I have an external RF transistor amplifier (AB1) using 2 MRF455
transistors in common emitter mode. I have a resistive T pad with an
approximate impedance of 50 ohms on the input side of the transistors.
The SWR on the input of the amp shows 1.3:1. When I use a 2 foot
jumper between the radio and the amp the amp puts out about 100 watts
max, but when I switch to a 9 foot jumper the amp puts out about 40
watts max. The receive stays the same no matter which jumper is used.
Why does changing the length of the jumper between the radio and the
amp make such a drastic change in how much the amp puts out?

Jim wrote:


I have an external RF transistor amplifier (AB1) using 2 MRF455
transistors in common emitter mode.

What type of cable and connectors are the two jumpers?

On Tue, 12 Jul 2005 21:56:22 -0500, Bob Liesenfeld
wrote:



Jim wrote:


I have an external RF transistor amplifier (AB1) using 2 MRF455
transistors in common emitter mode.

What type of cable and connectors are the two jumpers?

RG-58 cable with PL-259 connectors.

Jim wrote:




I have an external RF transistor amplifier (AB1) using 2 MRF455
transistors in common emitter mode. I have a resistive T pad with an
approximate impedance of 50 ohms on the input side of the transistors.
The SWR on the input of the amp shows 1.3:1. When I use a 2 foot
jumper between the radio and the amp the amp puts out about 100 watts
max, but when I switch to a 9 foot jumper the amp puts out about 40
watts max. The receive stays the same no matter which jumper is used.
Why does changing the length of the jumper between the radio and the
amp make such a drastic change in how much the amp puts out?

Are you sure the 9 foot jumper is good?

What frequency is this occurring at?

Is the SWR 1.3:1 for both jumpers? Measure the SWR, forward power, and
reflected power levels at the transmitter end for both jumpers and then
measure it at the amplifier end for both jumpers. See if everything stays
the same for both jumpers.

Do you have an RF probe you can use to measure the voltages at each end of
the link?

Unless you are using rg174 the 7 foot of extra cable should not make this
much difference unless the input impedance of the amplifer is not 50ohm
resistive. If it is not purely resistive then changing the cable length can
impact the SWR seen at the transmitter end significantly. This could cause
foldback in the RF amplifier in the transmitter thus affecting the output
of the amplifier. This would all be seen in the measurements of SWR,
forward power, and reflected power levels in the measurements above.

Let us know what you find.

tim ab0wr

tim gorman wrote:
. . .
Unless you are using rg174 the 7 foot of extra cable should not make this
much difference unless the input impedance of the amplifer is not 50ohm
resistive. If it is not purely resistive then changing the cable length can
impact the SWR seen at the transmitter end significantly. . .

Changing the cable length won't change the SWR on the cable regardless
of the kind of load impedance and, if the SWR meter is designed for the
cable's Z0, it won't change the SWR meter reading, either. Except, of
course, that cable loss will always lower the SWR -- but that shouldn't
be a significant factor with such short cables.

Changing the cable length *will* change the impedance looking into the
cable, whether or not the load is purely resistive. The only exception
to this is if the load is resistive *and equal to the line's
characteristic impedance* in which case the impedance looking in will be
Z0 for any length cable.

Transmitters will often put up with some mismatched impedances better
than others, even if the SWR is the same, and sometimes changing the
cable length between it and a mismatched load will cause it to see a
more or less favorable impedance. But if the SWR really is 1.3:1, I
doubt that's the cause of this problem.

I agree with the suggestion that the OP measure the SWR and if possible
the power at both ends with both cables. Something else is going on,
like maybe a bad cable or connector.

I don't think the OP said what frequency this is happening at. That
might give some additional clues.

Roy Lewallen, W7EL

"Roy Lewallen" wrote in message
...
tim gorman wrote:


I agree with the suggestion that the OP measure the SWR and if possible
the power at both ends with both cables. Something else is going on,
like maybe a bad cable or connector.

I am assuming this is VHF and if so I would say that there is a bad
connector at both ends - PL259s. These are not good enough connenctors for
use anywhere above HF, the are simply screened banana plugs. If this is a
ready made jumper for the CB market, the cable might not be anything like 50
ohms.

Mike

From: Roy Lewallen on Jul 13, 12:39 am


tim gorman wrote:
. . .
Unless you are using rg174 the 7 foot of extra cable should not make this
much difference unless the input impedance of the amplifer is not 50ohm
resistive. If it is not purely resistive then changing the cable length can
impact the SWR seen at the transmitter end significantly. . .


I agree with the suggestion that the OP measure the SWR and if possible
the power at both ends with both cables. Something else is going on,
like maybe a bad cable or connector.

I don't think the OP said what frequency this is happening at. That
might give some additional clues.

Good suggestions on simple checking of a cable AND connectors.

The six-decade-old design of the PL-259 is not the best on
"wiping" action of contacts on the sleeve (outer conductor
portion). It is a license-free standard design, standard
because it is relatively cheap. Cheap silver plating can
corrode fairly easily (rhodium flash over silver plating is
much better but costs more)

Yesterday my "ancient" HP-722 inkjet printer had no primary
power. Check of the AC cord, external supply, said that was
okay. Turns out that the coaxial connection to the back of the
printer had developed some kind of minor corrosion. Simple
unplugging and re-plugging that DC connector brought back
primary power.

