FET replacements in a Kenwood
Carlos Esteban
my Kenwood. I was thinking that if I replaced the JFET mixers and
MOSFET IF amps with something that has a lower noise figure, would
that make an appreciable difference?
I'm thinking of replacing the 2SK125s JFET mixers with 2SK152s and
3SK73 and 3SK74 IF amps being replaced by 3SK121s or BF998s.
Does anyone have any suggestions if this might be worthwhile?
gcd
I expect there is an RF amp before any mixer or IF amp. This stage will
determine the overall noise figure for the receiver.
Cheers
Greg
"Carlos Esteban" <carlos_...@hotmail.com> wrote in message
news:890178ba.03070...@posting.google.com...
Carlos Esteban
first mixer made of a pair of Jfets. I guess the design is relying on
purely the conversion gain as the first stage of amplification. I
guess I could add my own low noise preamp just after the attenuater to
help a little but I've since been told that the main problem was the
VCO and synthesiser stages would be a limiting factor.
I guess I will have to start poking around the Vco area and see if
there is anything that I can do about it and then come back to the RX
chain.
"gcd" <gcdn...@austarmetro.com.au> wrote in message news:<3f0a...@news.comindico.com.au>...
gcd
what model, I'll have a look at the schematic if I can find one. What
problem are you trying to fix / eliminate?
Cheers
Greg
"Carlos Esteban" <carlos_...@hotmail.com> wrote in message
news:890178ba.03071...@posting.google.com...
smithxpj
Esteban) wrote:
The trouble with simply doing a one-for-one FET swap is that the
existing biasing and circuit parameters might not be optimised for the
new version transistor. The expected performance could well be
marginal or less than the original condition. Fiddling around to find
the optimum operating parameters in anything other than a controlled
laboratory environment could be an exercise in 'laws of diminishing
returns'.
I once modified a Kenwood TS820 with a US origin RF and mixer
improvement kit, that included replacement of the FETS and a couple of
resistors and capacitors. A tuning tweak up was also needed. The mod
was obviously intended in the US to provided improved receiver
intermodulation amongst the 'kilowatt alleys'. In the lab, there *was*
a measurable improvement but in a realistic Australian on-air
situation...the improvement was probably academic.
Carlos Esteban
hash on RX, despite having a nice quiet wire dipole. My comms
receiver isn't getting this hash at all on the same antenna. I'm
assuming that the IF board has a problem, since I did have to replace
one 3SK73 in it on that particular board when I first got it. I did
have another TS430s for a very short time and managed to swap RF
boards, and that proved that my (RX) RF board is good.
Had other suggestions that the VCO might be noisy, or a capaciter in
the loop filter is dry. I did listen to the VCO with my comms RX and
can hear a slight hum on it (maybe 5-10% AM depth as a guess by ear),
but is otherwise OK - no hash like noise or jitter that I can hear.
The sensitivity is quite OK - weak signals can be received, just
theres a lot of noise between signal peaks. I've checked AGC action
too, which is fine.
Might just be the S meter is misaligned? Maybe this is just the way
Kenwoods are - have only had Yaesus in the past - though I can't
imagine this would be the case?
Any clues you could offer would be great.
"gcd" <gcdn...@austarmetro.com.au> wrote in message news:<3f0f...@news.comindico.com.au>...
Carlos Esteban
but I was hoping there'd be a FET with characteristics close enough to
do a one for one swap.
It sounds like i would be better to spend my time looking for the
source of this S3-S5 noise/hash. For the time being I just run the
attenuator on full time.
smithxpj <dont...@bother.com.au> wrote in message news:<cv91hv4jhvh8btdsb...@4ax.com>...
Phil Rice
hinted that use of the input attenuator reduced the hash. It sounds
like the hash is external to the radio. Does the hash disappear when
the antenna is removed? If so, no amount of internal modification will
make the slightest improvement (and may well make things worse).
Just my 2c worth,
73 de phil VK3BHR
Carlos Esteban
be.
Ive borrowed a friends 430 for a very short time and his does not have
that problem on my antenna - using the same power supply, same feeder,
same antenna, same day, side by side. 2 comms receivers also agree
that there is not this level of hash either.
But here's my 430 still with S3 to S5 hash on bands where my dipole is
resonant and this drops away when tuned to non-resonant freqs. The
hash does disappear when the antenna is unplugged or the attenuator is
on.
In my side by side tests above, I found that all 4 radios are about
equal with signal sensitivity, been able to resolve equally weak sigs
on all of them, but with my 430 compared to the other 'good' 430, I do
have to have the volume turned up about 40% more to get equal volume
out of it, and of course the S meter is always at S3-5 with the hash.
I'm starting to think it's just the S meter that is mis-aligned and
fooling me into thinking the hash is there - especially having to turn
the volume up to at least half way.
I have quickly tried swapping the RX board (contains the RF front end
and 1st mixer stage) and the first local osc VCO boards between the
430s, and the hash remained on my 430, so by doing that I deduce that
the RX and VCO boards in mine are OK, and that it must be the IF board
or control board that is causing the probs, or S meter misalignment?
Phil Rice <ri...@ironbark.bendigo.latrobe.edu.au> wrote in message news:<180720031049238352%ri...@ironbark.bendigo.latrobe.edu.au>...
gcd
OK, thats a lot more info than before, so some ideas should come in now.
