Background and design
Going back a few years, to at least the early 1970s, using separate transmitters and receivers was a fairly common practice among radio amateurs. By building "separates" you are duplicating some circuitry compared to a transceiver but it does give great flexibility. This receiver was originally built to replace a Drake R4A which had been used in conjunction with a simple crystal controlled transmitter. This receiver outperforms the old Drake by a significant margin and yet is fairly straightforward to build using off the shelf parts, modules and case. The details presented here are intended to assist others with previous building experience to construct a similar receiver, rather than being a step by step guide.
An IF frequency of 8.215 MHz was chosen based on having a pair of suitable filters and also a selection of crystals for that frequency. 8.215 MHz will mix with the minimum frequency of the synthesiser (10 MHz) to cover from 1.8 MHz upwards. "High side" oscillator injection is used on all bands. Although I have based my band pass filters on the 10 HF amateur bands, reception is possible anywhere on HF with suitable band pass filters, the exception being a small block around the IF frequency. Any IF filters in the 8 to 10.7 MHz range are are suitable for use in this receiver.
The performance is more than adequate with a two tone dynamic range of 94 dB at 2 KHz signal spacing, and an IF rejection of at least 80 dB (87 dB IF rejection measured on 7 MHz, possibly the worst band). The image rejection is at least 70 dB improving to better than 100 dB on 1.8 MHz. Only the top few amateur transceivers can better the dynamic range performance of this design. These figures could be improved at the expense of more complex band pass filtering and by using a stronger mixer with a higher level of oscillator injection. The minimum discernable signal (MDS) is -127 dBm (0.1 uV) on all bands without the pre-amp, and a few dB better with the pre-amp. Frequency stability is excellent. In my shack, with a stable ambient temperature, measuring against the 29 MHz harmonic of a GPS locked frequency standard there was less than 20 Hz drift during the first hour after a couple of minutes warm up, followed by less than 2 Hz drift over the next hour.
At the heart of the receiver are two modules that make assembly fairly straightforward. The IF and AGC board is the "Hybrid Cascode IF Amplifier" by Wes Hayward (W7ZOI) and Jeff Damm (WA7MLH) from the December 2007 edition of QST. The second module is a synthesiser by Cumbria Designs used with their mini counter frequency display. The synthesiser is capable of generating an output of 10 MHz to 250 MHz at a level of +10 dBm, although in this receiver the frequency is set for an upper limit of 38.215 MHz (corresponding to a receive range of 1.8 to 30 MHz). Unfortunately, checking with Cumbria in December 2013, they have been out of stock of these items for some weeks with a possibility of no longer supplying them. However, the design can utilise synthesisers from other suppliers, some of which cater for SSB/CW offset (although they may not be offsets that you can adjust), have RS232 capabilities and that can also switch band pass filters.
I etched two boards, the IF filter board and the AF pre-amp. The other boards were made using a series of "islands" or "pads" created with a 7mm diamond tipped tubular "drill" of the type used to drill ceramic tiles (mount your drill in a stand to avoid it wandering all over the board). The audio amplifier is a Velleman K4001 kit.
The case used is a "Unicase 2" by Metcase (www.metcase.com), part number M5502119, which measures 260 x 90 x 250mm (W, H, Depth). These cases are stocked in the UK by RS Components (769-4908). An aluminium sheet is fitted in the centre of the case, this is not supplied with the case. I bought a sheet 250mm square by 1mm thick, which is a perfect fit which just needs cutting to length (a sheet thicker than 1mm can be used, but would need to be narrower than 250mm to fit the side channels of the case). At 1mm thick the sheet would normally be considered too thin, however the bandpass filter mother board imparts sufficient stiffness into the sheet. The knobs, including a 31.8mm diameter tuning knob, are from eBay suppliers. The "signal" meter is from Maplin, with illumination by a white LED.
As a minimum the BFO needs to be enclosed in a screened box, this can be made from scraps of print board. The IF and filter boards are very sensitive to stray signals and could benefit from being enclosed too. Some of the other synthesisers have a tin plated screening can as an option, adding screening to the synthesiser is probably a good idea.
The front panel was produced in Corel Draw and printed on photo card, which in turn was glued to the aluminium panel with photo mount spray adhesive. A clear plastic self adhesive "book protector" film was stuck over the card to protect it.
Some test equipment is essential to the building of this receiver, the following will be needed:
An L/C meter to measure capacitors and inductors (an LC200A from eBay is adequate and costs around £25), Signal generator, spectrum analyser (or possibly a second receiver if an analyser is not available), low cost multi-meter and a frequency counter. An oscilloscope is useful to check supply lines for signal leakage from the BFO and noise on supply lines. A vector network analyser (VNA) could be used to align/test the band pass filters, not having one I used a signal generator and spectrum analyser.
For a view of the underside of the receiver click here, for an upperside view click here.
