DX-394 Reciever
Me
Just purchased a DX-394 rx and have read in this group about mods that
can be done to it to improve it. Could anyone provide me with a link to a
site where these mods are please.
Thanks.. Steve
craigm
modifications" yields this
http://www.mods.dk/mods.php3?model=dx-394&radio=radio_shack
among others.
"Me" <m...@udont.com> wrote in message
news:s86r8.11853$je5.1...@nnrp1.uunet.ca...
Me
Sorry about that, didn't mean to post this as I Started my own search
and didn't cancel the message. Guess I should have deleted this before I
started something else. Anyway, thanks for the site, was exactly what I've
been looking for.
Steve
Abb Naklie
Just be careful about the mods. Did you get a "B" version (it will be on the
back of your receiver) - it's an improved version over the none B. I had a
394 "B" and it was a ham friend that almost insisted I find one ("B"
version) and after checking a few Radio Shacks, I found one. It was a great
little receiver on it's own. A good point was made to me - don't do any mods
while it's still under warranty - something to think about. I was told the
"B" doesn't need them as bad as the original did.
Take Care
Abb
"Me" <m...@udont.com> wrote in message
news:s86r8.11853$je5.1...@nnrp1.uunet.ca...
Fcathell
implementing the mods it was very "hot" sensitivity wise and very stable on
SSB. Unfortunately, the selectivity is a bit too wide for SSB and there's
nothing you can do about that!
Frank
starman
Why can't the SSB selectiviy be changed?
*****
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Joe
http://www.webdeals.net/dx394.htm
Problem is the signal jumps around both filter via a VCC rail
The problem is caused by the fact that Tandy designers (or whoever) forgot
one of the CF1 bias decoupling capacitors. Anyone with a schematic will note
that the node at the junction of R58, R61, R62 and R64 should have a bypass
capacitor to ground from this point. Without it there is a resistive signal
path through R57, R58 and R61 that bypasses the CF1 filter and causes
broadband bleed through. All that is necessary is to solder a 0.047 to 0.2uf
ceramic capacitor directly across R64 and the crosstalk will be
significantly reduced.
For those of you without a schematic you can find R64 by removing the bottom
cover with the receiver knobs facing you. Find the silkscreen ID for D15.
This is a four lead surface mount, dual diode package. R64 is the third
component to the right of D15 and has 222 (2.2K) stamped on it (you may need
a magnifying glass!)
This resistor is also just below and slightly to the right of the third pin
(from the left) on CF1. It should be noted that with real strong adjacent
signals there still may be some crosstalk and reducing the RF gain will help
even more. Good luck and be careful - the parts are all tiny surface mount.
#2. Here is another DX-394 modification that will further reduce adjacent
channel crosstalk from strong shortwave stations (see my first mod of about
a week ago). This one may be more effective than the previous one and used
together, crosstalk is reduced to an insignificant level.
With the bottom cover removed and the front panel knobs facing you find the
silkscreen ID for Q13. It is a black surface mount transistor with three
leads (2 on one side, 1 on the other) midway toward the front of the unit
and left of center.
Resistor R91 is just below it and should be marked with 102 (1K). Bridge
this resistor with a 0.1uf or 0.22uf ceramic capacitor (keep leads short)
and you're done.
"Fcathell" <fcat...@aol.com> wrote in message
news:20020405191139...@mb-cp.aol.com...
Fcathell
(I've tried) is the way the BFO injection scheme is designed in the DX-394.
The USB/LSB ceramic resonators are offset from the IF center (455 kHz) by 3 khz
instead of the standard 1.5 kHz. The IF center frequency in this mode is then
shifted 1.5 kHz from center resulting in the correct 1.5 kHz difference freq.
but now the whole signal winds up being shoved up the passband skirts of any
filter more narrow than about 4 kHz. The audio is real tinny with severe
attenuation of the signal. In short, just a piss poor and unorthodox scheme.
More narrow filters will work for AM, however, because the IF bandpass is
exactly in the center in this mode.
Frank
starman
What were they thinking? :-)
*****
Fcathell
Beats me. Even my feeble electronic mind wouldn't have done it this way. A
simple tunable +/- BFO at 455 kHz would have been better than this. In fact,
assuming a stable L/C oscillator this old approach has some neat benefits!
Frank
Me
Anyone have a copy of the schematic diagram for the DX-394 radio? I've
searched Google and had no luck.
