EMRFD Message Archive 2962

Message Date From Subject
2962 2009-04-29 20:07:52 gt25psi2002 How to measure input/output impedance of any IF filter?
Dear Sirs,

How to measure input/output impedance of any IF filter? I've a few IF
filters restored from Swan 500cx stated 2.7kHz 5.5Mhz.

I prefer to use transformer impedance matching rather than transistor
buffer if possible.

Best regards,
Amin
2963 2009-04-29 21:24:26 John Kolb Re: How to measure input/output impedance of any IF filter?
You can get close by tuning a signal generator to the filter center
freq, and putting
a resistor in series between the generator and the filter. Adjust
the resistor value
until half the generator voltage is across the resistor and half
across the filter.
Measure the voltage with something that loads the test point as
little as possible -
Boonton RF voltmeter would be ideal. Test the filter in both
directions, as a few
have unequal in and out impedances. After the first test, check
again with the filter
terminated with the first pass values to see if termination affects
the measurement.

This will give a simple resistance number - doesn't say if the impedance has a
reactive component or not.

John

At 08:07 PM 4/29/2009, you wrote:
>Dear Sirs,
>
>How to measure input/output impedance of any IF filter? I've a few IF
>filters restored from Swan 500cx stated 2.7kHz 5.5Mhz.
>
>I prefer to use transformer impedance matching rather than transistor
>buffer if possible.
>
>Best regards,
>Amin
>
>
>
>
>------------------------------------
>
>Yahoo! Groups Links
>
>
>
2964 2009-04-29 21:25:47 John Kolb Re: How to measure input/output impedance of any IF filter?
When you know what equipment the filter came from, the impedance can
often be determined by looking at the schematic of the equipment.

John

At 08:07 PM 4/29/2009, you wrote:
>Dear Sirs,
>
>How to measure input/output impedance of any IF filter? I've a few IF
>filters restored from Swan 500cx stated 2.7kHz 5.5Mhz.
>
>I prefer to use transformer impedance matching rather than transistor
>buffer if possible.
>
>Best regards,
>Amin
2965 2009-04-30 03:53:48 gt25psi2002 Re: How to measure input/output impedance of any IF filter?
Dear John,

Thanks very very much for the advice.

The filters are from Swan 500cx transceivers which are tube type system.
I've seen the diagram but don't understand much.

Best regards,
Amin

2968 2009-04-30 07:19:23 john lawson Re: How to measure input/output impedance of any IF filter?
Amin. If they are the 5.595 Filters manufactured by Network Sciences that were in the Atlas transceivers, also, then there is a good chance that the  input and output impedances are 800 ohms.  I have some of those in my junkbox and got the information by calling them many years ago.  John, K5IRK 

2969 2009-04-30 07:36:02 gt25psi2002 Re: How to measure input/output impedance of any IF filter?
Dear John,

Thanks very much. This club is full of graceful people.

I understand that it was also manufactured by Network Sciences and it is
exactly at 5.50 mHz. It might be 800 ohms as well. Swan's filter is 6
poles while Altas's is 8 poles crystal filter. Atlas's filter was one of
the best then.

Anyway, does Network Sciences still exist?
I'm looking for their 16 poles filter manufactured specially for Swan
500.

Best regards,
Amin


2972 2009-04-30 10:13:52 john lawson Re: How to measure input/output impedance of any IF filter?
Hi Amin. Yes they are still in business...at least they have a website....see the link below. In additon you might want to do a google search. Several years ago I did one on all the different manufactures of Xtal Filters and found a bunch of information...It takes some time to search thru everything,  however I came away with some good information and Network Sciences did come up in the seach.  
 
    http://networksciences.com/
 
Have a good weekend and good luck on obtaining the needed information, 73, John K5IRK

---
2973 2009-04-30 17:18:17 Nick Kennedy Re: How to measure input/output impedance of any IF filter?
I don't understand this. The actual end-to-end resistance (or impedance) of
the filter won't necessarily equal the termination resistance it was
designed for, will it?

I like the idea, but can't convince myself that it's valid. I read some
time ago that the method for doing this is to plot the response with various
terminations untill you find one that "looks right". Seems tedious, so a
better method would certainly be preferred.

73--Nick, WA5BDU




You can get close by tuning a signal generator to the filter center freq,
and putting
a resistor in series between the generator and the filter. Adjust
the resistor value
until half the generator voltage is across the resistor and half
across the filter.
Measure the voltage with something that loads the test point as
little as possible -
Boonton RF voltmeter would be ideal. Test the filter in both
directions, as a few
have unequal in and out impedances. After the first test, check
again with the filter
terminated with the first pass values to see if termination affects
the measurement.

This will give a simple resistance number - doesn't say if the impedance has
a
reactive component or not.

John
>>

>
>
>


>


[Non-text portions of this message have been removed]
2974 2009-04-30 20:15:21 Wes Hayward Re: How to measure input/output impedance of any IF filter?
Hi Nick and gang,

You are absolutely correct. The input impedance of a crystal (or other) filter generally depends on the output termination. If that changes, the input also changes.

