Valve Radio Vs Solid State Radio

[Phil Allison]

"roughplanet"

Very handy for FM satellite reception where Doppler shift means that the
frequency appears to move constantly. Both my R-9000 & R-8500 have it & I
wouldn't want either of them without it.

** Must be low orbit, ham radio satellites - the usual commercial ones
are all geo-stationery.

I guess you would get about +/- 50 ppm frequency shift or about +/- 7 kHz
in 144 MHz, worst case.

BTW

Those two "radios" are Icom wide range jobs.



............ Phil

 

[Ayn Marx]

Very handy for FM satellite reception where Doppler shift means that the
frequency appears to move constantly. Both my R-9000 & R-8500 have it & I
wouldn't want either of them without it.

Why is there Doppler shift with FM satellite reception? Are you talking
about satellites that move?

 

[roughplanet]

roughplanet wrote:

Why is there Doppler shift with FM satellite reception? Are you talking
about satellites that move?

Yes.

ruff

 

[Joerg]

Hello Patrick,

Communications radios are terrible performers if you
want decent local station performance; they are deliberately
selective, and most without more than a maximum of 3 khz of AF bw.

Not really. I have a 6kHz filter in mine. Then you can tune it "to the
side" and get excellent sound. Of course, you'll have to have a good
speaker or even an amp connected.

Also, you can design your own filter. Most of them feature four or more
selectable filter sockets so there is usually a free one to have at it.

But the best is SSB reception of AM stations with a 6kHz filter. That
runs circles around any regular radio.

What is the measured and guaranteed AF response from the Denon
using AM and local stations, if the station modulation
or test signal has 20 kHz of flat level AF modulation?

Don't know but I bet it can't suppress and replace the carrier in bad
fading situation, like my comm receiver can.

Regards, Joerg

 

[Joerg]

Hello Phil,

** The point is that AFC creates an effect like that of a very broad IF when
this is not the case.

Yes, I see what you mean. But you would notice that because there would
be a "hysteresis" in frequency, you'd have to get closer to the station
frequency for it it pop in versus where you lost it.

** No - the AFC is counteracting your attempts to re-tune the radio - it
needs to be disabled when tuning.

I never use AFC. In radios where it can't be disabled I usually go in
there and do a little surgery.

In any case - a broad IF is *required* for low distortion on the FM band -
plus you will not hear adjacent channel signals along with the wanted one
since the FM "capture effect" prevents that.

Not if you want to listen to a weak station next to a strong one. A high
IP front end and a good crystal filter is the ticket here. My old
Kenwood did great but anything after that was, well, close to junk in
performance.

** So the dial says - but the local oscillator frequency has been pulled
in the opposite direction to the tuning dial.

No, I don't use AFC.

** Variable VHF oscillators stable to within +/- 20 kHz over the room temp
plus internal gear temp range are indeed a design challenge, especially in
the case of varicap ( non tuning gang) tuners. However, AFC compensates for
oscillator drift and mistuning for very little cost.

Little cost but often at the expense of not being able to listen to a
weak station on the next frequency.

Modern FM tuners ( since the 1980s) have their tuning frequencies derived
from a crystal time base so drift is eliminated.

This is one of those. But the lack of a reasonable IF filter does it in.

Regards, Joerg

 

[Joerg]

Hello Ruff,

Very handy for FM satellite reception where Doppler shift means that the
frequency appears to move constantly. Both my R-9000 & R-8500 have it & I
wouldn't want either of them without it.

Agree. But that is an application where you don't compensate for sloppy
engineering but for something that is happening because of the laws of
physics, where there is nothing other than an AFC that would fix it.

Regards, Joerg

 

[Joerg]

Hello Patrick,

Anyway AM radios are a compromise on selectivity needed for
long distance listening, and fidelity need for locals.
A wide bw RF section in an AM radio is harder to build,
but they only suits local station reception with stations at 45kHz minimum
apart.

Actually, I had one that did both well. It had a switch and that changed
the resistors and thus the Q in the IF section.

Why anyone would ever bother with the crackly hissy
interference ridden sound of distant AM station listening is beyond me.
They should all be made with a wider band width RF section.

