Regarding home made TRF - UK Vintage Radio Repair and Restoration Discussion Forum

Neil Purling · 10th Jul 2006, 1:57 pm

I made my radio with a mixture of circuits: The 'Electron Coupled' design from member heatercathodeshort, the audio stage adapted from Radio for Boys by E.N. Bradley (1952).
The audio stage is a 6J5 Triode feeding a 6V6.
I get the impression that it could be louder, as I seem to have some travel left on the volume control.
Now I was interested to substitute a 6SN7GT double triode and have the first section give a fixed gain of a factor of two or three. I have no idea of what sort of signal level the detector will give when it is on the point of oscillation.
Ideally I want the strong local signal to just drive the 6V6 just into distortion as the detector breaks into oscillation, assuming the volume control is at ful.

If Isoplethics ever come out with a range of plug in coils beyond the basic Long & Medium Wave ones that is going to be the cue for yet another makeover.
If I adapt the radio to cover Short Waves I would change the detector to a EF91 if that would be better to cover the Short Wave broadcast bands.
Well, that's summat for the longer evenings....

JHGibson · 10th Jul 2006, 4:55 pm

Neil, It would require 2 stages of 6J5 RC coupled audio amps, which have a gain of 15 each according to the gain tables in the back of my RCA Tube Manual, to fully drive a 6V6 from a regenerative detector. Or use the two halves of a 6SN7 in place of two 6J5's.

Don't forget to audio decouple the plate circuits of the audio amps or you may end up with motorboating oscillations.
John.

Neil Purling · 10th Jul 2006, 5:26 pm

Does your valve manual say anything about the sort of input level the 6V6 requires to be fully driven?
I am not sure what sort of output the detector gives but I guess at between 40-80mV, probably nearer 40mV.

JHGibson · 11th Jul 2006, 12:56 am

With a plate voltage of 250, the 6V6 requires 12.5 V RMS drive for max output of 4.5 watts at 8 percent distortion.
John.

Neil Purling · 11th Jul 2006, 10:06 am

The signal level required by the 6V6: 12 volts p-p, or dc measurement with a AVO?
I would like to boost the audio somewhat if I can.
Have any of those members who've built TRF radios been able to follow the signal through the set with a 'scope?
I need to know what sort of signal level a EF39/6K7 or a EF91 would give used as a detector. Then I know what sort of additional boost I need to arrange with the extra triode of a 6SN7GT.

Neil Purling · 11th Jul 2006, 3:57 pm

You can refer to my diagram I drew of the AF stage of my TRF radio from my copy of Radio for Boys.
If I do fit a 6SN7G Double Triode.
I was going to use a preset pot of 1Meg value on one stage and have the normal volume control on the other section.
Which triode should have the official vol control &which the preset?
Which would be best?

JHGibson · 11th Jul 2006, 4:04 pm

The drive to the 6V6 would really have to be measured by a vacuum valve voltmeter set to AC RMS, or with an oscilloscope. The impedance level at the control grid is high and a AC reading voltmeter like an AVO would load the cicuit badly and give a misleading too low a value.

Lacking this instrmentation, just crank the volume up until the sound out is really loud!
John.

Neil Purling · 11th Jul 2006, 5:55 pm

Lacking this instrmentation, just crank the volume up until the sound out is really loud!.
That is why I want to add the extra triode stage.
Where ought the volume control to work?
1st or 2nd half of the pre-amp?

JHGibson · 11th Jul 2006, 11:03 pm

It is better to have the volume control on the input of the first amp to prevent a really strong signal overloading this stage.

paulsherwin · 12th Jul 2006, 12:45 am

Quote:

Originally Posted by JHGibson

It is better to have the volume control on the input of the first amp to prevent a really strong signal overloading this stage.

This is how it's normally done, but it does mean any noise in the audio preamp goes straight to the output valve.

You should be able to steal all the info you need for the two stage audio amp from a typical radio circuit of the 40s and 50s.

The 6V6 isn't a high slope valve and you will get more output if you sub a European type like an EL33. Unfortunately these are now fairly rare and cost a lot more than a 6V6

Best regards, Paul

Neil Purling · 12th Jul 2006, 8:08 am

At the moment I believe that I am not driving the existing 6V6f ully. I think that four watts is quite enough for a radio. To fit a EL33 or whatever surely i'd need another o/p transformer. I think the present RS item is only rated up to 4W.

PJL · 12th Jul 2006, 6:49 pm

Can someone help me out here as I could be wrong but it strikes me that alot of TRF's work OK without the AF stage so I have a theory!

