Why should someone replace ALL the capacitors on old Tube eq

[Michael A. Terrell]

Foxs Mercantile wrote:

On 2/9/2017 2:08 PM, Michael Black wrote:
Collins used mechanical filters in their receivers, but not all
of them.

Collins introduced them in the 75A-4 and 51J-4 receivers back in
the early '50s.

Drake used a 50 KHz IF and tuned LC filters in their receivers up
until the mid '60s.

The alternative to Collins mechanical filters were multi-pole crystal
filters. Which was what everyone else was using.

One of the stranger things to come across was Henry Radio offered a
kit to install a Collins mechanical filter in the Drake 2B receiver.

Absolutely pointless, but it allowed you to win "dick waving" contests.

I recently saw a notice that Rockwell Collins has stopped production
of Mechanical Filters.


--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

 

[Jeff Liebermann]

On Mon, 13 Feb 2017 23:30:17 -0500, "Michael A. Terrell"
<mike.terrell@earthlink.net> wrote:

Jeff Liebermann wrote:
Sorta. The timer was basically a miniature electroplating bath, which
used a the current flow to move ions of something, from one end of a
glass cylinder to the other. A coulomb is 1 amp for 1 second and can
count both electrons and ions, as in the bath.
http://www.electrolytics.org/faradaysLaw.html
I have a box buried somewhere with the project notes which might have
the data sheet. Meanwhile, I think I may have found the patent, or
rather a later patent as the one I used was in about 1976:
https://www.google.com/patents/US6198701
I'll dig through the citations later...

http://oakbluffclassifieds.com/Funny-Listings/Very-Funny/Curtis-elapsed-time-indicators-520LNA-115-240-indachron-picture-3.jpg

Bingo:
<http://download.siliconexpert.com/pdfs/2010/11/28/5/7/11/117/crts_/manual/120-pc.pdf>
That's it except the one we were using was the PCB mounted model,
something like the 620PC model:
<https://rcfreelance.com/IC/MS3311-3/>
Curtis Instruments is still around. The mercury coulometer product
does not appear on their web site, but is mentioned in the company
history at:
<http://curtisinstruments.com/?fuseaction=Company.Anniversary>

Thanks much.


--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Jeff Liebermann]

On Mon, 13 Feb 2017 21:35:39 -0800, Jeff Liebermann <jeffl@cruzio.com>
wrote:

><http://download.siliconexpert.com/pdfs/2010/11/28/5/7/11/117/crts_/manual/120-pc.pdf>

More:

Mercury coulometer:
<https://en.wikipedia.org/wiki/Mercury_coulometer>

Patents galore:
<https://www.google.com/?tbm=pts#tbs=sbdo:1&tbm=pts&q=%22mercury+coulometer%22>

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Michael A. Terrell]

Jeff Liebermann wrote:

On Fri, 03 Feb 2017 13:39:02 -0500, clare@snyder.on.ca wrote:

On Fri, 03 Feb 2017 09:32:07 -0800, Jeff Liebermann <jeffl@cruzio.com
wrote:
In a previous life, I tried to design a "warranty timer" into a
product. Actually, it was suppose to accumulate and display the
amount of time that the unit had been powered on to help establish
maintenance intervals. In previous products, a mechanical
counter-timer was used, but for this version, it was deemed too big
and expensive.
http://www.alliedelec.com/images/products/Small/70132720.jpg
I found a company that made an electrochemical equivalent. It was
housed in a glass cylinder, similar to a common 3AG glass fuse. Inside
was some chemical solution. When a few volts of DC was applied,
electrolytic action caused one end to slowly turn dark, thus
indicating the amount of time that the DC was applied. Sorry, but I
couldn't find the vendor or an equivalent online. When the required
maintenance was performed, the indicator would be replaced as it could
not be reset.

Something like a coulometer???

Sorta. The timer was basically a miniature electroplating bath, which
used a the current flow to move ions of something, from one end of a
glass cylinder to the other. A coulomb is 1 amp for 1 second and can
count both electrons and ions, as in the bath.
http://www.electrolytics.org/faradaysLaw.html
I have a box buried somewhere with the project notes which might have
the data sheet. Meanwhile, I think I may have found the patent, or
rather a later patent as the one I used was in about 1976:
https://www.google.com/patents/US6198701
I'll dig through the citations later...