Depends on how interesting or unique the programming is.

The radios I have made for myself use double tuned two section
LC inputs before the frequency converter tube,
then have R strapped across the IFT coils to lower
the Q of the IF transformers, and one has variable IF coil spacing, and then I
get the
18 kHz of IF bw needed before one can get 9 kHz of audio bw.
Then I have a +/- 6 dB tine control, mandatory with AM radio
since HF audio content is variable.
Radio National here in the ACT on 846 kHz sounds magnificent.
Music programs are fine.

Great. Besides crystal or (very few) ceramic filter, the best IF filters
are the home-made versions.

I don't treasure any old AM sets as they were presented as standard.
All were crap.

The old Astor BPJ is quite good. Except for the cheesy plastic enclosure.

All need serious mods before one removes the horrendous thd levels
and bw limitations.

There isn't much that can be done about THD in a tube set but the IF can
be re-tuned. I have, on occasion, even kicked out filter pots and then
made my own filter on a daughterboard.

The effect of the invasion of Australia by cheap Japanese transistor
radios made people even more used to the worst possible sound.
I never ever purchased such crap.

I did :-(

Once.

If you want really good A reception, you have to search carefully
for tuners which guarantee at least 9 kHz of audio bw
for local station reception.
Unless a tuner specifies the AM audio bw id 9 kHz, assume is the usual
2 kHz so often found.

While some kits were offered in the past, I know of no standard sets available
right now
which do the business properly on AM.

You can get a comm receiver with several filter slots where at least one
is vacant. Then roll your own. It isn't easy if the IF is 9MHz or
something because the custom order crystals are freaking expensive. But
one solution is to mix down to where a reasonably priced but top
performance LC filter can be made and then mix back up. No big deal with
a couple of Mini Circuits mixers. There needs to be some space for this
daughter board though, so older comm receivers are more suitable.

There will come a day when we can do all that in a DSP without having to
be a millionaire and programming guru. But that time hasn't come yet.

To easily get 25 kHz bw from each of IFTs in an old tube set, one would really
need to have 2.5 MHz IF frequency rather than 455 kHz, the standard F for
over 60 years. And you'd want 3 IFTs, but this was something that
no accountant would ever allow a 1950's radio maker provide.

They did allow that. My parents had a Loewe Opta set like this,
selectable bandwidth and all. But in those days it must have cost more
than their stove did. Then again, it lived a good 40 years until a kid
tried to take it apart (not me...). The only thing that ever broke was
one of the dial light bulbs.

Regards, Joerg

 

[Phil Allison]

"Joerg"
Phil Allison

Yes, I see what you mean. But you would notice that because there would be
a "hysteresis" in frequency, you'd have to get closer to the station
frequency for it it pop in versus where you lost it.

** That is correct. One receiver I saw ( Goodmans 110 from the late
1970s ) allowed you to move the dial pointer half way along the band before
the AFC finally let go and the frequency snapped back to match to pointer
!!!!!

Not if you want to listen to a weak station next to a strong one.

** A very rare thing for anyone living in a major Australian city.

I understand Yanks are far more keen on that hobby activity.

A high IP front end and a good crystal filter is the ticket here.

** Errr, IP = "intercept point" ??

Short for: "third order, intercept point" ??

Jargon I can handle, but abbreviated acronyms for jargon is bit much.

No, I don't use AFC.

** Shame how that contradicts your comment above.

Little cost but often at the expense of not being able to listen to a weak
station on the next frequency.

** I see you like to ignore facts and post red-herrings.

( Nearly all radios have AFC disable switches )

This is one of those. But the lack of a reasonable IF filter does it in.

** I just checked my hi-fi FM tuner ( digital tuning) and found that with a
strong signal I can offset the frequency by +/- 200 kHz before the signal
deteriorates to a garbled noise.

This suggests that the IF filter is 400 kHz wide at the -60 dB points.

The IF filter for a hi-fi tuner is likely to be about 200 kHz wide at
he -3dB points.

I reckon that is OK performance.