The 6J5 seems to be incorrectly biased. The characteritics says grid should typically be -8V which would imply 2mA anode current from the cathode bias but 2mA across 220K = 440V so it is operating in that not too happy saturated region below the characteristic curve which could explain why the gain is low.

You can check the theory by looking at the anode volts which should be 100-150V to get a reasonable output. If it is low then you would need to increase the 3.9K bias resistor (and decoupler) to get it into a sensible working region.

I am no expert but I am not sure the choice of valve was right for a high gain AF stage.

PJL · 12th Jul 2006, 7:08 pm

On second thoughts I think you probably won't need to change the decoupler as the cutoff is related to the anode load not the bias resistance?

Neil Purling · 12th Jul 2006, 8:20 pm

I get between 125K and 150K depending on whether the value of the HT is 250V or nearer 300, for 2mA current flow.
I copied the design of the amplifier circuit from my 1952 edition of Radio for Boys. I believe Hunts Smoothing Bomb did the same.
The AF stage of both our sets is from the 'AC Mains Superhet'. This uses a EBC33 Double-Diode Triode for AVC, detector and audio pre-amp driving a 6V6. Me and H.S.B both substituted a 6J5 for the triode part of the EBC33.
Any further opinions on whether the 6J5 is biased correctly or not, according to the diagram with this thread?
If not, then what is appropriate?

PJL · 12th Jul 2006, 9:32 pm

Neil, I did say I could be wrong! I found a more comprehensive datasheet http://www.mif.pg.gda.pl/homepages/f.../093/6/6J5.pdf which suggests the 3.9K is a little low (4.7K or perhaps 5.6K depending on HT) not as much as I had thought - it will be running at <1mA anode current. As mentioned above the gain is 15 whilst the EBC33 acheives 23.

Neil Purling · 13th Jul 2006, 7:42 am

Well, that's quite a significant difference in gain.
While valve data sheets will quote the gain of the 6J5 or each section of the 6SN7GT as having a gain of 20, as a RC coupled class A1 amplifier it manages only 15 on average. I have looked at a resource of valve data sheets and I agree about the value of cathode resistor. It still needs changing, I wonder what sort of difference would be observable?

It justifies the usage of the 6SN7GT, with the combined gain of 30. Perhaps a twin gang volume control working on both grid circuits would be a good idea, if one could get such a thing.

Hunts smoothing bomb · 13th Jul 2006, 1:40 pm

Hi Neil,

You have used the electron coupled circuit, I have not tried this yet, my reciever uses the TRF design in the "One valve battery reciever" Radio For Boys. Could the EC circuit have a lower output as my set has stacks of volume.

Regards
Lee

Neil Purling · 13th Jul 2006, 6:35 pm

I was wondering if I could squeeze some more volume out of the circuit.
The design we both borrowed utilised a EBC33 triode part, and that offers more gain than the 6J5 we used. The volume does not increase after around 60% of the volume control travel.

You used a different regenerative detector. I do not know at present how much gain the EF39/6K7 RF Pentode gives, or if a EF91 would do any better.
I suppose there is only one way to find out.........
The only real way to find out would be to use a RF signal gen & scope in a bench test. I have neither item.

paulsherwin · 13th Jul 2006, 7:12 pm

Quote:

Originally Posted by Neil Purling

At the moment I believe that I am not driving the existing 6V6f ully. I think that four watts is quite enough for a radio. To fit a EL33 or whatever surely i'd need another o/p transformer. I think the present RS item is only rated up to 4W.

Neil, you are mixing up slope and power ratings. You won't need more than a couple of watts from a radio like this whatever output valve you use. The power at which the valve operates is determined by the bias, which is set by the cathode resistor - the bigger the resistor, the less power the valve will dissipate and the less audio output will be available if the valve is fully driven.

It is very likely you're not driving the output valve adequately at present. An EL33 is a higher slope valve than a 6V6, meaning it'll produce more output for the same input (simplifying a bit). It's got more gain. Many short superhets of the 40s and 50s didn't have an audio preamp stage and relied on a high slope output valve to produce enough output. A 6V6 type valve wouldn't have produced enough output in these designs because the slope would have been too low.

The same output transformer would be OK with an EL33 so long as the cathode bias resistor was adjusted to limit the valve current.

Best regards, Paul

Neil Purling · 13th Jul 2006, 8:53 pm

Well, I will stick with the 6V6 because whatever it's defeciencies they are very easiliy available. No I am not going to get into the audiofoolery thing:"Mine's a yellow print Mullard, so it must be the best".
What re-adjustment of the bias might get the best out of that output valve?
Then with the change of pre-amp valve this creature is going to have some balls.
Still, I can't see myself walking down the street with the chassis on one shoulder & the speaker on the other.
The simple reason being i'd need one of those minature generators in a rucksack and my sciatica would most definitely protest.

