HBR 14 using toroidal cores - UK Vintage Radio Repair and Restoration Discussion Forum

mpegjohn · 30th Jun 2014, 9:27 am

Hi,
I have managed to acquire a complete set of 85KHz IF coils, which will enable me to build an HBR receiver.
However, I would want this to cover the whole of the shortwave band 2-30MHz.
I planned to use reed relays to switch in coils for the RF and oscillator.
The original design has plug in coils for the amateur band only, that are air cored.

Would it be possible to design RF coils based on toroid's?
I had a look at RF transformer design in Langford Smith, and for the aerial transformer he says that you should use a low value of coupling coefficent, k. I can't see how you could do this using a toroidal core?

Can anyone help with this please?

Regards,
John

G6Tanuki · 30th Jun 2014, 11:04 am

Yes you can use toroids for RF tuned-circuits; the problem you're talking about is essentially because you can't have a 'loosely coupled' antenna winding on a toroid. You may have to accept 'tight' coupling and fit a panel-mounted antenna trimmer.

It's also harder to make toroids variable - since you have to add or remove turns. How are you planning to align it if you can't easily adjust the inductance?

Also beware of using toroids as local-oscillator coils - some of the toroid materials have thermal coefficients-of-induction that you really don't want anywhere near an oscillator. Better to use a ceramic-formered 'traditional' inductor with the turns wound under tension, for reasons of thermal stability.

mpegjohn · 30th Jun 2014, 11:39 am

Thanks for the reply.
How do I choose the number of turns for the aerial primary?
For the RF amp output transformer, Langford says you can resonate the primary above or below the lower or upper tuned frequency. If above, then this will cause regeneration, if below then degeneration.
I am guessing I want a small amount of regeneration, is that correct?

For the oscillator coil I was hoping to be able to use toroids, but after your comments maybe not.

G6Tanuki · 30th Jun 2014, 2:24 pm

Without knowing the nature of the aerial [its impedance] and the characteristics of the ferrite used in the toroid the number of turns (or more relevant, the ratio of aerial-coil turns to tuned-circuit turns) is difficult to guess. Are you looking for a 50-ohm low impedance antenna input (coax feed?) or a medium-impedance (300-400 Ohm) type?

I'm confused about your comments regarding tuning the primary of the RF output transformer - usually these circuits are non-resonant. You neither want regeneration or degeneration!

mpegjohn · 30th Jun 2014, 2:49 pm

Hi,

I will be feeding it with a low Z 50 Ohm input.

Langford says that "the primary should chosen to resonate with the stray capacity and sometimes even a fixed capacitance placed across the primary. This is chosen to resonate 1.2 to 1.5 times the highest freq tuned, so as to present an inductive load to the valve, and thus present a negative resistance, and a capacitance on the valve input, giving a degree of regeneration.

So is this how we choose the number of turns for the primary:

say an estimate of 20pF stray at 5MHz x 1.5 = 7.5MHZ. About 22uH.

What do you think?

G6Tanuki · 2nd Jul 2014, 7:48 pm

I really don't like making primary couplings resonant - it makes life so much more difficult and hard to align.

If you're coupling a 50-ohm antenna feeder to a first-stage RF tuned-circuit using a traditional vari-mu pentode I'd look at using something at most a 1:8 turns ratio. You don't want much of a transformer step-up at this point or you can easily deliver enough signal to cause cross-modulation. This is always more of an issue if AGC is applied to the first RF amp. "Variable-mu" isn't the universal solution to cross-modulation.

My personal preference is more like a 1:5 ratio, and to use a cascode front-end. If you must use a pentode go for something like a 6AK5 [or if you want historic an EF54] and control its gain via a variable cathode-resistor [don't connect the grid to the AGC line; derive the screen-grid voltage from a potential-divider network across the HT rather than a single dropper-resistor].

In any receiver aimed at handling modern amateur-band signal levels you always need to be achieving 'only just enough' gain on all the stages ahead of the main selectivity-elements. When faced with a crusher of a local signal [or a 500Kw broadcast station on 40-metres] you may need the first-RF stage to act as an attenuator not an amplifier.

mpegjohn · 3rd Jul 2014, 12:21 pm

Thank you,
I will take your advice and try the 1:5 ratio initially.

Thank you also for the help with the valve types.

I thought using toroids for the RF coils, I could trim them using trimmer caps, as I can't change the inductance any more than +/- 1 turn.

Any thoughts on toroidal cores for the RF stage?

