Home |
Search |
Today's Posts |
#1
|
|||
|
|||
Improving Imaging-12AX7
Hi
In Marantz M7 preamp the L and R signal go to a single output tube. Will using separate tube for L and R improve imaging? |
#2
|
|||
|
|||
A S Goh wrote: Hi In Marantz M7 preamp the L and R signal go to a single output tube. Will using separate tube for L and R improve imaging? Probably not beacuse the channel separation is over 40 dB up to 10 kHz perhaps. Above that stray C starts to let the channels talk to each other. I try to always have separate tubes for each channel. ARC SP11 preamp had 3 6DJ8 for the phono amp, so six triodes for 2 chanels. the channels shared tubes in each of 3 stages. I know, I just repaired one. Probably this is better than having cascaded preamp stages of one channel in a tube since interference is more likely. I like to have one whole twin triode tube with both its triodes devoted to only one amp stage. Patrick Turner. |
#3
|
|||
|
|||
If this gentleman has some experience with the welding iron, and if he's got
a 6.3V heater supply available, he can try changing 12AX7s with Russian 6N2p or NOS 6EU7s: these have a screening plate among the anodes of the two triodes, just like ECC88s, connected to the pin left free by the missing 12.6V heater connection. All remaining electric parameters are so similar to the 12AX7 that I suppose they were in fact the same tube with minor changes. Grounding that screen should reduce the parasitic capacitance among the two triodes (hence the "crosstalk) to levels well below the audible threshold, and it's an easy and cheap fix: 3 connections and a single 5$ tube to be swapped. Nevertheless I'd bet a leg, if I had one spare, that nobody will ever hear the difference... with headphones, maybe... not with loudspeakers. Ciao Fabio "Patrick Turner" ha scritto nel messaggio ... A S Goh wrote: Hi In Marantz M7 preamp the L and R signal go to a single output tube. Will using separate tube for L and R improve imaging? Probably not beacuse the channel separation is over 40 dB up to 10 kHz perhaps. Above that stray C starts to let the channels talk to each other. I try to always have separate tubes for each channel. ARC SP11 preamp had 3 6DJ8 for the phono amp, so six triodes for 2 chanels. the channels shared tubes in each of 3 stages. I know, I just repaired one. Probably this is better than having cascaded preamp stages of one channel in a tube since interference is more likely. I like to have one whole twin triode tube with both its triodes devoted to only one amp stage. Patrick Turner. |
#4
|
|||
|
|||
Thanks for the idea.
Russian tube may not be easy to get ECC88 may be able to find . Theoretically, separate tube can improve channel separation and therefor imaging, if I am not wrong What if I use separate tube and parallel the two triode for the one channel output? "Fabio Berutti" wrote in message ... If this gentleman has some experience with the welding iron, and if he's got a 6.3V heater supply available, he can try changing 12AX7s with Russian 6N2p or NOS 6EU7s: these have a screening plate among the anodes of the two triodes, just like ECC88s, connected to the pin left free by the missing 12.6V heater connection. All remaining electric parameters are so similar to the 12AX7 that I suppose they were in fact the same tube with minor changes. Grounding that screen should reduce the parasitic capacitance among the two triodes (hence the "crosstalk) to levels well below the audible threshold, and it's an easy and cheap fix: 3 connections and a single 5$ tube to be swapped. Nevertheless I'd bet a leg, if I had one spare, that nobody will ever hear the difference... with headphones, maybe... not with loudspeakers. Ciao Fabio "Patrick Turner" ha scritto nel messaggio ... A S Goh wrote: Hi In Marantz M7 preamp the L and R signal go to a single output tube. Will using separate tube for L and R improve imaging? Probably not beacuse the channel separation is over 40 dB up to 10 kHz perhaps. Above that stray C starts to let the channels talk to each other. I try to always have separate tubes for each channel. ARC SP11 preamp had 3 6DJ8 for the phono amp, so six triodes for 2 chanels. the channels shared tubes in each of 3 stages. I know, I just repaired one. Probably this is better than having cascaded preamp stages of one channel in a tube since interference is more likely. I like to have one whole twin triode tube with both its triodes devoted to only one amp stage. Patrick Turner. |
#5
|
|||
|
|||
Behold, A S Goh scribed on tube chassis:
Thanks for the idea. Russian tube may not be easy to get They (6N2P) are easier to find and cheaper than 12AX7A - check out Gintaras on Ebay - honest man, a little slow, but honest ;-) -- Gregg "t3h g33k" http://geek.scorpiorising.ca *Ratings are for transistors, tubes have guidelines* |
