"Michael Black" wrote in message
...

But when building with
an IC, you will have to actually build something around it in order to
get something useful.
SNIP
thirty or so years ago, when
people would write to the magazines and complain about so many ICs
being used, and about how the internal diagram of the IC was not shown.
A lot of that could be discounted as a transitional reaction, that
since ICs were new people were reacting to the newness rather than
an absolute reaction to ICs being "cheating".

Over 30 years ago I homebrewed a PDP-8 work-alike computer. It was based on
the PDP-8 instruction set. I've never seen PDP-8 electrical or detailed
logic diagrams. I used ICs, but none more complex than a 4-bit adder. The
7400-series was then too expensive for me, so I used a cheaper compatible
Signetic series in most cases. My choice to use ICs was based on a desire to
finish the project within a reasonable time - which I did. There would have
been too many parts going all-discrete. Of course discrete transistors were
used as lamp drivers, to drive the core memory select lines, and in the
voltage regulators.

At least thirty years on, it's hard to imagine that there are people
who haven't adapted.

Even using simple digital ICs, you have to know a little about the internals
of the family to avoid pitfalls - and to interface to other families. Analog
ICs require more understanding.

SNIP

On the other hand, there is validity in constantly thinking through
whether
something should be done with transistors or ICs.

Or with a $2 microcomputer. IMO, evaluating the trade-off between hardware
and software is just as important. Where should one draw the line?

My hobby RF experience has mostly been limited to VHF frequency converters.

OTOH I assembled the first FM-stereo broadcast station in the Washington DC
area, WHFS 102.3 MHz. The "components" were rather large: an HH Scott stereo
generator, an RCA "iron fireman" FM exciter, and a 1 kW power power
amplifier that had seen service a number of other places. I built a
frequency-multiplier / IPA between the exciter and the 4-400 finals, and the
power supplies including that for the finals. I built the stereo audio
console for the station; my memory is a little hazy after 40 years, but I
think the console used some audio ICs.

SNIP

But these are design decisions, not some rhetoric about how everything
must be made from scratch. Any time something is designed, it's important
not just to look at the way to do it, but at other alternatives, because
people often do get blocked by looking down only one path.

As someone else said, engineering is making what you want from the parts
that are available.

73 de bob w3otc

In , (Michael
Black) wrote:

Frank Gilliland ) writes:

Let me ask you a simple question: Suppose someone buys an SP-600, mounts some
chrome knobs, spray-paints his name on the front panel, then posts on the
newsgroup saying, "Hey, look what I built!" Would you call that "homebrew"?

I think there's a vast difference between that scenario, and building with
ICs. In your example, they are mere cosmetic work. But when building with
an IC, you will have to actually build something around it in order to
get something useful.

I am a bit surprised that you hold this opinion this late in the game.
Clearly, it was not an uncommon opinon thirty or so years ago, when
people would write to the magazines and complain about so many ICs
being used, and about how the internal diagram of the IC was not shown.
A lot of that could be discounted as a transitional reaction, that
since ICs were new people were reacting to the newness rather than
an absolute reaction to ICs being "cheating".

At least thirty years on, it's hard to imagine that there are people
who haven't adapted.

Adapted? Yes. I have built countless devices around IC's of just about every
shape and size. But I didn't design or build the -circuits- (i.e, the amps, the
oscillators, the comparators, the negative impedance converters -- the network
of active and passive components that make the device do what it does), I just
followed the instructions provided with the chip and connected the dots.

As others have pointed out, one can go down through a spiral to an
absolute level of "homebrew", but everything would be pretty bulky then.
Of course, early hams built their capacitors and all that, but it
was more necessity than some hard core belief. Once you could get
commercially available components, then they were used unless
a) someone was curious about making a capacitor or b) what was
exactly needed wasn't available.

The "level" of construction is the issue. It's one thing to homebrew the
components, but another to homebrew the circuits. I understand what you and
everyone else is saying, that "homebrew" can include the use of pre-packaged
circuits as components, that it's just a different "level" of construction. Some
might even think that building something from a kit is homebrew. I don't.
Somebody else "brewed" the circuit -- the kit-builder simply assembles the
components (connects the dots).

There are some borderline parts. It hardly makes sense to buy
a commercially made coil if you can wind one yourself, but that's
not because everyone should be making everything, but because if
someone isn't winding, they may not realize it is a simple thing,
and winding will save money.