Three decades ago I was involved in a seemingly "unsolveable"
problem in Navy-flown L-Band R&D system. Signals would just
cut out at altitude, said altitude varying depending on day
of flight test. System AND RF cables (to top and bottom
fuselage antennas) all checked out fine on the ground. Nothing
intermittent. Blade antennas were taken off and checked
okay, put back. All type N connectors, good ones. Trouble was
in an unlikely form of a "doubler plate."

"Doubler plates" are often used in retro-fitting antennas and
other things on aircraft, just a sheet of metal to re-enforce
strength of the metal skin. The doubler plate drawings had
clearance holes just too close to type N connector sleeve
outer diameter. Connectors mated, but NOT fully. As altitude
increases, temperature drops. The not-fully mated center
conductor pin just contracted until it lost contact at cold
temps. Enlarging the doubler plate clearance hole allowed
full mating, no shrinkage of contacts. Unlikely problem
solved at about quarter to 8 PM in a cold hangar. :-(

Sometimes the "unlikely" not-described-in-text things are to
blame.

wrote:
. . .
"Doubler plates" are often used in retro-fitting antennas and
other things on aircraft, just a sheet of metal to re-enforce
strength of the metal skin. The doubler plate drawings had
clearance holes just too close to type N connector sleeve
outer diameter. Connectors mated, but NOT fully. As altitude
increases, temperature drops. The not-fully mated center
conductor pin just contracted until it lost contact at cold
temps. Enlarging the doubler plate clearance hole allowed
full mating, no shrinkage of contacts. Unlikely problem
solved at about quarter to 8 PM in a cold hangar. :-(

Sometimes the "unlikely" not-described-in-text things are to
blame.

I was in Anchorage when color TV was first being broadcast there
(mid-'60s) and people were using outside antennas with coax feed for the
first time. The center pin of a male type F connector is just the center
conductor of the RG-59 cable, and on a cold day the center conductor
would shrink enough on a long cable run to pull the pin out of the
female connector.

Anybody who has spent some time as a technician (as I did) has a long
list of tales to tell about strange problems. Among my favorites are a
car with the battery installed backward (a Corvair -- car started and
ran, but gauges were funky and the radio blew), and a piece of copper
pipe with 300 ohms DC resistance. But those are just a couple. . .

Roy Lewallen, W7EL

" I have an external RF transistor amplifier (AB1) using 2 MRF455
transistors in common emitter mode. I have a resistive T pad with an
approximate impedance of 50 ohms on the input side of the transistors.
The SWR on the input of the amp shows 1.3:1. When I use a 2 foot
jumper between the radio and the amp the amp puts out about 100 watts
max, but when I switch to a 9 foot jumper the amp puts out about 40
watts max. The receive stays the same no matter which jumper is used.
Why does changing the length of the jumper between the radio and the
amp make such a drastic change in how much the amp puts out?
================================================== ==
If this is a 2m transmission system 2ft of coax equals 610 mm. Assuming a
coax velocity factor of 0.67 ,the electrical length is 1/ 0.67 * 610 equals
915mm which is about 'half a wavelength'.
Half a wave lenght of transmission line is an 'impedance repeater'
Hence if tx output- and amplifier input impedance are about the same ,there
is max power transfer.

Nine feet of coax for the above does NOT equal a multiple of half
wavelenghts hence the reduced power input to the amplifier and subsequent
lower amplifier output.

Just try to reduce the nine feet coax cable to about 8 ft which would be
about 4 half wavelengths . The result should be an amplifier output 40
Watts

You can readily make multiples of (electrical )half wavelength by using an
MFJ259B antenna analyer or even better a spectrum analyser..

An electrical half wave (or multiple half waves ) 'shorted' stub has a LOW
impedance at the other end .

Take a piece of coax which is slightly longer than a multiple of
(electrical) half wavelengths.
Fit a suitable connector at one end and connect to impedance measuring
instrument . Now penetrate the coax with a needle at the other end ,
observe the impedance.
The needle shorts the coax centre conductor with the braid.
Change frequency up or down finding the nearest freq where impedance is
minimal .
By penetrating the coax nearer the connected end you can find the length
where the impedance is minimal at the desired freq. That length is a
multiple of elctrical half wavelengths , and hence the optimal length of
your lead.

If you don't like the 'needle operation' , you can make an electrical
quarter wave '0pen' stub .
Again you start with a length of coax slightly longer , check the freq for
which the stub shows minimal impedance and subsequently cut tiny bits off
the stub until the instrument shows minimal impedance at the desired freq.

The lead you need between transceiver and amplifier is to be an EVEN
multiple of the remaining length of coax

I have done the above to optimise power transfer between a vintage Yaesu
FT290 ,2m transceiver (2.5W) and a 30 Watts power amplifier.

Note : The characteristics of open-ended and shorted transmission line
sections is well described in the ARRL Extra Class License Manual ( my copy
is from 1990)

Frank GM0CSZ / KN6WH

If this is a 2m transmission system 2ft of coax equals 610 mm. Assuming a
coax velocity factor of 0.67 ,the electrical length is 1/ 0.67 * 610
equals
915mm which is about 'half a wavelength'.
Half a wave lenght of transmission line is an 'impedance repeater'
Hence if tx output- and amplifier input impedance are about the same
,there
is max power transfer.

MRF455 is a 60W HF transistor- i.e. 30 MHz and down- from memory.
Dale W4OP