Module and circuit details
Below are diagrams in PDF format:
Band Pass Filters
There are 10 band pass filter boards selected by a 1 pole 12 way front panel switch, these are relay switched to minimise IMD. Each board is constructed on single sided PCB (avoids possible shorts to adjacent board), and individually tested/aligned before being added to the main board. The filter design was done using the freeware Windows program "Elsie", using a topology of "Mesh Capacitor-coupled band pass" and "Chebyshev" Family of 3rd order (uses 3 inductors and 5 capacitors per filter, actually produces "5th order filter"). The design was altered to use the nearest 5% capacitor value with inductors wound for the required inductance using a low cost L/C meter. For those not familiar with this program, beware of the default Q settings of capacitors and inductors being excessive (a value of 200 ~ 500 is more realistic) and also set the transmission to "Absolute", both these settings are found under the "Analysis" tab.
Capacitor and inductor values are critical, being off by a few percent can make a huge difference to performance. Unless you can source close tolerance capacitors, you will generally have to make up the required values by connecting 2 or 3 in parallel and measuring the value with an LC meter. The toroids should be wound for an extra turn and then fine adjusted by removing a turn or two and/or moving the wires on the core until you obtain the required value, estimating the value based on the number of turns will not produce repeatable results. For the higher bands, trimmers were used for the series capacitors (7 MHz and above). T50-2 cores were used from 1.8 to 10 MHz with T50-6 on the higher bands.
The image below is a spectrum analyser plot of the response of the 14 MHz band pass filter, measured between the in/out buss bars. The loss is 3dB.
The frequency response graphs produced by the Elsie program were pretty close to my measured results. See circuit diagram for my calculated values, however, the Elsie program is so easy to use that you can produce your own values easily enough. Do not try to cover too wide a bandwidth when designing the filters, as the image and IF rejection levels are better for narrower bandwidth filters. However, if you attempt to make too narrow a filter, you will find it gives excessive loss. Typical measured loss figures for these filters vary from less than 1 dB to 2.5 dB when measured directly between the input inductor and output capacitor on the board (ie without the relays). There is a series tuned circuit across the output of the band pass filter rail to ground which notches out the IF frequency, I used a 12 uH coil wound on a T37-6 toroid with a 65 pF trimmer capacitor.
Each band pass filter is "tack" soldered to a mother board, with the relay switched input/output soldered to a common bus bar.
Other bands could easily be accommodated as the filters are switched independently of the synthesiser, the only limitation being that you cannot receive near the IF frequency. Using this arrangement of filters, synthesiser and separate counter will allow operation to 50 MHz with appropriate filters. The synthesiser itself is capable of operation to more than 200 MHz.
Pre-amp and Attenuator board
These are fairly straightforward and conventional, they are both relay switched. There is 10 dB of attenuation and 12 dB of pre-amp gain available, selected by a centre off toggle switch. The pre-amp uses a 2N5109 bipolar transistor with heatsink.
Mixer/post mixer amplifier
A TUF-3 double balanced diode ring mixer is followed by a 2N5109 post mixer amplifier (heatsink needed). There is a 3 dB attenuator on the synthesiser output to drop the +10 dBm level to feed the mixer and to provide a good 50 Ohm termination to the mixer. The post mixer amplifier design has been optimised for 50 Ohm input impedance, several similar circuits have a poor input SWR.
IF filter board
I was able to source a pair of Yaesu filters at a reasonable price, these are nominally centered on 8.215 MHz and are (hopefully) of 500 Ohm impedance, however any filters in the 5 to 12 MHz range could be used, beware that 8.2 MHz is the lowest you could use with a 10 MHz synthesiser to cover the 160 metre band. The filters are switched using two dual pole change over relays for each filter, there is a slightly increased loss when using the CW filter, I didn't add additional amplification to overcome that loss as the receiver is sensitive enough without adding another stage. Room was left on the board for the future addition of an AM filter. There are "L match" coils and capacitors at the input and output of the filter board to match the 500 Ohm filter impedance to the 50 Ohms of the post mixer amplifier and IF amplifier.
IF amplifier and AGC board
This used a ready etched board offered by KA7EXM, However, as at April 2013 these boards are no longer offered. The original article (for those who are not ARRL members) can be downloaded from the KA7EXM web site (http://ka7exm.net/hycas/index.htm). The board is designed to be used with 50 Ohm filters, my IF filters are 500 Ohm so need a matching circuit comprising a 3 uH coil in series with a 100 pF capacitor to ground. An S-meter buffer was added to the AGC output terminal, this is a single transistor and feeds a simple "signal" meter, the circuit is the same as used in the HBR2000 by VE7CA. The meter was adjusted for 50 uV at half scale, unfortunately at signal levels below about -90 dBm the meter doesn't indicate unless the pre-amp is turned on, but that isn't important to me. The mute input is used to turn off the receiver when transmitting, it was not found necessary to also mute the AF pre-amp. There are no clicks or thumps when muting the receiver via this input.
