Tuesday, August 7, 2007

KD6VKF HF Handheld SSB QRP Radio ~ Design

Potential Handheld Circuit?


A search began for an off-the-shelf case for the prototype with these parameters. A likely off-the-shelf extruded aluminum case was found: Context Engineering Co P/N 2506 [E], which would fit a couple of 2.5” x 4.0” PCBs.  So this was the exact size of the board I would use

Handheld SSB radio!

So with the board dimensions, and a schematic in hand, a preliminary PCB layout began. I set out to see if a proto would really fit into the selected case and on these size boards. I drafted the schematic in to the PCB CAD software, and started to move components onto the virtual PCB. As this effort went on, some days it seemed likely and others it seemed like an absolute impossibility. Eventually- I managed to squeeze most of an entire transceiver, minus the VCO and RF interface, on that board layout.

After I completed the preliminary layout I realized there was a problem using this case; the radio board I came up with alone took up almost the entire 4” of the case.  This made mounting the controls and connectors I wanted on the end panel next to impossible; I would need to shrink the board by another half an inch or find another case.  Although reducing the layout by 1.25 sq. inches does not sound like much the design could not shrink much further without going to SMT or doing a lot of clever design work, either way it meant a new board layout was needed so it looked like I would have to come up with another chassis design to use this board. Then it occurred to me to buy a second extruded case and cut it in half.  This was an effective way to increase to about 6” of length as the extruded parts are all mirror images. It was one of those 4AM inspirations.

Case Layout–

So now, with plenty of room to spare, I could get the radio in there but what would it actually look like? As I sketched out the layout I thought it began to look more like a ‘90s “brick” VHF radio but the hand held concept would still be intact. This diagram also shows the maximum extended length using the two half extrusions as inserts.  The parts overlaps actually make a sturdy unit when connected together.






Off the shelf ~ Extruded alum chassis

Internal dimentions
Top Panel Layout

One of the first things you notice that most handhelds use concentric controls. I did a search for such units but came up empty. I could strip some out of non-working surplus devices but this makes the project hard to duplicate. So instead I ordered some of the smallest panel mounted controls, with switches for the VOL and SQL, I could find. The other major control is the DDS VFO encoder.  The unit I chose was actually intended for PCB mounting, but I tack soldered wires to the pins instead. Here is the layout I came up with and it seems to fit all the needed parts.

Model 610 ~Control Panel

6M - 10M (Model 610) Control Panel

Selecting the right parts for the panel was a key to this design layout.  Every component including the knobs were measured and checked for clearance.  The main issue found was the tabs for the controls and switches need to be bent at 90˚   for clearance inside. It’s tight but seems to work OK.


Preamp for whip antennea ~ if needed

I guess the whole idea of this project is to be able to demonstrate that a portable handheld HF station, under certain conditions, could be used to make transcontinental and possibly intercontinental contacts.  I still don’t know what the limits will be as the solar cycle climbs.  Internal batteries are the ultimate in portability. Although the obvious drawback of how much talk time can be had form the necessarily smaller cells can be overcome in part, by a larger external battery power pack attached by a short cable.  A 12V pack of AA, C, or even D cells is not too difficult to make and transport. A small 12V gel cell will certainly provide enough power for even days of intermittent operation. 

The antenna becomes the next part of the station to consider.  I have experimented with a Hamstick antenna [19] attached directly to the unit through a couple of adaptors. (Lakeview Cat. # 275 3/8” to UHF and a UHF Plug to BNC Plug adaptor Bomar Interconnect Part #: 3307505 [C] )  I got the whole assembly to resonate by attaching the antenna, adaptors, radial wire, and all,  to an antenna analyzer and making the appropriate length adjustment to the whip part of the antenna.  You can probably get a reasonable match by listing to the received background noise or tuning to a weak signal and adjusting the whip.

