Home |
Search |
Today's Posts |
#1
Posted to rec.audio.tubes
|
|||
|
|||
Sun to set on valve era, 'Twin Triode' AM Transmiter Nexus
After reading the recent "Sun to set on valve era" post to this group, I pursued
the link that was provided. http://www.whathifi.com/news/end-in-...io-4-long-wave That link lead me to do a little Googling from which I learned that the original Droitwich Long Wave transmitter used "series modulation" just like Flipper's little "'Twin Triode' AM Transmiter" that was extensively discussed in this group recently. I am curious if anyone knows of any other broadcast transmitters that used the series modulation system, besides the original Droitwich LW transmitter? I know that the superficially similar cathode modulation scheme was used in some Television Broadcast transmitters, and Ham radio transmitters. While cathode modulation is superficially similar to series modulation it is a completely different system of modulation. An interesting article from back in the day that discusses series modulation in considerable detail can be found here. http://ieeexplore.ieee.org/xpl/freea...number=1685887 -- Regards, John Byrns Surf my web pages at, http://fmamradios.com/ |
#2
|
|||
|
|||
Quote:
http://www.amwindow.org/tech/htm/wb9eckseriesmod.htm Cheers, John S |
#3
Posted to rec.audio.tubes
|
|||
|
|||
Sun to set on valve era, 'Twin Triode' AM Transmiter Nexus
In article ,
John L Stewart wrote: John Byrns;941268 Wrote: After reading the recent "Sun to set on valve era" post to this group, I pursued the link that was provided. http://www.whathifi.com/news/end-in-...io-4-long-wave That link lead me to do a little Googling from which I learned that the original Droitwich Long Wave transmitter used "series modulation" just like Flipper's little "'Twin Triode' AM Transmiter" that was extensively discussed in this group recently. I am curious if anyone knows of any other broadcast transmitters that used the series modulation system, besides the original Droitwich LW transmitter? I know that the superficially similar cathode modulation scheme was used in some Television Broadcast transmitters, and Ham radio transmitters. While cathode modulation is superficially similar to series modulation it is a completely different system of modulation. An interesting article from back in the day that discusses series modulation in considerable detail can be found here. http://ieeexplore.ieee.org/xpl/freea...number=1685887 See an example here- http://www.amwindow.org/tech/htm/wb9eckseriesmod.htm I really like this transmitter, it's the best home brew ham radio transmitter I have ever seen, at least for my taste, I give it a 9.5 on the 0-10 scale. The window on the front that allows the modulator tubes to be seen is a neat, although possibly semi useless feature. The schematic obviously only covers the basic circuit idea, and doesn't represent the transmitter as built, there are a lot of control switches that can be seen in the pictures that don't appear in the schematic. Most interestingly there are what appear to be about five relays on the underside of the chassis, I wonder what the function of these relays are, and if they are even relays, or are something else? If they are some kind of fancy power sequencing and protective system like in a real transmitter, I would think most of them would be on the power supply chassis, which isn't pictured. Anyone have any thoughts on these relays? My only quibbles with the design are that no efforts seem to have been taken to minimize the causes of slewing induced distortion at high audio frequencies, and it would be nice to have some feedback around the modulator tubes to linearize the tube characteristic, although as Flipper has pointed out there are problems implementing feedback around the modulator in this circuit. I'm still looking for examples of broadcast transmitters, besides the 1934 Droitwich Long Wave transmitter, that used series modulation? Or was series modulation a flash in the pan that was only used in the Droitwich LW transmitter? -- Regards, John Byrns Surf my web pages at, http://fmamradios.com/ |
#4
|
|||
|
|||
Quote:
The input triode is showing 10k on its cathode, probably is 1K. The 6LF6's, only 40V on their G2. Couldn'd find any 6LF6 specs showing G2 vs Plate max current. The sort of similar 6DQ5 manages 100 mA with only 50V on G2. The 6LF6 has a smaller cathode so I guess less current available. But he uses two of them so looks like 100% mod is possible. At the same time the 807 plate current is limited by this hookup should there be a problem. If the 807 is run ICAS there could be a little more than 100 mA plate current. All at 600 or so volts. That leaves the other 600 volts on the 6LF6's, dissipating 60 watts. The spec sheet sez each good for 40 watts. NFB is possible from the 6LF6 plates thru a HV Cap & suitable R to the input cathode R. Depends on how much FB needed. The cathode bypass cap would not be required. The result simply forms a feedback pair. My thoughts anyway. Cheers, John |
#5
Posted to rec.audio.tubes
|
|||
|
|||
Sun to set on valve era, 'Twin Triode' AM Transmiter Nexus
In article ,
