UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Amateur and Military Radio

Notices

Vintage Amateur and Military Radio Amateur/military receivers and transmitters, morse, and any other related vintage comms equipment.

Reply
 
Thread Tools
Old 25th Oct 2022, 10:31 pm   #1
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,458
Default Improving Selectivity In Vintage Receivers

A current thread discussing improving the performance of an Eddystone 840 identified the issue of selectivity as a key specific issue. This applies to many vintage receivers, generally those designed before SSB became a dominant mode.

There does not seem to be a "magic bullet" solution to this problem and various approaches were discussed. These included the possible use of Q multipliers and of the ceramic filters produced for use in various receivers with 455kHz and 10.7MHz IF's.

One member of the forum has already installed a mechanical filter in an HRO with success, but these are hard to find now. He had also tried the ceramic filters (cheap and cheerful, but not intended for use with valves) and was pleased with that outcome also.

Has anyone any thoughts or prior experience with this issue?

B
__________________
Saturn V had 6 millions pounds of fuel. It would take thirty thousand strong men to lift it an inch.
Bazz4CQJ is offline   Reply With Quote
Old 26th Oct 2022, 8:03 am   #2
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 12,410
Default Re: Improving Selectivity In Vintage Receivers

Small ceramic filters for valve receivers do exist, Trio-Kenwood used them in their 9R59 back in the 1960s.

I also remember that Toko made some midget mechanical filters with built in matching transformer intended for use in FET based circuits, these had high impedance inputs and outputs.

MFH41 and 61 series if my memory is correct.
__________________
"Acht Nul Noyyun Zwo Funnuf" -Magdeburg Annie.
G6Tanuki is offline   Reply With Quote
Old 26th Oct 2022, 10:16 am   #3
G3PIJpeter
Hexode
 
Join Date: Nov 2011
Location: Chippenham, Wiltshire, UK.
Posts: 285
Default Re: Improving Selectivity In Vintage Receivers

Another approach is to build an up / down converter from the 455 kHz IF in the receiver to the frequency of an HF ladder filter and then back down again to the IF. Many of us have shovelfulls of 3.51xxx TV crystals tucked away, from which a suitable selection can make a very respectable CW, SSB or AM passband. The up / down idea was described in a 1980s Radcom article and there were several others. However, if you do have a suitable Murata ceramic filter at 455, then it should be straightforward to fit it into an impedance matching amplifier (to make up the loss). Again, I seem to remember an article in Ham Radio.

And what about an outrider BC453? - there's more than one way of skinning a cat etc.

Peter
G3PIJpeter is offline   Reply With Quote
Old 26th Oct 2022, 10:57 am   #4
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 12,410
Default Re: Improving Selectivity In Vintage Receivers

The up/down converter is an intriguing idea. There was someone on EBay a while back selling modules containing 9MHz KVG filters at a good price. They were the things we all aspired to back a few decades.
Add an xtal oscillator to provide carrier for the up/down conversion and a pair of balanced mixers...

Then you can go on to add a product detector and a stabiliser for the LO HT to stop the effect where altering the RF gain alters the tuning..
__________________
"Acht Nul Noyyun Zwo Funnuf" -Magdeburg Annie.
G6Tanuki is offline   Reply With Quote
Old 26th Oct 2022, 12:14 pm   #5
Jon_G4MDC
Octode
 
Jon_G4MDC's Avatar
 
Join Date: Nov 2015
Location: Welwyn Garden City, Hertfordshire, UK.
Posts: 1,780
Default Re: Improving Selectivity In Vintage Receivers

I have always wanted to try a Murata CFM455J in an HRO. I bought two, have somehow lost one but at least there is one left to try. I want to do it using only passive components if possible.
Jon_G4MDC is offline   Reply With Quote
Old 26th Oct 2022, 12:22 pm   #6
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 21,117
Default Re: Improving Selectivity In Vintage Receivers

There are also the Collins and Kokusai mechanical filters which were used in valved receivers.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 26th Oct 2022, 2:13 pm   #7
G3PIJpeter
Hexode
 
Join Date: Nov 2011
Location: Chippenham, Wiltshire, UK.
Posts: 285
Default Re: Improving Selectivity In Vintage Receivers

I have found a couple of circuits for impedance matching of filters with centre frequency at the receiver IF - see attached.

The single emitter-follower circuit was given in an article about fitting a narrow CW crystal filter to Collins receivers. I imagine it would work OK with a ceramic filter.