What ICs have done is allow for a level of complexity that wasn't
available before them. Sure, there were PLLs described in the ham
magazines using tubes, but they were as complicated as a simple superhet
receiver. I can remember seeing tube based synthesizers, using multiple
crystals mixed together, and they were more complicated than a full
blown transmitter.

Superhets aren't that complicated, and neither is crystal mixing. Regardless,
crystal mixing can get expensive.

If you want to build up a whole synthesizer from transistors, it's
going to be terribly bulky. I suspect few will go to that trouble,
and instead making something simple but which won't give the performance
of a synthesizer.

Not true. Try linearizing a VCO and using a stepped voltage divider, then
beating the output against an OCXO for drift correction. It's an old method, but
it still works great. And it's a lot less complicated and expensive than using
multi-bit comparators, parallel ripple counters, a parallel-to-serial converter
(because there are very few parallel PLL chips anymore), BCD rotory switches or
a decimal-to-binary converter, phase detectors, prescalers, etc. And notice that
you still need that pesky reference oscillator!

There is so much that can be built nowadays that virtually
nobody would consider building in the tube era.

I agree. But I didn't say that every homebrew project must be built with tubes.

So I dismiss your hardcore view on this.

On the other hand, there is validty in constantly thinking through whether
something should be done with transistors or ICs. One shouldn't build
with ICs for the sake of building with them; if two transistors out of
a scrap VCR and some other components from it flash an LED perfectly
well, then what's the point of using an expensive and hard to get
IC that exists only to flash LEDs?

You're getting close....

If you don't lose anything in performance,
and only a little space, then you might as well use readily available scrap
transistors to build an IF strip, than spend money and time buying an
IC via mail order. If two transistors will supply a suitable prescaler
for Harry's project, then it likely is a good choice, because it's
easier to find transistors than prescaler ICs.

That was my point initially.

But these are design decisions, not some rhetoric about how everything
must be made from scratch. Any time something is designed, it's important
not just to look at the way to do it, but at other alternatives, because
people often do get blocked by looking down only one path.

And that, too, is a very good point. Notice that it works both ways -- I see a
lot of homebrewers getting stuck in an 'IC rut'.

For instance, as I write this it occured to me that it might be
easier for Harry to mix the VCO signal down to a frequency where
the average logic IC can work. You don't have to find a prescaler,
and the design's frequency steps won't be limited by the division of
that prescaler. There are various mixer schemes that will result
in the needed frequency. There may be reasons for not doing it this
way, but it may not even be explored because Harry hasn't given this
alternative any thought.

And the synth will be more accurate than using a prescaler, even a dual-modulus
prescaler. But I made the mistake of assuming that he knew the difference
between the two methods and had chosen the prescaler over down-conversion for
reasons unmentioned.





-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

In , (Michael
Black) wrote:

Frank Gilliland ) writes:

Let me ask you a simple question: Suppose someone buys an SP-600, mounts some
chrome knobs, spray-paints his name on the front panel, then posts on the
newsgroup saying, "Hey, look what I built!" Would you call that "homebrew"?

I think there's a vast difference between that scenario, and building with
ICs. In your example, they are mere cosmetic work. But when building with
an IC, you will have to actually build something around it in order to
get something useful.

I am a bit surprised that you hold this opinion this late in the game.
Clearly, it was not an uncommon opinon thirty or so years ago, when
people would write to the magazines and complain about so many ICs
being used, and about how the internal diagram of the IC was not shown.
A lot of that could be discounted as a transitional reaction, that
since ICs were new people were reacting to the newness rather than
an absolute reaction to ICs being "cheating".

At least thirty years on, it's hard to imagine that there are people
who haven't adapted.

Adapted? Yes. I have built countless devices around IC's of just about every
shape and size. But I didn't design or build the -circuits- (i.e, the amps, the
oscillators, the comparators, the negative impedance converters -- the network
of active and passive components that make the device do what it does), I just
followed the instructions provided with the chip and connected the dots.

As others have pointed out, one can go down through a spiral to an
absolute level of "homebrew", but everything would be pretty bulky then.
Of course, early hams built their capacitors and all that, but it
was more necessity than some hard core belief. Once you could get
commercially available components, then they were used unless
a) someone was curious about making a capacitor or b) what was
exactly needed wasn't available.

The "level" of construction is the issue. It's one thing to homebrew the
components, but another to homebrew the circuits. I understand what you and
everyone else is saying, that "homebrew" can include the use of pre-packaged
circuits as components, that it's just a different "level" of construction. Some
might even think that building something from a kit is homebrew. I don't.
Somebody else "brewed" the circuit -- the kit-builder simply assembles the
components (connects the dots).