<http://oakbluffclassifieds.com/Funny-Listings/Very-Funny/Curtis-elapsed-time-indicators-520LNA-115-240-indachron-picture-3.jpg>


--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

 

[Ralph Mowery]

In article <YuqdnZ0iC6ISDz_FnZ2dnUU7-X-dnZ2d@earthlink.com>,
mike.terrell@earthlink.net says...

I recently saw a notice that Rockwell Collins has stopped production
of Mechanical Filters.

Many of the newer radios are starting to use electronic circuits or more
likely software fro the filtering now.

Many newer radios do not have much RF circuity in them, mainly
microprocessors doing most of the work. This trend started about a
dozen years ago and is progressing more and more every year.
Instead of being limiated to just 2 or 3 filters (that often cost around
$ 100 each) you just turn a knob or move the mouse and set up any type
of filtering you want.

 

[Michael A. Terrell]

Jeff Liebermann wrote:

On Mon, 13 Feb 2017 23:30:17 -0500, "Michael A. Terrell"
mike.terrell@earthlink.net> wrote:

Jeff Liebermann wrote:
Sorta. The timer was basically a miniature electroplating bath, which
used a the current flow to move ions of something, from one end of a
glass cylinder to the other. A coulomb is 1 amp for 1 second and can
count both electrons and ions, as in the bath.
http://www.electrolytics.org/faradaysLaw.html
I have a box buried somewhere with the project notes which might have
the data sheet. Meanwhile, I think I may have found the patent, or
rather a later patent as the one I used was in about 1976:
https://www.google.com/patents/US6198701
I'll dig through the citations later...

http://oakbluffclassifieds.com/Funny-Listings/Very-Funny/Curtis-elapsed-time-indicators-520LNA-115-240-indachron-picture-3.jpg

Bingo:
http://download.siliconexpert.com/pdfs/2010/11/28/5/7/11/117/crts_/manual/120-pc.pdf
That's it except the one we were using was the PCB mounted model,
something like the 620PC model:
https://rcfreelance.com/IC/MS3311-3/
Curtis Instruments is still around. The mercury coulometer product
does not appear on their web site, but is mentioned in the company
history at:
http://curtisinstruments.com/?fuseaction=Company.Anniversary

Thanks much.

You're welcome. I have heard they stopped making them, after some
old ones caught on fire and did a lot of damage.


--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

 

[Michael A. Terrell]

Ralph Mowery wrote:

In article <YuqdnZ0iC6ISDz_FnZ2dnUU7-X-dnZ2d@earthlink.com>,
mike.terrell@earthlink.net says...



I recently saw a notice that Rockwell Collins has stopped production
of Mechanical Filters.

Many of the newer radios are starting to use electronic circuits or more
likely software fro the filtering now.

Many newer radios do not have much RF circuity in them, mainly
microprocessors doing most of the work. This trend started about a
dozen years ago and is progressing more and more every year.
Instead of being limiated to just 2 or 3 filters (that often cost around
$ 100 each) you just turn a knob or move the mouse and set up any type
of filtering you want.

I was involved in the development of the first successful DSP based
Telemetry receiver. It was the Microdyne RCB2000, which was around $80K
for a completely software configured dual diversity receiving system. It
went from RF to a 50 to 90 MHz analog IF, followed by the DSP, and
multiple FIR filters for the IF and Video systems. This was back in
2000-2001. The success of that product caused L3com to purchase
Microdyne, since their DSP based design never worked.


--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

 

[Jeff Liebermann]

On Tue, 14 Feb 2017 11:42:15 -0500, "Michael A. Terrell"
<mike.terrell@earthlink.net> wrote:

Jeff Liebermann wrote:
On Mon, 13 Feb 2017 23:30:17 -0500, "Michael A. Terrell"
mike.terrell@earthlink.net> wrote:

Jeff Liebermann wrote:
Sorta. The timer was basically a miniature electroplating bath, which
used a the current flow to move ions of something, from one end of a
glass cylinder to the other. A coulomb is 1 amp for 1 second and can
count both electrons and ions, as in the bath.
http://www.electrolytics.org/faradaysLaw.html
I have a box buried somewhere with the project notes which might have
the data sheet. Meanwhile, I think I may have found the patent, or
rather a later patent as the one I used was in about 1976:
https://www.google.com/patents/US6198701
I'll dig through the citations later...

http://oakbluffclassifieds.com/Funny-Listings/Very-Funny/Curtis-elapsed-time-indicators-520LNA-115-240-indachron-picture-3.jpg

Bingo:
http://download.siliconexpert.com/pdfs/2010/11/28/5/7/11/117/crts_/manual/120-pc.pdf
That's it except the one we were using was the PCB mounted model,
something like the 620PC model:
https://rcfreelance.com/IC/MS3311-3/
Curtis Instruments is still around. The mercury coulometer product
does not appear on their web site, but is mentioned in the company
history at:
http://curtisinstruments.com/?fuseaction=Company.Anniversary

Thanks much.