Thanks.. Steve
KG4FET
Take a look at radioShack.com under product manuals. Then click on
communications then Two Way & VHF radios. There is a complete manual with the
schematic there. You can print from the site or use the Fax Back feature as
well. Great Source for just about anything old or new as far as RadioShack
equipment goes both old and new.
Take Care and the best of 73
DE KG4FET Sandor
Steve
I've been to this site and all it shows is an exploded view and parts
list, no schematic. Thanks for the info just the same, still looking though.
Thanks..
"KG4FET" <kg4...@aol.comNOspam> wrote in message
news:20020409160640...@mb-cf.aol.com...
Tom Holden
I just got the manual and, from what I can see, the SSB/CW filter is centred
on 455 kHz and the BFO is switched between 451.5 and 458.5 kHz for USB/LSB.
After I realigned the BFO's so that the indicated frequency corresponded
with zero beat, it was my impression that the BFO's sit at appropriate
positions on the skirts of this relatively wide filter.
So, if a narrower SSB/CW filter was substituted, the carrier substitution
provided by the BFO would be way down the skirt and the detected audio
spectrum would be very high but narrower. For example, a 3kHz bandwidth
filter centred at 455 kHz would result in audio spectrum principally between
(455 +/- 1.5)-451.5 = 2 - 5 kHz, damned tinny! By changing the BFO's to 455
+/- 1.5 kHz, the spectrum would be (455 +/- 1.5)-453.5 = 0 - 3 kHz, more
like what we want.
Question is, where to get a suitable narrower filter?
73, Tom VE3MEO
"Fcathell" <fcat...@aol.com> wrote in message
news:20020406012237...@mb-ml.aol.com...
Dale Parfitt
Tom Holden wrote:
> Hi Frank,
>
> I just got the manual and, from what I can see, the SSB/CW filter is centred
> on 455 kHz and the BFO is switched between 451.5 and 458.5 kHz for USB/LSB.
> After I realigned the BFO's so that the indicated frequency corresponded
> with zero beat, it was my impression that the BFO's sit at appropriate
> positions on the skirts of this relatively wide filter.
>
> So, if a narrower SSB/CW filter was substituted, the carrier substitution
> provided by the BFO would be way down the skirt and the detected audio
> spectrum would be very high but narrower. For example, a 3kHz bandwidth
> filter centred at 455 kHz would result in audio spectrum principally between
> (455 +/- 1.5)-451.5 = 2 - 5 kHz, damned tinny! By changing the BFO's to 455
> +/- 1.5 kHz, the spectrum would be (455 +/- 1.5)-453.5 = 0 - 3 kHz, more
> like what we want.
>
> Question is, where to get a suitable narrower filter?
>
> 73, Tom VE3MEO
>
This confirms what saw some years back.
How about the Collins 455 filters- available directly from Collins. I have a
2.4KHz in my homebrew RX and SSB is wonderful- I use a 6KHz for AM but can use
the 2.4KHz if I offest the passband a bit to include more of the highs.
Dale W4OP
Dale Parfitt
Tom Holden wrote:
> Hi Frank,
>
> I just got the manual and, from what I can see, the SSB/CW filter is centred
> on 455 kHz and the BFO is switched between 451.5 and 458.5 kHz for USB/LSB.
> After I realigned the BFO's so that the indicated frequency corresponded
> with zero beat, it was my impression that the BFO's sit at appropriate
> positions on the skirts of this relatively wide filter.
>
> So, if a narrower SSB/CW filter was substituted, the carrier substitution
> provided by the BFO would be way down the skirt and the detected audio
> spectrum would be very high but narrower. For example, a 3kHz bandwidth
> filter centred at 455 kHz would result in audio spectrum principally between
> (455 +/- 1.5)-451.5 = 2 - 5 kHz, damned tinny! By changing the BFO's to 455
> +/- 1.5 kHz, the spectrum would be (455 +/- 1.5)-453.5 = 0 - 3 kHz, more
> like what we want.
>
> Question is, where to get a suitable narrower filter?
>
> 73, Tom VE3MEO
>
This confirms what i saw some years back in the 394.
Tom Holden
"Dale Parfitt" <par...@verizon.net> wrote in message
news:3D659320...@verizon.net...
>
[snip]
> This confirms what i saw some years back in the 394.
> How about the Collins 455 filters- available directly from Collins. I have
a
> 2.4KHz in my homebrew RX and SSB is wonderful- I use a 6KHz for AM but can
use
> the 2.4KHz if I offest the passband a bit to include more of the highs.