There are some borderline parts. It hardly makes sense to buy
a commercially made coil if you can wind one yourself, but that's
not because everyone should be making everything, but because if
someone isn't winding, they may not realize it is a simple thing,
and winding will save money.

What ICs have done is allow for a level of complexity that wasn't
available before them. Sure, there were PLLs described in the ham
magazines using tubes, but they were as complicated as a simple superhet
receiver. I can remember seeing tube based synthesizers, using multiple
crystals mixed together, and they were more complicated than a full
blown transmitter.

Superhets aren't that complicated, and neither is crystal mixing. Regardless,
crystal mixing can get expensive.

If you want to build up a whole synthesizer from transistors, it's
going to be terribly bulky. I suspect few will go to that trouble,
and instead making something simple but which won't give the performance
of a synthesizer.

Not true. Try linearizing a VCO and using a stepped voltage divider, then
beating the output against an OCXO for drift correction. It's an old method, but
it still works great. And it's a lot less complicated and expensive than using
multi-bit comparators, parallel ripple counters, a parallel-to-serial converter
(because there are very few parallel PLL chips anymore), BCD rotory switches or
a decimal-to-binary converter, phase detectors, prescalers, etc. And notice that
you still need that pesky reference oscillator!

There is so much that can be built nowadays that virtually
nobody would consider building in the tube era.

I agree. But I didn't say that every homebrew project must be built with tubes.

So I dismiss your hardcore view on this.

On the other hand, there is validty in constantly thinking through whether
something should be done with transistors or ICs. One shouldn't build
with ICs for the sake of building with them; if two transistors out of
a scrap VCR and some other components from it flash an LED perfectly
well, then what's the point of using an expensive and hard to get
IC that exists only to flash LEDs?

You're getting close....

If you don't lose anything in performance,
and only a little space, then you might as well use readily available scrap
transistors to build an IF strip, than spend money and time buying an
IC via mail order. If two transistors will supply a suitable prescaler
for Harry's project, then it likely is a good choice, because it's
easier to find transistors than prescaler ICs.

That was my point initially.

But these are design decisions, not some rhetoric about how everything
must be made from scratch. Any time something is designed, it's important
not just to look at the way to do it, but at other alternatives, because
people often do get blocked by looking down only one path.

And that, too, is a very good point. Notice that it works both ways -- I see a
lot of homebrewers getting stuck in an 'IC rut'.

For instance, as I write this it occured to me that it might be
easier for Harry to mix the VCO signal down to a frequency where
the average logic IC can work. You don't have to find a prescaler,
and the design's frequency steps won't be limited by the division of
that prescaler. There are various mixer schemes that will result
in the needed frequency. There may be reasons for not doing it this
way, but it may not even be explored because Harry hasn't given this
alternative any thought.

And the synth will be more accurate than using a prescaler, even a dual-modulus
prescaler. But I made the mistake of assuming that he knew the difference
between the two methods and had chosen the prescaler over down-conversion for
reasons unmentioned.





-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

In , "Michael A. Terrell"
wrote:

snip
... Homebrewing should be used to learn something,

I couldn't agree more.

and if you want to
remain at the lowest level, enjoy yourself, but don't ridicule others
who want to learn newer methods.

Which is the 'lower' level:

a) connecting the dots

b) learning and using the circuit concepts of a high-frequency bi-stable
multivibrator to build a flip-flop suitable for use as a prescaler

....?




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

In , "Michael A. Terrell"
wrote:

snip
... Homebrewing should be used to learn something,

I couldn't agree more.

and if you want to
remain at the lowest level, enjoy yourself, but don't ridicule others
who want to learn newer methods.

Which is the 'lower' level:

a) connecting the dots

b) learning and using the circuit concepts of a high-frequency bi-stable
multivibrator to build a flip-flop suitable for use as a prescaler

....?




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

Frank Gilliland wrote:

....snipped....


Must be the latitude. For the sake of this discussion, I put whipped up a
flip-flop that can do 240 MHz using 2N4401's. It took me about half an hour. You
don't need to saturate those puppies, you just need a little gain.


....snipped...

OK Frank, I'm interested to see a picture and a schematic of your circuit. Do
you have web pages like Harry's?

I can't help but think that you have been more than a little rough on Harry,
someone who obviously does "homebrew" his stuff. I wish I had the time
patience and ability to construct more stuff that actually worked myself ;-).
Perhaps I'm lazy but I'll choose a suitable IC over a descrete solution 9
times out of ten. I'm not sure why you have a downer on using IC's.

