Basics
Note the K3 firmware can be upgraded at the click of a mouse, features are liable to change on a weekly basis. Items that appear to be a problem one week may not be a couple of weeks later, check the Elecraft web site for the latest firmware changes. More recent changes to this page are shown in blue.
The Elecraft K3 follows in the footsteps of the K2,which is a highly successful 12 Volt transceiver. Unlike it's predecessor, the K3 is available either fully assembled, aligned and tested or as a kit comprising a set of pre-tested and aligned modules. The K3 is more desktop than backpack, measuring 10.7” wide, 10” deep and 4” high with a weight of 8.5 lbs with all options fitted. The K3 transmits and receives on the amateur bands from 160 to 6m including 60 metres, transmission is confined to being within the amateur bands only. On the 60m band there are no channelised limits, users with permission to operate on that band would need to be careful to select the correct frequency. An optional KBPF3 general coverage receive option provides receive from 500 KHz to 30 MHz, with the exception of a gap around the 8 MHz IF frequency.
There are several optional modules either currently available, or offered in the near future. These include internal 100 Watt PA, internal 100 Watt auto ATU, dual receive, higher stability TCXO, transverter interface, digital voice recorder, external frequency standard input and a selection of roofing filters.
Whereas the K2 started as a CW only transceiver which could be upgraded to SSB and basic "audio" data operation, the K3 includes CW, SSB, AM (true double sideband AM), FM and data (audio and direct keying) without needing to purchase additional options (FM needs firmware 1.81 or later). Note the K3 will need AM and FM bandwidth filters in order to use those modes, it doesn't include basic filters often found in amateur transceivers for AM and FM. There are front and rear panel microphone sockets, the front is configured in a similar manner to Kenwood 8 pin microphones and has a menu selectable gain range of high or low and selectable bias. The rear microphone socket is a 3.5mm jack and again can be configured for gain range and bias. There are line in and out sockets with transformer isolation (to avoid hum problems) and are intended for use with a PC for data or sound recording/playback. An inbuilt decoder/encoder for CW, RTTY and PSK displays received signals in a scrolling window on the front display panel, by keying via Morse paddles, RTTY and PSK transmissions can be sent directly from the K3 without any additional hardware or needing the use of a computer – while being quite novel, it is perhaps not the best way to operate data! Direct data sending and receiving appears to be possible via the RS232 serial cable, using the K3 as both transceiver and data modem. I haven't tested this feature as software available in April 2008 still struggles to cope with the mixed output from the K3 of frequency and received text. A firmware update is planned to stop the constant stream of frequency data from the K3 when operating in data modes.
There is a large and clear LCD display, the backlight is a soothing amber/yellow colour and can be adjusted from off to very bright. The display is easily viewed in complete darkness or full sunlight and can be also adjusted for viewing angle. There are green LED indicators for the multi-function knobs, the intensity of these LEDs can also be adjusted. The image above doesn't do justice to the LCD display and appears a little soft, one of these days I will get a really good image of my K3!
The antenna change over is via PIN diodes for completely silent operation, full break-in is provided together with semi for occasions such as when there is distraction from high band noise. There are no clicks or thumps when operating full break-in, coupled with the smooth side tone CW operation is very pleasant.
Tuning via the weighted tuning knob offers three main tuning rates and two display resolutions including display resolution to 1 Hz. Friction for the tuning knob is fairly crude and relies on adjusting the knob to rub against a felt washer. Having used the same method on a K2 for six years, I've not found it to be a problem or to need adjusting.
Several useful parameters can be displayed on the VFO B area of the display, these are selected by pressing the display button and rotating the VFO B knob. Available parameters are: time in 24 hour format, date in EU or USA format, PA and front panel temperature, phased locked loop voltage and audio output voltage from the rear panel. Note the display of all of these parameters required the menu to be set to “Tech mode” otherwise only the clock, supply Volts and Amps are displayed.
