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

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 19th Jul 2008, 10:01 pm   #1
Mike Brett
Hexode
 
Join Date: Dec 2005
Location: Wisbech, Cambridgeshire, UK.
Posts: 452
Default 80 meter superhet

Hello all
I have built a 80 meter superhet from various modules mainly from Radcom and homebrew circuits using modern components. It all works very well apart from a problem of low gain from the IF amps, so I have to use a signal booster on the antenna input. My best efforts have failed to resolve the problem so I am looking to replace the IF amp and product detector boards and also the AGC board.Does anybody know of a reliable circuit or even a kit that I could use. I have a IF frequency of 11.000 MHz, and the receiver runs on 12 volts DC.
Many thanks Mike
Mike Brett is offline  
Old 20th Jul 2008, 12:06 am   #2
Oldtimer74
Rest in Peace
 
Join Date: Apr 2008
Location: Worcester
Posts: 330
Default Re: 80 meter superhet

11 MHz seems a very high IF for an 80 metre receiver. Is that causing the problem of low gain I wonder? The selectivity is not going to be very high unless you are using a crystal or other type of filter at that frequency.
Pat
Oldtimer74 is offline  
Old 20th Jul 2008, 10:44 am   #3
Mike Brett
Hexode
 
Join Date: Dec 2005
Location: Wisbech, Cambridgeshire, UK.
Posts: 452
Default Re: 80 meter superhet

Hello
Thanks for the reply. Yes I had a very good 11.000 MHz crystal filter so that is why I went for the higher IF. As I said it all works very well apart from the fact of having to use the antenna booster. As you know it has the problem of not only increasing the wanted signal but also the background noise. I did wonder if fitting a pre amplifier before the mixer might help, but it would not cure the main problem.
Regards Mike
Mike Brett is offline  
Old 20th Jul 2008, 11:34 pm   #4
Skywave
Dekatron
 
Skywave's Avatar
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 6,495
Exclamation Re: 80 meter superhet

Your receiver may have a tunable RF amp. before the mixer. If so, its main function is to define the noise level and reduce 2nd. channel interference - not that that will be a problem to you since you are using an 11 MHz IF. Either way, the insertion of a "pre-amp" prior to the mixer is not the way to solve your low-gain problem. In fact, if the receiver does have a tuned RF stage - and assuming that it has a reasonable tuned gain - something like 10 to 15 times voltage gain - fitting a pre-amp. in front of it will seriously degrade the performance of the receiver - mainly due to cross modulation & similar unwanted effects.

The cause of your low gain is because you are using such a high IF. 11 MHz is far too high for a receiver for the 80m AmRad band. Yes, I understand why you chose that - because you happened to have a suitable crystal filter available - but I strongly suggest that you re-design your receiver for a more conventional IF of something like 450 kHz. With, say, 3 of maybe 4 tuned stages at 455kHz you should be able to get adequate selectivity for ssb transmissions. If you really want a high-performance RX, have you considered double-conversion: e.g. 1st. fixed IF at 1.6 MHz followed by 2nd. fixed IF at 100 kHz.

The high IF is the cause of your low-gain. Ditch it. You don't want a "pre-amp" at the front-end.

HTH.

Al / Skywave.
__________________
Never act on your first idea. Like solving a crossword clue, a better idea arrives tomorrow.
Skywave is offline  
Old 21st Jul 2008, 6:15 pm   #5
Sean Williams
Dekatron
 
Sean Williams's Avatar
 
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,447
Default Re: 80 meter superhet

At risk of wading in over my head here......

Why does the IF frequency matter?

If the IF is 11Mhz, and it has enough gain at that frequency, then why not use it?

I wonder if the mixer circuit is efficient, and has been redesigned for the new IF?

Granted, it is much easier to get good stable gain at lower frequencies, but I can't see a problem with the chosen IF in this design.

