Hi,

A bit off topic, but I figure you folks are the true theorists out
there, and would probably know about this.

Am thinking about purchasing a shortwave defined radio receiver from
WinRadio (http://www.winradio.com/home/g33ddc.htm).

Here is copy from part of their spec sheet:

"The receiver's robust front-end is equipped with an
ultra-high-linearity amplifier which results in exceptional
strong-signal performance. This already robust front-end is further
enhanced with a user-selectable preselector that can operate either in a
fully automatic or user-configurable mode. As many as 119 different
filter combinations can be constructed by the user (91 bandpass, 14
low-pass and 14 high-pass). The front-end employs 34 subminiature
electromechanical relays (rather than often used but distortion-prone
semiconductor switches) to ensure high dynamic range."

Was really surprised to see: The front-end employs 34 subminiature
electromechanical relays (rather than often used but distortion-prone
semiconductor switches) to ensure high dynamic range.

I am retired now, so haven't kept up withese things, or the state of the
art nowadays, but are they still using electromechanical relays for
front ends feeding a 16 bit A to D ?

Haven't semiconductor switches, caught up with reed relays for this
application by now, or do they "still" provide a lower "on" resistance,
etc. ?

Any thoughts would be most appreciated.

Thanks,
Bob

On 10/7/2010 11:58 AM, Bob wrote:

...
Any thoughts would be most appreciated.

Thanks,
Bob

I have a lot of SW radios and amateur gear. When you have used a SDR,
you are simply left with one question, "Why would anyone, anywhere, use
anything different?

I mean, it is still fun to fire up an old one and watch the tubes glow
.... nostalgia, I guess.

Regards,
JS

On 10/7/2010 11:58 AM, Bob wrote:

...
Haven't semiconductor switches, caught up with reed relays for this
application by now, or do they "still" provide a lower "on" resistance,
etc. ?
...
Thanks,
Bob

Forgot to respond to this part ... my whole experience with repairing
electronic gear has been, first suspect a mechanical problem (switch,
ground, etc.), then suspect a capacitor, then a resistor ... dead last
is a semiconductor ... it is taken for granted that properly engineered
equipment is the point of discussion ...

Regards,
JS

John Smith wrote:

I have a lot of SW radios and amateur gear. When you have used a SDR,
you are simply left with one question, "Why would anyone, anywhere, use
anything different?

Propigation is always good on the internet. With a fast (15mbit download)
internet connection why bother with a radio at all?

To be honest, I often don't. I skype chat, I call people on a VoIP phone,
I email, I use facebook, I watch TV from around the world, I listen to
radio from around the world.

TV and radio are especially nice because I can get the programs I want when
I have time to pay attention to them.

I also use my radios because I want to. I expect different things from them
and chalenges to get them.

To answer the original post, reed relays are relatively cheap and have very
high isolation between the poles. Electronic switches are available and cheap
but not with as good isolation (at least not at the price of the relays).

Geoff.


--
Geoffrey S. Mendelson N3OWJ/4X1GM
To help restaurants, as part of the "stimulus package", everyone must order
dessert. As part of the socialized health plan, you are forbidden to eat it. :-)

Bob wrote:
Was really surprised to see: The front-end employs 34 subminiature
electromechanical relays (rather than often used but distortion-prone
semiconductor switches) to ensure high dynamic range.

I am retired now, so haven't kept up withese things, or the state of the
art nowadays, but are they still using electromechanical relays for
front ends feeding a 16 bit A to D ?

Actually from a manufacturing point of view it sounds like they have taken
the matching board from a "100 watt" automatic tuner and converted it to
a preselector/filter board by chainging the programing on the controller
chip.

In this case the "the front end ...." stuff is marketing hype.

If they got enough of them as a close out, it would be a cheaper than
having to make one from scratch, no matter what was in it.

Geoff.

--
Geoffrey S. Mendelson N3OWJ/4X1GM
To help restaurants, as part of the "stimulus package", everyone must order
dessert. As part of the socialized health plan, you are forbidden to eat it. :-)

On Oct 7, 11:58*am, Bob wrote:
Hi,

A bit off topic, but I figure you folks are the true theorists out
there, and would probably know about this.

Am thinking about purchasing a shortwave defined radio receiver from
WinRadio (http://www.winradio.com/home/g33ddc.htm).

Here is copy from part of their spec sheet:

* * *"The receiver's robust front-end is equipped with an
ultra-high-linearity amplifier which results in exceptional
strong-signal performance. This already robust front-end is further
enhanced with a user-selectable preselector that can operate either in a
fully automatic or user-configurable mode. As many as 119 different
filter combinations can be constructed by the user (91 bandpass, 14
low-pass and 14 high-pass). The front-end employs 34 subminiature
electromechanical relays (rather than often used but distortion-prone
semiconductor switches) to ensure high dynamic range."

Was really surprised to see: The front-end employs 34 subminiature
electromechanical relays (rather than often used but distortion-prone
semiconductor switches) to ensure high dynamic range.

I am retired now, so haven't kept up withese things, or the state of the
art nowadays, but are they still using electromechanical relays for
front ends feeding a 16 bit A to D ?

