Softrock Lite 6.2
Adventures in Electronics and Radio
Elecraft K2 and K3 Transceivers
March 2008 Archive
30 March 2008
I've added a new page for the Z10000 Buffer Amplifier,
with pricing and ordering information. I'll take orders immediately for the
Z10000, but shipping may be delayed until mid-week.
I've also added a page for the Z10010 bandpass filter. I
am not taking orders for the Z10010, as I have not received the new printed
circuit boards and therefore cannot verify their correctness. The boards should
be here around mid-April.
29 March 2008
The postman left a couple of surprises for me today. Both
the parts orders I expected to arrive around the middle of next week showed up
today. Priority Mail is a real bargain for the speed of service.
I can't start full scale kitting yet because I have connectors
to assemble and also am short a few small parts that I will order Monday.
I assembled a Z10000-U today from the new PCB run, and
made performance measurements. Frankly, I'm a bit baffled, as the reverse
isolation is much better than I had measured for the first batch of boards in
which lead me to re-measure one of the first run Z10000 prototypes. It also
measured much better reverse isolation than I found a year and a half ago. I
believe I have an explanation, however.
Z10000 with connectors. Top view. Normally these pads
will be directly wired, but they are spaced for optional installation of
0.1" header pins.
Bottom view. The extra resistor on the input is a 49.9
ohm to permit accurate gain testing with 50 ohm test equipment.
I assembled the Z10000 for a nominal 6.7 dB gain, which is quite close to that
measured over the range 1 - 100 MHz. The slight rise as 100 MHz approaches is
due to a peaked high end frequency response. The nominal 6.7 dB gain is within
0.1 dB of the measured value at 8 MHz. The data is taken with an HP8752B
VNA, log frequency axis.
The reverse isolation measures 92 dB at 8.215 MHz, about 20 dB better than the
figure I quote, which was based on a connectorized prototype version of the
first PCB run nearly a year and a half ago.
I re-ran the forward and reverse gain of the prototype Z10000 from which I
based the 70 dB isolation value. The forward gain (with a 50 ohm through
termination on the input) closely matches the above data. It's actually about 20
dB better, measuring -115 dB isolation.
At this point, I started to question my test equipment and/or sanity. As a check,
I ran the reverse isolation with power disconnected to the connectorized
Compare this data with my data from a year and a half ago
for the same device.
March 2008 measurement - no power applied
September 2006 measurement
Interesting, no? The data matches quite well, with about a 5 dB offset. I now
think that my September 2006 data was taken with the power inadvertently removed
from the amplifier.
To verify this, I ran both amplifiers in forward and
reverse with a completely different test setup, an HP8640B signal generator and
an HP8557A spectrum analyzer. The data matches the VNA plots in both directions.
Hence, I conclude that the Z10000 is providing considerably greater isolation
than I thought.
Why would the amplifier measure much greater isolation
when powered up than unpowered? I assume this is because in the unpowered
state, all PN junctions within the AD8007 chip are un-biased and hence are, for
low signal levels, moderate value resistors, including parasitic or sneak path
junctions. When powered up, many of the sneak path junctions are
reverse biased and hence high impedance, which will greatly reduce unwanted
signal coupling within the AD8007.
I'm still cautious about revising the Z10000's
specifications to show the new reverse isolation figure without some independent
verification. Therefore, I'll make an early kit run of one or two Z10000 kits
available to purchasers that have the test gear to repeat my measurements
as an independent cross-check.
If you have the equipment and are interested, please drop
me an E-mail message and I'll respond. I can get the early kits out Monday.
28 March 2008
I received the Z10000 PCBs a couple days ago and
almost all the parts arrived today. I'm expecting the SMA bulkhead connectors
Monday, and a few other parts still in transit late in the week.
I plan to order the few remaining parts Monday so with
some luck everything will be here by April 4th or so. It will take me a
few days to kit up the parts, so don't expect the kits to go out until the 11th
I'll let prospective purchasers know when the kits ready
As a cost and shipping weight saving measure, I plan to
not provide a printed manual to accompany the Z10000's. The PDF manual is
available for download, and I'll work up a supplement to it to reflect how the
parts are packaged. I'm making changes in packaging as the prior method was much
too labor intensive.
Customers also ordering a Z10010 bandpass filter will
likely have a choice of waiting for the filters and having their order shipped
as a package or receiving two packages, as the filter PCBs won't be here until
mid-April at the earliest.
I have not mentioned it, but I can supply the Z10000
amplifiers wired and tested. So far no one has asked for that, but I'll
make it available as an option.
25 March 2008
I've ordered a small run of printed circuit boards for the
K2 and K3 IF bandpass filters, Z10010-K2 and Z10010-K3. I've bought enough
boards for 12 filters, which covers all the pending requests plus a few
spares. I'm using a different board manufacturer for these parts and delivery is
two to three weeks.
The Z10000 buffer amplifier boards have been shipped and I
ordered most of the electronics parts today. I have a few mechanical parts to
purchase and other odds and ends as well, but my mid-April shipping date
estimate should hold.