[Figure 16 Antenna Connector / Adaptor Arrangement]

Since these antennas are intended for mobile operation and are thereby necessarily a little on the robust side.  The resulting configuration puts a bit of stress on the unit’s BNC connector, but not too unreasonable for intermittent operation, but lighter weight materials should be a little easier on both the equipment and operator.   It feels a bit like holding on to a large surfcasting rod.  Antennas made of lighter weight materials such as a helically wound wire antennas on a lighter fiberglass rod or even the “tape measure” folding antennas used on military man pack radios (and even some OSCAR satellites )  literally made out of tape measure steel refills, may work. Note, with any type of whip antennae a tuned radial must be connected to the provided wing-nut.  This wire keeps the RF potential at the connector feed point to a minimum, reducing the effects of RF on the controls mounted on the same panel. (The maximum RF voltage potential should appear on the tuned radial instead) The technique I found that works is to lay the radial wire out first, attached to the radio, and hold the radio in one hand in front of the operator.  This forms a slanted/folded dipole arrangement that should yield NVIS type results with the operator’s back to the direction most likely to propagate. 

Since I doubt that a shortened whip antenna of any kind will yield the best results, they are also by nature have a narrow-bandwidth, so you probably won’t be able to have low SWR on the entire band.  The second approach is to use a ¼ or ½ wave dipole and erect it attached to appropriately insulated and elevated supports, (like trees, boulders, etc.) connected with a length of lightweight coax connected to the station. At these upper bands this arrangement would not take much wire, 10’s of feet, which can easily be carried on a small reel or spool.  Be aware of the orientation as these types of antennas, particularly the horizontal and slanted arrangements, are directional. The radial wire may also help the RF situation on the panel if it should develop.  The operator again holds the unit in front versus “telephone style” to use. A tripod mounted vertical with its own set of radials connected by cable as well may be another option to try.



More complex transistor SSB Radio




The LCD actually fits in between the slots.  The figure shows where the screw holes may go to attach the LCD to the case if needed.  As it turned out, the controller board was designed to fit in the slots and the LCD attaches to it so these holes were not needed. The other two things on the front were the speaker and microphone.  Also, I wanted the speaker on the bottom, to avoid the temptation to try to use the unit like a telephone. This layout shows a 1.6” speaker I found fit in place on the front 2” section.  The case should actually be able to take up to a 2-½” diameter speaker for a little more sound oomph.  The microphone simply sits behind the speaker or a small hole in one of the front sections.  The speaker area also contains the power and external speaker connectors.  The upper rear 2” section was used to hold the remaining RF power output circuits. Even with the ½-inch of space in the top of the case reserved for the controls, the LPA fits nicely on that small piece, using it as a heatsink as well.








Tuesday, July 31, 2007

KD6VKF HF Handheld SSB QRP Radio -Concept

SUBTITLE: “…Adding a form factor to circuits with this (almost) self-contained HF radio station.”



PART 1 – INTRODUCTION
This whole idea started way back before solar cycle 23!  I was working with some interesting circuits for the US Navy at the time including a new DDS chip that I knew would provide a stable VFO for such a unit.  (The now famous Analog Devices DDS series) It was one of those daydream projects that never fully materialized. I started to design and analyze various transistor RF circuits with the Radio Designer software [1]. This project fell behind as it was impossible to design a working miniature SSB transceiver after I began working with a struggling startup company full time.  I briefly hit on an idea during that period to simply repackage an existing transceiver into a hand held format.  I even bought a WM-20 [2] to kit bash but it was just too cool not to build it intact.  I eventually did interface the WM20 to a DDS VFO [3], in a 19” rack format - not a very hold-able format to be sure...  The solar max came and went and my dream of this radio faded too.

My interest was recently again peaked by a question from my dad; “…what’s the deal with these coast-to-coast ‘walkie-talkie’ phones? It sounds a lot like ham radio to me.”  After explaining some the cellular technology and marketing behind those phones I remembered this project once again.   Could my dream of a transcontinental handheld HF radio be revived?  It looked like I could use surface mount components (after all SOT transistors are real small) to make a cool mini version of some transistor only design?  I wanted to use components that could make this reproducible, no obsolete components (like the MC1350) and simple design controlled by a small microprocessor.  Then there is the upcoming solar max... Can it be done in time for Cycle 24, if it ever arrives, when the high ends of the bands are again active?



