Improving a Grid Dip Meter
Paul Burridge
Hi all,
I have a fairly oldish (say 1970) GDM which can also function
(theoretically at any rate) as a relative field-strength meter.
However, it's pretty 'deaf' so the later functionality of it is
impaired. The degree of dip when sussing out plain tuned-circuits is
also pretty lousy. Taking the back off shows the amplifying section of
this unit to consist of two elderly-looking tin-cased 2SB422
transistors which for some reason are soldered in circuit with full
inch-and-a-half leads. Seems a bit odd given the claimed top-end
frequency response of 280Mhz. Anyway, what, in the view of the panel,
are the most suitable up-to-date devices to replace 'em with in order
to improve sensitivity?
thanks
--
"What is now proved was once only imagin'd"
- William Blake, 1793
George R. Gonzalez
"Paul Burridge" <p...@notthisbitosiris1.co.uk> wrote in message
news:hkqg0vcbres0tmiep...@4ax.com...
If those are bipolar transistors, that would explain the poor dips.
You need a tube or FET in the oscillator signal to get a good dip.
Try the ARRL handbook, they used to have a FET dipper in there.
Paul Burridge
<gr...@attbi.com> opined thusly:
>If those are bipolar transistors, that would explain the poor dips.
>
>You need a tube or FET in the oscillator signal to get a good dip.
Thanks. I note that in the handbook it says that they are "PNP type
alloy-junction germanium transistors" whatever they are. The
oscillator stage uses a "PNP MESA type germanium transistor" - again
not a type I'm familiar with.
Something's definitely amiss with it anyway, since (when used as a
relative FSM) the tx's antenna needs to be practically physically
coupled to the meter's sensor coil!
>Try the ARRL handbook, they used to have a FET dipper in there.
I have one to hand and will have a gander. In the meanwhile, if
anyone's got a suitable device to suggest, I'd be interested to hear.
John Woodgate
<p...@notthisbitosiris1.co.uk> wrote (in <hkqg0vcbres0tmiephkrkqe01a1lj5i3
9...@4ax.com>) about 'Improving a Grid Dip Meter', on Tue, 24 Dec 2002:
>I have a fairly oldish (say 1970) GDM which can also function
>(theoretically at any rate) as a relative field-strength meter.
>However, it's pretty 'deaf' so the later functionality of it is
>impaired. The degree of dip when sussing out plain tuned-circuits is
>also pretty lousy. Taking the back off shows the amplifying section of
>this unit to consist of two elderly-looking tin-cased 2SB422
>transistors which for some reason are soldered in circuit with full
>inch-and-a-half leads. Seems a bit odd given the claimed top-end
>frequency response of 280Mhz. Anyway, what, in the view of the panel,
>are the most suitable up-to-date devices to replace 'em with in order
>to improve sensitivity?
Changing the transistors is most unlikely to produce any improvement,
assuming they are not faulty. To take advantage of modern devices, a
total re-design would be required.
--
Regards, John Woodgate, OOO - Own Opinions Only. http://www.jmwa.demon.co.uk
Interested in professional sound reinforcement and distribution? Then go to
http://www.isce.org.uk
PLEASE do NOT copy news posts to me by E-MAIL!
N. Thornton
Long leads are needed on geraniums. Try soldering them with short
leads and they're dead.
Regards, NT
carltons
big...@meeow.co.uk (N. Thornton) wrote:
> Paul Burridge <p...@notthisbitosiris1.co.uk> wrote in message
news:<hkqg0vcbres0tmiep...@4ax.com>...
> > Hi all,
> >
> > I have a fairly oldish (say 1970) GDM which can also function
> > (theoretically at any rate) as a relative field-strength meter.
> > However, it's pretty 'deaf' so the later functionality of it is
> > impaired. The degree of dip when sussing out plain tuned-circuits is
> > also pretty lousy. Taking the back off shows the amplifying section of
> > this unit to consist of two elderly-looking tin-cased 2SB422
> > transistors which for some reason are soldered in circuit with full
> > inch-and-a-half leads. Seems a bit odd given the claimed top-end
> > frequency response of 280Mhz. Anyway, what, in the view of the panel,
> > are the most suitable up-to-date devices to replace 'em with in order
> > to improve sensitivity?
> > thanks
I gather from the info given that the two transistors are part of the post
processing of the "grid dip". Germanium alloy transistors had Ft's way
too low to be usable at 260 MHz. I would guess that your active element
would be a 6CW4 or some other such tube. If my guess is correct, then
replacing the transistors can be done, but would probably not fix your
problem adequately. You would be better off taking the advise given you
already from others and try and salvage what you can from the present grid
dipper, such as the case, meter, supply, etc.
Good luck and happy trails.