30th Jun 2014, 9:27 am	#1
mpegjohn Hexode Join Date: Aug 2013 Location: Chichester, West Sussex, UK. Posts: 270	HBR 14 using toroidal cores Hi, I have managed to acquire a complete set of 85KHz IF coils, which will enable me to build an HBR receiver. However, I would want this to cover the whole of the shortwave band 2-30MHz. I planned to use reed relays to switch in coils for the RF and oscillator. The original design has plug in coils for the amateur band only, that are air cored. Would it be possible to design RF coils based on toroid's? I had a look at RF transformer design in Langford Smith, and for the aerial transformer he says that you should use a low value of coupling coefficent, k. I can't see how you could do this using a toroidal core? Can anyone help with this please? Regards, John

30th Jun 2014, 11:04 am	#2
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 14,007	Re: HBR 14 using toroidal cores Yes you can use toroids for RF tuned-circuits; the problem you're talking about is essentially because you can't have a 'loosely coupled' antenna winding on a toroid. You may have to accept 'tight' coupling and fit a panel-mounted antenna trimmer. It's also harder to make toroids variable - since you have to add or remove turns. How are you planning to align it if you can't easily adjust the inductance? Also beware of using toroids as local-oscillator coils - some of the toroid materials have thermal coefficients-of-induction that you really don't want anywhere near an oscillator. Better to use a ceramic-formered 'traditional' inductor with the turns wound under tension, for reasons of thermal stability.

30th Jun 2014, 11:39 am	#3
mpegjohn Hexode Join Date: Aug 2013 Location: Chichester, West Sussex, UK. Posts: 270	Re: HBR 14 using toroidal cores Thanks for the reply. How do I choose the number of turns for the aerial primary? For the RF amp output transformer, Langford says you can resonate the primary above or below the lower or upper tuned frequency. If above, then this will cause regeneration, if below then degeneration. I am guessing I want a small amount of regeneration, is that correct? For the oscillator coil I was hoping to be able to use toroids, but after your comments maybe not.

30th Jun 2014, 2:24 pm	#4
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 14,007	Re: HBR 14 using toroidal cores Without knowing the nature of the aerial [its impedance] and the characteristics of the ferrite used in the toroid the number of turns (or more relevant, the ratio of aerial-coil turns to tuned-circuit turns) is difficult to guess. Are you looking for a 50-ohm low impedance antenna input (coax feed?) or a medium-impedance (300-400 Ohm) type? I'm confused about your comments regarding tuning the primary of the RF output transformer - usually these circuits are non-resonant. You neither want regeneration or degeneration!

30th Jun 2014, 2:49 pm	#5
mpegjohn Hexode Join Date: Aug 2013 Location: Chichester, West Sussex, UK. Posts: 270	Re: HBR 14 using toroidal cores Hi, I will be feeding it with a low Z 50 Ohm input. Langford says that "the primary should chosen to resonate with the stray capacity and sometimes even a fixed capacitance placed across the primary. This is chosen to resonate 1.2 to 1.5 times the highest freq tuned, so as to present an inductive load to the valve, and thus present a negative resistance, and a capacitance on the valve input, giving a degree of regeneration. So is this how we choose the number of turns for the primary: say an estimate of 20pF stray at 5MHz x 1.5 = 7.5MHZ. About 22uH. What do you think?

2nd Jul 2014, 7:48 pm	#6
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 14,007	Re: HBR 14 using toroidal cores I really don't like making primary couplings resonant - it makes life so much more difficult and hard to align. If you're coupling a 50-ohm antenna feeder to a first-stage RF tuned-circuit using a traditional vari-mu pentode I'd look at using something at most a 1:8 turns ratio. You don't want much of a transformer step-up at this point or you can easily deliver enough signal to cause cross-modulation. This is always more of an issue if AGC is applied to the first RF amp. "Variable-mu" isn't the universal solution to cross-modulation. My personal preference is more like a 1:5 ratio, and to use a cascode front-end. If you must use a pentode go for something like a 6AK5 [or if you want historic an EF54] and control its gain via a variable cathode-resistor [don't connect the grid to the AGC line; derive the screen-grid voltage from a potential-divider network across the HT rather than a single dropper-resistor]. In any receiver aimed at handling modern amateur-band signal levels you always need to be achieving 'only just enough' gain on all the stages ahead of the main selectivity-elements. When faced with a crusher of a local signal [or a 500Kw broadcast station on 40-metres] you may need the first-RF stage to act as an attenuator not an amplifier.

3rd Jul 2014, 12:21 pm	#7
mpegjohn Hexode Join Date: Aug 2013 Location: Chichester, West Sussex, UK. Posts: 270	Re: HBR 14 using toroidal cores Thank you, I will take your advice and try the 1:5 ratio initially. Thank you also for the help with the valve types. I thought using toroids for the RF coils, I could trim them using trimmer caps, as I can't change the inductance any more than +/- 1 turn. Any thoughts on toroidal cores for the RF stage?