#6
|
|||
|
|||
A S Goh wrote: Thanks for the idea. Russian tube may not be easy to get ECC88 may be able to find . Theoretically, separate tube can improve channel separation and therefor imaging, if I am not wrong Theoretically, entirely separate triodes have the best separation of signals between input and output or between channels. ARC didn't worry too much though, and used each half of a twin triode for the same stage for two channels. They know it measures OK. What if I use separate tube and parallel the two triode for the one channel output? I have done this occasionally. In a revised preamp inside an old Quad 22 box I have paralleled 6DJ8 with 39k RL, wire wound, for low noise in a phono MM amp. But the noise was no less than for a decent 12AX7. There is very little cross talk between the tubes of each channel, that's for sure. It sounds OK, and that is the main thing. If the capacitance between two 12AX7 anodes is less than 5 pF, and the circuit impedance is 35k, and there are two gain stages, then you have a high pass filter with 5 pf and 70k, and the pole is at 46 kHz. While this might not seen too bad, you will get some cross talk at 5 kHz, where the signal cross talk would be at -20 dB. Using 6DJ8, with plate circuit Z of say 10k, the pole is at 160 kHz, if there was the same 5 pF between anodes and including between anode leads or other surrounding components. To get the best separation for hi gain high Ra twin triodes, use 6EU7 as Fabio suggested. These are not as common as 12AX7. Separate triodes are not enough though; you have to use separate source switches, and build each preamp channel as a monoblock within the one chassis. The dual gang volume control also want to be well separated. Ganged volume controls with pots well apart are not made, so you have to place a metal disc shield around the pot terminals. I have never seen a dual gang pot with the 6.3mm shaft running through each gang situated 150mm apart. It is possible to use two pots with a chain drive from one to the other, since such small chains and sprockets are available from hobby stores; a toothed belt on pulleys is also OK. Source switches can have their wafers well seperated, since there are wafer assemblies available with a 150 long shaft and bolts to hold a distant wafer. An easy test for channel cross talk is to run a square wave into one channel, with the other channel input terminated with same impedance as the live channel and look at the output signal of the dead channels. Often you will see the vertical HF part of the square wave appear a s a series of pulses, since they can be transmitted by stray capacitance. To avoid it, mono is the only sure way. For virtually complete channel separation, you must work hard to get it; its not a problem confined to the tubes; its in all the other bits and in the layout as well. Patrick Turner. "Fabio Berutti" wrote in message ... If this gentleman has some experience with the welding iron, and if he's got a 6.3V heater supply available, he can try changing 12AX7s with Russian 6N2p or NOS 6EU7s: these have a screening plate among the anodes of the two triodes, just like ECC88s, connected to the pin left free by the missing 12.6V heater connection. All remaining electric parameters are so similar to the 12AX7 that I suppose they were in fact the same tube with minor changes. Grounding that screen should reduce the parasitic capacitance among the two triodes (hence the "crosstalk) to levels well below the audible threshold, and it's an easy and cheap fix: 3 connections and a single 5$ tube to be swapped. Nevertheless I'd bet a leg, if I had one spare, that nobody will ever hear the difference... with headphones, maybe... not with loudspeakers. Ciao Fabio "Patrick Turner" ha scritto nel messaggio ... A S Goh wrote: Hi In Marantz M7 preamp the L and R signal go to a single output tube. Will using separate tube for L and R improve imaging? Probably not beacuse the channel separation is over 40 dB up to 10 kHz perhaps. Above that stray C starts to let the channels talk to each other. I try to always have separate tubes for each channel. ARC SP11 preamp had 3 6DJ8 for the phono amp, so six triodes for 2 chanels. the channels shared tubes in each of 3 stages. I know, I just repaired one. Probably this is better than having cascaded preamp stages of one channel in a tube since interference is more likely. I like to have one whole twin triode tube with both its triodes devoted to only one amp stage. Patrick Turner. |
#7
|
|||
|
|||
Hi Patrick
