Elecraft K3 Review

[g4ilo]

The outdated DB9 Serial / RS-232 connection on the back was my first complaint when I was looking at the specs. Regardless of all the "reasons" posted by the Elecraft fanatics on their email list, an RS-232 port is still a bad idea. The "reasons" typically given are that USB chipsets are difficult to maintain drivers for, a lot of hams still have PC's with serial ports (really??), etc.

Look, the RS-232 standard is from 1969. It is obsolete. My teenagers don't even know what a serial cable is. There is so much more that could be done with a USB bus connection, if they would only stop to think of the possibilities. Radio manufacturers need to stop taking the lazy, easy route with PC interfaces and use up to date standards.

As one of the "fanatics" who defended the use of the RS-232 connection on the Elecraft list I have to say that I think the suggestions that the K3 should have a USB port are narrow minded. The USB is a consumer interface intended for devices designed specifically to use with a PC. RS-232 is a standard that has stodd the test of time and provides the user with far more flexibility while still not preventing the connection to a USB port should that be required.

By providing an RS-232 port Elecraft has made it possible to interface with non-PC devices, for example a PIC based keyboard that could allow RTTY or PSK31 use without a computer using the modems built in to the K3. It could also interface with Ethernet-RS232 adapters - think remote control over the internet. Why lose that functionality just to save people the bother of using a $10 USB adapter instead of a straight serial cable that probably cost about the same?

I upgraded my shack PC this year and have just installed a 4-port serial card. Yes they are still available, even for PCs that only have the latest PCI-E slots like mine. I wonder if USB will still be around in 40 years time?

 

[Moleculo]

RS-232 is a standard that has stodd the test of time and provides the user with far more flexibility while still not preventing the connection to a USB port should that be required.

By providing an RS-232 port Elecraft has made it possible to interface with non-PC devices, for example a PIC based keyboard that could allow RTTY or PSK31 use without a computer using the modems built in to the K3. It could also interface with Ethernet-RS232 adapters - think remote control over the internet. Why lose that functionality just to save people the bother of using a $10 USB adapter instead of a straight serial cable that probably cost about the same?

RS-232 did stand the test of time, but it's lifespan has ended. It is no longer a standard that is supported by PC manufacturers. A USB port can be used for all of the examples you cited: USB keyboards - check, Ethernet USB adapters- been out for years, Remote control over a USB port - a no brainer. Icom and TenTec rigs already allow for several of these scenarios using USB ports. Elecraft should to. The reason for not offering the interface as a USB port is because if you make it RS-232, you offload the programming support for other devices to someone else. Is it a big deal to me? Not really. I just think it's a little odd for a rig that is positioning itself in the Elite class.

 

[Hutch]

G4ILO might I suggest you get familiar with some actual modern technology, because not even one of the things you list is only available to RS 232.

Caves stood the test of time but eventually we moved on, its time to move on.

 

[Moleculo]

Macros

One of the firmware updates that was released after the rig was introduced gave us the capability to create macros that can be stored on the rig or triggered from a PC. With these macros you can get the status of various things on the screen as well as the current status of the many functions. You can also toggle functions and command the rig to perform certain tasks. The rig can store eight macros where you can store them on any of 10 button functions (PF1, PF2, and M1-M4 as either tap or hold). You can also store 8 more macros assigned to software buttons in the K3 Utility software, as well as free-form input and send macros that you could type or paste from another text file.

You create and test macros using the supplied K3 utility by typing in the K3 custom macro language. The Programmer's Reference has the complete explanation, but commands are generally two or three characters followed by the instruction. It's cryptic but effective, and you'll have to keep the reference handy to do them. Here's a sample macro that is commonly used to enter the config menu, toggle ESSB, and then exit the menu: MN096;SWT11;MN255;

Here's another longer macro example from the programmer's reference that switches the rig to an AM broadcast station, sets the passband width, turns the attenuator on and the preamp off:
FA00001550000;MD5;FA00001550000;BW0400;IS 9999;RA01;PA0;

As I'm sure you can tell, you can get pretty creative with these macros and save yourself many key presses. If you're a programmer, you could also embed them in simple scripts incorporating boolean logic to really have complete computer control of the rig.