............... Phil

 

[Patrick Turner]

Hello Patrick,


Not really. I have a 6kHz filter in mine. Then you can tune it "to the
side" and get excellent sound. Of course, you'll have to have a good
speaker or even an amp connected.

Most comm radios I have repaired have lousy AM, but
quite reasonable FM.

Also, you can design your own filter. Most of them feature four or more
selectable filter sockets so there is usually a free one to have at it.

I am not a radio expert.

But the best is SSB reception of AM stations with a 6kHz filter. That
runs circles around any regular radio.

Most folks wouldn't know about this, and not want to get a comm radio
for the lounge.

6 kHz isn't a great deal of bw.

Don't know but I bet it can't suppress and replace the carrier in bad
fading situation, like my comm receiver can.

Of course not.

The original poster wanted good local reception i thought,
and for that an AM radio doesn't have to have the complexity
of a comm set, especially if tubes are to be used.
There is the synchrodyne type of circuit,
which can be done with a chip, but I have
not seen an easy to make tomorrow kit.


Patrick Turner.

 

[Patrick Turner]

Hello Phil,


Yes, I see what you mean. But you would notice that because there would
be a "hysteresis" in frequency, you'd have to get closer to the station
frequency for it it pop in versus where you lost it.


I never use AFC. In radios where it can't be disabled I usually go in
there and do a little surgery.

AFC isn't normally needed in a well built tube AM tuner,
especially if they have a temp compensated C in the oscillator circuit.

In FM sets with all tube tuners and ratio detectors,
AFC is easy with the detector DC applied to a 1/2 twin triode
at the input to make a reactance tube, thus a small DC offset change
prevents the oscillator from drifting much, because
little C change is needed to tune a large F change at 100MHz.

Not if you want to listen to a weak station next to a strong one. A high
IP front end and a good crystal filter is the ticket here. My old
Kenwood did great but anything after that was, well, close to junk in
performance.

Most old AM/FM receivers did not have and did not need
crystal filters for local stations.

No, I don't use AFC.

Tube sets drift a bit without AFC.

Patrick Turner.

 

[Patrick Turner]

Hello Patrick,


Actually, I had one that did both well. It had a switch and that changed
the resistors and thus the Q in the IF section.

Sure, but adding R across 455 kHz IFTs has limitations since
skirt selectivity is reduced too much with R = too low a value.

Kenwood tube sets used a switched tertiary on IFT1 to broaden its bw.

I have one of the IFT windings on a slide, to vary the critical coupling.

The adding an R method is crude.

Depends on how interesting or unique the programming is.

Programme material away from Radio National
is abysmal, imho, talk back, repetitive dance & rock and roll shite
from the 70s and 80s and all the fucking adds.
RN in Oz is the govt owned radio network
that occurs around OZ on AM and FM stations.

Great. Besides crystal or (very few) ceramic filter, the best IF filters
are the home-made versions.

I have a murata 12 kHz wide filter somewhere but prefer
LC IFTs. IF at 2.5 MHz would be nice.

The old Astor BPJ is quite good. Except for the cheesy plastic enclosure.

I have heard plenty of plastic cased crap.

There isn't much that can be done about THD in a tube set but the IF can
be re-tuned.

IN my set the AVC is applied to an RF amp ahead of the
frequency converter which has fixed bias,
and the IF amp is a sharp cut off pentode with current FB from unbypassed cathode R.

The detector is done by a diode and RC fed from a CF after the
2nd IF sec.
The common main sources of thd have been drastically reduced.

I have, on occasion, even kicked out filter pots and then
made my own filter on a daughterboard.


I did :-(

Once.


You can get a comm receiver with several filter slots where at least one
is vacant. Then roll your own. It isn't easy if the IF is 9MHz or
something because the custom order crystals are freaking expensive. But
one solution is to mix down to where a reasonably priced but top
performance LC filter can be made and then mix back up. No big deal with
a couple of Mini Circuits mixers. There needs to be some space for this
daughter board though, so older comm receivers are more suitable.

For the ordinary person, nothing much is available.

There will come a day when we can do all that in a DSP without having to
be a millionaire and programming guru. But that time hasn't come yet.