Thanks for your explanation. You are a great contributor to rats "Patrick Turner" wrote in message ... A S Goh wrote: Thanks for the idea. Russian tube may not be easy to get ECC88 may be able to find . Theoretically, separate tube can improve channel separation and therefor imaging, if I am not wrong Theoretically, entirely separate triodes have the best separation of signals between input and output or between channels. ARC didn't worry too much though, and used each half of a twin triode for the same stage for two channels. They know it measures OK. What if I use separate tube and parallel the two triode for the one channel output? I have done this occasionally. In a revised preamp inside an old Quad 22 box I have paralleled 6DJ8 with 39k RL, wire wound, for low noise in a phono MM amp. But the noise was no less than for a decent 12AX7. There is very little cross talk between the tubes of each channel, that's for sure. It sounds OK, and that is the main thing. If the capacitance between two 12AX7 anodes is less than 5 pF, and the circuit impedance is 35k, and there are two gain stages, then you have a high pass filter with 5 pf and 70k, and the pole is at 46 kHz. While this might not seen too bad, you will get some cross talk at 5 kHz, where the signal cross talk would be at -20 dB. Using 6DJ8, with plate circuit Z of say 10k, the pole is at 160 kHz, if there was the same 5 pF between anodes and including between anode leads or other surrounding components. To get the best separation for hi gain high Ra twin triodes, use 6EU7 as Fabio suggested. These are not as common as 12AX7. Separate triodes are not enough though; you have to use separate source switches, and build each preamp channel as a monoblock within the one chassis. The dual gang volume control also want to be well separated. Ganged volume controls with pots well apart are not made, so you have to place a metal disc shield around the pot terminals. I have never seen a dual gang pot with the 6.3mm shaft running through each gang situated 150mm apart. It is possible to use two pots with a chain drive from one to the other, since such small chains and sprockets are available from hobby stores; a toothed belt on pulleys is also OK. Source switches can have their wafers well seperated, since there are wafer assemblies available with a 150 long shaft and bolts to hold a distant wafer. An easy test for channel cross talk is to run a square wave into one channel, with the other channel input terminated with same impedance as the live channel and look at the output signal of the dead channels. Often you will see the vertical HF part of the square wave appear a s a series of pulses, since they can be transmitted by stray capacitance. To avoid it, mono is the only sure way. For virtually complete channel separation, you must work hard to get it; its not a problem confined to the tubes; its in all the other bits and in the layout as well. Patrick Turner. "Fabio Berutti" wrote in message ... If this gentleman has some experience with the welding iron, and if he's got a 6.3V heater supply available, he can try changing 12AX7s with Russian 6N2p or NOS 6EU7s: these have a screening plate among the anodes of the two triodes, just like ECC88s, connected to the pin left free by the missing 12.6V heater connection. All remaining electric parameters are so similar to the 12AX7 that I suppose they were in fact the same tube with minor changes. Grounding that screen should reduce the parasitic capacitance among the two triodes (hence the "crosstalk) to levels well below the audible threshold, and it's an easy and cheap fix: 3 connections and a single 5$ tube to be swapped. Nevertheless I'd bet a leg, if I had one spare, that nobody will ever hear the difference... with headphones, maybe... not with loudspeakers. Ciao Fabio "Patrick Turner" ha scritto nel messaggio ... A S Goh wrote: Hi In Marantz M7 preamp the L and R signal go to a single output tube. Will using separate tube for L and R improve imaging? Probably not beacuse the channel separation is over 40 dB up to 10 kHz perhaps. Above that stray C starts to let the channels talk to each other. I try to always have separate tubes for each channel. ARC SP11 preamp had 3 6DJ8 for the phono amp, so six triodes for 2 chanels. the channels shared tubes in each of 3 stages. I know, I just repaired one. Probably this is better than having cascaded preamp stages of one channel in a tube since interference is more likely. I like to have one whole twin triode tube with both its triodes devoted to only one amp stage. Patrick Turner. |
#8
|
|||
|
|||
Hi Patrick