Mode and band selection are done by pressing band and mode rocker switches. In addition, frequencies can be directly entered. The front panel is well laid out, although being right handed I find the width control to be on the wrong side of the main tuning knob, I've managed to avoid moving the VFO when adjusting the width but for me it is too close. Filters can be manually selected however the width control overrides the manual selection and automatically switches the appropriate filter for the selected bandwidth. There has been minor criticism of Elecraft for not allowing true manual selection of filters independent of the width control, having used the K3 for both casual rag chewing and contest operation, I can see no reason for the criticism and suspect it came from those who hadn't used a K3.
There is a cute “centre zero” tuning indicator which can be used on both data and CW, the tuning indicator replaces the dB over S9 part of the “S meter” indicator. I find the tuning indicator very useful for netting CW stations and far better than using a spot tuning beat frequency. Both a spot and auto netting feature are available on the K3, so users have the best of all options and no excuse for not being able to net onto a station very accurately. On RTTY, the K3 DSP filtering operates with two passbands for improved filtering. The tuning indicator shows both tones of a received RTTY signal and is easy to use.
The rear panel has one or two antenna sockets depending on whether the ATU is fitted, these antenna sockets are selectable from the front panel and are protected by static discharge resistors and spark gap/gas discharge tubes. Also on the rear panel are transformer isolated audio sockets for hum free connection to a PC for either recording or audio data transmission using programs such as MMTTY.
1/4” jack sockets are provided on the rear panel for a pump handle Morse key and paddles. The internal keyer includes 8 message buffers of 250 characters each, both iambic mode A and mode B are selectable from the menu. Both of the Morse input sockets can be used simultaneously, for example you can have a contest keyer such as the K1EL Winkey connected to the hand/computer key input with a speed set on the computer of 30 wpm and also have your paddles connected to the paddle input with the K3 speed set to 20 wpm. This combination allows you to use 30 wpm for most stations in the contest but swiftly repeat your call or details at a slower speed via the paddles.
Remote control and logging are available via a conventional 9 pin RS232 connector on the rear panel, most of the Kenwood commands can be used for programs that do not include a K3 option. The contest program by N1MM can key the K3 without needing any additional hardware and uses the RTS and DTR lines to operate PTT and keying, these parameters are set in the K3 menu.
Up to 5 first IF filters can be fitted, ranging in bandwidth from 200 Hz to 12 Khz. The additional loss of narrow filters is compensated for by increasing the IF gain, this adjustment is available via the menu. Elecraft refer to these filters as roofing filters as they are the first filter in the receiver. A later option will include variable bandwidth filters. Unlike the K1 and K2 where the builder had to make their own filter, the K3 filters are conventional ready made units. The 8 pole filters are made by Inrad, while they share the same IF frequency as some Yaesu filters also made by Inrad, they are not inter-changeable.
The serial number is written in the firmware, while there probably isn't a big market in stolen K3 transceivers, it's reassuring to know the serial number may have been removed from the rear panel but can still be read from the menu.
Design features
The K3 is a double conversion transceiver with intermediate frequencies of 8.215 MHz and 15 KHz. A high level active mixer and strong post mixer amplifier contribute to a dynamic range exceeding 100 dB, even with closely spaced signals. The K3 uses high side oscillator injection on bands up to 10m and low side injection on 6m.
Modulation, demodulation and AGC is accomplished by the DSP board and operates at 15 KHz. On receive signals are digitised in a 24 bit ADC before being processed with the 32 bit floating point DSP processor. After processing another 24 bit ADC converts the signal to audio. On transmit audio is converted with a 24 bit ADC then processed by the DSP. Noise blanking is done both within the DSP and the IF and is configurable to use either DSP, IF or both.
One nice feature of the receiver is the audio effects or “AFX”, when enabled the stereo audio of the K3 gives a delayed signal to one side of the headphones which sounds like the signal originates from a loudspeaker nearby rather than the more usual “middle of your head” source when using headphones. The delay is adjustable and works on both speech and Morse.