Haven't semiconductor switches, caught up with reed relays for this
application by now, or do they "still" provide a lower "on" resistance,
etc. ?

Any thoughts would be most appreciated.

Thanks,
Bob
Excuse me, Bob, but no where in the description do they mention reed
relays. I really doubt they are using reed relays because of the open
capacitance of the contacts, plus, reed relays often stick shut
because of fatigue, static electricity, and if left closed for a long
time, metal migration will stick them closed.

Bet they are using regular relays, but in miniature form.

Paul

Haven't semiconductor switches, caught up with reed relays for this
application by now, or do they "still" provide a lower "on" resistance,
etc. ?



Semiconductor switches have issues with intermodulation, etc. and power
consumption. For wideband RF, you're not going to be using cheap 4066
CMOS muxes.. if you need to handle strong signals, that implies decent
diodes with a fair amount of current.

High quality sealed relays are inexpensive, very reliable, etc.

On Thu, 07 Oct 2010 14:58:38 -0400, Bob wrote:

Hi,

A bit off topic, but I figure you folks are the true theorists out
there, and would probably know about this.

Am thinking about purchasing a shortwave defined radio receiver from
WinRadio (http://www.winradio.com/home/g33ddc.htm).

Here is copy from part of their spec sheet:

"The receiver's robust front-end is equipped with an
ultra-high-linearity amplifier which results in exceptional
strong-signal performance. This already robust front-end is further
enhanced with a user-selectable preselector that can operate either in a
fully automatic or user-configurable mode. As many as 119 different
filter combinations can be constructed by the user (91 bandpass, 14
low-pass and 14 high-pass). The front-end employs 34 subminiature
electromechanical relays (rather than often used but distortion-prone
semiconductor switches) to ensure high dynamic range."

Was really surprised to see: The front-end employs 34 subminiature
electromechanical relays (rather than often used but distortion-prone
semiconductor switches) to ensure high dynamic range.

I am retired now, so haven't kept up withese things, or the state of the
art nowadays, but are they still using electromechanical relays for
front ends feeding a 16 bit A to D ?

Haven't semiconductor switches, caught up with reed relays for this
application by now, or do they "still" provide a lower "on" resistance,
etc. ?

Any thoughts would be most appreciated.

Thanks,
Bob
Really good relays are available at very low cost today. Like less
than a dollar each! Between the Low Pass filter and the inboard
antenna tuner my TenTec Jupiter there are about 30 relays.

Other than overrall circuit size and switching speed I see no
advantage for semiconductors as compared to relays.

As far as "Why Ham Radio instead of internet?" :
I worked Japan last night (not my first time) on PSK-31/20 Meters with
40 watts and off the side of the Beam. The feeling was not unlike my
first Ham Radio contact 50 years ago....

It is my kind of getting high!
John Ferrell W8CCW

On Thu, 07 Oct 2010 17:23:15 -0700, Jim Lux wrote:

Haven't semiconductor switches, caught up with reed relays for this
application by now, or do they "still" provide a lower "on" resistance,
etc. ?



Semiconductor switches have issues with intermodulation, etc. and power
consumption. For wideband RF, you're not going to be using cheap 4066
CMOS muxes.. if you need to handle strong signals, that implies decent
diodes with a fair amount of current.

High quality sealed relays are inexpensive, very reliable, etc.

Just a note ...
The signal that will make a switching diode go nonlinear will be Strong
signals (mostly AM/FM/TV) that are not even in the desired band. Several
receiver manufacturers even place a 20db BCB BP filters as the first
thing after the antenna connection and before the attn circuit (also
diode controlled)to prevent these signals from even reaching the first
BPF diode switching network. A bad thing for people interested in BCB-DX.

I think relays in the front end of any RX is a good thing. Relay
reliability has improved allot in the last 30 years too so longevity
should not be an issue in a receiver.

Bob,

you made an easy question and a difficult (for me) question.

The easy question.

Haven't semiconductor switches, caught up with reed relays for this
application by now, or do they "still" provide a lower "on" resistance, etc. ?

The problem of using diode-based switches is not in their higher "on"
resistance, but in the fact that, even if strongly forward biased, they do not
behave linearly, like a pure resistance. So, in presence of strong in-band or
out-of-band signals, they can create intermodulation products. Increasing the
diode forward biasing improves the situation, but the intermodulation
performance of a mechanical contact anyway remains superior. Accordingly, some
of the latest state-of-the-art transceivers (e.g. ICOM IC-7700) boast the use of
mechanical switches instead of diode-based switches.

The difficult question.

I am retired now, so haven't kept up withese things, or the state of the art
nowadays, but are they still using electromechanical relays for
front ends feeding a 16 bit A to D ?

I would re-formulate your question in this way: which is the point of avoiding
the use of diodes OUTSIDE the A/D converter, if the A/D converter chip anyway
includes switching diodes INSIDE it. This question is to be put to somebody who
is familiar with the internal electrical design of A/D converter chips, and
knows whether their internal structure is such to cause intermodulation products
in presence of strong signals.

73

Tony I0JX
Rome, Italy