John L Stewart wrote: John Byrns;941347 Wrote: In article , John L Stewart wrote: - John Byrns;941268 Wrote: - After reading the recent "Sun to set on valve era" post to this group, I pursued the link that was provided. http://www.whathifi.com/news/end-in-...io-4-long-wave That link lead me to do a little Googling from which I learned that the original Droitwich Long Wave transmitter used "series modulation" just like Flipper's little "'Twin Triode' AM Transmiter" that was extensively discussed in this group recently. I am curious if anyone knows of any other broadcast transmitters that used the series modulation system, besides the original Droitwich LW transmitter? I know that the superficially similar cathode modulation scheme was used in some Television Broadcast transmitters, and Ham radio transmitters. While cathode modulation is superficially similar to series modulation it is a completely different system of modulation. An interesting article from back in the day that discusses series modulation in considerable detail can be found here. http://ieeexplore.ieee.org/xpl/freea...umber=1685887- See an example here- http://www.amwindow.org/tech/htm/wb9eckseriesmod.htm- I really like this transmitter, it's the best home brew ham radio transmitter I have ever seen, at least for my taste, I give it a 9.5 on the 0-10 scale. The window on the front that allows the modulator tubes to be seen is a neat, although possibly semi useless feature. The schematic obviously only covers the basic circuit idea, and doesn't represent the transmitter as built, there are a lot of control switches that can be seen in the pictures that don't appear in the schematic. Most interestingly there are what appear to be about five relays on the underside of the chassis, I wonder what the function of these relays are, and if they are even relays, or are something else? If they are some kind of fancy power sequencing and protective system like in a real transmitter, I would think most of them would be on the power supply chassis, which isn't pictured. Anyone have any thoughts on these relays? My only quibbles with the design are that no efforts seem to have been taken to minimize the causes of slewing induced distortion at high audio frequencies, and it would be nice to have some feedback around the modulator tubes to linearize the tube characteristic, although as Flipper has pointed out there are problems implementing feedback around the modulator in this circuit. I'm still looking for examples of broadcast transmitters, besides the 1934 Droitwich Long Wave transmitter, that used series modulation? Or was series modulation a flash in the pan that was only used in the Droitwich LW transmitter? I see a couple of oddities in the schematic, perhaps nothing more than typos. The input triode is showing 10k on its cathode, probably is 1K. I have seen 12AU7s with cathode resistors as high as 8.2k, however in this application a high value cathode resistor doesn't seem correct, I hope that you are right and that it was only a ³typo². The 6LF6's, only 40V on their G2. Couldn'd find any 6LF6 specs showing G2 vs Plate max current. The sort of similar 6DQ5 manages 100 mA with only 50V on G2. The 6LF6 has a smaller cathode so I guess less current available. But he uses two of them so looks like 100% mod is possible. At the same time the 807 plate current is limited by this hookup should there be a problem. I'm not following your point about cathode current? Even an 807, which presumably has a much smaller cathode than the 6LF6, cold serve as a modulator in this circuit if cathode current were the only issue. I believe that the reason for the paralleled 6lF6s has nothing to do with cathode current, but rather with the fact that these pure class A modulation schemes, both the series modulation scheme as well as the traditional Heising parallel modulation scheme, require that the modulator tubes have a dissipation rating of about 4 times the rating of the modulated RF tube. For example look at the Droitwich Long Wave transmitter ³Abridged Technical Data² where the ratio between the modulator dissipation under carrier conditions, and the modulated RF amplifier dissipation, is something on the order of X5.3. I assume that the choice of 40 volts for the screen voltage was to optimize the operating point of the modulator tubes, so that the control grid voltage is approaching zero volts at peak positive modulation. If the 807 is run ICAS there could be a little more than 100 mA plate current. All at 600 or so volts. That leaves the other 600 volts on the 6LF6's, dissipating 60 watts. The spec sheet sez each good for 40 watts. I doubt that the 807 plate current is as much as 100 mA. It is difficult to guesstimate what the actual operating conditions of the 807 might be in this circuit, as the information given on the web page is very sketchy. The author calls it a 25 watt transmitter, but is this the DC input to the final RF amplifier, the gross RF output power as a tube manual would specify, or is it the net RF output power as a broadcast transmitter would be rated. For example the 807 data sheet shows an operating point, for CCS, with a plate voltage of 475 volts, plate current of 83 mA, and a gross RF power output of 27.5 watts. The author also says that if the transmitter is run at ³15 