Just my thoughts on the matter.....
__________________
Engineers make things work and have spare bits when finished
Sean Williams is online now  
Old 21st Jul 2008, 6:22 pm   #6
Mike Brett
Hexode
 
Join Date: Dec 2005
Location: Wisbech, Cambridgeshire, UK.
Posts: 452
Default Re: 80 meter superhet

Hello all
Thanks for all your comments and ideas, food for thought.
Regards Mike
Mike Brett is offline  
Old 22nd Jul 2008, 1:02 pm   #7
milairuk
Pentode
 
Join Date: May 2005
Location: Chelmsford, Essex, UK.
Posts: 150
Default Re: 80 meter superhet

Quote:
Originally Posted by Sean Williams View Post
At risk of wading in over my head here......

Why does the IF frequency matter?

If the IF is 11Mhz, and it has enough gain at that frequency, then why not use it?
I agree with Sean concerning the choice of IF, but have you considered the loss in the crystal filter ? Is it being correctly terminated at the input and output with the correct impedances ? If you are using coupled windings then just try shorting the filter out - the selectivity will be horrible but you might see a large increase in signals ??

John
milairuk is offline  
Old 22nd Jul 2008, 2:02 pm   #8
dominicbeesley
Octode
 
dominicbeesley's Avatar
 
Join Date: Nov 2004
Location: Hebden Bridge, West Yorkshire, UK.
Posts: 1,754
Default Re: 80 meter superhet

Would a dual conversion be worth considering? With 11 at the front then a conventional 450kc/s couple of stages?

Dom
dominicbeesley is offline  
Old 23rd Jul 2008, 12:37 am   #9
Skywave
Dekatron
 
Skywave's Avatar
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 6,495
Arrow Re: 80 meter superhet

Dominic: May I be a little bit blunt here, without giving offence?

Re: your Q. above . . . in a word . . . no. It isn't worth the hassle.

The only reason for going down the dual-conversion route on a comms. RX is to get good selectivity by virtue of a low IF (which must be the 2nd. IF) and, at the same time, good image rejection - by virtue of a high 1st. IF.
Typical figures for fixed IFs with this arrangement are: 1st. IF in the range 1.6 to 1.8 MHz; 2nd. IF 50 kHz to 100 kHz.

Agreed, 11 MHz will meet the criterion of a "high 1st. IF" and, therefore, is a mathematically sound approach. But the gain per stage compared with a 455 kHz IF will be a lot lower. You'll need several stages of amplification at 11 MHz using double-wound IFTs to get about the 80dB (or thereabouts) of gain that will be needed. (This addresses the point raised by Sean above. I'm tempted to delve into the relevant gain calculations over all this, but I'll leave that - for those interested - to refer to the classic works by Langford-Smith, F.E. Terman, M.G. Scroggie, Everett & Anner et alia.)

However, with regard to the Q. raised immediately above this Post, the really relevant Q. that has to be answered first is:
for a comms. RX for the 80m. AmRad band (3.5 - 3.8 MHz), is a dual-conversion RX necessary? To which, IMHO, the answer is no; it is simply over-kill for that freq. band.

Classic comms. receivers - when operating on 80m. - for the most part amount to straighforward superhets with one* tuned RF stage (pre-mixer). Either single-conversion to 455 kHz or double-conversion* (1.6 Mhz / 85 kHz). The more exotic ones with tunable IFs are a possibility for consideration - but that moves this Topic away from the 11 MHz filter - which is at the heart of this Thread.

* RXs that employ two RF amps. and / or double-conversion also cover much higher frequencies. Hence the two RF and/or 2x conversion: needed for input RF above 13 MHz (approx).


Mike: I don't think you'll go far wrong if you simply follow some well-tried routes that the designers of these aforesaid receivers have followed. Some of the Eddystone receivers spring to mind as examples worthy of study.

Yes, I do understand that your 11 MHz filter is a tempting beast that is simply begging for use - I have several such 'animals' here myself - but the blunt fact is that it is simply an inappropriate component for your particular application.

