MKARS80
The MKARS80 is a relatively simple but ingenious mono-band 80m QRP transceiver kit. It was developed by Steve Drury, GL6ALU, at the Milton Keynes Amateur Radio Society, hence the name MKARS80.
Detailed information on the MKARS80 can be found on Steve’s site: http://www.radio-kits.co.uk/.
I bought four MKARS80 from Steve as I intended to experiment with them and gain more experience building transceiver kits.
MKARS80 Mk1
The first MKARS I made was assembled strictly by the book. I used the excellent illustrated step by step assembly instructions that can be downloaded from Steve’s site. I used the supplied metal box to accomodate the PCB and made a slightly different design for the front panel inscriptions. Also, I made an aluminum handle to carry the rig, a handle that did double duty as a stand to tilt the rig to a more comfortable operating position.
From kit .....
.... to transceiver !
The little rig worked fine first time after switch-on and I was surprised by its sensitivity. However, I found that there were some things that needed improvement. First, there was the tuning. The main tuning is performed by a polyvaricon. Since the local oscillator frequency is higher than the signal frequency, the polyvaricon has to be turned the wrong way; turning it clockwise decreases the frequency on the LCD and vice versa. Most of us are accustomed to the opposite. I assume this has to be a typically British custom, comparable to driving on the wrong side of the road and I thought about changing it. Not the driving, just the tuning...
Besides being the wrong way in my view, I find the tuning procedure time-consuming since one has to use two knobs; the polivaricon for coarse tuning and pot for fine tuning. I decided to improve that at the same time. Third, LCD readability is insufficient in sub-optimal lighting conditions, so I felt I should change the LCD for a back-lit version. Also, I thought about another and more traditional housing with the controls at the front and connectors at the back.
VARACTOR TUNING
For the tuning, I started to experiment with a varactor I had handy, the Philips BB112. Soon I found out that it could easily tune the entire band and thus replace the polyvaricon. So I removed the polyvaricon and replaced it by a simple circuit with the BB112, a few C’s and R’s and a 10-turn pot. With it I could now tune the entire band the right way and with even twenty times more precision than with the polyvaricon, since the polyvaricon tunes the band in half a turn whereas the pot does it in 10 turns. The simple schematic can be found here: Tuning_MKARS80_1.1.GIF.
If you wish to make a PCB for the circuit, please find the design for the copper side here: Tuning_MKARS80_COP.GIF and the component side here Tuning_MKARS80_SCRN.GIF . The board dimensions are 22mm (h) x 20 mm (w), about the footprint of the original polyvaricon.
Although there is nothing new to varactor tuning, I was the first to apply this type of tuning to the MKARS80 and I still feel it is an improvement.
BACKLIT LCD
The next step was the LCD. I did not find a back-lit LCD that was pin compatible with the original one but I did not bother. I was now at a point where the rig would not fit the original housing any longer because of the 10-turn pot and since I was not too fond of the original housing anyway I decided to find an entirely new one. Fortunately, googling my way through the various offers of metal box suppliers on the internet, I found a suitable aluminum box at a wholesale company at only 20 kilometers away from my home. It so appeared that this box was going out of production and they had only five left. They sold the lot to me for 2,50 Euro each. Was I happy. Now I was free in the layout of the front panel.
My first MKARS80 with carrying handle ....
.... also serving as a stand .....
.... to improve the operating position.
I designed a front panel for the MKARS80 on my computer using Corel Draw and printed it on matte photo paper. I cut out the holes for the controls, the LCD and the LED. Once that was finished I sprayed the now ready bezel with quick drying clear matte PU varnish to make the front more scratch proof. The bezel was now ready be glued to the metal front of the box once the rig was ready. So far so good. Now, I had to make multi-wire cables to connect the LCD on the front to the PCB on the bottom of the box and also cables for the individual controls. I decide to use connectors on the PCB side, so that at least one side of a cable could be removed easily for experiments or repair.
... and now with the LCD and the controls in place , waiting for the new bezel.
The extension cables with connectors, I later provided the two LCD cables with an earthed shielding in order to prevent “birdies” in the receiver.
The simple varactor tuning circuit, her depicted in prototype form, takes the place of the polyvaricon.
A second connector for the microphone is installed underneath the MKARS80 and an angular mike plug is used as not to accidentally pull the plug during a QSO.....
A few days later, the MKARS 80 Mk 2 was ready to go on the air. I used a large tuning knob that I provided with a glue-on finger dimple to make tuning across the band even easier.
