John, if you really want to measure impedance today just
get one of those MJF impedance bridges. They have done
all the compensation for you and put it in a computer built in
to the unit and it reads out the frequency and impedance
on a LCD display.
I have a MFJ-269 and it works from 1.8 to 170 MHz and also
the 430 MHz band. It has an N-connector on the input.
I have checked it with a Mini-Circuits 50 Ohm termination
good to 1 GHz. I also have very good HP calibrated 50 Ohm
loads with APC7 connector.
Regards, Jim W7ANF
-----Original Message-----
From: John - G3JAG <patents@dx0man.prestel.co.uk>
To: Dr. Gerald N. Johnson, P.E. <geraldj@ames.net>
Cc: tentec@contesting.com <tentec@contesting.com>; Michael Melland
<badger@vbe.com>
To: <tentec@contesting.com>
Date: Saturday, March 04, 2000 4:30 AM
Subject: Re: [TenTec] PT-340 noise bridge vs. SWR meter
>
>Probably much less than an extra inch .... the best article on this
>topic was perhaps the one in Ham Radio quite a few years back. The
>authors found that toroid lead length was remarkably critical and
>showed what they called "inductive rotation" because the error was
>frequency dependant.
>
>Their final design provided very simple and adjustable compensation; it
>gave a bridge with excellent accuracy over the HF bands. I have never
>seen any later reference to this article, which is pretty amazing
>considering the work that went into writing it. Maybe its the "not
>invented here" syndrome.
>
>I have the mag somewhere in my junk room, but so far I cannot locate
>it, or I would be able to give the date/page info.
>
>John G3JAG
>
>On 04-Mar-00 Dr. Gerald N. Johnson, P.E. wrote:
>>
>> It only takes an inch or so of extra lead length in the bridge to
>> make a
>> noise bridge read wrong. What you need is a precision terminator.
>> They
>> are commercially made by Bird, GR, M/A.
>>
>> You can make a decent load, but its a bother. You need a 49.9 or 51.1
>> ohm 1% resistor, a connector an a few inches of coax. With a half
>> watt
>> resistor, I use type N connector and some material from R-8 coax. To
>> prepare the resistor, take off the braid intact from the coax and
>> pull
>> the center conductor from the insulation, preserving that insulation.
>> Drill the opening left from the center conductor to just fit over the
>> resistor. Then using a pencil sharpener carve that insulation to a
>> point. cut a tapered section of insulation the length of the body of
>> the
>> resistor. put it over the body of the resistor. stretch a couple
>> inches
>> of braid into a cone to fit over that insulation. Solder the pointed
>> end
>> of the cone to the resistor lead right at the end of the cone of
>> insulation. At the fat end of the insulation flare out the braid and
>> mount it in the connector, trimming the ungrounded end of the
>> resistor
>> short and soldering it to the coax connector center pin. This could
>> probably be built in a RG-59 adapter to screw into a UHF connector,
>> but
>> I don't depend on UHF connectors for measuring anything. This
>> construction should be good to a Ghz or so, and so not be different
>> from
>> DC to 30 MHz.
>>
>> Then you can see which measuring instrument is wrong.
>>
>> SWR bridges can be in error because of strays too.
>>
>> An alternative dummy load, is a long roll of coax. Say 100 feet of
>> RG-174 or 500' of RG-58 terminated with a simple 51 ohm resistor. The
>> loss of either length of coax is enough that two passes through the
>> coax
>> is as good as a precision termination.
>>
>> 73, Jerry, K0CQ
--
FAQ on WWW: http://www.contesting.com/tentecfaq.htm
Submissions: tentec@contesting.com
Administrative requests: tentec-REQUEST@contesting.com
Problems: owner-tentec@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|