10th Jul 2006, 1:57 pm	#1
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Regarding home made TRF I made my radio with a mixture of circuits: The 'Electron Coupled' design from member heatercathodeshort, the audio stage adapted from Radio for Boys by E.N. Bradley (1952). The audio stage is a 6J5 Triode feeding a 6V6. I get the impression that it could be louder, as I seem to have some travel left on the volume control. Now I was interested to substitute a 6SN7GT double triode and have the first section give a fixed gain of a factor of two or three. I have no idea of what sort of signal level the detector will give when it is on the point of oscillation. Ideally I want the strong local signal to just drive the 6V6 just into distortion as the detector breaks into oscillation, assuming the volume control is at ful. If Isoplethics ever come out with a range of plug in coils beyond the basic Long & Medium Wave ones that is going to be the cue for yet another makeover. If I adapt the radio to cover Short Waves I would change the detector to a EF91 if that would be better to cover the Short Wave broadcast bands. Well, that's summat for the longer evenings.... Attached Thumbnails Attached Images

10th Jul 2006, 4:55 pm	#2
JHGibson Retired Dormant Member Join Date: Apr 2003 Location: Salt Spring Island, BC, Canada Posts: 368	Re: Regarding home made TRF Neil, It would require 2 stages of 6J5 RC coupled audio amps, which have a gain of 15 each according to the gain tables in the back of my RCA Tube Manual, to fully drive a 6V6 from a regenerative detector. Or use the two halves of a 6SN7 in place of two 6J5's. Don't forget to audio decouple the plate circuits of the audio amps or you may end up with motorboating oscillations. John.

10th Jul 2006, 5:26 pm	#3
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Re: Regarding home made TRF Does your valve manual say anything about the sort of input level the 6V6 requires to be fully driven? I am not sure what sort of output the detector gives but I guess at between 40-80mV, probably nearer 40mV.

11th Jul 2006, 12:56 am	#4
JHGibson Retired Dormant Member Join Date: Apr 2003 Location: Salt Spring Island, BC, Canada Posts: 368	Re: Regarding home made TRF With a plate voltage of 250, the 6V6 requires 12.5 V RMS drive for max output of 4.5 watts at 8 percent distortion. John.

11th Jul 2006, 10:06 am	#5
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Re: Regarding home made TRF The signal level required by the 6V6: 12 volts p-p, or dc measurement with a AVO? I would like to boost the audio somewhat if I can. Have any of those members who've built TRF radios been able to follow the signal through the set with a 'scope? I need to know what sort of signal level a EF39/6K7 or a EF91 would give used as a detector. Then I know what sort of additional boost I need to arrange with the extra triode of a 6SN7GT.

11th Jul 2006, 3:57 pm	#6
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Re: Regarding home made TRF You can refer to my diagram I drew of the AF stage of my TRF radio from my copy of Radio for Boys. If I do fit a 6SN7G Double Triode. I was going to use a preset pot of 1Meg value on one stage and have the normal volume control on the other section. Which triode should have the official vol control &which the preset? Which would be best?

11th Jul 2006, 4:04 pm	#7
JHGibson Retired Dormant Member Join Date: Apr 2003 Location: Salt Spring Island, BC, Canada Posts: 368	Re: Regarding home made TRF The drive to the 6V6 would really have to be measured by a vacuum valve voltmeter set to AC RMS, or with an oscilloscope. The impedance level at the control grid is high and a AC reading voltmeter like an AVO would load the cicuit badly and give a misleading too low a value. Lacking this instrmentation, just crank the volume up until the sound out is really loud! John.

11th Jul 2006, 5:55 pm	#8
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Re: Regarding home made TRF Lacking this instrmentation, just crank the volume up until the sound out is really loud!. That is why I want to add the extra triode stage. Where ought the volume control to work? 1st or 2nd half of the pre-amp?

11th Jul 2006, 11:03 pm	#9
JHGibson Retired Dormant Member Join Date: Apr 2003 Location: Salt Spring Island, BC, Canada Posts: 368	Re: Regarding home made TRF It is better to have the volume control on the input of the first amp to prevent a really strong signal overloading this stage.

12th Jul 2006, 8:08 am	#11
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Re: Regarding home made TRF At the moment I believe that I am not driving the existing 6V6f ully. I think that four watts is quite enough for a radio. To fit a EL33 or whatever surely i'd need another o/p transformer. I think the present RS item is only rated up to 4W.