The two-transistor circuit is from the RSGB digest of Pat Hawker's 'Technical Topics' articles. The collector resistance R1 is made equal to the filter input impedance; the two R2 resistances are each twice the filter output impedance. I have yet to come across a filter where input and output impedances are not essentially the same.

Can anyone find a circuit for the up / down IF converter? G3JIR wrote a comprehensive set of articles on the design of ladder filters in Radcom - something else to dig out.

Peter
Attached Thumbnails
Click image for larger version

Name:	Filter impedance matching-1.jpg
Views:	133
Size:	59.1 KB
ID:	266896   Click image for larger version

Name:	Filter impedance matching-3.jpg
Views:	123
Size:	85.8 KB
ID:	266897  

Last edited by G3PIJpeter; 26th Oct 2022 at 2:23 pm.
G3PIJpeter is offline   Reply With Quote
Old 26th Oct 2022, 2:17 pm   #8
G3PIJpeter
Hexode
 
Join Date: Nov 2011
Location: Chippenham, Wiltshire, UK.
Posts: 285
Default Re: Improving Selectivity In Vintage Receivers

Quote:
Originally Posted by G3PIJpeter View Post
G3JIR wrote a comprehensive set of articles on the design of ladder filters in Radcom -
Peter
Here's a link - https://warc.org.uk/?page_id=387

Peter
G3PIJpeter is offline   Reply With Quote
Old 26th Oct 2022, 2:28 pm   #9
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 21,117
Default Re: Improving Selectivity In Vintage Receivers

The first circuit in Peter's attachment doesn't work. Zout from an emitter follower is a lot lower than the emitter resistor value. The transistor in an emitter follower works as a feedback amplifier, turning on as much collector-emitter current as necessary to make the emitter voltage try to follow the base voltage. So the circuit tries to be a voltage source of low output impedance.

In a better analysis, the impedance of whatever is applied to the circuit (in parallel with the bias resistors) is reduced by a factor of Hfe. Then add 'little Re' which is 25 Ohms divided by the quiescent current in mA.

An emitter follower with 2.2k in series on the way to the filter would be a much better attempt at matching. It also makes obvious the losses implicit in simple matching.

That website has a very good collection of DIY resources for crystal filters. Jack Hardcastle was a nice chap and I remember having regular chats with him at the Rochdale QRP doos. Also worth looking for a WW article by a Mr Pochet, and Wes Hayward published some articles in the ARRL and in one of his RF design books.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 26th Oct 2022, 2:47 pm   #10
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 12,410
Default Re: Improving Selectivity In Vintage Receivers

Emitter followers working into loads that have potentially rather variable impedance / frequency curves can be problematic, you can easily enter the world of negative resistance and things can oscillate at unexpected frequencies..
This is why, specially at microwave frequency, you may find a 3dB or 6dB resistive pad (often a classic 3-resistor pi network) between stages. You lose gain but gain stability!!
__________________
"Acht Nul Noyyun Zwo Funnuf" -Magdeburg Annie.
G6Tanuki is offline   Reply With Quote
Old 26th Oct 2022, 3:28 pm   #11
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,458
Default Re: Improving Selectivity In Vintage Receivers

I come to this topic specifically in the context of working with my HRO. I don't have any "favourite" approaches in mind, but I have built up a small collection of ceramic filters, but I realised yesterday that most of them are 3kHz, with just a few 4kHz. I'll need to read the above new posts and think about the significance of some of those.

In terms of using the ceramics filters, I would be happy to implant a transistor or three in to the HRO if that helped impedance matching and/or compensation for the insertion loss. Taking a look on Google a couple of days ago, I came across two features by this old guy radiofun 232 (who has quite a few fans on this forum https://www.vintage-radio.net/forum/...d.php?t=195411)

I found these two videos on ceramics; https://www.youtube.com/watch?v=lVE5IM32VcA part 1 & https://www.youtube.com/watch?v=mTqXWB6RSOA part 2, (Part 2 is the best) and I think there may be other videos by that guy which are relevant (certainly to us less-experienced guys).

Peter's article does provide a lot of encouragement to take a serious look at ceramics. I was not aware that some had been produced with valves.

B
__________________
Saturn V had 6 millions pounds of fuel. It would take thirty thousand strong men to lift it an inch.
Bazz4CQJ is offline   Reply With Quote
Old 26th Oct 2022, 4:34 pm   #12
turretslug
Dekatron
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,202
Default Re: Improving Selectivity In Vintage Receivers

I recall Brush-Clevite as an early name in ceramic filters of the valve era- though the name tends to conjure up images of locomotives rather than precision electronics!