On transmit there is a low power amplifier for output levels to 12 Watts and an optional internal high power amplifier for levels above 12 Watts. With power levels below 12 Watts, the PA is bypassed with a relay for improved efficiency.
When the high power amplifier is fitted two quiet fans mount on the rear panel to provide sufficient cooling for a 100 Watts carrier output of up to 10 minutes duration. The fans are almost inaudible even when running at full speed and make a pleasant change from the vacuum cleaner sound of many modern transceivers. PA temperature is constantly monitored by the main processor in the K3, under high temperature conditions the power output is reduced under processor control.
The synthesiser uses a combination of direct digital synthesis and a phase locked loop for low phase noise.
Both on transmit and receive there are 8 band graphic equalisers with +/- 16 dB adjustment at each frequency. Unlike an external equaliser, adjusting these is not something you would want to do too often. I left the receive one with all the levels set to zero. On transmit I boosted the higher frequencies and cut the lower ones to try and improve my deep voice. The equaliser cannot be switched on/off except by clearing the settings or setting each tone, maybe in the future this will be mode sensitive as using an equaliser is sensible with a microphone but not when using audio tones for data modes.
Assembly
Most buyers will probably choose the modular kit which is a “no solder” kit requiring only a few basic hand tools to assemble. The term “no solder” isn't quite correct as the power lead uses Anderson Power Pole connectors which need to be soldered to the supplied cable, however this only takes the most basic of soldering skills and an ordinary 25 Watt iron.
My K3 comprises the 100 Watt PA, auto ATU, general coverage receiver module, higher stability TCXO, 6 KHz 8 pole, 2.8 KHz 8 pole filter and 400 Hz 8 pole filter. The 2.8 KHz filter was supplied instead of the stock 2.7 KHz 5 pole unit. There was a $30 deduction for not taking the 2.7 KHz filter, there was not a similar deduction for not supplying the stock TCXO.
The modules, panels, instructions and various nuts, bolts and washers arrived very neatly packed in a single large cardboard box. There is no need to do a full inventory check of all the minor hardware items, although it helps to check the basic modules and panels are correctly supplied.
For assembly it is essential to provide an anti-static mat and wrist strap as most of the modules are static sensitive. Good lighting and good screwdrivers are also essential if you are to avoid a slipped screwdriver damaging surface mount components. A basic multi-meter is required to measure resistance to ground when fitting modules, a cheap digital multi-meter is more than adequate for the purpose.
The actual assembly is straightforward and well documented in a step by step manner. The only problems I encountered were mounting the front panel to the main unit which required a ten minute tea break before a second attempt, there are four multi-way sockets to mate together when fitting the panel and they are tedious to fit, however they connected correctly at the second attempt. The other issue was the apparent lack of two correct length bolts for the LCD plastic cover, another builder also mentioned the same bolts being too long. Elecraft supplied replacements without question in a few days. Only using two instead of four bolts didn't stop me using the K3.
Assembly of the basic K3 with the auto ATU, but without the 100 Watt PA, took around 5 hours. The later addition of the PA took a further hour and involved partial dismantling of the transceiver.
Alignment of the K3 requires no test equipment, although the power measurement can be fine tuned if (and only if) you have an accurate power meter -note the power calibration has already been done at the factory. The alignment and calibration is straightforward and takes about an hour, the transceiver should be aligned and tested as a ten Watt transceiver before adding the 100 Watt PA. The master oscillator can be easily set to frequency against an external standard, unlike many other transceivers this can be done from the menu without removing panels or needing tools.
