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 26th Jul 2006, 10:45 pm   #41
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: Regarding home made TRF

Hi Al,
Quote:
Originally Posted by G8DLH
I'm a bit puzzled by that comment, Frank. Just out of interest, what do you have in mind, may I ask?
Balanced low noise [1], high linearity, low VHF RX mixers for one. Also if you attempt to 'back date' some of the recent developements in semiconductor RF circuits to using valves, you often end up looking for the 'hottest' valve you can find. E180Fs and E288CCs being fairly uncommon the more down to earth frame grid valves, like E88CC, EF183 and EF184, are excellent substitutes.

I have a pet theory that valves are much better than their reputation when it comes to phase noise in oscillators etc. You just have to use modern circuit ideas and stop the nasty habit of counting valve envelopes.

Frank N.

[1] Well, relatively speaking, at least.
YC-156 is offline  
Old 26th Jul 2006, 11:11 pm   #42
Skywave
Dekatron
 
Skywave's Avatar
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 6,621
Default Re: Regarding home made TRF

Thanks Frank.

"low VHF RX mixers for one". I hadn't thought of that.

Re: phase noise in oscs. Agreed: that has been my experience.

Although there is this tendency these modern days to keep everything as small as possible and get maximum functionality per cubic centimeter, the fact that valves (especially those with good RF performance - e.g. those you mention) will readily interface to high impedance circuits, tends to be overlooked. And they will stand more abuse (deliberate or otherwise) before going self-destruct! (As any time-served Radio Amateur will know! )

Regards,

Al / G8DLH
__________________
Never act on your first idea. Like solving a crossword clue, a better idea arrives tomorrow.
Skywave is offline  
Old 27th Jul 2006, 9:26 am   #43
Chris_C
Hexode
 
Chris_C's Avatar
 
Join Date: Jan 2003
Location: Darlington (DL3) North East U.K.
Posts: 393
Default Re: Regarding home made TRF

Hi All,

Please let's keep to Neil's original subject matter, fascinating though the discussion could be on frame grid valves
__________________
Chris C G8TJR
Chris_C is offline  
Old 5th Aug 2006, 10:45 pm   #44
Neil Purling
Octode
 
Join Date: Jul 2005
Location: Hull, East Yorkshire, UK.
Posts: 1,998
Default Re: Regarding home made TRF

Am I correct in believing that for a regenerative detector one should have a varaiable mu RF pentode? Therefore a EF183 might work.
However I am sticking with a EF39/6K7. I am also going to try making another radio for Long Wave reception.
This new radio will have a coil with more turns; approx 240 in total tapped at 20 turns for the cathode of the EF39.

I was going to try a ECL80 for the audio o/p stage as I have a o/p transformer to suit this valve. I was however going to use only the pentode portion and drive it with a seperate triode, (one half of a 6SN7GT).
The ECL80 is rated at about 1.5W audio o/p, right? Any one any ideas as to what sort of signal level one would have to put on the grid of the pentode to get that o/p?
Neil Purling is offline  
Old 6th Aug 2006, 7:17 am   #45
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: Regarding home made TRF

Quote:
Originally Posted by Neil Purling
Am I correct in believing that for a regenerative detector one should have a varaiable mu RF pentode?
No on both accounts.

Every valve you will encounter is variable mu to some degree or other. The changes you get from varying the anode or screen grid voltage on a non vari-mu valve is quite sufficient to control regeneration.

Nor does the valve have to be a type intended for RF operation. Audio valves will do quite nicely.

As I mentioned previously, then with regenerative receivers less is more. It is an all too common mistake to use too 'hot' valves for the detector, and doing so is a major contributor to many of the ailments, which can plague the species. You will in particular not get better sensitivity, smoother regen control nor less hum from using a 'strong' RF valve, quite the contrary.

Frequency pulling, another common problem associated with regens in the presence of strong signals, is at least in part directly associated with the vari-mu characteristics of a valve.

My preferred choice for a detector below 10MHz (Ie. for most any sane design ) is the EF86 audio valve. I am presently putting the final touches on my little 'reference' design for an experimental testbed, even though I have been forced by external events to postpone the work for now.

Frank N.

Edit: The pentode in the ECL80 requires around 5Vrms for full power output, somewhat dependent on screen grid voltage and anode load impedance. But 5V or perhaps a bit less should be a reasonable target to aim for.