12th Jul 2006, 6:49 pm	#12
PJL Dekatron Join Date: Sep 2005 Location: Seaford, East Sussex, UK. Posts: 5,997	Re: Regarding home made TRF Can someone help me out here as I could be wrong but it strikes me that alot of TRF's work OK without the AF stage so I have a theory! The 6J5 seems to be incorrectly biased. The characteritics says grid should typically be -8V which would imply 2mA anode current from the cathode bias but 2mA across 220K = 440V so it is operating in that not too happy saturated region below the characteristic curve which could explain why the gain is low. You can check the theory by looking at the anode volts which should be 100-150V to get a reasonable output. If it is low then you would need to increase the 3.9K bias resistor (and decoupler) to get it into a sensible working region. I am no expert but I am not sure the choice of valve was right for a high gain AF stage.

12th Jul 2006, 7:08 pm	#13
PJL Dekatron Join Date: Sep 2005 Location: Seaford, East Sussex, UK. Posts: 5,997	Re: Regarding home made TRF On second thoughts I think you probably won't need to change the decoupler as the cutoff is related to the anode load not the bias resistance?

12th Jul 2006, 8:20 pm	#14
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Re: Regarding home made TRF I get between 125K and 150K depending on whether the value of the HT is 250V or nearer 300, for 2mA current flow. I copied the design of the amplifier circuit from my 1952 edition of Radio for Boys. I believe Hunts Smoothing Bomb did the same. The AF stage of both our sets is from the 'AC Mains Superhet'. This uses a EBC33 Double-Diode Triode for AVC, detector and audio pre-amp driving a 6V6. Me and H.S.B both substituted a 6J5 for the triode part of the EBC33. Any further opinions on whether the 6J5 is biased correctly or not, according to the diagram with this thread? If not, then what is appropriate?

12th Jul 2006, 9:32 pm	#15
PJL Dekatron Join Date: Sep 2005 Location: Seaford, East Sussex, UK. Posts: 5,997	Re: Regarding home made TRF Neil, I did say I could be wrong! I found a more comprehensive datasheet http://www.mif.pg.gda.pl/homepages/f.../093/6/6J5.pdf which suggests the 3.9K is a little low (4.7K or perhaps 5.6K depending on HT) not as much as I had thought - it will be running at <1mA anode current. As mentioned above the gain is 15 whilst the EBC33 acheives 23.

13th Jul 2006, 7:42 am	#16
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Re: Regarding home made TRF Well, that's quite a significant difference in gain. While valve data sheets will quote the gain of the 6J5 or each section of the 6SN7GT as having a gain of 20, as a RC coupled class A1 amplifier it manages only 15 on average. I have looked at a resource of valve data sheets and I agree about the value of cathode resistor. It still needs changing, I wonder what sort of difference would be observable? It justifies the usage of the 6SN7GT, with the combined gain of 30. Perhaps a twin gang volume control working on both grid circuits would be a good idea, if one could get such a thing.

13th Jul 2006, 1:40 pm	#17
Hunts smoothing bomb Octode Join Date: Oct 2005 Location: Wimborne, Dorset, UK. Posts: 1,407	Re: Regarding home made TRF Hi Neil, You have used the electron coupled circuit, I have not tried this yet, my reciever uses the TRF design in the "One valve battery reciever" Radio For Boys. Could the EC circuit have a lower output as my set has stacks of volume. Regards Lee __________________ Lee

13th Jul 2006, 6:35 pm	#18
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Re: Regarding home made TRF I was wondering if I could squeeze some more volume out of the circuit. The design we both borrowed utilised a EBC33 triode part, and that offers more gain than the 6J5 we used. The volume does not increase after around 60% of the volume control travel. You used a different regenerative detector. I do not know at present how much gain the EF39/6K7 RF Pentode gives, or if a EF91 would do any better. I suppose there is only one way to find out......... The only real way to find out would be to use a RF signal gen & scope in a bench test. I have neither item.

13th Jul 2006, 8:53 pm	#20
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,086	Re: Regarding home made TRF Well, I will stick with the 6V6 because whatever it's defeciencies they are very easiliy available. No I am not going to get into the audiofoolery thing:"Mine's a yellow print Mullard, so it must be the best". What re-adjustment of the bias might get the best out of that output valve? Then with the change of pre-amp valve this creature is going to have some balls. Still, I can't see myself walking down the street with the chassis on one shoulder & the speaker on the other. The simple reason being i'd need one of those minature generators in a rucksack and my sciatica would most definitely protest.