You're welcome. I have heard they stopped making them, after some
old ones caught on fire and did a lot of damage.

I don't see what might have burned. They were very low power.
According to the data sheet, 4uA at 5V for the model 120PC. My
guess(tm) is that the mercury used in the indicator was deemed
environmentally incorrect.

It would interesting to resurrect the idea, but using something less
hazardous, in order to build a warranty timer. Offering a warranty by
hours of use, instead of years after manufacture, would be an
attractive sales gimmick.



--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Michael A. Terrell]

Jeff Liebermann wrote:

On Tue, 14 Feb 2017 11:42:15 -0500, "Michael A. Terrell"
mike.terrell@earthlink.net> wrote:

Jeff Liebermann wrote:
On Mon, 13 Feb 2017 23:30:17 -0500, "Michael A. Terrell"
mike.terrell@earthlink.net> wrote:

Jeff Liebermann wrote:
Sorta. The timer was basically a miniature electroplating bath, which
used a the current flow to move ions of something, from one end of a
glass cylinder to the other. A coulomb is 1 amp for 1 second and can
count both electrons and ions, as in the bath.
http://www.electrolytics.org/faradaysLaw.html
I have a box buried somewhere with the project notes which might have
the data sheet. Meanwhile, I think I may have found the patent, or
rather a later patent as the one I used was in about 1976:
https://www.google.com/patents/US6198701
I'll dig through the citations later...

http://oakbluffclassifieds.com/Funny-Listings/Very-Funny/Curtis-elapsed-time-indicators-520LNA-115-240-indachron-picture-3.jpg

Bingo:
http://download.siliconexpert.com/pdfs/2010/11/28/5/7/11/117/crts_/manual/120-pc.pdf
That's it except the one we were using was the PCB mounted model,
something like the 620PC model:
https://rcfreelance.com/IC/MS3311-3/
Curtis Instruments is still around. The mercury coulometer product
does not appear on their web site, but is mentioned in the company
history at:
http://curtisinstruments.com/?fuseaction=Company.Anniversary

Thanks much.

You're welcome. I have heard they stopped making them, after some
old ones caught on fire and did a lot of damage.

I don't see what might have burned. They were very low power.
According to the data sheet, 4uA at 5V for the model 120PC. My
guess(tm) is that the mercury used in the indicator was deemed
environmentally incorrect.

It would interesting to resurrect the idea, but using something less
hazardous, in order to build a warranty timer. Offering a warranty by
hours of use, instead of years after manufacture, would be an
attractive sales gimmick.

It was the AC versions that were reported to have caught fire, just
like some of the modular IEC power connector/line filters when the
European made film capacitors failed. Certain brands of these modules
are replaced in every piece of equipment by some people who collect old
test equipment.


--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

 

[Clifford Heath]

On 15/02/17 09:51, Ralph Mowery wrote:

In article <SLWdncqGJo3OHz7FnZ2dnUU7-anNnZ2d@earthlink.com>,
mike.terrell@earthlink.net says...
Have you tried Software Defined Radio? Free software and under $10
for the USB based hardware will get you started.
I have been playing with them for about 2 years. They seem to work very
well and depending on the one you actually use, they go from about 25
MHz to 2 GHz. The sensitivity for them is about 1/2 of a microvolt for
FM. The things that can be done and free software is amazing to me.

The do all the common modes of ssb, am,fm, cw, and some other things
such as decoding pagers and even the transponders on airplanes.

Here is a link to some more info on them.

http://www.rtl-sdr.com/about-rtl-sdr/

Look for some software called sd sharp while you are at it.

The weaknesses are intermodulation due to insufficient RF band-pass
filtering, and quantization noise due to insufficient resolution in the
A/D converters. RTL2832 has ENOB of around 7... hopeless.

There's no escaping the need for a proper RF front-end for any serious work.

 

[Clifford Heath]

On 15/02/17 08:52, Michael Black wrote:

On Tue, 14 Feb 2017, Ralph Mowery wrote:
In article <YuqdnZ0iC6ISDz_FnZ2dnUU7-X-dnZ2d@earthlink.com>,
mike.terrell@earthlink.net says...
I recently saw a notice that Rockwell Collins has stopped production
of Mechanical Filters.