Steve WB4CZR
David Snowdon
offered a "Tunnel-dipper" in the '80s, and it had a bad reputation because
of the low level of RF it emitted.
David Snowdon, VA3DKS
--------------------------
In article <motsteve-251...@c-66-176-144-254.se.client2.attbi.com>,
Paul Burridge
opined thusly:
>>I gather from the info given that the two transistors are part of the post
>>processing of the "grid dip". Germanium alloy transistors had Ft's way
>>too low to be usable at 260 MHz. I would guess that your active element
>>would be a 6CW4 or some other such tube. If my guess is correct, then
>>replacing the transistors can be done, but would probably not fix your
>>problem adequately. You would be better off taking the advise given you
>>already from others and try and salvage what you can from the present grid
>>dipper, such as the case, meter, supply, etc.
I shall do so, thanks. Having compared the remarks on this group with
an inspection of the internals leaves me with two inescapable
conclusions: the circuitry and components are *crap*; the hardware is,
OTOH, first class so it's a case of ripping out the guts and starting
again with an up-to-date circuit. Thanks, chaps.
p.
carltons
p...@notthisbit.osiris1.co.uk wrote:
I think that you are absolutely right in your conclusions. However, it is
abundently clear that there should be a new and better way of doing this
function without too much expense. The original design used an oscillator
and it was coupled loosely to an other circuit which should be "tuned" to
the same frequency of the measuring instrument. A "dip" in grid current
would indicate this resonance and give you an idea as to where the
external circuit was tuned. The idea is sound, but I believe the
implementation to be faulty in that the circuit be measured would pull the
test instrument off freq. rendering the reading off by the length of the
pull. If the field being used for measurement could be gauged and the
absorbsion of same be measured with no change in its forced frequency,
then this would be useful. An instrument to be used to fit all loads
doesn't seem very practical here. I believe a design to fit a particular
range of loads makes more sense. Okay design engineers, the problem is
now crudely stated. I guess now the title of the original post will be
very precise, save the reference to the tube element now not very likely
to be used in the design.
Happy Thoughts,
Steve WB4CZR
Mike Monett
[...]
> I think that you are absolutely right in your conclusions. However, it is
> abundently clear that there should be a new and better way of doing this
> function without too much expense. The original design used an oscillator
> and it was coupled loosely to an other circuit which should be "tuned" to
> the same frequency of the measuring instrument. A "dip" in grid current
> would indicate this resonance and give you an idea as to where the
> external circuit was tuned. The idea is sound, but I believe the
> implementation to be faulty in that the circuit be measured would pull the
> test instrument off freq. rendering the reading off by the length of the
> pull. If the field being used for measurement could be gauged and the
> absorbsion of same be measured with no change in its forced frequency,
> then this would be useful. An instrument to be used to fit all loads
> doesn't seem very practical here. I believe a design to fit a particular
> range of loads makes more sense. Okay design engineers, the problem is
> now crudely stated. I guess now the title of the original post will be
> very precise, save the reference to the tube element now not very likely
> to be used in the design.
>
> Happy Thoughts,
>
> Steve WB4CZR
Actually, a GDO is an extremely valuable lab tool, but it not considered
a precision instrument. It only gives indications of where resonances
occur so you can investigate further. One of the significant advantages
is finding unexpected resonances that can affect circuit behavior.
At resonance, the external circuit reflects a resistive component to the
oscillator tank, which reduces the Q but has little effect on the
frequency. Reducing the Q reduces the voltage across the tank, and is
indicated by the dip in the output voltage. Probably the biggest effect
on the tank frequency comes from adding stray capacity to ground and
setting up circulating currents in nearby metal. These effects can tend
to cancel each other, depending on how the coil is wound and its
orientation to the circuit under test.
Once a resonance is located, you can reduce the coupling error to a
minimum to improve the frequency accuracy, and add an external counter
such as the Radio Shack Frequency Counter, P/N 22-305, to eliminate the
dial calibration error.
Adding a headphone to the GDO can help locate parasitic oscillations that
are within the frequency range of the GDO. These can be difficult to
locate, since they may be above the bandpass of your scope, and the added
load of a scope probe can kill the oscillation.
Mike
John Woodgate
wrote (in <motsteve-291...@c-66-176-144-254.se.client2.attbi.c
om>) about 'Improving a Grid Dip Meter', on Sun, 29 Dec 2002:
>The idea is sound, but I believe the
>implementation to be faulty in that the circuit be measured would pull the
>test instrument off freq. rendering the reading off by the length of the
>pull.