 

[Moleculo]

CW Thoughts

As you can probably tell so far from the review, I'm mainly a voice/phone and soundcard modes operator. My CW is pretty poor these days, but I still wanted to see what it is like to operate this mode on the K3.

I can really see why the hard core CW guys like this rig. I quickly found that the receiver is sensitive enough that I had to back way down on the RF gain, which got rid of most of the background hash while still allowing weaker signals to come right through. This is where all those technical test results that Elecraft publishes at elecraft K2 HF Tranceiver Performance Numbers really show off. The rig's ability to focus in on one station and block out all others is really quite impressive.

In CW mode, you can also turn on Dual Passband CW Filtering. The name of this function sounds similar to Icom's "Twin Passband Tuning", but it behaves differently. When you activate the Dual Passband on the K3, the filter graphic changes and looks like this:

What it does is let you set a narrow filter on the station that is right in the center so you can focus there, while still allowing other signals to pass through on the wider setting, but attenuated by 20db. This allows you to be aware of other stations in the passband without them dominating your receiver, which could be particularly useful in contesting or DXing.

The rig also has a couple of interesting CW tuning aids. Rather than try to describe what they do in print, I made a short video to demonstrate:



One last note on CW: Just as the rig can decode PSK31 and several data modes on the screen, it can also decode CW. I gave this a try and it seemed to work pretty well.

 

[Moleculo]

One other CW note:

I set the rig to full break-in mode and was pleasantly surprised to not hear any relay clicks at all when alternating between transmit and receive. It's a nice change from the 746 Pro to have QSK be nice and quiet except for the sound of the key.

 

[Robb]

Having taken the cover off of my Kenwood TS-2000 - as I never can break such a habit - I noticed the sturdy cast aluminum frame surrounding the circuit boards. Indeed, the Icom IC-718 also had a similar frame. The Kenwood radio may be small; but is very compact and considerably heavy for its size.

I noticed that the K3 has no real frame. It is a box with a handle and no underlying frame to support it. Does it concern you that this radio is a little bit too light or physically weak? Since the radio is connected by boards plugged in to the main board; does torsional twist concern you? Do you feel that it is tough enough for setting up a remote station that may see rigorous abuse in such circumstances?

 

[Moleculo]

Having taken the cover off of my Kenwood TS-2000 - as I never can break such a habit - I noticed the sturdy cast aluminum frame surrounding the circuit boards. Indeed, the Icom IC-718 also had a similar frame. The Kenwood radio may be small; but is very compact and considerably heavy for its size.

I noticed that the K3 has no real frame. It is a box with a handle and no underlying frame to support it. Does it concern you that this radio is a little bit too light or physically weak? Since the radio is connected by boards plugged in to the main board; does torsional twist concern you? Do you feel that it is tough enough for setting up a remote station that may see rigorous abuse in such circumstances?

The sides, rear, and front panel shield all act as part of the structure. There is also a frame stiffener that runs across the middle of the chassis shown here:

When you install the top and bottom cover, the whole unit is very structurally sound and rigid. I don't feel any type of torsion, nor do I see any concerns of ruggedness with the way the chassis is designed.

 

[N0NB]

Excellent series of reviews, Tim. I feel like I know more about this radio before I've even touched one than any other I've purchased. As I'm reading the operating manual over the course of the past several days I'm simply amazed at some of the features they have developed and managed to incorporate into the K3. I suspect that even with all of my research and study that there will still be a learning curve. The next couple of months ought to be fun.

 

[Moleculo]

N0NB, thanks for the kind words.

