Ferrite Rods for MW/LW - Design Compromise?

[kalee20]

I've recently been doing some experiments to compare performance of frame aerials against ferrite rod aerials, and I'm finding that dual-band (MW/LW) is something of a compromise.

There's not much 'out there' on ferrite aerial design - F Langford-Smith was just a few years too early to have covered, and I've not found anything online about twin-band ferrite rod engineering theory. So, naturally, UKVRR is my 'go-to' place

I've looked at what's used by manufacturers who have already invented this particular type of wheel. Practice seems to be to use a compact MW winding, with turns-number so that it tunes near one end of the rod, and a LW winding, also compact, near the other end. Switching is that the LW is shorted on MW, and the two are in series-aiding on LW.

However, there is interaction with MW coil position on LW (easy to see) and LW coil position on MW (a bit more subtle). It seems to me, MW tracking and alignment can never quite be as good as a dedicated, single-winding designed for MW only. Normal procedure is to adjust position of MW coil on MW, fix in position, then switch to LW and adjust LW coil. But, even with coils at opposite ends, moving the (shorted) LW coil is going to have some effect on MW, which means that this should be revisited. Add the fact that parallel trimmers also need tweaking at opposite ends of the tuning scales, it's going to be a complex process - and one which can never work as well as two, separated, independent aerial rods.

I've confirmed this using a twin-winding ferrite aerial salvaged from a scrap Murphy U198 radio to replace the frame aerial in an Ever Ready Model K. I've got more thoughts and results, but didn't want to make this thread-starter too long.

Comments invited and appreciated!

[Boulevardier]

Kalee what a fascinating vintage topic to investigate! We tend to take the working and previous optimisation of old stuff for granted... I look forward to an interesting thread!

Mike

[Herald1360]

Ultimately it's probably no more of a compromise than those already needed for tracking setting adjustments. If the resultant performance of the set is "fit for purpose" there would have been no point in spending more for perfection.

[music-centre]

I have got a couple of Dynatron Nomad sets with the Aerials separate so perhaps they found the same during design?
Steve.

http://www.radiomuseum.org/r/dynatro..._tp11tp_1.html

[Boulevardier]

I have never seen a MW/LW radio with two ferrite rods like that before, but perhaps I just don't get around enough! Does anyone know of any others?

Mike

[Ed_Dinning]

Hi Peter, in all things ferrite it might be worth consulting Snelling.
There may be some detail in the excellent Mullard/ Philips app note of the era if any are still in existence

Ed

[kalee20]

Originally Posted by Herald1360

Ultimately it's probably no more of a compromise than those already needed for tracking setting adjustments. If the resultant performance of the set is "fit for purpose" there would have been no point in spending more for perfection.

Yes, I can see that - what I'm trying to ascertain is, is there a compromise and if so, how much?

The subject for my trials is an Ever Ready Model K. This has a single winding frame aerial which is tuned directly for MW. On LW, a 1.96mH inductor (small, so virtually zero pick-up in its own right) is added in series to get the appropriate total inductance for tuning.

I've been replacing with a 160mm long, 10mm dia, ferrite rod, with various winding configurations.

Using a winding with appropriate inductance for MW, I get good results (better than frame).

Using a winding with appropriate inductance for LW, ditto.

Using a winding which tunes for MW, and including an extra winding capable of being added in series for LW, then LW performance with the two coils in series is comparable with the single-winding - as to be expected. But, MW results are interesting.

Sliding the extra winding (unconnected, open-circuit) onto the rod, detunes the MW winding. That's not surprising: even though my LW coil is single-layer, low-capacitance, the presence of a coupled winding, with its few pF shunt capacitance, has an effect. MW can be peaked-up again by sliding the MW coil nearer the end of the rod, thus lowering inductance which compensates for the added capacitance - or by reducing the MW parallel trimmer. But, the peak signal is less than with MW coil only. The unused LW coil is definitely absorbing some signal.