BAND SWITCHING
Although tuning was now more precise and the tuning knob turning the right way, I still had to turn two knobs to set the desired frequency. To overcome this I decided to make the main tuning even more precise and do away with separate fine tuning. So I split the 3,5 to 3,8 MHz band into three 0,1 MHz portions, selectable by a three-position switch. The upper and lower limits of the individual bands should be controlled by 12-turn trimmer pots, two for each band. I made a small board with these six trimmer pots and was able to adjust the lower and upper band limits quite easily. The fine tuning pot on the front panel was replaced by the three position switch. The main tuning pot, when tuning a 100 kHz band segment, makes 10 turns, so 10 KHz per turn. This is more accurate than what the fine tuning knob did before. I was afraid that I would notice frequency jumps when tuning with the wire wound 10-turn pot, making tuning insecure, but this was not the case.
You will find the schematic for the three-segment-band at the right.
The entire circuit, excluding the 10-turn pot and the band switch, can be accomodated on a small PCB that has about the same footprint as the original polyvaricon and actually replaces it. The PCB is also suitable for the simple varactor tuning circuit discussed above.
The revised version of the MKARS80 Mk2, called Mk3, added are the three band segment mod, a new front bezel design and a push button (A.T.) for tuning the remote automatic antenna tuner. Pushing the button generates a tone and thus RF output that the AT needs for tuning.
Click on the picture to watch a short Quicktime movie showing how easy, quick and precise varactor tuning with a 10-turn pot actually is, compared to the original polyvaricon and fine tuning. Band partitioning makes it even more comfortable.
AGC
I thought the MKARS80 had now come to a point where there was no more room for improvement. Still however, I hoped that there would be a simple way to add an AGC, maybe even an S-meter, without having to alter the original PCB. The MKARS80 does not have any AGC circuit at all, so the operator has to adjust the RF gain manually. Although this is not a problem for most QSOs, the lack of an AGC becomes annoying when one monitors a QSO between a strong and a weak station. The strong station can easily blow your speaker whereas you may miss on the weak station if you forget to turn up the RF gain in time. Also, one cannot use headphones, since using them may damage your ears permanently when a strong station comes in unexpectedly. So I studied the schematic again in order to see if I could make a simple AGC with a minimum number of components. The result is shown here: AGC_PIN_MKARS80.GIF. It uses a cheap standard diode as a PIN-diode. Bias voltage for the diode is derived from a simple rectifying circuit connected to the speaker output. A strong signal makes the PIN-diode conduct and this brings its impedance down. Thus the signal to the first RF stage is reduced. This mod, although not a full-fledged AGC, increases listening comfort, but still do not use headphones!
CASE
AS I remarked above the case of my present MKARS80’s as shown above is no longer in production, so giving details here would be pointless.
S-METER
The only wish I still have regarding the MKARS80 is some form of an S-meter, calibrated if possible. Ideally, it should be displayed on the LCD as a bar or as a digit and take the place of the battery voltage indicator which is of no use to me as I operate my MKARS on mains. I found out that I would have to drastically alter the PIC code and this, I know, would beyond my capabilities. Any volunteers?
First design of the MKARS80 Mk2 ....
The ready circuit and the naked PCB, measuring abt. 22 x 25 mm
PAøKLT DDS VFO. Neat and compact design, easy to operate, good back lit LCD, low price, what more would one require? That we do not need to modify the MKARS80 for it!
N3ZI DDS VFO with its large LCD. The latter can be exchanged for a standard serial interface LCD with backlight. Good optical encoder included.
Suitable for the MKARS80 without having to modify the latter!
The second DDS VFO I bought and built is the N3ZI DDS2. I ordered the kit with the SMD chip pre-installed as I did not want to ruin it because of my sometimes somewhat shaky hands (just age, not booze). This VFO provides a very clean sine wave and what is more, it has no problem with a “negative” IF. Just tell the VFO that you are using minus 10 MHz as IF and the display shows minus 3,5 to 3,8 MHz as working frequency, there is really nothing to it. No modification of the MKARS80‘s BFO is required!
The kit comes with a very large LCD without backlight, much too big for my MKARS80 and not very readable so you need to buy another LCD, with backlight.
A nice optical encoder is included, but this one has a shaft diameter of 6,3 mm, an exception on the continent so if you live there be prepared to buy another encoder also. Doing so will save you much time trying to find a fitting and suitable tuning knob. These additional purchases make the kit, which already has a rather stiff price at near 100 USD (say 70 Euros), even more expensive compared to the PAøKLT kit (at about 48 Euros). The latter kit has more options built in, more memories and personally I find it easier to operate. Also, the PAøKLT VFO has a much wider frequency range and it is physically smaller.
DISPLAY FORMAT
Another mod I wanted to perform was the frequency readout. In its original form the MKARS80 displays six characters for frequency, thus suggesting a frequency accuracy down to the last Hertz. This is somewhat pretentious for little rig like this with a free running VFO (although it is stabilized by a Huff & Puff circuit). After all, transceivers using PLL and synthesizer techniques, costing ten to forty times more, usually only display frequency to the last ten Hertz and in many older transceivers only to the last hundred Hertz. For me, a display with a hundred Hertz accuracy would be adequate and if this could be changed I would at the same time put some decimal points at the proper places. With help from Steve, G6ALU, and a few other MKARS80 users I was able to change the code in the MKARS80’s PIC. The result is shown in the pictures at the right. In addition I made a few other changes, like arrows in the display to indicate which way to turn the tuning knob to stay on frequency should the VFO drift.