I sympathise with Harry's difficulty in getting hold of parts in Sweden as it
is getting difficult to find (hobbiest) RF parts in the UK.

The hobby of radio and electronics construction is dying in the UK, so we
should be happy if some people build only kits, otherwise the kit suppliers
will go out of business too.

regards...

--Gary

Frank Gilliland wrote:

....snipped....


Must be the latitude. For the sake of this discussion, I put whipped up a
flip-flop that can do 240 MHz using 2N4401's. It took me about half an hour. You
don't need to saturate those puppies, you just need a little gain.


....snipped...

OK Frank, I'm interested to see a picture and a schematic of your circuit. Do
you have web pages like Harry's?

I can't help but think that you have been more than a little rough on Harry,
someone who obviously does "homebrew" his stuff. I wish I had the time
patience and ability to construct more stuff that actually worked myself ;-).
Perhaps I'm lazy but I'll choose a suitable IC over a descrete solution 9
times out of ten. I'm not sure why you have a downer on using IC's.

I sympathise with Harry's difficulty in getting hold of parts in Sweden as it
is getting difficult to find (hobbiest) RF parts in the UK.

The hobby of radio and electronics construction is dying in the UK, so we
should be happy if some people build only kits, otherwise the kit suppliers
will go out of business too.

regards...

--Gary

"Harry - SM0VPO" ) writes:
By the
way, the TDA7000 is also obsolete and is soon to be joining the
ZN414/5/6
and TCM3105 chips in the depths of obscurity.

BR Harry - SM0VPO

Do they have really cheap FM broadcast band only pocket radios over there,
that tune with one button (and have a second button to reset to the bottom
of
the band)?

They're everywhere here, and I've seen them for as low as $1.99 Canadian.
Curious, I bought one at that price. No cermic filter, so it's using
an oddball scheme.

Michael VE2BVW

Hi Michael,

Yes, we do have those wireless sets. The cheapest I have found is 69 crowns
(US$9) which is still quite expensive. But then again, the TDA70xx ICs are
also quite expensive, or obsolete. I am seeking a source of the chip outside
Europe for my kits.

BR Harry

"Harry - SM0VPO" ) writes:
By the
way, the TDA7000 is also obsolete and is soon to be joining the
ZN414/5/6
and TCM3105 chips in the depths of obscurity.

BR Harry - SM0VPO

Do they have really cheap FM broadcast band only pocket radios over there,
that tune with one button (and have a second button to reset to the bottom
of
the band)?

They're everywhere here, and I've seen them for as low as $1.99 Canadian.
Curious, I bought one at that price. No cermic filter, so it's using
an oddball scheme.

Michael VE2BVW

Hi Michael,

Yes, we do have those wireless sets. The cheapest I have found is 69 crowns
(US$9) which is still quite expensive. But then again, the TDA70xx ICs are
also quite expensive, or obsolete. I am seeking a source of the chip outside
Europe for my kits.

BR Harry

Frank Gilliland ) writes:

... as I write this it occured to me that it might be
easier for Harry to mix the VCO signal down to a frequency where
the average logic IC can work. You don't have to find a prescaler,
and the design's frequency steps won't be limited by the division of
that prescaler. There are various mixer schemes that will result
in the needed frequency. There may be reasons for not doing it this
way, but it may not even be explored because Harry hasn't given this
alternative any thought.

Michael VE2BVW

Hello Michael. Down-conversion has been considered, but quickly dismissed.
The sysnthesiser I need is a tool to be used for a wide variety of projects:
projects that include modulation.

Down conversion of frequency preserves any applied modulation. The prescaler
and subsequent dividers filter out modulation so that a true phase lock may
be achieved. This is especially true with WBFM where the total divide rate
must be greater than 10000 to achieve a phase lock. NBFM, with a minimum
modulating freq of, say, 250Hz, still needs a divider to divide by more than
60.

Down conversion could make a simple synth if modulation were never to be
used. I intend the project area to be both TX + RX, and to become a
"building block" for a variety of circuits.

Anyway, I have received a lot of response. My original question was "is
there a good cheap source of prescalers" (but in a long-winded way) and this
I have received. Many thanks to you and all who have given me positive
suggestions. Unfortunately I will not be melting down horses hooves to make
the glue to bond the home-beaten copper to a board to make the PCB. The only
thin board available would be ready-made plywood, and that would not be
"homebrew" ;-)

BR Harry