>
> Dale W4OP
>
find them. My guess is that they are pretty costly. I paid only US$85 for
the DX-394 - maybe it's not too smart to throw a lot more at it! Kiwa has
some 3 kHz filters that might work but even they run $50-75 plus the new BFO
crystals and I'm not really sure what happens with the frequency display.
Tom VE3MEO
Fcathell
"tinny" that it's unintelligible. The passband will be shoved so far up the
filter skirts that a 2.5 kHz filter centered on 455 kHz just won't work.
Frank
Tom Holden
passband where we want it in relation to the BFO?
What I'm less sure of is what the freq display will be; i.e., maybe it's
programmed with an offset of +/- 3.5 kHz in SSB/CW mode and that can't be
changed - net result being an error of 2kHz if we shift the BFO freq by that
much. On my 'A' version, I adjusted the BFO trimmers to zero-beat WWV with
the freq displaying 15.000 MHz. Listening and observing the audio spectrum
on CoolEdit as I tune through a carrier confirms that the stock filter and
BFO freq's with this adjustment are correctly aligned. That implies that for
the same freq displayed, the LO's must be different for each of three modes:
AM, LSB/CW1 and USB/CW2; for the same carrier frequency, at 2nd IF the
carrier must come out at 455, 458.5 and 451.5 kHz, respectively, for zero
beat (I might have the LSB and USB reversed if the preceding conversions
flip the sidebands). Since the displayed frequency is the same for all
three, there must be a different programmed offset between the displayed
freq and the LO's for each mode. I don't know how to change that.
The implication may be that the BFO's should not be changed and that the
only way to apply a narrower SSB/CW filter is that separate filters would be
required for USB and LSB with centre freq's offset by half the passband from
the BFO freq plus approx 0.5kHz for inteligibility optimisation . So for a
sharp 2 kHz SSB filter, the USB would be centred at 451.5+1+0.5 and the LSB
at 458.5-(1+0.5).
All of that is getting too complicated and expensive so that the outcome
will be the same regardless of the rationale - I probably won't even try to
get a narrower SSB filter!
Regards,
Tom
"Fcathell" <fcat...@aol.com> wrote in message
news:20020823005309...@mb-fr.aol.com...
Fcathell
394 to the "standard" 455+1.5kHz and 455-1.5kHz would do it because the
receiver actually also shifts the PLL center freq.+/- 1.5 kHz from 455 kHz so
the BFO/center freq. differential is still 1.5kHz. This is essence of the 394
design flaw. Why in the hell did they shift the IF center freq when in SSB/CW?
It would seem with this scheme that you could use a single 455 kHz BFO
resonator and still get what you wanted only with USB and LSB reversed (???)
Frank
Frank
Tom Holden
shifts when you change mode on a given frequency; rather the LO's shift so
that what passes through the IF centre frequency is a higher or lower slice
of the converted r.f. spectrum, relative to the AM mode. That's how it
slices off the upper or lower sideband instead of both. I suppose that's the
same as saying the IF centre frequency has shifted relative to the carrier
but it's the carrier that was moved.
I guess it was cheaper or more feasible to have two BFO's with 1 filter
serving both USB and LSB modes than to have a single BFO and 2 filters, 1
for each sideband. That's a good design compromise for a budget receiver but
limiting for us 'hot rodders'. One could probably muck around in it to force
it to stay in AM mode as far as LO freq is concerned so that a single BFO
and separate USB and LSB filters could be used for SSB/CW and the freq
display would be accurate.
Same conclusion, though. It's probably better to spend the time and effort
needed for this mod on making money to buy a better receiver or just enjoy
what we've got!
BTW, I finally did your CERAMIC FILTER MOD FOR AM you posted some years ago
but with a variation - I added a switch so that I can 'enjoy' both the stock
wide and narrow filters on all modes. I posted a comment on www.mods.dk and
will write up a how-to sometime.
Regards,
Tom VE3MEO
"Fcathell" <fcat...@aol.com> wrote in message
news:20020823123502...@mb-bh.aol.com...
Fcathell
signal pumping of the AGC and subsequent loss of gain on the desired signal as
opposed to true interference. I had a disable switch for the AGC that helped
this in many cases, however, the RF gain had to be (obviously) reduced. I had
fun optimizing this receiver but sold it shortly thereafter and started
"hacking" another receiver. It's a damn shame that just a few changes in the
original design would have made this a fairly decent receiver. Even in the
"stock" condition the one I had ("A" version) was sensitive and rock stable.