DIGITAL VFO
The MKARS80‘s Huff & Puff method used to stabilize the VFO frequency works quite well, but it cannot compare to a digtal DDS VFO, of course. The latter, in addition to high stability and accuracy, also offers the possibility to store frequencies and to choose between various increments for the frequency steps of the tuning knob. So I decided to experiment with a few of these digital VFO’s.
The first one that came to my attention was the DDS of PAøKLT. I bought and built it, which was not too difficult although it has some SMD components. It worked fine, but it is not suitable for the MKARS80 without a modification to the latter since it cannot compensate for a situation whereby the VFO frequency is lower than the IF. I tried to work round this by using high end injection so a VFO at 13,5 - 13,8 MHz instead of 6,2 - 6,5. This works but of course this method makes that we receive the upper side band instead of LSB. The only way out here is to change the BFO frequency. Why change hardware if one can change software, was my first thought. It should be easy to change the DDS microprocessor code to take account of “negative” frequencies, but this code is not available to the user and the maker of the kit has no plans to change it for the benefit of MKARS80 or BITX20 users.
At first I thought the PAøKLT VFO would still make a nice signal generator on the workbench, but later I decided to modify one of my MKARS80’s BFO for high end injection. The frequency of the BFO should be brought down from 10.000.000 Hz to 9.997.000 Hz. This proved not all too difficult. An inductor in series with the quartz crystal plus a higher value capacitor was all that was needed, but a trace on the PCB had to be cut for this mod.
Of course, the MKARS80’s own VFO had to be silenced and this can be done, for instance, by removing the VFO coil. Since the PAøKLT has its own (back lit) LCD, one does no longer need the MKARS80’s own display and the microprocessor can be removed, too. The only disavantage of this is that one loses information on the battery voltage.
The output signal of the DDS VFO is injected at the testpoint marked VFO on the MKARS80 PCB, via a 100 nF capacitor. Of course, the DDS VFO needs to be programmed with the proper values for the MKARS80. The drawing at the right gives all the details.
In hindsight, I am satisfied with the PAøKLT VFO and with the support I received from PAøKLT himself. Operating the VFO is straightforward and a real pleasure. One has VFO A and VFO B and each has 16 (tunable) memories. With its backlit standard LCD it is good value for money.
TUNING KNOB AND FINGER DIMPLE
I have had a few questions about the main tuning knob and the finger dimple. What make is it and where did I buy it? Here is my secret:
The tuning knob is from a German manufacturer called OKW. In the UK, for instance, you will find it at Conrad’s: http://shop.conrad-uk.com/.
The part number for the knob is 717578 and the matching cap is 717632. The picture of the knob you will see on the Conrad-page is not that of the actual knob, a well-known and irritating problem with Conrads. Please note that the knob is for pots with a shaft diameter of 6mm, it does not fit 1/4 inch shaft pots!
The finger dimple is available as a ready glue-on item at http://www.fingerdimple.com/. Buy a few, they always come in handy!
The settings for the PAøKLT DDS VFO plus the schematic for the MKARS80’s BFO mod. Click the picture to open it in a new window. The values seem a little odd but I have tested various ones and these shown here work best. The reason is that the theoretical values for the frequency settings have to be corrected for central frequency of the crystal filter, otherwise the signal is attennuated. Do not forget to calibrate the DDS VFO first, as per the instructions supplied with it, or the values become even odder.
The circuit diagram and alignment instructions, it’s simpler than it looks!
Click on the picture to enlarge it.
I fitted two MKARS80’s with the 3-band board and after a year or so of testing I am still very satisfied with this mod. Tuning is quick, easy and pleasant. I changed a parameter in the PIC in order to make the Huff & Puff lock faster, within a second. The lock situtaion is indicated by a plus sign on the LCD, as shown in the pictures below.
Since I was at it anyway, I made more changes to the firmware, please see the heading “Display Format” further down the page.
The feel of the tuning is now similar to that of a “normal” commercial rig, not quite as good, but good enough.
DRAKARS80 :-)
One of the plans I have is to execute an MKARS80 in Drake TR4-style, just to give me that touch of mysterious nostalgic feeling that I sometimes miss when operating a modern rig.
I drew an impression of a “DRAKARS80” with CorelDraw.
The transceiver is housed in a LC640 metal box providing a tight fit to the contents, there is no room for a digital VFO, alas. Who cares, the orginal Drake 4 series did not have digital VFO’s either.
A blue back-lit LCD I still have laying around somewhere and a blue LED to indicate the modulation shall enhance the Drake-impression.
HEATHKARS80 :-)
And here is a virtual MKARS80 in Heathkit style.... and so on....