No, with *loose* coupling and the GDO tuned for minimum grid current,
there is no pulling. The circuits are synchronous.
carltons
<j...@jmwa.demon.co.uk> wrote:
> I read in sci.electronics.design that carltons <mots...@attbi.com>
> wrote (in <motsteve-291...@c-66-176-144-254.se.client2.attbi.c
> om>) about 'Improving a Grid Dip Meter', on Sun, 29 Dec 2002:
> >The idea is sound, but I believe the
> >implementation to be faulty in that the circuit be measured would pull the
> >test instrument off freq. rendering the reading off by the length of the
> >pull.
>
> No, with *loose* coupling and the GDO tuned for minimum grid current,
> there is no pulling. The circuits are synchronous.
Very loose I would imagine and probably not a dramatic effect to note.
Steve
J M Noeding
wrote on Tue, 31 Dec 2002 01:26:34 GMT :
>John Woodgate <j...@jmwa.demon.co.uk> wrote:
>
>> I read in sci.electronics.design that carltons <mots...@attbi.com>
>> wrote (in <motsteve-
>> om>) about 'Improving a Grid Dip Meter', on Sun, 29 Dec 2002:
>> >The idea is sound, but I believe the
>> >implementation to be faulty in that the circuit be measured would pull the
>> >test instrument off freq. rendering the reading off by the length of the
>> >pull.
>>
>> No, with *loose* coupling and the GDO tuned for minimum grid current,
>> there is no pulling. The circuits are synchronous.
>
>
>Very loose I would imagine and probably not a dramatic effect to note.
>
>Steve
it is better to use a signal generator with a coax cable to a probe
consisting of 2-5 turns dependent on frequency, and measure the
voltage drop on the output from signal generator when 'dip' occurs.
You could also use a potmeter to vary the impedance between the
generator and voltage detector with coaxcable
It was described in QST May 1986
Hayward "Simplified Scalar Network Analyzer" p.14-20
(with attachments for various measurements)
(Feedback Aug p.40)
"Beyond the dipper(using signal generator) W7ZOI QST"
Just keep the HP651B for the purpose since it is easier with a
continous variable generator than a thumbweel tuned SG.
Jan-Martin
LA8AK
http://www.laud.no/la6nca/LA8AK/
--
remove ,xnd to reply
Mike Monett
[...]
> it is better to use a signal generator with a coax cable to a probe
> consisting of 2-5 turns dependent on frequency, and measure the
> voltage drop on the output from signal generator when 'dip' occurs.
> You could also use a potmeter to vary the impedance between the
> generator and voltage detector with coaxcable
>
> It was described in QST May 1986
> Hayward "Simplified Scalar Network Analyzer" p.14-20
> (with attachments for various measurements)
> (Feedback Aug p.40)
> "Beyond the dipper(using signal generator) W7ZOI QST"
>
> Just keep the HP651B for the purpose since it is easier with a
> continous variable generator than a thumbweel tuned SG.
>
> Jan-Martin
> LA8AK
> http://www.laud.no/la6nca/LA8AK/
>
> --
> remove ,xnd to reply
Very interesting, Jan. Did the authors mention any problems with
reflections on the coax due to the mismatched termination?
Mike
John Woodgate
om>) about 'Improving a Grid Dip Meter', on Tue, 31 Dec 2002:
that there is only on grid-current minimum.
carltons
This makes more sense, Jan-Martin. I guess my problem with the whole
thing revolves around the fact that the concept of grid dipping is good
for approximate use, but not too good if you have limited resources (i.e.
equipment and advisors) and are trying to build a project. I used a gdo
in the Navy which was considered at the time to be state of the art (I
think it was the military version of the James Millen gdo), but my results
did not provide much help in determining inductance values. I got all
kinds of dips and peaks everywhere, some of which were not too dramatic in
depth. I would suspect that much of the problem was due to outside
interference at frequencies near the measurement. I ended up building a
bridge and those results were a difference of day and night. Moral of the
story is that the right tool will do the job, not to mention the fact that
doing the project is very educational in itself. :-)
Steve WB4CZR
J M Noeding
<z06hg...@sneakemail.com> wrote:
>J M Noeding wrote:
>
>> it is better to use a signal generator with a coax cable to a probe
>> consisting of 2-5 turns dependent on frequency, and measure the
>> voltage drop on the output from signal generator when 'dip' occurs.
>>
>> It is described in QST May 1986
>> Just keep the HP651B for the purpose since it is easier with a
>> continous variable generator than a thumbweel tuned SG.
>>
>Very interesting, Jan. Did the authors mention any problems with
>reflections on the coax due to the mismatched termination?
>
>Mike
The described solution has detector near the coil, and I also did
this, but a friend of mine used it without this extra detector and
relied only on the detector inside MFJ259.
I wouldn't know how important it is to use the original version of
detector or it might just as well be on the SG output. One suggestion
is to use a series capacitor such that you could feed the detected DC
voltage back to an indicator at the SG
Jan-Martin
LA8AK