Frank
Tom Holden
it? I'm thinking that switching the 7 volt line onto TP2 on the AGC line via
a 9 - 10k resistor should do it, plus maybe a diode inserted into the AGC
line ahead of TP2 to block current being drawn by the AGC transistor.
However, that would require a closed contact for AGC defeat; I'd rather use
an open contact for defeat and closed for enable so I could use a centre-off
DPDT switch for both AGC speed and defeat.
BTW, it would be great if you participated in
http://groups.yahoo.com/group/RADIOSHACKDX394 . Membership is now 160. There
are a number who are doing your mods.
Regards,
Tom VE3MEO
"Fcathell" <fcat...@aol.com> wrote in message
news:20020823180430...@mb-fa.aol.com...
Fcathell
By the way if you haven't done the noise blanker mod it's worth it because once
done the 394 has a very effective NB. You can actually set the threshold level
by adjusting R180. I guess I'll have to check out the Yahoo 394 group.
Frank
Tom Holden
instructive. Thanks for the rapid exchanges.
Re AGC Defeat mod: There is what I think is a protective auto attenuator on
the r.f. input (C160, C166, D1) that is activated by the voltage on the
emitter of Q9, which also drives the S-meter. D1 is 2 diodes in series so
they would turn on and ground the capacitors when the emitter voltage rises
to 1.4V. I was thinking that that should remain functional when AGC is
defeated; grounding the base of Q9 to defeat AGC would also kill that
feature. So I was trying to come up with an OR between the AGC and the AGC
Defeat signals. Whichever is the greater voltage has control; at max gain
with no signal, AGC voltage appears to be about 4.5V at TP2 dropping by
about 2V on very strong signals. So AGC Defeat should deliver 4.5V. Both
pass through diodes to the controlled amplifiers - perhaps have to adjust
AGC driver resistor to compensate for diode drop.
Re Noise Blanker mod: I did do it but my results weren't very significant.
Having just received the service manual, I've been marking up the schematic
with the mods and discovered what I think is a typographical error in your
instructions. They say to connect the cathode of the Schottky diode to the
collector of Q32 to allow "capacitor C182 to reset (discharge) in between
noise pulses". I discovered that Q32 collector is tied directly to the 7V
rail so my diode doesn't do anything. C182 is on the collector of Q34 and
that's where I think you meant.
Regards,
Tom
"Fcathell" <fcat...@aol.com> wrote in message
news:20020823213105...@mb-ch.aol.com...
Fcathell
anode to emitter. Someone must have made a typo because my original notes do
say Q34. Also your analysis of the AGC operation appears correct and grounding
the base of Q9 does defeat the AGC but also affects the S-meter.
Frank
Frank
Joe
Best is just to slow the AGC, dont forget to loose the lousy speaker and
double the coupling cap,
Yank a fat sounding speaker out of a Sony TV
"Fcathell" <fcat...@aol.com> wrote in message
news:20020824010527...@mb-fg.aol.com...
Pete KE9OA
division in Costa Mesa, California.
Pete
Fcathell <fcat...@aol.com> wrote in message
news:20020823005309...@mb-fr.aol.com...
Tom Holden
add-on: operate in AM mode so that the carrier is centred in the i.f.
passband and take that out to the external DSP-based sideband filter, BFO,
product or synchronous detector, maybe even generate AGC there. Somebody
else posted an inquiry about accessing the DX-394 455 kHz IF for this
purpose - hope he is successful in developing something; I know very little
about doing this stuff in software. :(
73, Tom VE3MEO
"Pete KE9OA" <n.giana...@worldnet.att.net> wrote in message
news:cwJ99.32096$Ke2.2...@bgtnsc04-news.ops.worldnet.att.net...
Fcathell
like I did for $125 on E-Bay, fix it, modify the hell out of it with new
filters, revamped tone circuit, etc. and you don't have to worry about
re-designing the 394. But then again it's probably more fun "squeezing" that
394!
Frank
Tom Holden
for 4 decades! And the better half is away so I'm escaping the usual chores.
So far my investment has been pretty light financially but the hours are
incredible. I've just blown the weekend working on the AGC - yeah, I'm
sloooow. Yes, Frank, I like to tinker when the risks are low and I enjoy
this Internet sharing of ideas and experience. I've written a 'paper' on the
AGC mods - would you be willing to review it before I post it?
73, Tom
"Fcathell" <fcat...@aol.com> wrote in message
news:20020825171858...@mb-bk.aol.com...
Fcathell
did. I do have a junked 394 here and some parts have been removed. If you
need any replacement parts let me know and I'll see if they are still in the
thing.
Frank