I suspect that the original question was incorrectly stated. Often we come upon a crystal filter and want to use it, but don't know how to terminate it to obtain a desired response. Wishing to discover the termination impedance, we merely try a variety. John Lawson, K5IRK, outlined how this is done and that method has always worked well for me. That is, terminate the filter initially in 50 Ohms at each end and take a look the response. Then add resistance at each end. This will generate additional loss during measurement, but that's not an issue so long as it is expected. Equal terminations are most common. The "proper" terminati
2976 2009-05-01 09:08:22 Jim Kearman Re: How to measure input/output impedance of any IF filter?
2979 2009-05-01 16:06:38 Alan Melia Re: How to measure input/output impedance of any IF filter?
I suspect thaat a better way to phrase the question would be "how do you
find the correct terminating impedance for a (crystal) filter"

Well I dont know whether it is the "right way" to do it but I sweep the
filter (slowly) with no termination. This gives a very peaky looking
response but can give a good idea of the design pass band of filter quickly.
I then apply a load (usually keeping the input drive at 50 ohms or below)
and adjust the load (cermet pot works quite well) to minimise the ripple and
loss. I can usually find a workable load value, but it is time
consuming.....still what is time to true amateurs :-))

If you have the time you don't even really need a sweeper to do this, just a
levelled signal source and a detector. .....Oh, and lots of patience and a
fair bit of perseverance :-))


Alan G3NYK
2984 2009-05-05 18:39:12 gt25psi2002 Re: How to measure input/output impedance of any IF filter?
Dear Sirs,

Thanks very much for all the advices and info. It is quite tough to get
the right answer without a scope and tries.

It seems that I need to buy one (scope) but still wondering how with
limited budget now - already in the list to buy about 2k USD.

Anyway, I still have to buy it because a lot more homebrews and
experiments need to be carried out. Anybody please advise me features
and specifications that is suitable for homebrew builder.

Best regards,
Amin
2985 2009-05-06 00:17:04 kerrypwr Re: How to measure input/output impedance of any IF filter?
If your budget is very very tight, consider an AD8307 power meter instead of a 'scope.

I have a roomful of testgear but the most-used, by far, is my homebrew AD8307 meter.

There is a design in EMRFD; a more-extended design that I used for my instrument is at;

http://www.arrl.org/qex/020506qex003.pdf

My meter is in the Photos section;

http://groups.yahoo.com/group/emrfd/photos/album/1974945304/pic/553525822/view?picmode=&mode=tn&order=ordinal&start=1&count=20&dir=asc

You can do very good measurements with a power meter and a good signal generator.

A 'scope and other nice things can come later.
2986 2009-05-06 04:29:20 Harold Smith Re: How to measure input/output impedance of any IF filter?
I would second that. Since I built my version of the AD8307 power meter I
use it much more than I had originally thought I would. And when you get a
scope, you can build a swept oscillator and, with it as the sourse, the
scope as the display and the power meter as the detector you can have a
scalar network analyzer, a very fast, easy way to look at filters.

de KE6TI, Harold

2987 2009-05-06 05:47:50 gt25psi2002 Re: How to measure input/output impedance of any IF filter?
Dear Sir,

I'm very interested in the instrument advised by your goodself. The
problem is that I'm not sure whether I can get all parts down here or
not.

Please advise me if there is anybody produces it as a kit or assembled
or even finished product for sales.

Best regards,
Amin

2990 2009-05-06 09:38:44 Alan Melia Re: How to measure input/output impedance of any IF filter?
I am not so sure I agree about the "order of preference" of test equipment,
but everone has their own ways :-))

I regard a scope as the most versatile piece of test equipment you can own.
You do not need spend a fortune, you do NOT need high frequency performance,
though it is nice if you can afford it. A 20MHz scope will do all you need
in the IF area. If you build a simple (diode) RF probe you can use to
1000MHz or above!! But the problem with narrow filter tests is that it will
probably not display a complete passband at slow sweep speeds. You do need
an expensive digital storage facility for that. Then there are the scopes
that plug into a PC (I have a 50MS/sec unit from Pico Tech bought on eBay
for about $100 which will display a spectrum to 25MHz as well as operate as
a 25MHz scope.)

For an "economy" test bench.......you must I suppose have a general coverage
HF receiver. In that case you could build a noise source on the lines of
those used in noise bridges. Ah the filters are not narrow enough you may
say....but put the AGC off and connect the audio output to a sound card and
run one of the free audio spectum analysers ...Spectrum Lab for instance.
With a run-of -the-mill on-board soundcard you should be able to plor filter
responses up to 20kHz bandwidth. No expensive HP equipment required at all!

I am not knocking test equipment, like you all I like (its really an
adiction !) using nice test gear, but it sometimes blinds us to the fact
that we have vast capability at our disposal in the form of a general
coverage receiver and a PC.

Best wishes
Alan G3NYK
2991 2009-05-06 14:52:02 kerrypwr Re: How to measure input/output impedance of any IF filter?
There seem to be kits available; Google "AD8307 power meter" and similar terms.

I don't know where you are, Amin, but the important parts are available from Futurlec;

http://www.futurlec.com/index.shtml

They are a mail-order firm and I'm sure they would send items to you if you can arrange payment (PayPal, for instance).

They have the AD8307 (in the easy-to-work-with DIP package) and the CX101A LCD meter that I used. Other components are pretty basic but they have op-amps, resistors etc if you can't get them locally.

The W7ZOI (EMRFD) meter is simple to build; the Kopski meter is a little more elaborate but the analog/digital feature is very useful.
2992 2009-05-06 20:46:19 gt25psi2002 Re: How to measure input/output impedance of any IF filter?
Dear Sir,

I'm from Malaysia. I don't know what is the minimum order they accept
with Paypal account.

Anyway, let me check if anyone has produced a kit similar to your design
first.

Best regards,
Amin

2993 2009-05-07 05:54:40 Corey Minyard Re: How to measure input/output impedance of any IF filter?
gt25psi2002 wrote:
> Dear Sir,
>
> I'm from Malaysia. I don't know what is the minimum order they accept
> with Paypal account.
>
> Anyway, let me check if anyone has produced a kit similar to your design
> first.
>
Futurlec is located in several places, including Thailand. I know
because I ordered something that happened to come from Thailand (to the
USA) and it took a *long* time :). They didn't have much of a minimum
order, either.

-corey

> Best regards,
> Amin
>
>