(A couple of random ideas: depending on its bandwidth, a better use could be as a a filter-block between a 144-146 MHz fixed-tuned converter feeding a comms receiver acting as a tunable IF, tuning 10 - 12 MHz. Or possibly in an FM tuner - IF strip).

I am sincerely regretful if all that lot sounds a little bit hard - but I am trying to save you from what could be a lot of wasted effort.

Regards,

AL / Skywave.
__________________
Never act on your first idea. Like solving a crossword clue, a better idea arrives tomorrow.

Last edited by Skywave; 23rd Jul 2008 at 12:56 am. Reason: General tidy-up; a complicated Post!
Skywave is offline  
Old 23rd Jul 2008, 1:40 pm   #10
dominicbeesley
Octode
 
dominicbeesley's Avatar
 
Join Date: Nov 2004
Location: Hebden Bridge, West Yorkshire, UK.
Posts: 1,754
Default Re: 80 meter superhet

Hi Al,

Fair enough, it was only an idea - and like most of mine possibly a daft one. I'd thought I'd seen receivers that had 10Mhz front ends.

Don't forget though Al, sometimes we (I) like to just try things out that are "different" in a spirit of experimentation / determination to make the thing in the junk-box work, with no expectation that it will be the utmost tool for the job.

For instance I'm also building an 80m Rx and so far it is pretty hopeless but I'm learning new stuff all the time...it's now on it's third break to bits and start again! I know it will never be as good as a modern digital rig but it will use up some bits of junk I've got lying about and will give many times more pleasure in use. The first time I managed to get Shannon Volmet to tune and stay tuned for more than 10seconds at a time was a major milestone!

So please be patient with us, I know we sometimes must seem like complete idiots to you guys who are experts, but we all have to start somewhere and this is after all a hobby not a profession for most of us.

Dom
dominicbeesley is offline  
Old 23rd Jul 2008, 3:03 pm   #11
dominicbeesley
Octode
 
dominicbeesley's Avatar
 
Join Date: Nov 2004
Location: Hebden Bridge, West Yorkshire, UK.
Posts: 1,754
Default Re: 80 meter superhet

PS: Your excellent advice is always appreciated, I know you've got me out of a lot of confusion in the past!
dominicbeesley is offline  
Old 23rd Jul 2008, 5:18 pm   #12
Skywave
Dekatron
 
Skywave's Avatar
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 6,495
Arrow Re: 80 meter superhet

Hi Dominic.

It took me about 2 hours to write that last Post; I do realise - as I did at the time - that it could seem a little heavy-handed; the edits were many.

For some of us though - who have been around a bit and have the battle-scars to show - it gets difficult to know just how to fine-tune a message that (a) on the one hand is truly helpful (and not just to the recipient, either - the Messages on the Forum can be read by anyone) and try to pass on the benefit of accumulated hard-won advice but (b) on the other hand, not to sound patronising /authoritarian / dogmatic / and reactionary towards new ideas.

It would be very easy to simply not bother. But I'm not 'wired' that way.

By all means experiment - I do all the time - at Work & at Play. I have made several Posts on this very Forum asking Q's that to some Members must seem elementary. But, IMHO, experiments need to be based on firm foundations & clearly understood principles - preferably provable with a bit of Math & Physics - unless you really have got time to waste and accept that one's rate of learning is probably going to be painfully slow & frustrating.
Mine certainly was in the Early Days.

Analogy: thrashing around in the dark, without a torch, looking for a black cat - that you don't know is there.

Al / Skywave.
__________________
Never act on your first idea. Like solving a crossword clue, a better idea arrives tomorrow.
Skywave is offline  
Old 23rd Jul 2008, 5:54 pm   #13
dominicbeesley
Octode
 
dominicbeesley's Avatar
 
Join Date: Nov 2004
Location: Hebden Bridge, West Yorkshire, UK.
Posts: 1,754
Default Re: 80 meter superhet

No worries, I had no problem with your post, I was just trying to apologise for what must sometimes be frustrating stupidity.