Sliding the extra winding (unconnected, short-circuit) onto the rod, also detunes the MW winding. Again not surprising: with a shorted winding on the same rod, although LW self-capacitance effects are now irrelevant, there can only be zero field in the shorted winding so a portion of the ferrite rod is now, effectively, chopped off - so inductance of the MW winding is reduced. Again, MW can be peaked-up again by sliding the MW coil towards the middle of the rod, thus raising inductance back to originally. But again, the peak signal is somewhat less than with MW coil only. The shorted LW coil is either absorbing, or shielding, some signal, and the nearer to the MW coil it is, the more the MW performance is clobbered.

It does seem that a compromise exists. The fact that Dynatron felt it necessary to use separate rods, says that they wanted to get the best out of their circuit, and the compromise wasn't good enough for them.

Ed - good suggestion to look in Snelling. I have a copy! At work ATM so Tuesday!

[Radio Wrangler]

A new one on me. I've never had to design a LW-MW ferrite rod aerial. If the winding not in use is shorted, then it's going to look like a shorted turn which needs to be as far away as possible from the other for fear of killing the inductance and killing the Q of the one in use. So the flux getting into the active winding from one end at least must be leakage flux from the more central parts of the rod so the permeability of the rod will get in on the act. and you don't want the windings too close to the ends because you want to collect flux and pass it through as many of the turns as possible.

Lots of compromises, it seems.

Philips applications notes, seem likely.
How about Philips Technical Review of the appropriate era?
I've just looked and found nothing useful in 'Mullard Transistor Audio and Rdio Circuits' the AM receiver seems to be single band and no talk of the antenna.

Lots of firms have done transistor radios, but Philips are unusually chatty about it because they want to flog components as well as sets

David

[Julesomega]

I blagged a Nomad expecting a really good and sensitive receive on both bands but the aerial circuits just won't align no matter what I do with the coils or trimmers (bangs head on wall etc.)
Steve, how do yours perform? I'm thinking of substituting the (single) aerial and tuning cap from a scrap radiogram just to see if I can get any sense out of it. I've already aligned the IFTs to perfection but there's still no sign of Europe N° 1 - or anything else except Droitwich and Moorside Edge

[G0HZU_JMR]

On the BBC website there are various research papers about ferrite rod antennas. The link below seems to be the most relevant

http://downloads.bbc.co.uk/rd/pubs/w...les/WHP164.pdf

There are other papers that show ferrite rod antennas being used at HF
http://downloads.bbc.co.uk/rd/pubs/w...les/WHP091.pdf

I think the calculator used in the above paper is available here:

http://downloads.bbc.co.uk/rd/pubs/w...calculator.pdf

...also ferrite rods are discussed at VHF!

http://downloads.bbc.co.uk/rd/pubs/reports/1977-11.pdf


There appear to be some useful reference links at the end of the first paper although I've not checked them out.

[Boulevardier]

Originally Posted by G0HZU_JMR

On the BBC website there are various research papers about ferrite rod antennas. The link below seems to be the most relevant

http://downloads.bbc.co.uk/rd/pubs/w...les/WHP164.pdf

There are other papers that show ferrite rod antennas being used at HF
http://downloads.bbc.co.uk/rd/pubs/w...les/WHP091.pdf

I think the calculator used in the above paper is available here:

http://downloads.bbc.co.uk/rd/pubs/w...calculator.pdf

...also ferrite rods are discussed at VHF!

http://downloads.bbc.co.uk/rd/pubs/reports/1977-11.pdf


There appear to be some useful reference links at the end of the first paper although I've not checked them out.

Thanks for those excellent links. Fourth link (ferrite rods at VHF) is particularly interesting (to me, at any rate!)

Mike

[kalee20]

Thank you too, Jeremy!

I have some measurement results for various configurations, but might not be able to post until tomorrow.

[ionburn]

Originally Posted by Radio Wrangler

Philips applications notes, seem likely.
How about Philips Technical Review of the appropriate era?
I've just looked and found nothing useful in 'Mullard Transistor Audio and Rdio Circuits' the AM receiver seems to be single band and no talk of the antenna.

Lots of firms have done transistor radios, but Philips are unusually chatty about it because they want to flog components as well as sets

David

You are quite right here. I have some issues which have been useful to me. Try Vol 16 No:7 which appears relevant to me (twin ferrite rods). There are other relevant papers on local interference and noise too.

I cannot see where I got them from but do have a pdf if problems.