Many of the newer radios are starting to use electronic circuits or more
likely software fro the filtering now.

Many newer radios do not have much RF circuity in them, mainly
microprocessors doing most of the work. This trend started about a
dozen years ago and is progressing more and more every year.
Instead of being limiated to just 2 or 3 filters (that often cost around
$ 100 each) you just turn a knob or move the mouse and set up any type
of filtering you want.

That's why in the past, some receivers went down to a 50KHz IF, if you
wanted a selection of filters, it was cheaper using LC circuits down
there than a bunch of crystal or mechanical filters at some higher
frequency.

So long as there is nothing 100KHz away from your desired signal,
and no two signals spaced 50KHz apart that intermodulate in your
RF front end. Imaging is a killer. Double conversion can help, but
nothing beats RF selectivity.

Some of the modern receivers are taking advantage of the availability
of devices with much better linearity and headroom, which reduces
the intermod - and they're using I/Q conversion to defeat imaging -
but they cannot achieve the performance of a good communications
receiver.

Software hacks love SDR because it makes them feel like RF experts,
but they're not.

Clifford Heath

 

[Clifford Heath]

On 15/02/17 11:08, Ralph Mowery wrote:

In article <58a393d4$0$10197$c3e8da3$e408f015@news.astraweb.com>,
no.spam@please.net says...


That's why in the past, some receivers went down to a 50KHz IF, if you
wanted a selection of filters, it was cheaper using LC circuits down
there than a bunch of crystal or mechanical filters at some higher
frequency.

So long as there is nothing 100KHz away from your desired signal,
and no two signals spaced 50KHz apart that intermodulate in your
RF front end. Imaging is a killer. Double conversion can help, but
nothing beats RF selectivity.

Some of the modern receivers are taking advantage of the availability
of devices with much better linearity and headroom, which reduces
the intermod - and they're using I/Q conversion to defeat imaging -
but they cannot achieve the performance of a good communications
receiver.

Software hacks love SDR because it makes them feel like RF experts,
but they're not.


Are mixing the $ 10 dongles and real communication receivers that use
microprocessors instead of the crystal and mechanical filters ?

Sorry, but that sentence is not English, and I have no idea what you mean.

I have a ham transceiver that uses the microprocessor instead of fixed
filters. I can assure you that it will seperate signals that are within
a KHz or less without any problems.

So do I. They can be remarkably good, but not "amazing" in any sense.

 

[Michael Black]

On Tue, 14 Feb 2017, Michael A. Terrell wrote:

Foxs Mercantile wrote:
On 2/9/2017 2:08 PM, Michael Black wrote:
Collins used mechanical filters in their receivers, but not all
of them.

Collins introduced them in the 75A-4 and 51J-4 receivers back in
the early '50s.

Drake used a 50 KHz IF and tuned LC filters in their receivers up
until the mid '60s.

The alternative to Collins mechanical filters were multi-pole crystal
filters. Which was what everyone else was using.

One of the stranger things to come across was Henry Radio offered a
kit to install a Collins mechanical filter in the Drake 2B receiver.

Absolutely pointless, but it allowed you to win "dick waving" contests.


I recently saw a notice that Rockwell Collins has stopped production of
Mechanical Filters.

That's now at least a year old, maybe two. They'd continue to sell what
they had in stock, but that was it.

I think I mentioned that some Japanese company was making mechanical
filters in the sixties (word is that they had some foam in them that goes
bad over the decades), which showed up in Lafayette ham receivers and CB
sets, but for some reason, that didn't continue to be a source of
mechanical filters.

There wsa an article in CQ magazine in the early sixties where someone
described a receiver they built, and they even built the mechanical
filter. Not sure if they gave enough information to duplicate the filter
at home, or if anything too out of the ordinary was needed.

I think ceramic filters took care of the cheap mechanical filters, leaving
COllins for the really good receivers.

Michael

 

[Michael Black]

On Tue, 14 Feb 2017, Ralph Mowery wrote:

In article <YuqdnZ0iC6ISDz_FnZ2dnUU7-X-dnZ2d@earthlink.com>,
mike.terrell@earthlink.net says...



I recently saw a notice that Rockwell Collins has stopped production
of Mechanical Filters.

Many of the newer radios are starting to use electronic circuits or more
likely software fro the filtering now.