The first items to configure are the IF filters, these need setting for their position on the main board, bandwidth, loss and frequency offset. The configuration of the filters is mode dependant, so needs doing for every mode. Like the K2, the K3 has an automatic VCO alignment feature which takes a couple of minutes to complete. The only other alignment is the optional power meter setting followed by a semi automatic power gain where the user sets the power level on the front panel firstly to 5 Watts, then 50 Watts. Alignment is automatic following a press of the "TUNE" button, the K3 then sets and stores the gain figure. The gain adjustment procedure is then applied to each band in turn. Finally the oscillator is set on frequency by setting the display to the known frequency of a reference, e.g. the 5th harmonic of a 10 MHz standard on 50 MHz, the menu item "REF CAL" is selected, the "SPOT" button is pressed, you then tune the main VFO knob until you hear a beat of one or two cycles. Setting the frequency is very simple if a reference standard is available in the shack, setting the frequency against a broadcast station or time signal transmitter is slightly more difficult but achievable.
Use
The first job that has to be done once the basic operation and alignment has been completed is to update the firmware, or operating system, in the K3. There are easy to use "updater" programs available for Microsoft Windows, Macintosh and Linux. The program connects via ftp to the Internet and automatically downloads the latest firmware version to the computer. Updates initially seemed to appear every few hours which was annoying as it wasn't possible to check the rig fully between updates! Fortunately, more recent updates have reduced to typically less than one per week. Some of the updates may include both the MCU (main control unit) and DSP firmware, in these cases the firmware files should be upgraded together. There are no incremental upgrades and all upgrades downloaded by the firmware loading tool are saved to your computer hard disk, so if something ceases to work after an upgrade you can re-install an earlier version by deleting or moving the appropriate files on the PC so the install tool doesn't see them as being the latest version. The configuration of the K3 can also be saved to your PC for later re-loading, it is not normally necessary to save and reload the configuration when updating firmware. If all else fails it is possible to return the K3 to an “as received” condition by powering up while holding the SHIFT/LO knob depressed.
The K3 took a little getting used to, needing a few references to the manual. Basic operation was straightforward, however a few areas were a little puzzling. There were problems such as not being aware that unless the “Tech mode” is set to “ON” in the menu, none of the extra display items such as PA temperature are displayed on the front panel. There is no need to use any of the menu settings for normal operation, unlike some competitors, the K3 can be driven by the buttons and rotary controls on the front panel. Another irritation was the reference to features in the manual that were not currently implemented in firmware, however the latest situation on firmware features can be found by looking on the Elecraft web site for the "owner's manual errata" which is often a better source of information than the brief notes included with the firmware updates.
Band conditions were very noisy when I first connected the K3 to an antenna, often signals were hard to distinguish giving the impression of a problem with the K3. I started to think the DSP was behaving like feeding a picture into a shredder and then trying to paste the pieces together. However comparisons between the K3 and other receivers in the shack showed the problem to be entirely down to band conditions. with the K3 sounding better on CW than the superb K2 and also sounding better on SSB than my Drake R4A, which prior to the K3 I had regarded as sounding very pleasant. There are no signs of holes being made in the received audio by AGC overshoot from noise pulses as has been reported with other DSP IF transceivers. There is a parameter in the menu for AGC pulse suppression that needs to be set to "nor" (normal) otherwise the receiver momentarily dies everytime someone turns a light on in the house. Why this option is included in the menu is a mystery, as setting it to off really makes the receiver seem poor.
Transmit signal reports were excellent with no reported issues. I used the internal keyer with Schurr Profi paddles for CW and a Kenwood MC-43S fist microphone for SSB.
I have operated with the K3 in the CQWW and BERU CW contests, using mainly 40 metres during the dark evenings with an Acom 1000 linear and quarter wave vertical antenna. Overloading of the receiver was never an issue and comparisons against the K2 gave the K3 a clear receive advantage. The linear required no special setting up, connection being nothing more than a single screened phono to phono lead and a PL259 lead for the RF. There is no ALC provided from the Acom amplifier and no need for the transmit enable lead to be fed back to the K3 from the amplifier. Since version 1.65 of the firmware the TUNE button can be configured to output any preset RF level from 0.1 to 120 Watts for amplifier or ATU tuning purposes, this makes amplifier tuning a breeze. Full QSK is no problem with the K3 and Acom 1000, although I tend to use semi break in with a fairly short delay. Spotting, logging and keying with the N1MM program worked perfectly during the contest.