They did allow that. My parents had a Loewe Opta set like this,
selectable bandwidth and all. But in those days it must have cost more
than their stove did. Then again, it lived a good 40 years until a kid
tried to take it apart (not me...). The only thing that ever broke was
one of the dial light bulbs.

In Oz, the average AM tube radio of 1955 had about 1/4 the parts count
of a 1955 Grundig; I don't know about Loewe.

The best AF bandwidth from a Radiola to range radio gram
made in 1958 was 2 kHz. I have two such chassis,
and have modded one to give 7.5 kHz, and also some bass,
and it is heaps better.
A total redesign and rewire was required.

Patrick Turner.

 

[Joerg]

Hello Patrick,

Most comm radios I have repaired have lousy AM, but
quite reasonable FM.

The ones I dealt with never had FM. The pro-receivers are usually 10kHz
to 30MHz. I did see some that went all the way to a GHz but I wasn't too
impressed.

Most folks wouldn't know about this, and not want to get a comm radio
for the lounge.

It's usually their wives who'd object to that ;-)

6 kHz isn't a great deal of bw.

For one sideband it's plenty. In the US stations must not emit anything
past their slot which is 9kHz wide. In other countries it is 10kHz but
that also won't allow more than 5kHz audio BW. Else the radio
authorities would come out and shut the place down.

The original poster wanted good local reception i thought,
and for that an AM radio doesn't have to have the complexity
of a comm set, especially if tubes are to be used.
There is the synchrodyne type of circuit,
which can be done with a chip, but I have
not seen an easy to make tomorrow kit.

Kits aren't around much anymore these days, at least not in the US. For
good local AM I'd still vote for the tube set. Plus, a glass of Merlot
just looks nicer in front of that versus some plastic transistor thingie.

Regards, Joerg

 

[Joerg]

Hello Patrick,

The adding an R method is crude.

Agree, but at least this Loewe radio had a BW selection at all. It did
help some, especially on shortwave.

Programme material away from Radio National
is abysmal, imho, talk back, repetitive dance & rock and roll shite
from the 70s and 80s and all the fucking adds.
RN in Oz is the govt owned radio network
that occurs around OZ on AM and FM stations.

It is a bit better in the US since we have so many different ethnic
groups. Occasionally you can catch interesting music on AM, especially
from countries that otherwise only run propaganda stuff on SW. Once
emigrated the people then are free of those bounds and can share their
real cultural heritage.

IN my set the AVC is applied to an RF amp ahead of the
frequency converter which has fixed bias,
and the IF amp is a sharp cut off pentode with current FB from unbypassed cathode R.

Way to go. I never understood why AGC in "contemporary" radios doesn't
regulate the first amp first. In my own RF designs I usually even go
before the first stage using a PIN diode attenuator. That way you don't
have to be running up and down the linear range. Of course, in a
consumer device the extra Dollar for the PIN section wouldn't be in the
cards.

In Oz, the average AM tube radio of 1955 had about 1/4 the parts count
of a 1955 Grundig; I don't know about Loewe.

When I compare the Astor BPJ of 1959 with the Philips Gamma of similar
vintage there isn't much difference if you discount the parts the
Philips contains for FM and band switching. Performance is similar, too.
The Astor is a tad better in AM reception, probably because it has a
much larger ferrite rod. The Philips wins in versatility because it has
more taps on the power transformer so I could easily swing it to 120V.
The Astor only offers a 200V tap so it needs to run off the 230V rail
which I only have upstairs.

The Philips also boasts the better enclosure, nicely polished wood. Then
again, the Astor is much more compact and portable.

I have a post-WW2 radio where a true minimalist approach had been
attempted. It is a Jotha Liliput. Plastic case, pretty shoddy
construction for my taste. I got it to run again but the performance,
well, leaves a lot to be desired. It is a regenerative single tube set.

Regards, Joerg

 

[Patrick Turner]

Hello Patrick,


The ones I dealt with never had FM. The pro-receivers are usually 10kHz
to 30MHz. I did see some that went all the way to a GHz but I wasn't too
impressed.


