UPDATED May, 2010

Part 3 of the Second Receiver for an IC7700

A compensated splitter for feeding a software receiver

Jeff VE1ZAC

Device Communications

I have been having lots of fun getting the SDR-IQ running in conjunction with my IC7700 and DXLab software. The coms with the SDR, the IC7700 and the PC have been sorted out thanks to a tip by another user. The IC7700 has two CI-V channels… one via a DB9 RS232 style port and the other via the CAT line that ICOM and other manufacturers use to hook various RF gadgets together. The SDR talks to the rig via its own CI-V adapter to the IC7700 and the PC goes direct to the IC7700 via the RS232 port. Both ports are transparent as far as commands go. This combination has been working perfectly for several weeks. It allows any of the modes of the SDR to be controlled by the IC7700 and/or the DX Lab suite applications via Commander.

Operating modes

I am very enamored with the spectrum display that follows my rig tuning and allows a really beautiful look at signals close by or anywhere else, for that matter. You can look else where for band openings in other bands than the one you are parked in. or hunt out pile up openings. It’s great! If you turn on the demodulators in the software, you can actually listen and track one or two signals as you move the main rig around the band.

CWSkimmer

WARNING ! Do not try this if you are prone to addictions! (hah hah) Seriously, if you haven’t tried this amazing piece of software, I suggest you head on over to the DXAtlas site and download this software. It is free to sue for 30 days. After just 5 days, I am convinced it is a bargain if you register. ($75 to Alex at DXAtlas, Afreet Software). This software works with ordinary sound cards and can scan the audio pass band of the receiver you feed it with, but frankly, this is like walking compared with an F18! If you have a software receiver, you really want to try this out. Skimmer grabs the full command set of the SDR-IQ (and other SDRs as well) and operates it as a standalone receiver. Great, you say, so what? Well, the ‘what’ is that Skimmer actually ‘Operates’ for you and decodes every CW signal it can latch onto in the SDR pass band. In SDR-IQ mode, this is between 50 and 190 kHz. Alex has implemented a decoding scheme using Bayesian statistics and filtering. Pretty sophisticated stuff. And it works ! After a few minutes of listening, Skimmer begins to populate a database with call signs and other useful DX features. It looks for things like a ‘599’ being transmitted, to help you locate the station in a pile up that the DX station is working. After 20 or 30 minutes on 20M CW, I have seen the database grow to over 100 entries.

And there is more. There is a built in telnet feature which if turned on, allows you to hook up to Skimmer as a spot source. All of its QSOs and data are available to your DX software and full inter action with your transceiver. I have been using this for a few days with DXLab, and it works perfectly. It’s pretty amazing to have this thing sitting there listening and reporting on stations for you… just like a second op. What a great assistant! (I can see why there is a move to declare Skimmer use as assisted contest operating.)

The antenna splitter with a gain stage

My trial splitter device as reported in Part 1 works very well. It has about 3 dB losses though and I occasionally find that a bit annoying, especially on the higher bands. Further, since I know longer have any interest in operating a second transceiver as a second receiver, I don’t need all the bits to prevent accidental transmitter input. So, time to toss the experimental unit and build a new one. This time I will put it in a proper box on top of the rig, include some switches to allow easy switching to another receive only antenna, and include a wide dynamic range gain stage to compensate for the splitter losses. This will have a switch to defeat the gain stage, if not needed. (My splitter ideas were gleaned from http://www.qrp.pops.net/swl-ant-split.asp ) I have swept this unit with a spectrum analyzer and tracking generator right up to 60 MHZ with satisfactory performance.

For a gain stage, I wound up back at the same site and found a wonderful broadband gain stage with low noise transformer feedback. See http://www.qrp.pops.net/preamp.asp and check out the broad band Norton design using a 2N5109. ( 2N5109s are available from Mouser Electronics http://ca.mouser.com/Home.aspx . I got a handful a few years ago for just this kind of use) I setup my amp for about 6 dB gain. In practice, it just seems to push the splitter transformer losses back to 0 DB. There is potential for increased IMD issues with some gain stages. This particular one is set at a very low gain, has a very low noise factor and can be switched out when needed. That should take care of any issues, if they arise.

Construction

I use Eagle Cad (the free version) to make small circuit boards with my little Sherline CNC mill. There is a free G-Code output module available via the Yahoo PCB-GCode group. This works very well. Here is the schematic and board layout for a single sided board.

( If anyone wants my Eagle Cad files or the G-Code outputs, email me and I will send them to you) My board includes the splitter transformers and pads to allow external switch connections.

Jumpers J1 and J2 are shown to allow ferrite beads to be installed on the collector lead, as shown. T1 and T2 are the splitter transformers, and 1,N,and M are three windings on a binocular ferrite core for the amplifier. My power is taken from the 8V regulated pin on the ACC2 plug on the IC7700.

Here is a board layout view and a shot of my little homebrew CNC setup. I use Mach3 on an old laptop to run the system. The modified PC case has the stepper driver inside and uses the PC power supply. Note the large red ‘Panic’ button on the desk. I have not had to use it yet !

I engage or disengage the splitter from the IC7700 antenna controls. This unit takes care of all the features needed to dress up using an SDR with the IC7700. ( I know, the labels are on crooked) Actually, it would work the same for just about any transceiver setup. After seeing how well these standalone SDRs work, there doesn’t seem to be any real need to have one look at limited spectrum from an IF port on a transceiver. Especially when you can range around the entire HF band directly with the SDR.

May, 2010 UPDATE on this splitter: There was an inherent problem with the gain stage used in this splitter. The amplifier dies below 1.8 MHz. I occasionally like to tune around in the BCB and LF bands and this means I would have to live with the approximate 5 to 6 dB loss of the splitter, or go through the drill of switching the external receive loop out. That also meant I couldn’t use my K9AY, my best low band receive antenna or the phase controlled verticals I just setup. Also, since this article first released, I have discovered the joys of high intercept low noise Norton style preamps and have built 5 of the units now.

I had intended to replace the preamp in the splitter with one of these and finally got around to it. The swept loss of the splitter is about 5 to 6 dB over the range of 50 kHz to 35 MHz. I elected to use a 10 dB gain preamp with a 4 db pad at the output to increase it’s isolation characteristics. The unit swept at + .5 dB across the entire range as a result. Essentially, I plan on leaving the splitter in place as a default arrangement.

Here is a shot of a pair of the units. One is 10 dB, the other about 18 dB for the phased vertical project. These are push pull maps with 2N5109’s , homemade heat sinks and the transformer feedback promoted in the Norton original article and by Dallas Lankford in his website postings . ( Link available on this website front page).

On some of the units I used 75 mix ferrite toroids to make the input, output and feedback transformers to insure a very broad bandwidth. On these particular units I used 73 mix ferrite binocular cores with no apparent lack of frequency range. Windings are a combination of #30 Kynar wire, magnet wire and a few stray bits of solid hookup wire. I tested the finished amp from 50 KHz to 30 MHz with no dips apparent anywhere. The pairs of .1 uFd capacitors are to provide low ESR at the LF beacon band, and up.

If you need very high quality preamps for splitters or receive antenna projects, I highly recommend these preamps. If you don’t want to build them, you can buy them from KIWA.

I have these preamps built into this splitter project, a K9AY gain stage, as well as three of them inside the phase control for the short vertical phased array.

A Postscript: I discovered that DXLab and Commander like to running, before you engage CW Skimmer. For some reason, there are com port conflicts if done the other way around. I will look into this when the opportunity arises, but in the meantime, the order of starting up is easy to deal with.

I hope you have as much fun with an SDR as I have. A truly useful addition to any shack.