The K3 really works well on data modes used via sound cards, I've tested RTTY with hard wired connections to the 15 pin aux socket using MMTTY (single transistor interface from the MMTTY help files) and also on PSK31, Olivia and MFSK-16 via Ham Radio Deluxe and Digital Master 780, the latter two modes are particularly effective. On audio FSK data modes the only connections needed are a pair of stereo 3.5mm plug to plug screened leads between your K3 and PC sound card, the isolated line input/output sockets on a K3 really make for easy connections - especially with Digital Master 780 which keys the K3 via the RS232 lead as well as displaying the actual frequency of PSK31, Olivia, RTTY and MFSK signals. On mode "FSK D", in addition to being able to externally key the transceiver in the conventional manner, the Morse paddles can also be used for "PC free" RTTY operation.. this takes a little getting used to, but does work.
Amtor and to a lesser extent Pactor, had proven tedious to use with the K3. There is quite a long delay through the DSP which necessitated initially using an external PTT "off" delay circuit to allow the audio from the microphone to exit the antenna socket before the PTT was released. Later K3 firmware includes a variable PTT release command (PTT RLS) giving 0 - 30 mS of delay. The are two settings for Amtor and Pactor that seem to both work equally well, listed below are my settings as used with an SCS PTC-IIex modem and both work fine on Amtor and Pactor I/II/III:
Mode "Data A" (USB), rear panel line sockets. PTT RLS 12 mS, synchronous mode enabled (-s displayed). Or PTT RLS 20 mS, synchronous mode not enabled.
Operating on RTTY and PSK31, which involves the transmitter operating at a 100% duty cycle, doesn't seem to cause excessive heating of the K3 PA. The cooling fans only run for a short time after returning to receive and only sound to be running at a very quiet quarter or half speed. Being able to monitor the PA temperature via the front panel is very useful and reassuring.
The hundred or so frequency memories are not very useful, on some transceivers it is possible to rotate a knob and listen on each stored frequency in turn, or indeed scan selected memories. The K3 with the current firmware (2.13) offers only very basic memory to VFO writing and no viable quick memories. Kenwood transceivers (TS570 and 480) have quick memories that I often use, maybe one day Elecraft will offer such a feature. To illustrate the problem with using memories (i.e. using two button pushes against one with a Kenwood), the following is extracted from the current manual:
"To store a frequency memory, tap V > M, then: tap M1 -M4 to save a per-band
quick memory; or tap 0 -9 to save a general-purpose quick memory; or rotate VFO A to
select from memories 0-99, then tap V > M again to save. Tap M > V to recall."
Elecraft are aware of the above memory limitations and have promised improvements in a later firmware release!
Measurements
While I have quite a few items of test equipment, much of it is not officially calibrated and therefore it is not appropriate to be too specific on measurements in many areas. I do have a very accurate GPS derived frequency standard, a calibrated power meter which is better than 5% from 160 to 6m and a calibrated spectrum analyser. My signal generator is reasonably accurate in comparison with the calibrated spectrum analyser, but cannot be confirmed accurate at low signal levels as the analyser will only measure down to around - 90 dBm.
Receive figures are:
My previous table included MDS (minimum discernable signal) , which is a modern measurement that seems to have little significance in the real world as it's not realistic to copy a signal by ear that is producing only a 3 dB signal to noise ratio. The above SINAD figure (signal including noise and distortion) is more like a real world situation, the figure improves significantly if the DSP bandwidth is reduced as there is obviously "more signal and less noise", however for the sake of comparison with fixed bandwidth receivers I left the figures for a common CW bandwidth of 400 Hz. I had to use USB as opposed to CW as the SINAD measurement was only possible at a tone frequency of 1 KHz and the maximum sidetone on the K3 is 800 Hz, using USB rather than CW makes no difference to the measurement. The performance using the KBPF3 general coverage receive module hasn't been thoroughly tested, however sensitivity below 1 MHz drops off and is roughly 20 dB less sensitive at 500 KHz than at 1.9 MHz.