At this point, I've covered most of the features of the rig and options that I've purchased. The K3 also has an optional board that gives you a separate RX antenna along with an IF out and transverter interface. You need this if you are going to order a panadapter like the P3. There is also a 2 meter 10 watt PA, but I don't really need a 2 meter radio that only does 10 watts. A digital voice recorder is available, which would be useful to contesters, and if you like bling, you can get stainless steel covers.

If you really want to go full bore, you can order the second receiver and second set of roofing filters. This changes your A/B VFO's to true dual receivers. What is REALLY neat is that you can use different antenna combinations on each receiver to achieve diversity receive. No other receiver that I'm aware of gives you the ability to do diversity receive at this price point. At this point I'm debating on whether I want the P3 panadapter next or the additional receiver

Next up I'll cover some of the configuration settings that I changed to suit my personal operating preferences. Perhaps you'll find them useful, as well.

 

[Moleculo]

I wanted to make sure and follow up my AVG power output comments that I made earlier in this review because I spent some more time testing. Several private emails confirmed my suspicion that others were able to achieve better results than I was seeing.

Tonight after playing around with the rig some more, I was actually able to get the AVG power output real close to 60 watts on on specific syllables or words by playing with the audio response, audio gain input levels, etc. Typical voice response produced about 40-45 watts AVG. This is considerably better than I reported earlier, all things considered. What's different about the K3 is that it's AVG power output seems very sensitive to a lot of variables, especially your EQ settings. The K3 CMP setting quickly reached a point of diminishing returns for me, unlike what others are reporting, but you probably have to with what works for your voice/mic combination. I just can't run that much compression before the audio sounds like trash to me.

This definitely isn't a plug and play rig. It is probably easier to achieve certain results with other rigs, but what is nice about the K3 is the ability to tweak almost everything to your heart's content.

 

[Moleculo]

DSP Noise Reduction Follow Up

I've been experimenting some with the DSP Noise reduction feature quite a bit, especially because it works differently than most other radios that I'm used to. I've found through experimentation that the NR settings of 1-1 through 4-4 really work better for CW and 5-1 through 8-4 work better for SSB. As luck would have it, yesterday I read a post on the Elecraft mail list confirming my suspicion that this is indeed the case. I'm not sure when the manual will be updated, but it does not indicate this important setting differentiation at this time. Regardless, I find myself running with the 5-1 setting the majority of the time.

My Configuration Changes

Here is a few of the configuration changes I made to suit my preferences. You may want to look at these as a starting point or just ignore them completely .

• VFO CRS - This changes the rate of frequency adjustment when you press the COARSE button (it's mode specific). The default for SSB was 0.1 (I think). I changed it to 1.0 Khz to move it whole, even frequency steps when COARSE is selected.
• VFO CRS ROUNDING - This is also set by pressing "1" on the VFO CRS menu and makes the rig immediately round to the nearest 1 Khz when turning the dial. It took me a while to find this option, but if you set COARSE to 1Khz, you probably want to do this also.
• MEM 0-9 - Set to BAND SEL. This makes these memory slots act like band change buttons where they go to the last frequency used on the band you programmed them for.
• SQ MAIN - Set to SUB POT. I don't have a second receiver so I set the SUB dial on the MAIN receiver to act as a squelch dial. I like it!

BTW, did I mention that this rig has a 2 tone generator built in for SSB testing? Very cool.

 

[Robb]

I noticed that you you pointed out the different wattage readings with the K3 as compared to the Icom 746. Did you find a way to get a similar input into the power amp stages in each radio to give a relative/fair evaluation difference in the power output of each? I imagine that would be pretty hard to accomplish. Unless you could find a way to measure the the average output voltage of each preamp from both radios using a dual tone sine wave.

Many Kenwoods and Icoms (dunno about Yaesu in this regard) have an accessory input plug on the back of the radio that bypasses the radios on board mic preamp. Does the K3 have such an option? Can you you run your W2IHY gear in this fashion? Or do you have to go thru the front mic plug only?

BTW - Have you had a chance to run the K3 into your linear amp yet?
Did it require any special wiring requirements or relays?