Interfacing modern ceramic filters with all their appeal of high, tailored effectiveness in a small, convenient, inexpensive package to traditional valve IF architecture is something I'd wondered about pretty much since I'd had an interest in vintage radio, but like Bazz, I felt that there wasn't much out there that was really convincing or comprehensive in approach. I'm aware that the filter makers recommend wideband resistive termination on input and output in order to maintain published response characteristics out-of-band, but they often offered matching transformers with their more expensive offerings and Toko et al offered IFTs with low impedance secondaries for typical filters. It's noticeable that the AM sections of modern (i.e. solid state!) tuners frequently have resistive matching between such coupling windings and filter inputs, with the filter output going to transistor base with bias resistors chosen to give appropriate loading. I wonder just what supply voltage 10mm (and smaller) format IFTs would withstand.... At least the 840 series sets run things at quite low HT, even with 230V mains.

I'm slightly shocked that the first circuit in post #7 got through to publication, really- surely that output impedance booboo would have occurred to someone en route? Maybe it was a drafting error and the 2k2 should have been a series resistor, with the emitter resistor separate but accidentally omitted. It's certainly not the first time I've looked at a magazine circuit and thought, "Er, what?"!
turretslug is offline   Reply With Quote
Old 26th Oct 2022, 4:49 pm   #13
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 12,410
Default Re: Improving Selectivity In Vintage Receivers

Brush Clevite did some of the first ceramic filters, under the trademark of 'Transfilters' back in the early 1960s, they were used in the likes of the Heath Mohican solid state radios.

Along with Kokusai and Collins filters, both long since obsolete and unobtainable.

Must admit that I am looking at this thread with interest, I have a R209 receiver with a 465KHz IF-out socket, which would mean additional outboard filters and detector circuits more appropriate for SSB are easily added.

(extracting I and Q components then feeding them into the left and right sides of a PC soundcard for a bit of DSP, why not??)
__________________
"Acht Nul Noyyun Zwo Funnuf" -Magdeburg Annie.
G6Tanuki is offline   Reply With Quote
Old 26th Oct 2022, 5:14 pm   #14
G3PIJpeter
Hexode
 
Join Date: Nov 2011
Location: Chippenham, Wiltshire, UK.
Posts: 285
Default Re: Improving Selectivity In Vintage Receivers

Quote:
Originally Posted by turretslug View Post
... I'm slightly shocked that the first circuit in post #7 got through to publication, really- surely that output impedance booboo would have occurred to someone en route?
That 75A-4 additional filter article was in Ham Radio Magazine - they ought to have known better because it was rather at the 'heavy' end of the spectrum. The magazine seemed to reach the point where it had said everything there was to say and so closed down. I have cut out and filed many articles from a stack of inherited past numbers - a useful resource.

The approach I took to adding an extra IF filter to an HRO (during rebuild) was to improve the existing 'barn door' response for SSB. Whether the ripple was not ideal or the skirts not going down as far as they should due to my bodging does not really matter to me. The result was a vast improvement achieved simply. There was also sufficient gain in hand not to worry about making good the 12+ dB loss through the filter.

However . . . I have just started the rebuild of another HRO basket case and the only filter available is a 3 kHz Murata CFXxxx?, so I shall be interested to see what this thread finally decides about how to shoehorn this type of device into a 1930s receiver design.

Peter
G3PIJpeter is offline   Reply With Quote
Old 26th Oct 2022, 5:19 pm   #15
Jon_G4MDC
Octode
 
Jon_G4MDC's Avatar
 
Join Date: Nov 2015
Location: Welwyn Garden City, Hertfordshire, UK.
Posts: 1,780
Default Re: Improving Selectivity In Vintage Receivers

G6T I have done that - sort of. A digital TV dongle with a feed into the IF chip Balanced +/- inputs for the I channel. I fed it from the HRO filter box via twisted pair taken from the crystal filter transformer / tuned circuit.

It worked but I'm not a fan of the external Q-fiver approach especially when it's a PC. I considered an Internal RazPi instead. Then I thought about reporting myself to the police.

It was lovely to have all those modes, filters, notches and recording abilities. It's main disdvantage was the delay. Audio coming out was about 1 second behind real time so tuning the HRO with the knob was just horrible. It was OK using the Dig IF frequency to tune of course.