watts², somewhere around 115% peak positive modulation can be reached. This suggests that at 25 watts this transmitter is capable of nowhere near 100% modulation, perhaps maxing out at something like 67% modulation on positive peaks. NFB is possible from the 6LF6 plates thru a HV Cap & suitable R to the input cathode R. Depends on how much FB needed. The cathode bypass cap would not be required. The result simply forms a feedback pair. As Flipper has pointed out, negative feedback from the series modulator plate decreases the negative modulation capability of the circuit, presumably this holds even when a series capacitor is used, although to a slightly lesser extent. After thinking about it a bit, negative feedback from an RF rectifier, as in the typical Doherty AM broadcast transmitter, would probably be the best way to apply negative audio feedback to this transmitter circuit. Anyway, just a few more thoughts on the subject. -- Regards, John Byrns Surf my web pages at, http://fmamradios.com/ |
#6
|
|||
|
|||
I'm not following your point about cathode current? Even an 807, which
presumably has a much smaller cathode than the 6LF6, cold serve as a modulator in this circuit if cathode current were the only issue. I believe that the reason for the paralleled 6lF6s has nothing to do with cathode current, but rather with the fact that these pure class A modulation schemes, both the series modulation scheme as well as the traditional Heising parallel modulation scheme, require that the modulator tubes have a dissipation rating of about 4 times the rating of the modulated RF tube. For example look at the Droitwich Long Wave transmitter ³Abridged Technical Data² where the ratio between the modulator dissipation under carrier conditions, and the modulated RF amplifier dissipation, is something on the order of X5.3. I assume that the choice of 40 volts for the screen voltage was to optimize the operating point of the modulator tubes, so that the control grid voltage is approaching zero volts at peak positive modulation. If the 807 is run ICAS there could be a little more than 100 mA plate current. All at 600 or so volts. That leaves the other 600 volts on the 6LF6's, dissipating 60 watts. The spec sheet sez each good for 40 watts. I doubt that the 807 plate current is as much as 100 mA. It is difficult to guesstimate what the actual operating conditions of the 807 might be in this circuit, as the information given on the web page is very sketchy. The author calls it a 25 watt transmitter, but is this the DC input to the final RF amplifier, the gross RF output power as a tube manual would specify, or is it the net RF output power as a broadcast transmitter would be rated. For example the 807 data sheet shows an operating point, for CCS, with a plate voltage of 475 volts, plate current of 83 mA, and a gross RF power output of 27.5 watts. The author also says that if the transmitter is run at ³15 watts², somewhere around 115% peak positive modulation can be reached. This suggests that at 25 watts this transmitter is capable of nowhere near 100% modulation, perhaps maxing out at something like 67% modulation on positive peaks. NFB is possible from the 6LF6 plates thru a HV Cap & suitable R to the input cathode R. Depends on how much FB needed. The cathode bypass cap would not be required. The result simply forms a feedback pair. As Flipper has pointed out, negative feedback from the series modulator plate decreases the negative modulation capability of the circuit, presumably this holds even when a series capacitor is used, although to a slightly lesser extent. After thinking about it a bit, negative feedback from an RF rectifier, as in the typical Doherty AM broadcast transmitter, would probably be the best way to apply negative audio feedback to this transmitter circuit. Anyway, just a few more thoughts on the subject. -- Regards, John Byrns Surf my web pages at, http://fmamradios.com/[/quote] The modulator in this case is single ended & thus must run Class A where efficiency is none too good. The total HV appears to be 1200 volts so I would guess 600 for each of the modulator & RF Final. Look here for a good summary of 807 specs- http://frank.pocnet.net/sheets/049/8/807.pdf If we take the CCS numbers at 600 volts the plate current looks like 100 mA & RF output is 40 Watts. 807 plate dissipation is then 20 watts. That means we will need 20 Watts + of audio to fully modulate, not possible for a SE 807. Thus we see 6LF6's & their fat cathodes paralleled to get the required audio power. Just 40 volts on the screens gets enough current & I think prevents large fault currents in the 807. To the 6LF6's the RF Final appears as a load of (600 / 0.1) R (6K). If we had the 6LF6 plate vs G2 curves we could determine if that is optimum for the conditions set. There are quite a few possibilities for a 300 volt version of this thing for those who would like a low power experimental circuit. Cheers, John |
#7
|
|||
|
|||
Quote:
http://frank.pocnet.net/sheets/049/8/807.pdf If we take the CCS numbers at 600 volts the plate current looks like 100 mA & RF output is 40 Watts. 807 plate dissipation is then 20 watts. That means we will need 20 Watts + of audio to fully modulate, not possible for a SE 807. Thus we see 6LF6's & their fat cathodes paralleled to get the required audio power. Just 40 volts on the screens gets enough current & I think prevents large fault currents in the 807. To the 6LF6's the RF Final appears as a load of (600 / 0.1) R (6K). If we had the 6LF6 plate vs G2 curves we could determine if that is optimum for the conditions set. There are quite a few possibilities for a 300 volt version of this thing for those who would like a low power experimental circuit. Cheers, John[/quote] The modulator power required as I had indicated earlier will fall short of 100% modulation. I had ignored the 8 mA G2 current for the 807. As well, the RF Final does not have 100 % efficiency in converting the modulator power into AM sidebands, so more audio is necessary. Modulator Audio Power needs to be 1/2 the DC input to the RF Final. That would be (600 X 0.108)* (1/2) or 32.4 Watts. Hope hat makes sense. John |
#8
Posted to rec.audio.tubes
|
|||
|
|||
Sun to set on valve era, 'Twin Triode' AM Transmiter Nexus
In article ,
John L Stewart wrote: John L Stewart;941765 Wrote: I'm not following your point about cathode current? Even an 807, which presumably has a much smaller cathode than the 6LF6, cold serve as a modulator in this circuit if cathode current were the only issue. I believe that the reason for the paralleled 6lF6s has nothing to do with cathode current, but rather with the fact that these pure class A modulation schemes, both the series modulation scheme as well as the traditional Heising parallel modulation scheme, require that the modulator tubes have a dissipation rating of about 4 times the rating of the modulated RF tube. For example look at the Droitwich Long Wave transmitter ³Abridged Technical Data² where the ratio between the modulator dissipation under carrier conditions, and the modulated RF amplifier dissipation, is something on the order of X5.3. I assume that the choice of 40 volts for the screen voltage was to optimize the operating point of the modulator tubes, so that the control grid voltage is approaching zero volts at peak positive modulation. - If the 807 is run ICAS there could be a little more than 100 mA plate current. All at 600 or so volts. That leaves the other 600 volts on the 6LF6's, dissipating 60 watts. The spec sheet sez each good for 40 watts.- I doubt that the 807 plate current is as much as 100 mA. It is difficult to guesstimate what the actual operating conditions of the 807 might be in this circuit, as the information given on the web page is very sketchy. The author calls it a 25 watt transmitter, but is this the DC input to the final RF amplifier, the gross RF output power as a tube manual would specify, or is it the net RF output power as a broadcast transmitter would be rated. For example the 807 data sheet shows an operating point, for CCS, with a plate voltage of 475 volts, plate current of 83 mA, and a gross RF power output of 27.5 watts. The author also says that if the transmitter is run at ³15 watts², somewhere around 115% peak positive modulation can be reached. This suggests that at 25 watts this transmitter is capable of nowhere near 100% modulation, perhaps maxing out at something like 67% modulation on positive peaks. - NFB is possible from the 6LF6 plates thru a HV Cap & suitable R to the input cathode R. Depends on how much FB needed. The cathode bypass cap would not be required. The result simply forms a feedback pair.- As Flipper has pointed out, negative feedback from the series modulator plate decreases the negative modulation capability of the circuit, presumably this holds even when a series capacitor is used, although to a slightly lesser extent. After thinking about it a bit, negative feedback from an RF rectifier, as in the typical Doherty AM broadcast transmitter, would probably be the best way to apply negative audio feedback to this transmitter circuit. Anyway, just a few more thoughts on the subject. The modulator in this case is single ended & thus must run Class A where efficiency is none too good. The total HV appears to be 1200 volts so I would guess 600 for each of the modulator & RF Final. The schematic shows the total HV to be 1,100 volts, unfortunately it doesn't indicate what the voltage is at the modulator plate. Look here for a good summary of 807 specs- http://frank.pocnet.net/sheets/049/8/807.pdf If we take the CCS numbers at 600 volts the plate current looks like 100 mA & RF output is 40 Watts. 