Thanks for your explanation. You are a great contributor to rats "Patrick Turner" wrote in message ... A S Goh wrote: Thanks for the idea. Russian tube may not be easy to get ECC88 may be able to find . Theoretically, separate tube can improve channel separation and therefor imaging, if I am not wrong Theoretically, entirely separate triodes have the best separation of signals between input and output or between channels. ARC didn't worry too much though, and used each half of a twin triode for the same stage for two channels. They know it measures OK. What if I use separate tube and parallel the two triode for the one channel output? I have done this occasionally. In a revised preamp inside an old Quad 22 box I have paralleled 6DJ8 with 39k RL, wire wound, for low noise in a phono MM amp. But the noise was no less than for a decent 12AX7. There is very little cross talk between the tubes of each channel, that's for sure. It sounds OK, and that is the main thing. If the capacitance between two 12AX7 anodes is less than 5 pF, and the circuit impedance is 35k, and there are two gain stages, then you have a high pass filter with 5 pf and 70k, and the pole is at 46 kHz. While this might not seen too bad, you will get some cross talk at 5 kHz, where the signal cross talk would be at -20 dB. Using 6DJ8, with plate circuit Z of say 10k, the pole is at 160 kHz, if there was the same 5 pF between anodes and including between anode leads or other surrounding components. To get the best separation for hi gain high Ra twin triodes, use 6EU7 as Fabio suggested. These are not as common as 12AX7. Separate triodes are not enough though; you have to use separate source switches, and build each preamp channel as a monoblock within the one chassis. The dual gang volume control also want to be well separated. Ganged volume controls with pots well apart are not made, so you have to place a metal disc shield around the pot terminals. I have never seen a dual gang pot with the 6.3mm shaft running through each gang situated 150mm apart. It is possible to use two pots with a chain drive from one to the other, since such small chains and sprockets are available from hobby stores; a toothed belt on pulleys is also OK. Source switches can have their wafers well seperated, since there are wafer assemblies available with a 150 long shaft and bolts to hold a distant wafer. An easy test for channel cross talk is to run a square wave into one channel, with the other channel input terminated with same impedance as the live channel and look at the output signal of the dead channels. Often you will see the vertical HF part of the square wave appear a s a series of pulses, since they can be transmitted by stray capacitance. To avoid it, mono is the only sure way. For virtually complete channel separation, you must work hard to get it; its not a problem confined to the tubes; its in all the other bits and in the layout as well. Patrick Turner. "Fabio Berutti" wrote in message ... If this gentleman has some experience with the welding iron, and if he's got a 6.3V heater supply available, he can try changing 12AX7s with Russian 6N2p or NOS 6EU7s: these have a screening plate among the anodes of the two triodes, just like ECC88s, connected to the pin left free by the missing 12.6V heater connection. All remaining electric parameters are so similar to the 12AX7 that I suppose they were in fact the same tube with minor changes. Grounding that screen should reduce the parasitic capacitance among the two triodes (hence the "crosstalk) to levels well below the audible threshold, and it's an easy and cheap fix: 3 connections and a single 5$ tube to be swapped. Nevertheless I'd bet a leg, if I had one spare, that nobody will ever hear the difference... with headphones, maybe... not with loudspeakers. Ciao Fabio "Patrick Turner" ha scritto nel messaggio ... A S Goh wrote: Hi In Marantz M7 preamp the L and R signal go to a single output tube. Will using separate tube for L and R improve imaging? Probably not beacuse the channel separation is over 40 dB up to 10 kHz perhaps. Above that stray C starts to let the channels talk to each other. I try to always have separate tubes for each channel. ARC SP11 preamp had 3 6DJ8 for the phono amp, so six triodes for 2 chanels. the channels shared tubes in each of 3 stages. I know, I just repaired one. Probably this is better than having cascaded preamp stages of one channel in a tube since interference is more likely. I like to have one whole twin triode tube with both its triodes devoted to only one amp stage. Patrick Turner. |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
6eu7 or 12ax7? what to to use... | Vacuum Tubes | |||
Does anyone know of this challenge? | High End Audio | |||
Stereo imaging affecting EQ | Pro Audio | |||
Imaging, soundstage, 3D | High End Audio | |||
Low voltage(15V) 12AX7 tube driver schematic for real? | Vacuum Tubes |