Set to display wideband (say 1MHz) it could also show deficiencies in coil box alignment. You could see the RF stage response by watching the noise floor change as you tuned across the band. One of my coil packs for 80 m was very badly behaved until it allowed me to sort it out.

Last edited by Jon_G4MDC; 26th Oct 2022 at 5:32 pm.
Jon_G4MDC is offline   Reply With Quote
Old 26th Oct 2022, 6:29 pm   #16
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,458
Default Re: Improving Selectivity In Vintage Receivers

One question it may be useful to discuss up front, what kind of bandwidth are we aiming for? Does anyone want to end up with 2 or 3 switchable options?

Most of the ceramic filters seem to be specified in terms of either 3 or 6dB points and span 2 to 15kHz. I assume, however, that putting one 3kHz ceramic filter in to the HRO will not result in a 3kHz bandwidth?

Also, bearing in mind the comment of radiofun about the ceramic filters tending to get loaded and having their resonant frequencies reduced, I wonder if trying a 460kHz ceramic filter in a set with a nominal 455 IF could have merit?

B
__________________
Saturn V had 6 millions pounds of fuel. It would take thirty thousand strong men to lift it an inch.
Bazz4CQJ is offline   Reply With Quote
Old 26th Oct 2022, 8:07 pm   #17
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Improving Selectivity In Vintage Receivers

A few years ago, I measured one of the little yellow 2 pin 455kHz resonators on a VNA and exported a two-port model. I had a go at making a USB ladder filter to suit the 450kHz IF of the 840. It took eight resonators to get the performance below. It's fairly reasonable for an SSB filter but it would be a labour of love to actually make one. Every resonator is identical as I used the same model for all 8 resonators. In reality, they won't be identical so this would require some select on test caps to get it aligned.

The filter below only uses series resonators. I think it could be improved if I used some shunt elements, but this was just a first attempt.

The impedance of the filter is about 1200Ω.
Attached Thumbnails
Click image for larger version

Name:	450kHz_SSB_BPF.jpg
Views:	71
Size:	70.8 KB
ID:	266908  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 26th Oct 2022, 8:26 pm   #18
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Improving Selectivity In Vintage Receivers

To show I've done this kind of thing before, see below for a cheapo SSB filter built using 9.83MHz crystals from RS that cost 25p each.

This was designed using a two-port model of the crystal and I let the computer optimise the ladder network for an SSB bandwidth. I then built it using the real crystals.

The plot below is taken of the real filter using the real crystals. There are 8 crystals in the filter and so it cost about 2 total for the crystals plus the cost of some ceramic SMD caps.

The plot below is across a 20kHz span. This isn't bad for a fairly decent SSB filter that cost maybe 3 using new parts.
Attached Thumbnails
Click image for larger version

Name:	9830ssb.gif
Views:	78
Size:	18.5 KB
ID:	266911  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 26th Oct 2022, 9:53 pm   #19
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,458
Default Re: Improving Selectivity In Vintage Receivers

The discussion is interesting, but is very speculative.

Referring back to the ideas of our Dutch friend (radiofun232), it seems to me that his idea of maybe 1 or 2 BJT followed by a FET with a sprinkling of ceramic filters is a "minimal-effort" thing to try. If it didn't prove to be a great outcome, it could well be very informative. Anyone want to critique that approach, ideally in a way the at the 'hams' can understand ?

B
__________________
Saturn V had 6 millions pounds of fuel. It would take thirty thousand strong men to lift it an inch.
Bazz4CQJ is offline   Reply With Quote
Old 27th Oct 2022, 3:29 am   #20
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,458
Default Re: Improving Selectivity In Vintage Receivers

Does the design of the AR88 crystal filter (see attached) suggest away forward? A single crystal provides switchable selectivity of either 7, 3, 1.5 and 0.4kHz at -6dB.

It does not look like it would be hard to re-jig the HRO crystal filter to make it in to an AR88 filter (even adding an extra valve would be easy), or cobble something similar for something like an 840?

B
Attached Files
File Type: pdf AR88 filter.pdf (151.5 KB, 68 views)
__________________
Saturn V had 6 millions pounds of fuel. It would take thirty thousand strong men to lift it an inch.
Bazz4CQJ is offline   Reply With Quote
Reply

Thread Tools



All times are GMT. The time now is 6:24 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2023, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.