[ionburn]

I have had a quick look and, although I could not find my original source, I find it in the usual place - see page 181 -

https://worldradiohistory.com/Archiv...eview-1954.pdf

[kalee20]

Many thanks for replies!

Ed Dinning: Yes, there is a chapter in E C Snelling's 'Soft Ferrites' book. But, it doesn't cover twin-winding, dual-band configurations.

Ionburn: That's my evening reading sorted. Thank you very much - good, solid-looking theory.

Jeremy: Links good, but again, single-band only is considered.

David: Yes it seems compromises exist.

For best performance, the winding should be near the centre of the ferrite rod (confirmed in experimental results in Jeremy's links).

For least interaction - especially if one winding is S/C - the coils should be well apart, ie near the ends.

Can't have both

And, not too near the centre is also necessary, so as to have some adjustment available. But that's a detail.

My results are as follows, using an Ever Ready Model K, orienting for maximum pickup; peaking the aerial trimmer for best possible tracking; and measuring the (dc) AVC voltage as indicator of signal strength. Trials for original (single-band) frame aerial; my own-wound ferrite rod aerials; and a ferrite rod aerial salvaged from a Murphy U198 radio.

MW Performance (801kHz local station)

Original frame aerial 1.01V
Home-wound 160mm ferrite rod (MW coil only) 1.49V
Home-wound 160mm ferrite rod, twin-coil MW/LW, LW coil O/C 0.43V
Home-wound 160mm ferrite rod, twin-coil MW/LW, LW coil S/C 1.1V
Murphy ferrite rod, MW only (LW coil removed) 1.1V
Murphy ferrite rod, twin-coil MW/LW, LW coil O/C 0.6V
Murphy ferrite rod, twin-coil MW/LW, LW coil S/C 1.0V

LW Performance (198kHz Droitwich)

Original frame aerial (with original series inductor for LW) 0.9V
Home-wound 160mm ferrite rod (LW coil only) 1.59V
Home-wound 160mm ferrite rod (MW coil + series non-receiving inductor) 1.1V
Home-wound 160mm ferrite rod, twin-coil MW/LW 1.50V
Murphy ferrite rod, twin-coil MW/LW 0.5V

So it does seem that best signal reception is by separate rods, optimised for each band. Obviously, they'd need to be well separated (confirmed by trial).

One note on interpretation of results: The voltages can’t be converted to db easily, thus 1.49V does not represent a 3.38db improvement on 1.01V. It’s a bigger improvement. Why? Because the 1.49V is obtained after going through amplification which has already been throttled back an unknown amount, from the 1.01V figure.

[Radio Wrangler]

It certainly looks like shorting the longwave coil is better for MW than leaving it open.

Allied with the coils needing to be well towards opposite ends, I suspect that the shorted coil acts as a flux block and the working coil thinks it's on a shortened rod. The further away the other coil, the longer the remaining length of effective rod.

David

[kalee20]

Originally Posted by Radio Wrangler

I suspect that the shorted coil acts as a flux block and the working coil thinks it's on a shortened rod. The further away the other coil, the longer the remaining length of effective rod.

That's a good way of putting it David - thank you!

[turretslug]

This thread has reminded me of a couple of recent threads regarding French domestic radios that used small (within the set) rotatable frame aerials on plastic? formers whose MW and LW windings were at right angles to each other- I wonder if any maker ever tried MW and LW ferrite rods at right angles to each other for minimising interaction? It would be space-consuming- but their were German (notably, presumably others) table radios that accommodated swivelable ferrite rods within their cases.

It's interesting that the ferrite rod overall proved capable of better reception than the original frame- I'll confess that I thought of them as a size/convenience compromise as sets got smaller.

[kalee20]

It is interesting. Though, my ferrite aerial is wound with Litz; the frame aerial, which has a longer length of wire anyway, is single-strand copper wire.

It may therefore have a lower overall Q and perform worse, simply because of this. However, I'm not about to go into that much detail! I'm mainly interested in getting optimum results from ferrite!

[Radio Wrangler]

Smaller diameter turns and magnified flux density in the ferrite rod compared to the frame, So likely more flux cutting the wire in an orientation which generates eddy currents circulating perpendicular to the length of the wire, so Litz helps on the ferrite rod, but less so on the frame.

David