Many newer radios do not have much RF circuity in them, mainly
microprocessors doing most of the work. This trend started about a
dozen years ago and is progressing more and more every year.
Instead of being limiated to just 2 or 3 filters (that often cost around
$ 100 each) you just turn a knob or move the mouse and set up any type
of filtering you want.

That's why in the past, some receivers went down to a 50KHz IF, if you
wanted a selection of filters, it was cheaper using LC circuits down there
than a bunch of crystal or mechanical filters at some higher frequency.

It is amazing. Even the low end shortwave portables are using ICs that do
a bit of RF handling, and then down to where it can convert to digital.
SO you have ICs that offer more than a given receiver provides, the
modification now consisting of changing the firmware, or bypassing the
internal microprocessor, and controlling it through another computer.

And of course the high end receivers are going the same way, except they
use better circuitry. Build the circuit once, and then you can do endless
things like changing selectivity or adding different types of
demodulators.

Michael

 

[Michael A. Terrell]

Michael Black wrote:

That's why in the past, some receivers went down to a 50KHz IF, if you
wanted a selection of filters, it was cheaper using LC circuits down
there than a bunch of crystal or mechanical filters at some higher
frequency.

It is amazing. Even the low end shortwave portables are using ICs that
do a bit of RF handling, and then down to where it can convert to
digital. SO you have ICs that offer more than a given receiver provides,
the modification now consisting of changing the firmware, or bypassing
the internal microprocessor, and controlling it through another computer.

And of course the high end receivers are going the same way, except they
use better circuitry. Build the circuit once, and then you can do
endless things like changing selectivity or adding different types of
demodulators.

Have you tried Software Defined Radio? Free software and under $10
for the USB based hardware will get you started.


http://www.ebay.com/itm/182111674573

DVB-T-DAB-FM-RTL2832U-R820T-Tuner-Mini-USB-RTL-SDR-ADS-B-Receiver-Stick-DH-/



--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

 

[Ralph Mowery]

In article <SLWdncqGJo3OHz7FnZ2dnUU7-anNnZ2d@earthlink.com>,
mike.terrell@earthlink.net says...

Have you tried Software Defined Radio? Free software and under $10
for the USB based hardware will get you started.


http://www.ebay.com/itm/182111674573

DVB-T-DAB-FM-RTL2832U-R820T-Tuner-Mini-USB-RTL-SDR-ADS-B-Receiver-Stick-DH-/

I have been playing with them for about 2 years. They seem to work very
well and depending on the one you actually use, they go from about 25
MHz to 2 GHz. The sensitivity for them is about 1/2 of a microvolt for
FM. The things that can be done and free software is amazing to me.

The do all the common modes of ssb, am,fm, cw, and some other things
such as decoding pagers and even the transponders on airplanes.

Here is a link to some more info on them.

http://www.rtl-sdr.com/about-rtl-sdr/

Look for some software called sd sharp while you are at it.

 

[Michael A. Terrell]

Ralph Mowery wrote:

In article <SLWdncqGJo3OHz7FnZ2dnUU7-anNnZ2d@earthlink.com>,
mike.terrell@earthlink.net says...



Have you tried Software Defined Radio? Free software and under $10
for the USB based hardware will get you started.


http://www.ebay.com/itm/182111674573

DVB-T-DAB-FM-RTL2832U-R820T-Tuner-Mini-USB-RTL-SDR-ADS-B-Receiver-Stick-DH-/

I have been playing with them for about 2 years. They seem to work very
well and depending on the one you actually use, they go from about 25
MHz to 2 GHz. The sensitivity for them is about 1/2 of a microvolt for
FM. The things that can be done and free software is amazing to me.

The do all the common modes of ssb, am,fm, cw, and some other things
such as decoding pagers and even the transponders on airplanes.

Here is a link to some more info on them.

http://www.rtl-sdr.com/about-rtl-sdr/

Look for some software called sd sharp while you are at it.

That is the software that I'm using, and I've had the hardware since
2014.


--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

 

[Michael A. Terrell]

Clifford Heath wrote:

On 15/02/17 09:51, Ralph Mowery wrote:
In article <SLWdncqGJo3OHz7FnZ2dnUU7-anNnZ2d@earthlink.com>,
mike.terrell@earthlink.net says...
Have you tried Software Defined Radio? Free software and under $10
for the USB based hardware will get you started.
I have been playing with them for about 2 years. They seem to work very
well and depending on the one you actually use, they go from about 25
MHz to 2 GHz. The sensitivity for them is about 1/2 of a microvolt for
FM. The things that can be done and free software is amazing to me.