I also received a response from Fair-Rite to my question
concerning a change in Type 43 Material. More details at the
Estimating Q of Ferrite Cores
I mentioned before that our house was burglarized on on
February 28th, and we've been dealing with insurance companies, alarm companies
and the like since then.
If you are in Northern Virginia, I can recommend two
companies I've deal with over the last month.
Alarm systems: Digital Security, 204 S King St
Leesburg, VA 703-742-6060. Digital Security is a small company, but highly rated
by the readers of Washington Consumer
Checkbook, and I was impressed with the work they did. We went with a wired
alarm system, and the quality of installation was excellent. Dave Huthwaite, the
salesman who came out to quote the job, turned up wearing a tool belt and
tee-shirt and lead the installation team, doing much of the work himself. Dave
is extremely knowledgeable about the alarm business and if you are in the market
for a home alarm system, I recommend their services. Digital Security does not
have a web presence.
Anderson Window: The burglar entered by breaking
one of the living room Anderson windows. It turns out that only authorized
Anderson dealers can order replacement parts, with a three week lead time.
(The window in question is 22 years old, installed when our house was built.)
The replacement window came in today and was installed by Window Smith, a
division of Smitty's Building Supply Co., 7898 Notes Dr, Manassas VA, 703
530-1253. Dennis Morrison, the service technician, did an excellent job
replacing the broken sash and explaining to me how the Anderson windows are
constructed. www.windowsmith.net is
their web site.
Washington Consumer Checkbook. I mentioned that
Digital Security is highly rated by readers of Washington Consumer Checkbook,
but not everyone in the Washington DC area may be familiar with the WCC. WCC is
a non-profit, independent, organization that rates services in the Washington DC
area. (Sister organizations serve other cities.) Ratings and reviews are
obtained from their readers and statistically evaluated. WCC publishes a
quarterly magazine and subscribers may also access ratings and reviews on-line
at their web site http://www.checkbook.org/.
I've been a subscriber for many years and find their ratings accurate and very
22 March 2008
I've added a new page showing how one may use the
published Complex Permeability tables to more accurately determine the
inductance and estimated Q of inductors wound on ferrite material. The page
includes measured data with 43 and 61 material and an identical air-equivalent
core. Click here to view the
page or use the navigation table at the left margin.
18 March 2008
I've build a 400 KHz bandpass coupled resonator today,
centered at the K3's IF frequency, 8215 KHz. Details available by clicking
here or via the K3 IF Bandpass Filter
link in the navigation bar.
If there is sufficient demand, I will consider a run of
these bandpass filters, either in the K2 version or the K3 version. These will
be assembled, aligned and tested units, with a sweep response plot. Kits will be
available only to purchasers with adequate test equipment to build and tune
Whether you need a bandpass filter such as these between a Softrock receiver and your K2 or K3, I will leave to you to determine.
16 March 2008
I've tinkered a bit with the web site structure, moving
most Z90/91 material to sub-pages under the
Prior Products link.
Nothing has been deleted, just rearranged.
I've added a page showing B-H curve measurements for a
Fair-Rite 43 material ferrite core. Click
here to view the B-H page or follow
the link at the left margin.
14 March 2008
I've decided to go forward with the second run of
Z10000 buffer amplifier kits, with delivery around the middle of April.
The PCB will be identical with the first Z10000 run, with
a fix or two in silk screening. In looking at the parts cost over the last few
days, I now realize that that there was a major cross-subsidy between the Z90/91
and the accessory Z10000 buffer amplifier. The options and pricing for the
second Z10000 run are:
- Base kit�printed circuit board, all electronic parts
to build the broadband (universal) version of the PCB, including a complete
set of gain-setting resistors. Price is $24.95.
- K2 filter parts�Parts necessary to provide a bandpass
filter shaped response, centered around 4.9 MHz for the K2's IF. The filter
parts are installed on the printed circuit board. $9.95.
- Internal Installation parts�Parts necessary if you
wish to install the Z10000 inside a receiver. Although these parts are
intended for the K2, they can be adapted to work in most receivers. These
include a pre-assembled SMA female bulkhead connector with 2 ft length of
RG178 miniature Teflon coaxial cable, 3 pin male & female header, 8 pin
female header, fish paper, stand-off and wire. Price is $14.95.
- All inclusive kit�All three above. Price is $44.95.
- Jumper cable, male SMA to male BNC, approx. 3 ft (90
cm) length, RG-174 cable. $9.95.
These prices include shipping via first class mail within
the US. International shipping is available at extra charge. Based on the size
and weight of the kit, international first class air-mail shipping should be
around US$4.00 to 5.00 to most European countries. International purchasers
should contact me via E-mail for an exact price quotation. Payment may be made
by PayPal or check.
The price does not include a printed manual. You are
responsible for downloading the assembly manual from this site.
If you have an IF output port on your receiver/transceiver
and just want a Z10000 to provide isolation between your Softrock or other
panadapter and the receiver, all you need is the base kit. The Z10000 PCB might
be installed in the same enclosure as your Softrock.