It's usually their wives who'd object to that ;-)


For one sideband it's plenty. In the US stations must not emit anything
past their slot which is 9kHz wide. In other countries it is 10kHz but
that also won't allow more than 5kHz audio BW. Else the radio
authorities would come out and shut the place down.

Here the appear to allow 9 kHz of AF bw.
Oz is a big place, and its easier to spread out all the stations.

Kits aren't around much anymore these days, at least not in the US. For
good local AM I'd still vote for the tube set. Plus, a glass of Merlot
just looks nicer in front of that versus some plastic transistor thingie.

I couldn't find an AM radio quite good enough for me so I
made my own.
I first tried a synchrodyne using tubes, but I couldn't get the
diode ring demodulator tranny to work, and then when i tried a
self oscillationg 6BE6 mixer tube for the AF I still got poor
selectivity and other problems.
Afaiac, superhets are king.

Patrick Turner.

 

[Patrick Turner]

Hello Patrick,


Not much though if their designers were clever and had enough budget.

AM sets below 1,600 kHz don't drift much with a tempco cap
in series with the oscilator tuning gang section.

But the FM sets drift a bit, but the AFC using detector DC applied
to a triode to vary C on the oscillator LC tank works very well and once
tuned
a set won't drift. Set up right, it can run without a disable switch as it
is in my set,
and once one of two tuning meters indicates strongest limiting and the other

indicates 0VDC as a centre reading for the ratio detector, the set would
stay tuned for years if left on.

They used variable caps with dissimilar metals in them that compensated
for temperatur expansion. These old engineers also knew how to gauge the
drift of an inductor and compensate for that with capacitor tempco.

Seriously, I have met young engineers who didn't even know that
capacitors had a temperature coefficient.

The best radio I ever saw was an old (tubed) Drake short wave receiver
whose oscillator barely drifted 100Hz. It doesn't have to be this good
but there should be no problem keeping an FM tuner to within a few kHz
without an AFC.

The point of using FM is that tuning does not have to be spot on,
and few kHz off the station F doesn't matter.

Patrick Turner.

 

[Patrick Turner]

Hello Patrick,


Agree, but at least this Loewe radio had a BW selection at all. It did
help some, especially on shortwave.

And on short wave the selectivity is very high since the IF coils are
kept apart, so high selectivity is easy, although you have a higher insertion loss,
and need more gain.

The AF bw on short wave reception is usually quite poor.

adding R across such LC transfromers merely lowers the Q,
but does not take advantage of the flat topped pass band character as one brings the coils

of the IFT closer together, but not too close, lest
one get a double tuned peak, which occurs
when LC coils are over coupled in a tuned IFT.

It is a bit better in the US since we have so many different ethnic
groups. Occasionally you can catch interesting music on AM, especially
from countries that otherwise only run propaganda stuff on SW. Once
emigrated the people then are free of those bounds and can share their
real cultural heritage.

I tire quickly of the ethnic stations. Mostly talk and babble in a foriegn lingo
between the occasional bit of music, and none of it as nice as something
written by Bethoven or Mozart.
I don't even like a digeridoo; sounds like a washing machine with a problem.
Stepan Grappelly is almost as bad on jazz violin.....

Way to go. I never understood why AGC in "contemporary" radios doesn't
regulate the first amp first.

Noise reasons, cost reasons....

In my own RF designs I usually even go
before the first stage using a PIN diode attenuator. That way you don't
have to be running up and down the linear range. Of course, in a
consumer device the extra Dollar for the PIN section wouldn't be in the
cards.


When I compare the Astor BPJ of 1959 with the Philips Gamma of similar
vintage there isn't much difference if you discount the parts the
Philips contains for FM and band switching. Performance is similar, too.
The Astor is a tad better in AM reception, probably because it has a
much larger ferrite rod.

But ferrite rods have a high Q, and enough to cause side band limiting,
and hence AF bw limiting, even without the IFTs.
The set needs to track well with ferrite RF input LC.