Product detector and BFO
For the BFO I cheated and used one crystal with a "VXO" style of circuit to pull the crystal to the appropriate offsets for CW, LSB and USB. Note the use of a basic three pole 4 way switch to select USB/LSB/CW, I tried switching the BFO capacitors with PIN diodes but found the adjustment range poor compared to direct switching. Note the need for short leads to minimise stray RF radiation. The BFO is followed by a low pass filter as there were significant harmonics into the high VHF range when checked on a spectrum analyser. While measuring for stray RF with an oscilloscope, I noticed a higher than expected level of oscillator injection into the SBL-1 mixer and so added a 3 dB attenuator to the output of the low pass filter.
Both the supply rail and the AM BFO disable line need substantial decoupling, I used 100 uH RF chokes in series and a combination of a 1000 pF feedthrough capacitor, 0.022 uF disc ceramic and 22 uF radial lead electrolytic capacitors. An output for an AM detector is taken from the IF side of the SBL-1 mixer. The BFO transistor is turned off when AM is selected. The 1K output level pot is to match audio levels between AM and SSB/CW, when I add the AM filter and detector. It is essential the BFO is shielded to avoid oscillator leakage into the IF amplifier. Shielding was done using pieces of print board and was fairly easy although a little time consuming to construct a complete shield. The image of the upper side of the receiver shows the board prior to soldering the top cover in place, the top cover needs holes to allow for alignment.
This circuit is based on a design from Experimental Methods in RF Design (ARRL) and is reproduced here with permission. The op amp is a low noise type, do not try to use common 741 op amps here. There is no particular attempt to filter the audio, although the higher frequencies are attenuated to some extent, for a graph of the overall response, courtesy of the circuit modelling software "Tina", click here. While not shown in the image of the upper side of the receiver, a low level audio input is catered for in order to provide a side-tone from a companion transmitter.
Audio output amp
A Velleman K4001 kit amplifier was used, the output coupling capacitor was replaced by one of 100 uF to reduce the bass content of the audio and a precautionary 0.01 uF disc ceramic capacitor added across the input terminal to reduce the risk of RF pickup. The amplifier is specified as 0.05% distortion and a maximum output of 3.5 Watts RMS into 4 Ohms.
Synthesiser and counter
These are standard items from Cumbria Designs. The synthesiser is based on an Si570 device outputting +10 dBm into 50 Ohms. The advantage of using a separate counter module is that it allows me to use any offset I program into the counter, the downside is the extra cost and the tendency of the display to flick between the 10 Hz digits. A Bourns 64 pulse per revolution optical shaft encoder is used in place of the supplied mechanical encoder. I do not use a "push" switched encoder, the front panel "F/S" button is wired in place of the encoder switch.
Once left on a frequency for around one minute, the frequency is internally stored in the synthesiser allowing the receiver to be powered up on the last used frequency.
In order to mute the receiver in the conventional manner of grounding a socket on the rear panel, it is not really necessary to use anything more than a pull up resistor with decoupling of the mute input line. However, I added a single 2N3904 NPN transistor as an emitter follower...
Each band pass filter board was tested/aligned before adding them to the mother board. Those with trimmer capacitors were adjusted for maximum signal towards the higher end of each band, there was no need to stagger tune them. A final adjustment was done with the boards soldered in place. The IF notch filter was adjusted for minimum signal on 8.215 MHz with the 7 MHz band pass filter selected. A signal generator and spectrum analyser were used for alignment, if an analyser isn't available at a push a second receiver could be used.
The BFO was initially set to the predicted offset frequencies using a frequency counter connected to the output of the low pass filter, followed by a final adjustment with the screening cover in place using a smart phone running a free audio spectrum analyser program (AKLite, an iPhone App) to measure an offset from a signal generator, a 1 KHz audio tone in the case of LSB/USB and 750 Hz for CW.
The front panel counter can be adjusted using a known accurate signal, the synthesiser need not be accurately set.
The overall gain and noise level from this receiver gives very pleasant audio, the gain of the AF pre-amp produces just the right amount of audio. The AGC works very smoothly even if the S-meter lacks the ability to measure low level signals. There are one or two minor birdies, but none that cause any issues. Adding screening to the IF and filter boards (and possibly the synthesiser and counter) would have been a good idea.
I haven't added DC wetting to the relays, so far I do not have problems with intermittent contacts, but others have reported issues when using relays without a few milliamps of DC current passing through the contacts.
Thanks to Dave, GM4EVS, for the neat diagrams and circuit analysis and to W7ZOI/WA7MLH, for the IF amplifier design. Also thanks to Markus, VE7CA, for the circuit ideas associated with the IF filters and IF amplifier, in particular the S-meter circuit and the filter matching. Click here for a link to his HBR-2000 transceiver web page.
The circuit diagrams were drawn using the Windows software sPlan 7. Click here for a link to their site.
Elsie can be downloaded from here.
Please do not build this design without checking the values for yourself, in particular the band pass filter components. The information is presented here to assist experienced constructors who may wish to build something similar, it is not suitable for novice constructors.