Last edited by YC-156; 6th Aug 2006 at 7:27 am.
YC-156 is offline  
Old 6th Aug 2006, 12:39 pm   #46
Neil Purling
Octode
 
Join Date: Jul 2005
Location: Hull, East Yorkshire, UK.
Posts: 1,998
Default Re: Regarding home made TRF

In which case then i'd need both triodes in the 6SN7GT.
Neil Purling is offline  
Old 6th Aug 2006, 2:27 pm   #47
quantum
Heptode
 
quantum's Avatar
 
Join Date: Mar 2003
Location: Birmingham, West Midlands, UK.
Posts: 785
Default Re: Regarding home made TRF

Quote:
Originally Posted by YC-156 View Post
As I mentioned previously, then with regenerative receivers less is more. It is an all too common mistake to use too 'hot' valves for the detector, and doing so is a major contributor to many of the ailments, which can plague the species. You will in particular not get better sensitivity, smoother regen control nor less hum from using a 'strong' RF valve, quite the contrary.
True, I remember reading in a 1930s textbook on the subject of regeneration that low gain audio valves are preferred, as the efficiency of such a stage depends on achieving optimum regeneration and not the gain of the valve, and with low gain audio valves it is easier to control the rate of regeneration.

I've used an EF37A as a regen detector with success before, but only used an EF86 as an anode bend detector, where it performed perfectly satisfactorily - the only issue I'd have in using an EF86 in a regen circuit is to make sure that you mount it carefully and perhaps ensure it is away from the speaker to make sure there is no danger from microphony.

I'd have thought that one stage of audio after the regen detector driving the output valve would be enough for local station reception - 'conventional' wisdom would suggest that if you are going to have another stage this should be an RF stage before the regenerative detector.
quantum is offline  
Old 6th Aug 2006, 4:22 pm   #48
Neil Purling
Octode
 
Join Date: Jul 2005
Location: Hull, East Yorkshire, UK.
Posts: 1,998
Default Re: Regarding home made TRF

The only design of TRF I have seen so far with a RF amplifier was one using EF91's as RF amp, detector and AF stage. I guess that the age of the design was when there were large numbers of these valves ex-equipment, and you could get some off your local TV engineer very easily. However the circuit uses commercial coils that are no longer available.

Maybe it's a better idea to stick with a lower gain valve on its own. I know that the EF39 works well. The EF91 did work but it was not as good in this role where regeneration was involved. I found it to be microphonic as well.
I suspect that the EF39 would be fine used on a low frequency.
Any opinion on the tuning coil?
The lowest frequency of the Long Wave band is around four times lower than the low frequency end of Medium Wave. For Medium wave I had 65 turns, tapped at 5 turns. In winding a coil four times larger I tapped it at 20 turns. The point where both coils are tapped is around 1/13th of the total number of turns.
Neil Purling is offline  
Old 6th Aug 2006, 6:10 pm   #49
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: Regarding home made TRF

Quote:
Originally Posted by Neil Purling View Post
The lowest frequency of the Long Wave band is around four times lower than the low frequency end of Medium Wave. For Medium wave I had 65 turns, tapped at 5 turns. In winding a coil four times larger I tapped it at 20 turns.
You may have to make some slight corrections to your new coil data.

A (very) rough formula for the inductance of an airwound coil is

I (uH) = D * N^2 / ( 100* (0.43 + L/D ))

Where D and L are the Diameter and Length in centimetres respectively, N is the number of turns. What this says is that the inductance increases with the square of the number of turns, but decreases linearily with the overall length of the winding.

If you didn't have the effect from the non-zero diameter, then you would get four times the inductance by increasing both the number of turns and the length four times.

Depending on how critical you think you need to know the inductance, you may have to play with the dimensions a bit to get exactly four times the current inductance.

Quote:
Originally Posted by Neil Purling View Post
The point where both coils are tapped is around 1/13th of the total number of turns.
That does sound a bit high. 1/20th to 1/30th is probably closer to optimum, perhaps even smaller, depending on how low losses your setup has.

On LW you may run into the small snag that a good regen detector will have too low bandwidth for good reproduction of an AM signal. If this happens, then you will notice this because the highs may be attenuated to some degree.

Frank N.
YC-156 is offline  
Old 6th Aug 2006, 8:10 pm   #50
Neil Purling
Octode
 
Join Date: Jul 2005
Location: Hull, East Yorkshire, UK.
Posts: 1,998
Default Re: Regarding home made TRF

Now I understand why coils are usually wound in layers.
But if I were to wind a coil with the turns on top of each other in a smaller space: Is there a particular way I should wind it?
I will give the coil I have made a try. If I can get Radio 4 then it's OK!
Maths was never my strong point at school, and that formula you quoted is a bit much for me.
Neil Purling is offline  
Old 6th Aug 2006, 9:37 pm   #51
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: Regarding home made TRF

Quote:
Originally Posted by Neil Purling View Post
But if I were to wind a coil with the turns on top of each other in a smaller space: Is there a particular way I should wind it?
The short answer is that you probably shouldn't!

To have low losses, which you want for your regen detector, the longwave coil would have to be wound using the 'honeycomb' or basket weave, which is virtually impossible to do correctly by hand. My suggestion would be to either make one long single layer coil, possibly using a larger diameter coil form. Or, at your discretion, simply use several coil forms of similar yet different diameters, each wound with a single layer of wire, and put one inside the other.