If you have an Elecraft K2, then you likely will find the
all inclusive kit the best option. Or, you can purchase just the base kit
and install it inside your K2 following the suggestions in the Z10000
If you have a receiver or transceiver other than a K2, but
without an IF output port, then you will likely wish to purchase the base kit
and the internal installation parts. You would not need, nor find useful,
the K2 filter parts.
One caveat must be made now�my measurements on a
Softrock 6.2 Lite receiver show local oscillator leakage (crystal frequency / 4)
at about -40 dBm, an extremely strong signal level. (Click
here for details.) The Z10000 has a reverse gain (isolation) around -70 dB,
depending on frequency. (Details are in the Z10000 documentation, available by
here.) Thus, the Z10000 will reduce the Softrock's leakage level to around
-100 to -110 dBm, assuming proper attention is paid to shielding and bypassing.
However, a -110 dBm signal is still quite audible when injected into a
receiver's IF stage. Hence, either additional isolation will be necessary, or
you will wish to intentionally shift the Softrock's center frequency to one side
or the other of your receiver's IF bandpass filter.
Some transceivers, such as Kenwood's TS940, employ a
separate panadapter signal path and hence are less subject to this leakage
problem than others. In the Elecraft K2, for example, you will wish to offset
the Softrock several KHz from the nominal center.
08 March 2008
I've revised the intermodulation measurement results at
the 6 dB hybrid coupler page, and
replaced the original data with the revised results. I've also added a
conversion loss measurement for the Minicircuits ZHL-3A broadband amplifier.
I've also rearranged the material and added page bookmarks
for easier navigation.
06 March 2008
I've added performance data at the
6 dB hybrid coupler page for:
- Minicircuits ZFSC-2-1W combiner
- Minicircuits ZFRSC-2050 combiner
- RCA VH476 CATV splitter
I've also included intermodulation performance
measurements of the Wilkinson lumped combiner, two home brew combiners and the
05 March 2008
Although the 6 dB hybrid couplers I've been recently
tinkering with have more than decent isolation and bandwidth performance, at
least as long as a small binocular core is used, the question has been raised by
Geoff, GM4ESD, as to whether the ferrite core will be a contributor to
intermodulation distortion and whether the level of IMD will prevent properly
measuring very high IP3 devices, such as the Mode H mixer developed by G3SBI. A
Mode H mixer can have an IP3 in the +40 to 50 dBm range and hence
requires a test signal with an IP3 at least 6 dB and preferably 10 dB greater.
Although larger ferrite cores will certainly alleviate the
IP3 problem, a different approach is to remove the ferrite totally and go with a
lumped Wilkinson splitter/combiner using air core inductors.
The main concerns a Wilkinson combiner present are (a)
relatively poor port-to-port isolation; and (b) narrow bandwidth.
The narrow bandwidth issue can be resolved by building a
separate coupler for each amateur band. I found close to 30 dB isolation in the
Wilkinson combiner I built today for the 14 MHz band, which is 10 dB or more
above my expectation.
I've added the details to the
6 dB Hybrid Combiner page.
04 March 2008
I received a telephone call today to discuss how
well the Hakko FT-800 thermal
wire stripper has worked for me.
When I wrote my review of the FT-800, I had owned it for a
week or two, and now that I've had several months experience with it, I can only
say that I wish that I had purchased it years ago. It's an excellent product
that does exactly what it should do; strips wire without leaving a nick behind.
It will leave a small collar of melted insulation with some insulation types,
but this is generally not a problem.
I've also used the FT-800 with Teflon insulated wire, both
coaxial cable and normal hookup wire. It does a great job with Teflon. Of
course, the temperature setting must be increased, to about 8 on the 0-10 dial.
And, when hot enough to thermally strip Teflon insulation, the heating element
glows a dull red.
I've also found it useful to keep a small wire brush,
either brass or stainless steel, at hand to clean the stripping blades after
use. Small pieces of insulation stick to the blades and will smoke and leave a
deposit unless cleaned. I bought a small stainless steel brush, illustrated
below, from McMaster-Carr, stock no.
33335K2, for this purpose. The brush cost $9.55 in single lots. A normal plastic
or animal hair bristle brush isn't happy when used on a hot surface. And the
stripper blade is rather small, so the trick is to find a small, soft
metal bristle brush. The 33335K2 brush has a Delrin handle and is about the size
of a normal pencil. Although the stainless steel bristles are stiffer than I
would have liked, it does a good job at cleaning the blades.
03 March 2008
I've added a quick intermodulation test of one of the 6 dB
02 March 2008
I've expanded the 6 dB
hybrid combiner page to reflect additional core types.
01 March 2008
As usual, I've moved the February 2008 page to archives,
reachable by clicking here, or via the navigation
table at the top of this page.
I've added a new page,
6 dB hybrid combiner, with some data on a couple of hybrid combiners I built
over the last couple days. These are simple devices and work quite well for
combining two signal generator outputs for intermodulation testing.