it was available new. You can still find used 222’s for less than the price of a new SW200, but the likelihood is that it will have been in constant service for many years, and close to needing new electrolytics (if not already needing them), and perhaps some other servicing. Conversely, the SW200 is brand spanking new and ready to go. The one area where the 222 is potentially superior to the SW200, apart from it’s physical ruggedness, is the tight filtering. Licensed broadcasters will find that a compliance processor designed specifically for them, such as the 222, will allow their on-air signal to have as much high-frequency response as possible within the limits of the law. The 5KHz cut-off filter option in the SW200 will keep licensed operations within the law, but at the expense of a few of the higher frequencies. Once again, if you’re a Part 15’er, this isn’t a consideration, as you’ll be using the wider default filter on the SW200 anyway.

I guess what I was trying to say in the previous paragraph is that, like the well-known song, in my opinion, the SW200 and the Rangemaster go together like a horse and carriage, or love and marriage. They are the perfect intrepid Part 15 AM broadcasting duo. I have no doubt the SW200 will play well with a Procaster or other similar transmitter too. The only Part 15 AM transmitter I own though, is the Rangemaster, which is why all my comments have focused on it. As the Procaster comes with simple on-board processing, many users will be happy with that. For those who want to develop their processing a little more though, the SW200 would be a good choice.

I have been very happy with my Schlockwood SW200 so far. It is on 24/7, and even during the times when I am not monitoring and have the AM receiver turned down, I can glance over at the front panel, and see all those LEDs changing color and blinking, which they do constantly. Very gratifying! Please let me know if you plan on purchasing one, or actually do, and what you are using it for. Enquiring minds here want to know!

 

More details, and ordering information on the SW200 can be found at –

The Schlockwood Laboratory

*TIS = Travelers Information Service. Several different usages fall under this umbrella, including stations in cities and airports that broadcast local emergency information, as well as installations in National Parks, broadcasting visitor information.

Advertisement