Another great project designed by Peter Rhodes, G3XJP.
This PIC controlled, automatic ATU is designed to be remotely controlled and placed at the far end of the feed line - where it will do most good. Local builders are: Kevin (VK3CKC), Ian (VK3ZZG), Ken (VK3FKEN), Michael (VK3ZPY), Luke (VK3HJ) and one spare, possibly for Club use. The project is strictly home brew and you will need access to the construction article that was published in RadCom, September 2000 through to January 2001. Copyright is held by the RSGB and appropriate acknowledgement is made to the RSGB.
Circuit boards are not available commercially or from this Club as such. You will have to create the enjoyment and satisfaction of making your own. This is a fairly simple process and the article provides instructions on at least one method of doing that. The main, large board is single sided while the smaller logic board is double-sided. You will also need access to the
PIC-A-STAR forum so that you can ask Peter (G3XJP), who has graciously agreed, to provide a copy of his PIC software when you get to that point. See the PIC-A-STAR link in the navigation panel for more information about that project.
Features:
The basic ATU configuration is an L-match with more than 250,000 relay-switched L, C and Z combinations. In addition, it has 64 series, L or 2048 shunt C possibilities in 'straight through' mode. It is rated at 200W for most antennas and substantially more for some. It will match end-fed, coax-fed and (with a balun) balanced antennas - at any frequency in the 9 amateur HF bands only - to 50 ohms. Optional outputs also allow you to automatically switch antennas as a function of which band you are on.
There are no expensive relays or HV capacitors to buy. Cheaper relays are modified for the task and you make your own HV capacitors.
Restore From Before:
In normal operational use, all you do is transmit the first letter of your callsign on SSB - or a CW dot - and your pre-stored matching solution for that frequency will be applied.
Match From Scratch:
Requires about 15 seconds of steady carrier at about 5-10W at any given frequency. The ATU has the potential to remember up to 1,000 potentially different solutions. This works out to about one every 5kHZ over the whole amateur HF spectrum.
Quiet Matching:
When matching from scratch, PicATUne uses a 'quiet tune' approach to attenuate the radiated signal by some 25dB, saving stress on your Tx, ATU and fellow amateurs.
RF Deck Board Construction - Part 1:
As the first of our units is still being constructed, you will have to return to this page to see how it all comes together.

The ATU consists of two boards - the RF Deck and a Logic Board. The RF Deck board is quite large and will not print on a single A4 sheet at full size. It is printed on two pages in the RadCom article and you will have to make at least one composite image from these. You must ensure that you print the image without scaling and measure an appropriate component to ensure that mounting holes will line up. Only then can you be sure that you have it at the right size.
You can follow the pcb making instructions in the article but I went the UV process. After considering the combined images of the larger board it was obvious that I would have to cut it in order to make it fit the 150mm x 300mm Kinsten pcb blank that I was going to use. Two of these blanks are required for making the composite board and the blanks are the largest that will fit in my plastic trays used for developing and etching. An ideal place to cut the composite image was through the gap in the large pads for the five largest HV capacitors. This means that there are only five copper tracks that will need to be cut. Three are related to the central relays and the remaining two are related to the edge of the board. These tracks can be bridged later when the finished main board halves are put together. The two outermost of the central relays are fitted on one board and the other board is trimmed to fit snugly in place. Soldering completes the assembly - including ensuring that gap crossing tracks are made continuous.The above process preserves the original dimensions/spacing and results in a very sturdy, composite board. Araldite or something similar will be used to complete the finished board manufacture if considered necessary.
Note that I have left plenty of bare gap on each board for trimming back for a snug fit for the relay pins.
The displayed image is something of a 'suck-it-and-see' prototype for the local construction group and there is a thin copper track missing around part of the board perimeter (mistaken for a board outline) which I will have to address later. All holes should be drilled as described in the article. Note that they are not all the same size. The documentation advises to initially drill all holes at 0.8mm. This will be the finished size for some and a pilot hole for others.
RF Deck Board Construction - Part 2:

The image at the right shows a view from the track side of the composite board. The six central relays have been soldered in and the board is now quite sturdy.
A relatively easy way of cutting the boards, including the long slot under the main coil (a nibbling tool was additionally used here), is to use an angle grinder with one of those new 1mm or so cutting wheels. These are only a couple of dollars and cut boards in a flash. I suggest that you practice on a piece of scrap first as you don't really want to practice on your new masterpiece.
Don't be tempted to solder all the relays or other components in at this stage. You will need to allow for some working space before you can do that. Also, the holes as shown at the top of the image are not drilled until you have made and installed the main coil. It is acceptable to solder the 6 central relays at this stage to hold your board together but it can be done later. I soldered mine in at this stage to ensure that the board could be easily made in the first place.
Once you have got this far and your final board has been defined, follow the instructions in the article and make the HV capacitor clamps.
Proceed slowly with the polystyrene drilling. A drill press on a slow speed is best here and a little at a time. If you use a hand drill and the drill bit snags, you could ruin at least one clamp at a time. If yours is a composite board and you've soldered the 6 central relays, temporarily bolt C7 and C11 clamps across the gap to strengthen the board for subsequent handling.
Relay Modification:
This part of construction is quite fiddly but is not very difficult. There are 21 relays in all but
only 19 of them require this modification which is to remove unwanted contacts. I suggest that you put 2 relays aside to prevent accidental modification.
 |
The relays - before modification (left) and after modification (right). They ordinarily have 2 sets of changeover contacts. The normally closed, fixed contacts are to be removed. |
 |
Firstly, place small screwdrivers between the body of the relay and the catch of the plastic cover so that the catches are freed. The screwdrivers should sit there on their own. Using a small pair of needle nosed pliers on the pin of the contact to be removed, pull the relay clear of its cover. |
 |
The contact to be removed needs to be heated so that it will pull though the plastic that holds it in place. Solder will aid the heat flow process. While holding the relay down, apply a hot soldering iron under the contact to be removed and apply some upward pressure. The contact will be forced up and out of position when it becomes hot enough. Be sure to remove any plastic that might become dislodged in the process as it will impede the cover. Re-fit the plastic relay cover. |
The Finished Board:

Well, not quite. All the relays and a few other components have been installed. The coil has also been epoxied in place. The two relays at the bottom of the image are the unmodified ones that were put aside to avoid accidental modification.
The Logic Board:

The double-sided logic board (one side has a complete ground plane) requires a bit of care in duplicating. There is not much clearance between many of the relay-controlling transistor tracks and these will have to be carefully inspected -
BEFORE any soldering. Any suspect clearances can be opened up with either a scribe or a scalpel.
Kinsten board, developer and ammonium perculphate etchant were again used for this board. The image was first copied to a laser transparency making sure that the print was definitely 1:1. The protective film layer was then peeled from one side of the fresh board only and this side of the board was exposed using the artwork on the laser film. The UV protective film was then removed from the other side of the board and the board was was placed in the developer without any further exposure and developed in the normal way. The next stage was the usual etching.
The image shown here is the track side of a freshly etched board after a rough removal of the etch resist with steel wool.
Be prepared to spend a lot of time drilling and building this board as a bit of care is required. Follow the printed instructions closely - especially regarding drill size. If you use larger drills, you will remove the donuts around the component holes.
Most of the ground plane component holes need slight countersinking -
on the ground plane side. This can be done with a sharp, large drill twisted between the fingers. Check the component holes that
DO NOT have to be countersunk as some component leads have to be soldered to the ground plane as well as to an underside track. Countersinking holes for these connections make this more difficult. There are also a few component leads that do not go through the board at all and are simply soldered directly to the ground plane.
The Enclosure:
The construction article calls for polystyrene as the material for the enclosure. I decided to use what I had on hand - a product called Handiclear and the applied label stated that it was clear acrylic. This is also what I used for the HV capacitor clamps and the sheet measured 900mm x 600mm as purchased at Hume & Iser, Bendigo, some years ago for another project. It comes with a plastic film attached to both sides for protection during handling.
Cutting the sheet is easy enough and requires a fine tooth saw, jigsaw with plastic compatible blade, circular saw with fine toothed blade or a trimming knife - according to the label. I used a jigsaw with a not necessarily plastic compatible blade. I progressed slowly and carefully and it worked well - regardless. Be careful to support the plastic when you get near the end of a cut to prevent breakage if you use something similar.
Try as I might, there was some slight variation in the plastic side widths. They were not all exactly 80mm as required. This was addressed by clamping the sides together in a vise to form a block and dressing them all together with the angle grinder and cutting disc mentioned earlier. The top and bottom of the box can wait until later.
The adhesive I purchased, with fingers crossed, was called Selleys Plastics Glue - from Fitzpatrick's Home Hardware in Eaglehawk. This is a two part product consisting of a primer pen and adhesive. Following the provided instructions produced a first class bond within minutes - on scrap. However, progressing to gluing the case, two of the joins were quite weak for some unknown reason. I then tried some 'beyond its use by date' PVC Pipe Cement (primer and cement) on some scrap. This worked quite satisfactorily although you have to hold the join together for 5 minutes or more. I haven't re-glued the case yet as the scrap join is still under test.
Created 2007 by
Kevin Crockett. Comments and suggestions may be e-mailed to