 

[Moleculo]

I noticed that you you pointed out the different wattage readings with the K3 as compared to the Icom 746. Did you find a way to get a similar input into the power amp stages in each radio to give a relative/fair evaluation difference in the power output of each? I imagine that would be pretty hard to accomplish. Unless you could find a way to measure the the average output voltage of each preamp from both radios using a dual tone sine wave.

I don't have the equipment necessary to do that and I'm not sure what useful information it could tell you anyway. Remember, most of the numbers I was quoting was for SSB only. You're able to calibrate the PAs for 10 and 50 watts in CW mode so that you get exactly that. The AM figures I gave for the K3 were significantly better than anything I've seen from the Icom 746 Pro.

Many Kenwoods and Icoms (dunno about Yaesu in this regard) have an accessory input plug on the back of the radio that bypasses the radios on board mic preamp. Does the K3 have such an option? Can you you run your W2IHY gear in this fashion? Or do you have to go thru the front mic plug only?

The K3 has a front mic socket, rear mic input that is a 1/8" jack and a Line In. You can use any of them. I could run the W2IHY gear into any of them if I wanted to. I doubt you would want to use Line In for voice modes, though.

BTW - Have you had a chance to run the K3 into your linear amp yet?
Did it require any special wiring requirements or relays?

I just hooked up the amp last night. Nothing special is required to key the Ameritron AL-800H by the K3.

 

[Moleculo]

I've been away from the radio for a few days, but I'm back to playing with the K3 and discovering more features.

AGC Customization

The K3 is one of the few rigs to offer complete control of the AGC to suit your needs or preferences. Tonight I was able to find a frequency on 75 meters that had some distant stations barely above the noise level combined with some local strong signals coupled with fairly noisy band conditions to test out the AGC functions. Given that this was my first experience using a rig that allowed this much AGC customization, I was pleasantly surprised and impressed with the capabilities.

In addition to the typical FAST/SLOW toggle, you also have the ability to turn the AGC off. These are the other settings that are customizable:

-AGC Decay Characteristic: toggles between linear (typical) and "soft" decay. The "soft" setting is supposed to reduce IMD and is specifically for helping in pileups so you don't have to turn the AGC completely off

-AGC Hold: This applies to the Slow AGC setting and specifies the number of seconds that the AGC level is held after the signal drops. Also useful for working weak DX in pileups. I haven't had an opportunity to experiment with this setting yet.

-AGC Pulse: Turns on or of the noise pulse rejection. It is normally on.

-AGC Slope: The higher the setting, the more aggressive the AGC is on higher signals. As you approach the higher values, all signals will tend to start to sound like they're at the same level. This by definition, is what AGC is supposed to achieve, but it doesn't sound natural to everyone. I prefer it set so that stronger signals sound a little louder than weaker stations - it just sounds more natural to me. I ended up turning the default value down in value a bit.

-AGC Threshold: A rig specific value that determines how weak of a signal activates the AGC. Higher values means it takes a stronger signal to activate the AGC.

- AGC Fast/Slow decay rate: separate settings for the Fast/Slow settings respectively that determines how fast the decay works on each setting.

The neat thing about all of these settings is that the flexibility allows you to tweak them to affect your receiver as band and noise conditions change. You can set the AGC to work for your preferences at your location and then as the seasons change tweak them again. Setting the AGC settings at extremes also yields some interesting results that could be very useful for working that rare weak DX. For example, you could set the Threshold to maximum, while setting the Slope to minimum and then play with the RF Gain, Pre Amp, and Attenuator and see what affect that has on the really weak stations.

There really isn't a "right" way to set all these AGC items - it all depends on your preferences and what you're trying to accomplish. But the fact that you can customize them is really neat, and I can tell after playing with it for a bit tonight how valuable this capability is. I already tweaked both the Slope and Threshold to suit my listening preferences and now that I know how to do it, I know I'll get a lot of use out of the feature set.