My thoughts when I read this is an AGC problem in the IF stage. A quick
visit to mods.dk agrees that the AGC has a few problems in the 430, not this
specific issue, but a lot of other AGC issues.
I'd check to make sure your rf gain control potentiometer is working
correctly and then check the operation of the agc in the if stages.
Thats a start at any rate.
Cheers
Greg
"Carlos Esteban" <carlos_...@hotmail.com> wrote in message
news:890178ba.03071...@posting.google.com...
Carlos Esteban
yes, I initially thought it was an AGC problem too, but AGC action
checks out - I can hear it working between syllabels on SSB and the RF
gain control acts as it should, and AGC voltages as per the service
manual are 'ballpark'.
I found out last night after a very long session that the problem is
intermittant - the hash goes away for short periods of time, but only
after the thing has been on for more than a few hours. I know this
suggests dry joint, but I would have thought that would expand /
contract much sooner than a few hours. It could still be an active
device I think, after all, I did have to replace one of the IF amps
when I first started to restore this - maybe another is working OK but
a bit noisy? Which means that I've come full circle - if it is a
noisy FET, I may as well replace it with one that is a little better
than originally there, but as pointed out that may will stuff up
biasing etc, so I guess it would be best to use the original spec item
- if it does turn out to be a noisy FET.
"gcd" <gcdn...@austarmetro.com.au> wrote in message news:<3f19...@news.comindico.com.au>...
gcd
hmmm is all I can say. hmmmmmmm
Don't pull those FETs just yet. 1st disconnect and isolate the 2nd IF and
feed a signal directly in from a sig gen. See if the same thing still
happens. Also do the same for 1st IF , removeing the signal in and the
VCO/PLL.
The idea there is to see if it can first be isolated to either the 1st IF ,
2nd IF or perhaps if one of the LO or the VCO is the source of the noise.
You have already swapped out a VCO/PLL and the RF board (also contains the
1st IF?) so that reduces your investigation to the IF board so that reduces
the suspect areas a lot.(where you started this thread).
The fact that it's intermittant makes me think it's not a semiconductor
problem.
I agree with the strong possibility of a dry joint, and again the isolation
of modules will help locate it. I'd also be inclined to inspect any
potentiometers on the board in the agc circuit . If these are dirty they may
be changing the bias on the agc amps as a result and being seen as as noise.
I'd also check any interboard connectors and make sure they are clean.
Last but certainly not least I'd check the power supply/regulator area and
make sure there are no issues there. Noise here can go through the agc and
also affect the bias etc of the IF stages.
Running out of ideas after that but I still don't believe it's an activce
device.
Hope this is of some help
Cheers
Greg
"Carlos Esteban" <carlos_...@hotmail.com> wrote in message
news:890178ba.03072...@posting.google.com...
gcd
hmmm is all I can say. hmmmmmmm
Don't pull those FETs just yet. 1st disconnect and isolate the 2nd IF and
feed a signal directly in from a sig gen. See if the same thing still
happens. Also do the same for 1st IF , removeing the signal in and the
VCO/PLL.
The idea there is to see if it can first be isolated to either the 1st IF ,
2nd IF or perhaps if one of the LO or the VCO is the source of the noise.
You have already swapped out a VCO/PLL and the RF board (also contains the
1st IF?) so that reduces your investigation to the IF board so that reduces
the suspect areas a lot.(where you started this thread).
The fact that it's intermittant makes me think it's not a semiconductor
problem.
I agree with the strong possibility of a dry joint, and again the isolation
of modules will help locate it. I'd also be inclined to inspect any
potentiometers on the board in the agc circuit . If these are dirty they may
be changing the bias on the agc amps as a result and being seen as as noise.
I'd also check any interboard connectors and make sure they are clean.
Last but certainly not least I'd check the power supply/regulator area and
make sure there are no issues there. Noise here can go through the agc and
also affect the bias etc of the IF stages.
Running out of ideas after that but I still don't believe it's an activce
device.
Hope this is of some help
Cheers
Greg
"Carlos Esteban" <carlos_...@hotmail.com> wrote in message
news:890178ba.03072...@posting.google.com...
Carlos Esteban
I managed to get another TS 430s to swap boards to help isolate the
problem. I found that if the IF board was replaced, the noise dropped
by half, and then if I replaced the control board, the 'noise'
disappeared.
So, I am dealing with two problems here.
Some fault finding has borne fruit, though. On the IF board (the
faulty one) L3 was found mis-aligned. This apprently needs to be
detuned off peak according to the service manual. Tuning as per
instruction restored the IF board to correct reception in the 'good'
TS 430s.
The other problem of the control board is not yet solved. I have gone
over every solder join in the applicable sections of this board (ie
ignored the 'digital' sections) but this has not helped. But at least
with this one last fault remaining, all I am left with is a constant
S2 noise, as opposed to the 'good' TS 430s normal S1 noise floor.
Much more acceptable.
Hopefully the next time I write, it will be good news that everything
has been fixed.
R's
C
"gcd" <gcdn...@austarmetro.com.au> wrote in message news:<3f1f...@news.comindico.com.au>...
gcd
good to hear there has been some success. Sounds like you're determined to
get that rig going :)
I'm still confident that you will not find any bad FETs in there :)
Cheers
Greg
"Carlos Esteban" <carlos_...@hotmail.com> wrote in message
news:890178ba.03080...@posting.google.com...