The sensitivity on 50 MHz has been the subject of some debate on the Elecraft Reflector and there is a suggestion of Elecraft offering an external pre-amp to improve the sensitivity. By comparison, my Kenwood TS-480SAT is 8 dBm more sensitive on 50 MHz. Note -126 dBm equals a level of 0.112 uV pd. Whether the above sensitivity really makes much difference in the real world may be academic, it depends on your local noise levels. The image rejection figure I measured is some 8 dB below published specification and also a similar amount below the figures measured by others, a re-alignment of the bandpass filters on 6m didn't make any difference to the image rejection or sensitivity figures.
Two tone dynamic range testing was only possible on 14 MHz as I only have one signal generator, the other being a well buffered 14 MHz fixed crystal oscillator based on the design for dynamic range testing from the book "Solid state design for the radio amateur". Both these were combined in a hybrid coupler and fed via a variable attenuator to the K3. The factory figures give a 100 dB dynamic range at 5 KHz spacing and 95 dB for a 2 KHz spacing, both using a 400 Hz (8 pole) filter. My measurements give a two tone dynamic range at 2 KHz signal spacing of 100dB with the pre-amp off. These figures are for a 400 Hz bandwidth with the 8 pole 400 Hz roofing filter, similar high dynamic range figures exceeding 100 dB at close signal spacing were also obtained by the ARRL (review in April 2008 QST), two other amateurs and also by Rob Sherwood the well known receiver tester, these tests were independent of each other and on different K3s.
The K3 would normally be calibrated for an S9 signal with an input of 50 uV with the pre-amplifier turned on. As most of my operating is on 80m and 40m where I don't use the pre-amplifier, I set the S-meter for 50 uV without the pre-amplifier. There are easy to use menu settings for S-meter calibration. Above S9, the S-meter tracked signal level increases remarkably. In dBm levels 50 uV = -73 dBm, an input of -53 dBm read S9 +20 dB, -33 dBm indicated S9 + 40dB and -13 dBm indicated S9 + 60 dB. Below S9 the accuracy dropped slightly but was still close to 6 dB/S unit. The results (which in reality are not important) are significantly better than any other S meter I've tested. The S meter can be configured to show the same reading regardless of the pre-amp or attenuator being selected.
At 100 Watts the power output measured on several bands was lower than expected. The figures at the 5 Watt and 50 Watt calibration levels are very close to the indicated power, the 100 Watt level may be something later firmware could address by including a third calibration level. The lower power output at 28 MHz may be due to the SWR meter indicating reflected power on that band which is something internal to the K3, I've not investigated the matter further at this point.
There were no high power peaks when transmitting on low power and the waveform of CW transmissions were smooth and clean when observed on an oscilloscope. The rise and fall times were originally fairly soft with a 8 mS rise and 5 mS fall time. A later change to the value of capacitor C222 (to 100 nF) reduces the rise time to 5 mS. The time delay between the linear keying line going low and RF appearing from the antenna socket is adjustable from 8 mS to 20 mS in 1 mS steps (needs C222 at 100 nF and MCU firmware 1.81 or later). The oscilloscope trace shown below was triggered by the linear keying Voltage going low and shows a single Morse "dit" at 20 words per minute, power output 10 Watts, this was taken after changing C222 to 100 nF and with a TX delay of 8 mS (TX DLY 008 in the menu).
Change over times for ARQ modes are approximately 20 mS when using the "Synchronous data mode". This timing is suitable for use on Amtor and Pactor modes.
Power output when operating at low Voltage was also tested. For a supply of 11 Volts measured at the power supply with a power level set at the K3 of 100 Watts, an output power of 85 Watts was measured on the 20m band. At 5 and 50 Watts the power output closely matched the set power on the K3. Operating from batteries should not be an issue with a K3.