The do all the common modes of ssb, am,fm, cw, and some other things
such as decoding pagers and even the transponders on airplanes.

Here is a link to some more info on them.

http://www.rtl-sdr.com/about-rtl-sdr/

Look for some software called sd sharp while you are at it.

The weaknesses are intermodulation due to insufficient RF band-pass
filtering, and quantization noise due to insufficient resolution in the
A/D converters. RTL2832 has ENOB of around 7... hopeless.

There's no escaping the need for a proper RF front-end for any serious
work.

Or for good RF test equipment to design and build it. I replaced my
old Boonton 42B and 92B analog meters with their 4200B and 9200A digital
meters. I also have a Fluke 8920A digital True RMS meter that works to
20 MHz. This is some of the same type of equipment that I used on the
production floor at Microdyne.


--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

 

[Ralph Mowery]

In article <58a393d4$0$10197$c3e8da3$e408f015@news.astraweb.com>,
no.spam@please.net says...

That's why in the past, some receivers went down to a 50KHz IF, if you
wanted a selection of filters, it was cheaper using LC circuits down
there than a bunch of crystal or mechanical filters at some higher
frequency.

So long as there is nothing 100KHz away from your desired signal,
and no two signals spaced 50KHz apart that intermodulate in your
RF front end. Imaging is a killer. Double conversion can help, but
nothing beats RF selectivity.

Some of the modern receivers are taking advantage of the availability
of devices with much better linearity and headroom, which reduces
the intermod - and they're using I/Q conversion to defeat imaging -
but they cannot achieve the performance of a good communications
receiver.

Software hacks love SDR because it makes them feel like RF experts,
but they're not.

Are mixing the $ 10 dongles and real communication receivers that use
microprocessors instead of the crystal and mechanical filters ?

I have a ham transceiver that uses the microprocessor instead of fixed
filters. I can assure you that it will seperate signals that are within
a KHz or less without any problems.

 

[Michael Black]

On Wed, 15 Feb 2017, Clifford Heath wrote:

On 15/02/17 08:52, Michael Black wrote:
On Tue, 14 Feb 2017, Ralph Mowery wrote:
In article <YuqdnZ0iC6ISDz_FnZ2dnUU7-X-dnZ2d@earthlink.com>,
mike.terrell@earthlink.net says...
I recently saw a notice that Rockwell Collins has stopped production
of Mechanical Filters.

Many of the newer radios are starting to use electronic circuits or more
likely software fro the filtering now.

Many newer radios do not have much RF circuity in them, mainly
microprocessors doing most of the work. This trend started about a
dozen years ago and is progressing more and more every year.
Instead of being limiated to just 2 or 3 filters (that often cost around
$ 100 each) you just turn a knob or move the mouse and set up any type
of filtering you want.

That's why in the past, some receivers went down to a 50KHz IF, if you
wanted a selection of filters, it was cheaper using LC circuits down
there than a bunch of crystal or mechanical filters at some higher
frequency.

So long as there is nothing 100KHz away from your desired signal,
and no two signals spaced 50KHz apart that intermodulate in your
RF front end. Imaging is a killer. Double conversion can help, but
nothing beats RF selectivity.

And nobody was dropping from HF to 50KHz, except in the very early days of
the superhet. SOme of the receivers were modern enough that they had a
"roofing filter" in the HF range or even above it.

Some of the modern receivers are taking advantage of the availability
of devices with much better linearity and headroom, which reduces
the intermod - and they're using I/Q conversion to defeat imaging -
but they cannot achieve the performance of a good communications
receiver.

There's nothing magic about using phasing to do conversion and getting rid
of the image. It was certainly talked about in microwave circles fifty
years back. But nobody seemed to apply it to low frequencies, though the
RF phasing network probably put some limits on it. But most hobbyists
thought in terms of the phasing method for generating SSB, and were unable
to make the leap to downconversion.

The Drake R8, did it come out about 2000? used the phasing method to go
from an IF around 45MHz, straight down to 50KHz. Yes, the ultimate
selectivity wasn't right after the first mixer, but you weren't likely to
get an SSB-width filter at 45MHz either, so you'd still need to
downcovernt

Software hacks love SDR because it makes them feel like RF experts,
but they're not.

I'm more interested in a hybrid, the variable selectivity and ability to
do different types of demodulation, rather than a magic IC that pretends
traditional RF issues are no longer a problem.

Michael