As long as the layers of copper wire are spaced a few mm apart this will be about as good and compact a coil as you can make for longwave on your workdesk, assuming you don't 'cheat' by using ferrite cores or some similar modern inventions.

If you somehow tries to just make a multi-layer coil, the interwinding capacitance will be unacceptably large for achieving low losses, resulting in worse performance from your receiver than would otherwise be the case.

If you tell me A) what coil form diameter you intend to use, B) wire diameter, C) what frequency range you would like to cover and D) what capacitor range you have in your air variable tuning cap, then I will be happy to crunch the numbers for you.

Frank N.
YC-156 is offline  
Old 7th Aug 2006, 7:13 am   #52
Neil Purling
Octode
 
Join Date: Jul 2005
Location: Hull, East Yorkshire, UK.
Posts: 1,998
Default Re: Regarding home made TRF

The coil form is 38mm in diameter.
Frequency range for Long Wave is usually 150 - 270 Khz.
Tuning capacitor is 500pf. I have no idea of what it's minimum capacitance is.

There are commercially available coils (just), from Isoplethics, but they have a seperate reaction winding and there's no tap for the detector's cathode.
They do have Octal plug in coil forms, which I would like to use, of the diameter mentioned above.

Is the usage of a ferrite rod aerial going to be necessary?
The only large diameter material I can think of is plastic drain pipe, from out of the nearest skip.

Last edited by Neil Purling; 7th Aug 2006 at 7:15 am. Reason: Extra comment
Neil Purling is offline  
Old 7th Aug 2006, 8:39 am   #53
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: Regarding home made TRF

What wire diameter do you intend to use, please? I suggest thinner rather than thicker, since you will need rather a lot of it.

Frank N.
YC-156 is offline  
Old 7th Aug 2006, 9:55 am   #54
Neil Purling
Octode
 
Join Date: Jul 2005
Location: Hull, East Yorkshire, UK.
Posts: 1,998
Default Re: Regarding home made TRF

The spool of wire I have is 36 SWG.
Neil Purling is offline  
Old 7th Aug 2006, 2:12 pm   #55
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: Regarding home made TRF

If you want to sweep 150 to 270 KHz using a 500pF variable capacitor, you will need a coil inductance of 1,6mH (1600uH), plus an additional 200pF of capacitance in parallel with the main tuning capacitor. Some of that additional capacitance will come from parasitic capacitance in the coil though.

Using a 38mm diameter coil form and 36SWG (Imperial) wire, which is 0.2743mm, you will need

N : 377
L (uH): 1602
Length (cm): 11.3
Wire (m): 45.7
R (Ohm): 19.67
Diameter : 3.8

377 closely wound windings for a total length of 11.3cm using a total amount of 46m of copper wire. This will yield a coil with a loss (Q) factor of only 75 at 150KHz. Not terribly impressive, but it is hard to get much higher than that unless you can use a ferrite core (which you won't).

Not shown above is the fact that I have set the winding pitch to be 0.3 mm to allow for a bit of fudge factor for a handwound coil.

If you had a slightly larger diameter coil form, which at least around here building suppliers stocks in small lengths, you would, say, need 247 windings for a 5cm diameter coil form.

Another option would be to make two coils, each covering their segment of LW with a bit of overlap. Not only would the coil losses be lower, you would also benefit from a better band spread as you tune across the band.

500pF variable plus 600pF fixed and a 1.025mH coil would cover around 150 to 205 KHz:

N : 256
L (uH): 1025
Length (cm): 7.7
Wire (m): 31.0
R (Ohm): 13.36
Diameter : 3.8

500pF variable plus 500pF fixed and a 665uH coil would cover 195 to 275KHz:

N : 180
L (uH): 669
Length (cm): 5.4
Wire (m): 21.8
R (Ohm): 9.39
Diameter : 3.8

Yep, I have written a small computer program, which spits out these figures.

Frank N.
YC-156 is offline  
Old 7th Aug 2006, 8:25 pm   #56
Neil Purling
Octode
 
Join Date: Jul 2005
Location: Hull, East Yorkshire, UK.
Posts: 1,998
Default Re: Regarding home made TRF

The commercially available coils I am aware of mean I would have to redesign the circuit of the detector.
Presently I have re-wound the coil for the electron-coupled detector with but one 65 turn winding for the detector. Aerial coupling is two turns of wire as a loose coupler arranged around the coil form. The presence of the 100 turn winding still exerted a loading effect even as it was completely disconnected.
The detector works much better now.
Neil Purling is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 3:14 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 - 2020, vBulletin Solutions, Inc.
Copyright ©2002 - 2020, Paul Stenning.