Radio Control

Meter Disp functions

RF Block Diagramme
CONCEPT
The concept for this design would be a single band transceiver; the RF elements could be re-populated and re-program the controller to make different band versions on all of the HF bands, 20 meters and up, hence Model 620!  [5] This transceiver is designed for USB operation so it is compatible for the upper HF band plans. The radio for this project may seem by some almost as an afterthought but the point of this effort was intended to be the re-packaging of some “proven” existing design.  It turned out to be a bit of a cut & paste design, so you may see one or two of your favorite circuits included. I tried to give all the credit where it is due for these circuits; some are just public domain by now so often reproduced. There is a lot of design work involved in designing even a simple transceiver from the ground up.  As well as all of the mechanics of packaging a working transceiver in this form factor.   It turned out; 3 separate PCBs would be needed; RF Board, Controller Board (initially with a DDS-60 [6] daughter board), and PA Board.   This project presents some unique problems and challenges, not only stuffing a QRP transceiver into a pre-defined form factor, but some lessons in RF compatibility as well.  The control panel shares not only the majority of the user controls but the antenna RF connector. Then the issue of making the unit totally self contained with batteries – a whole slew of problems as well. 





DESIGN

With the basic features nailed down, the next question was: how big?  It would be possible to squeeze the radio into this recognizable format using thru-hole components, or would I need to use “the dreaded” [7] Surface Mount Technologies? (SMT) package sizes for some of the passive components available are so small that it is not practical to expect to hand solder and assemble the 100’s of components needed for this design.  My personal size limitation for PCB work is 0602, even then it is very difficult to manage to install and solder these small parts by hand.  You may be surprised that ICs do not have the same limitations.  I have personally successfully hand soldered some of the finest pitch ICs available using a technique using solder wick to soak up the excess solder bridging the pins, but I knew this design would probably also need to use a number of discrete components as well.

So how does SMT work for these nearly microscopic parts? Typically, in SMT a paste mask is developed, along with the PCB, which is basically a thin sheet of metal has holes (called apertures in PCB terminology) for all the pads on a board, the paste is spread, kind of like peanut butter, over the mask and board sandwich and when the mask is removed the paste remains only on the pads. Then all the small components are placed and temporarily held by the sticky solder paste.  After all the components are placed; the board is heated, in a temperature controlled oven, until the solder paste is melted and then it is cooled  down, effectively soldering all components in place at once!

This process is not too practical for “one-off” prototypes as the mask needs to be developed and machined, not to mention the extra soldering equipment involved. On the other hand, by using the larger 0805 and larger component sizes; although hand soldering is possible, the actual board space used is achievable using through-hole components, if they are mounted vertically.   [8] The problem with standing axial components is the extra lead lengths cause unwanted inductance, but if you open up many “older” radios they successfully use this technique.  As it turned out; the project ended up as a hybrid, using the thru-hole components except where SMT was the only alternative.

 I decided to give an initial PCB go to see how many of the  radio components could be squeezed into a small board, using thru-hole components, so at the very least it could be used as a drilling and placement guide for a hand built proto. (I call that the “pretty ugly” technique) even if an actual PCB was not made from it. The board size itself was determined by measuring several VHF handhelds and scanners I had on hand. Some of the older models dimensions seemed likely to be workable so I used these measurements as a starting point. Also, for the chassis, I liked the extruded models with slots for boards and end caps as they also tend to have rounded, more comfortable looking sides, and provide a better control and antenna mounting surfaces as they are perpendicular to the rest of the chassis. [9]


My project is not the original HF handheld either; Mizuho Radio had several handheld models in production [4]. They were hard to find in the US, but since the last production ('02) sold out this may be a good sign there could be some demand for this project...