The bandwidth of the K3's transmission is very narrow when operating at 20 wpm, the spectrum analyser plot is shown below (for a 50 wpm plot on a larger scale file size 57 KB, click here - it opens in a new window). These plots are essentially the same both prior to and after changing C222 for the improved CW waveform envelope.
The frequency of a K3 fitted with the optional TCXO can be set to within 1 or 2 Hz at 50 MHz. Frequency accuracy was maintained to better than 20 Hz at 50 MHz when the transceiver was either warm or cold and may be even more accurate once the firmware allows compensation based on the K3 front panel temperature.
The spectral purity within all the amateur bands was within specification, however on 5 MHz the spurious levels failed to reach specification when transmitting below the current lower band limit and was approx -46 dBc at 5.151 MHz..
The image below is the two tone spectrum of the K3 using the in-built two tone test oscillator.
Interfacing issues
Several users, or perhaps prospective users, have complained of the lack of negative ALC input and also the apparent lack of being able to inhibit the K3 from transmitting until a linear or other external device has signaled it is ready for transmission. From Version 2.02 firmware the ALC has changed from +ve to -ve input, although this needs a hardware modification within the K3.
Purchasing
The K3 is only currently available in the UK by mail order, although there are distributors in Germany and Japan. There was a six month wait for those who placed initial orders, the delivery situation is improving and should eventually result in the K3 being available from stock in the same manner as the other Elecraft items.
The only viable delivery option to the UK is via UPS, the alternative US Postal Service “air mail” doesn't include insurance for items of greater value than $650. With the K3 costing upwards of $1700 that's not worth the risk. UPS delivery cost $171 for their express service which takes between 3 and 5 working days. 17.5% VAT and a handling charge of £11 is payable to UPS, in my case on delivery but others have reported having to pay prior to delivery (prices paid in November 2007). The K3 should be free of import duty in Europe (both kit and assembled versions), several buyers of K2 kits in the past were stung for import duty and had quite a battle to be refunded. The K3 is CE marked, so assembled units shipped to Europe are not an issue.
Full details, including manuals, are available at: http://www.elecraft.com/
Issues still of concern at early July 2008, firmware (MCU) 2.15
In no particular order, there are some VFO and memory issues that have become irritating after some 6 months of use...
1. Entering a frequency directly on the keypad "hijacks" the nearest amateur band, for example you are on 7 MHz and enter 15.1 MHz. When you subsequently step up and down through the bands, the 20m band (14 MHz) has mysteriously become 15.1 MHz.
2. Channelised operation is still not fully implemented,there is a messy workaround where it is possible to prefix memory alphatags with a * then recall one and rotate the main VFO knob to "tune" through those prefixed with a *, these must all be on the same band. Other transceivers have a VFO/MEM switch to toggle between the two and allow you to rotate a knob to switch through stored frequencies in memory, listening to each one as you step through or even scanning.
3. It's not possible to fully store and recall "split" frequencies, both "A" and "B" VFO frequencies can be recalled but the split button needs to be re-pressed.
The above issues are fairly minor and may be resolved in later firmware.
Conclusion
The K3 represents one of the finest transceivers available, and at a lower price than many competing transceivers. It caters for casual operators, contest operators and QRP enthusiasts with genuinely high performance for code and speech modes. The K3 is still being developed and is subject to regular firmware updates to add additional features, and fix minor bugs. Regardless of minor irritations and features yet to be implemented, it still is about as good as it gets for an Amateur HF transceiver and has around 30 dB greater two tone dynamic range than a Yaesu FT2000* at 2 KHz signal spacing based on my measurements and those of at least three other testers.
Rob Sherwood (Sherwood Engineering) currently puts the K3 at the top of his list - http://www.sherweng.com/table.html
*The FT2000 figures were published in the March 2008 RSGB magazine RadCom, FT2000D review by Peter Hart G3SJX.
|