Dom
dominicbeesley is offline  
Old 23rd Jul 2008, 6:53 pm   #14
Mike Brett
Hexode
 
Join Date: Dec 2005
Location: Wisbech, Cambridgeshire, UK.
Posts: 452
Default Re: 80 meter superhet

Hello all again
Very good of you all to take all this trouble for me. I see what you mean about the 11.000 MHz filter being not the perfect tool for the job, but I have made some progress. Regarding the point raised about matching the impendencies at each end of the crystal filter. My original mixer was a diode ring mixer, I spent lots of time trying to get that to work but in the end I gave up and changed it for a 602 mixer. At first that too was no good untill somone sugested I put a toroid transformer in between the mixer and the crystal filter. It was only then that I started to hear the 80 meter band. Regarding the suggestion that I try bypassing the filter to see what happens, I have already tried that and the result is deafening. Of course its mostly broadcast stations, but this leads me to the question does it need a transformer on the output too. If it does then this is where I am at a loss as to what would be suitable. This receiver was built as an experimental set from the begining, that was the whole idea, that is why it is built on the modular principle, it is a test bed for RF circuits. I might have to try another route one day but its all valuable experience in the end.
Regards Mike
Mike Brett is offline  
Old 7th Aug 2008, 9:24 pm   #15
SM5EUF/5
Diode
 
Join Date: Jul 2008
Location: Finspong, Sweden
Posts: 1
Default Re: 80 meter superhet

You need impedance mathing at both ends of the xtal filter.
Do you have any idea of the filter impedance ?
What kind of filter is it ? Homebuilt ?
A diode mixer is a low impedance device and often xtal filters need about 500ohm drive and load impedance. So you need a toroid that transforms 50 to 500ohms. This means a turns ratio of 3.33.
Note that the inductance of the toroid must be so high that the inductive reactance is 4-5 times the needed impedance, i.e. for 500ohms and 11MHz around 30uH.
It is also possible to use an LC-filter to match from 50ohms to the filter impedance.

Also note that a diode mixer needs quite a bit if drive to overcome the threshhold voltage of a diode of around 0.6V. Too low drive power gives a lot of loss in the mixer.

Regards,
Urban
SM5EUF/5 is offline  
Old 7th Aug 2008, 11:02 pm   #16
renato_enca
Tetrode
 
renato_enca's Avatar
 
Join Date: Feb 2007
Location: Lisbon, Portugal
Posts: 51
Default Re: 80 meter superhet

Please be aware that if the filter is homebrew and has more than 1 crystal (eg. a crystal ladder) then you may have serious problems with frequency mismatching between crystals, due to tolerances in manufacture. One aproach that may be pretty expensive but that's the way to do with crystal ladders is to buy 10 or 20 crystals of the same frequency and from the same factory, preferentially. Then, with an 74ls04 oscillator and a frequency meter you select 3 or 4 crystals that are pretty close to each other (better to be less than 100Hz). Unfortunately, as said, you must have a frequency meter, and the crystals of that particular frequency may be too expensive. For the poor guys like me, dual-conversion (may be cumbersome) or 455KHz (quick and easy) are definitely the ways to go.

You will REALLY (and I mean: ABSOLUTELY) need a good impedance matching between the filter and the rest of the receiver and a REALLY good frequency matching between the crystals, or you will have a "crystal barrier" than a crystal filter.

About the mixers, please note (I don't actually believe that you hadn't noted) that SA602 mixers have 1.5kHz input/output impedances.

Quote:
Also note that a diode mixer needs quite a bit if drive to overcome the threshhold voltage of a diode of around 0.6V. Too low drive power gives a lot of loss in the mixer.
Don't be afraid: use a diode ring mixer made with BAR28 schottky diodes (0.2V voltage drop). Those marvels can even pick-up DRM stations quite amazingly!!

Last edited by renato_enca; 7th Aug 2008 at 11:08 pm.
renato_enca is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 7:01 pm.


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 - 2019, vBulletin Solutions, Inc.
Copyright ©2002 - 2019, Paul Stenning.