But the ferrite picks up the magnetic RF rather than the electostatic,
and gives less hum.
My set with a long wire antenna and dual
input LC circuit hums real badly right on 846 kHz which is the Radio National station,
and its because of some really bad source of local interference, presumably
coming in on the mains wiring.
I have a second set with a ferrite rod mounted above the one I normally use
for times when whoever is causing the interference has their
interfering gear turned on.
Sometimes I hear when they switch on.

The Philips wins in versatility because it has
more taps on the power transformer so I could easily swing it to 120V.
The Astor only offers a 200V tap so it needs to run off the 230V rail
which I only have upstairs.

The Philips also boasts the better enclosure, nicely polished wood. Then
again, the Astor is much more compact and portable.

I have a post-WW2 radio where a true minimalist approach had been
attempted. It is a Jotha Liliput. Plastic case, pretty shoddy
construction for my taste. I got it to run again but the performance,
well, leaves a lot to be desired. It is a regenerative single tube set.

Yuk.

Patrick Turner.

 

[Steve Batt]

Hello Patrick,


The ones I dealt with never had FM. The pro-receivers are usually 10kHz to
30MHz. I did see some that went all the way to a GHz but I wasn't too
impressed.


It's usually their wives who'd object to that ;-)


For one sideband it's plenty. In the US stations must not emit anything
past their slot which is 9kHz wide. In other countries it is 10kHz but
that also won't allow more than 5kHz audio BW. Else the radio authorities
would come out and shut the place down.


Kits aren't around much anymore these days, at least not in the US. For
good local AM I'd still vote for the tube set. Plus, a glass of Merlot
just looks nicer in front of that versus some plastic transistor thingie.

Regards, Joerg

As I said before, check out the Tivoli Audio range of am/fm radios

Steve

 

[Patrick Turner]

As I said before, check out the Tivoli Audio range of am/fm radios

Steve

The manuals on line for the Tivoli Audio AM/FM radios don't include the AF
response data.

So I will assume the products are junk.

Ain't it grand the way idiots advertise their products,
but they don't tell you what you could expect.

Patrick Turner.

 

[Joerg]

Hallo Patrick,

My set with a long wire antenna and dual
input LC circuit hums real badly right on 846 kHz which is the Radio National station,
and its because of some really bad source of local interference, presumably
coming in on the mains wiring.

Often this hum is caused when the local AM signal is strong enough to
get into the mains wiring, being partially demodulated across a
rectifier diode in some power supply. Then the AM signal plus the newly
added audio components of 50Hz, 150Hz and so on reaches your receiver.
In the olden days engineers knew that there should be a few nF across
each diode section to "spoil" its RF performance. But now most forego
that, either out of ignorance or to save a few pennies.

Sometimes I hear when they switch on.

That way you could figure out what it is. But it's likely going to be
water under the bridge because they probably won't let you solder a few
caps across the diodes.

Regards, Joerg

 

[Patrick Turner]

Hallo Patrick,


Often this hum is caused when the local AM signal is strong enough to
get into the mains wiring, being partially demodulated across a
rectifier diode in some power supply.

Some power supply in someone's house in my street.
The noise continues after I turn off my set and the audio fades slowly down
due to a large cap in my radio supply.
So I know its something external.
A battery portable shows pick up of the hum on 846 kHz as well.

Then the AM signal plus the newly
added audio components of 50Hz, 150Hz and so on reaches your receiver.
In the olden days engineers knew that there should be a few nF across
each diode section to "spoil" its RF performance. But now most forego
that, either out of ignorance or to save a few pennies.

Maybe its some SMPS somewhere.

That way you could figure out what it is.

Not a chance, without being intrusive and un-neighbourly.

But it's likely going to be
water under the bridge because they probably won't let you solder a few
caps across the diodes.

I placed a few ferrites around the mains input active and neutral,
and it made no difference.
I don't like placing a cap across my mains.

I also earthed the ground at the power point in the kitchen to a water pipe immediately
under where the radio sits, which is also just inside a wall where the metal meter box is.
But elsewhere in the house I get hum with valve sets with
a wire antenna and old style RF input coil with a small ferrite slug adjust.

Patrick Turner.