807 plate dissipation is then 20 watts. The datasheet you linked to above says that the maximum CCS rating for the 807 operating as a plate modulated Class C RF power amplifier is 475 Volts and 83 mA. That means we will need 20 Watts + of audio to fully modulate, not possible for a SE 807. Thus we see 6LF6's & their fat cathodes paralleled to get the required audio power. Just 40 volts on the screens gets enough current & I think prevents large fault currents in the 807. I understand the need to parallel two 6LF6s to get enough plate dissipation capability to generate the required class A audio power, that requires multiple big fat anodes. I still don't get your point about the 6LF6's "fat cathodes"? To the 6LF6's the RF Final appears as a load of (600 / 0.1) R (6K). If we had the 6LF6 plate vs G2 curves we could determine if that is optimum for the conditions set. Quite right. There are quite a few possibilities for a 300 volt version of this thing for those who would like a low power experimental circuit. Cheers, John The modulator power required as I had indicated earlier will fall short of 100% modulation. I had ignored the 8 mA G2 current for the 807. As well, the RF Final does not have 100 % efficiency in converting the modulator power into AM sidebands, so more audio is necessary. Modulator Audio Power needs to be 1/2 the DC input to the RF Final. That would be (600 X 0.108)* (1/2) or 32.4 Watts. Hope hat makes sense. John Yes, neglecting the "fat cathode" stuff, it makes sense, except that splitting the total high voltage 50-50 across the modulated RF amplifier and the modulator will not provide for 100% positive modulation peaks, most of these designs end up with a considerably higher percentage of the total HV across the modulator than across the RF amplifier to achieve decent positive modulation, while accounting for the finite non zero saturation voltage of the modulator. -- Regards, John Byrns Surf my web pages at, http://fmamradios.com/ |
#9
|
|||
|
|||
The modulator in this case is single ended & thus must run Class A where
efficiency is none too good. The total HV appears to be 1200 volts so I would guess 600 for each of the modulator & RF Final.[/i][/color] The schematic shows the total HV to be 1,100 volts, unfortunately it doesn't indicate what the voltage is at the modulator plate. My Schema copy appears to be 1200 volts but in any case not much to change the basic idea. Would be nice if he showed the 6LF6 plate volts. Look here for a good summary of 807 specs- http://frank.pocnet.net/sheets/049/8/807.pdf If we take the CCS numbers at 600 volts the plate current looks like 100 mA & RF output is 40 Watts. 807 plate dissipation is then 20 watts. The datasheet you linked to above says that the maximum CCS rating for the 807 operating as a plate modulated Class C RF power amplifier is 475 Volts and 83 mA. Yes, in my rush to get out to the bicycle I overlooked that. Tubes not a high priority with me these daze. The bicycle is just short of 2500 miles since first of May this past Spring. Then there is the Fall cleanup underway now, rather extensive on a property this size. That means we will need 20 Watts + of audio to fully modulate, not possible for a SE 807. Thus we see 6LF6's & their fat cathodes paralleled to get the required audio power. Just 40 volts on the screens gets enough current & I think prevents large fault currents in the 807. I understand the need to parallel two 6LF6s to get enough plate dissipation capability to generate the required class A audio power, that requires multiple big fat anodes. I still don't get your point about the 6LF6's "fat cathodes"? Back in the day we used a lot of tubes with 'fat cathodes'. Nothing more than a large emitting surface, read High Perveance. Enables large currents while using low plate or screen voltages as we see in most all of TV's horizontal sweep tubes. I managed to coax 1.5 Amp pulses into 50 ohms from 6DQ5's while working on a core memory research project around 1960. To the 6LF6's the RF Final appears as a load of (600 / 0.1) R (6K). If we had the 6LF6 plate vs G2 curves we could determine if that is optimum for the conditions set. Quite right. There are quite a few possibilities for a 300 volt version of this thing for those who would like a low power experimental circuit. Cheers, John The modulator power required as I had indicated earlier will fall short of 100% modulation. I had ignored the 8 mA G2 current for the 807. As well, the RF Final does not have 100 % efficiency in converting the modulator power into AM sidebands, so more audio is necessary. Modulator Audio Power needs to be 1/2 the DC input to the RF Final. That would be (600 X 0.108)* (1/2) or 32.4 Watts. Hope hat makes sense. John Yes, neglecting the "fat cathode" stuff, it makes sense, except that splitting the total high voltage 50-50 across the modulated RF amplifier and the modulator will not provide for 100% positive modulation peaks, most of these designs end up with a considerably higher percentage of the total HV across the modulator than across the RF amplifier to achieve decent positive modulation, while accounting for the finite non zero saturation voltage of the modulator. I had thought there a need to somehow get more audio at the expense of volts on the RF Final but did not pursue that further. NTL, I think a low voltage version possible, say 250-300 volts total. The extra heater transformer still a problem but a lot less $$$ than a modulation transformer otherwise needed. And with a bit of care looks like HIFI is possible as noted by one of the posters over at that link. For now I will leave it to others to suggest a lineup. I've got my own in mind. Now I gotta get on the tractor. Cheers, John -- Regards, John Byrns Surf my web pages at, http://fmamradios.com/[/quote] |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
'Twin Triode' AM Transmiter | Vacuum Tubes | |||
'Twin Triode' AM Transmitter | Vacuum Tubes | |||
CV5843 twin triode? | Vacuum Tubes |