I've been thinking of building a 6SN7 LTP for a balanced input line stage using a choke as the plate load. I haven't seen much discussion of this approach. Anyone tried it?
What I'm considering is a Lundahl LL1667 which has two windings, one for each plate. I plan to use LED biasing, setting the bias point around 7ma per triode. Lundahl indicates that this choke is available with different air gaps to increase inductance for lower DC current. I'm hoping I can get up to 200H per winding based on my DC current requirements.
Based on simulations, this will start to roll off below about 200 Hz, but it's only down a small amount. I have a separate SS line stage for my subwoofer outputs which crossover to my mains at about 180Hz, so I think the low-frequency roll-off shouldn't be a problem. Anything else I should worry about?
The LTP plates will DC couple to a balanced mosfet source follower stage, so the next stage load will be negligible (except for the modest capacitive load).
I'm also open to other tube suggestions. I am thinking about the 6SN7 because I've heard that the new-production Linlai Global Elite tubes sound quite nice and I like the idea of being able to get new tubes.
What I'm considering is a Lundahl LL1667 which has two windings, one for each plate. I plan to use LED biasing, setting the bias point around 7ma per triode. Lundahl indicates that this choke is available with different air gaps to increase inductance for lower DC current. I'm hoping I can get up to 200H per winding based on my DC current requirements.
Based on simulations, this will start to roll off below about 200 Hz, but it's only down a small amount. I have a separate SS line stage for my subwoofer outputs which crossover to my mains at about 180Hz, so I think the low-frequency roll-off shouldn't be a problem. Anything else I should worry about?
The LTP plates will DC couple to a balanced mosfet source follower stage, so the next stage load will be negligible (except for the modest capacitive load).
I'm also open to other tube suggestions. I am thinking about the 6SN7 because I've heard that the new-production Linlai Global Elite tubes sound quite nice and I like the idea of being able to get new tubes.
6N6P will have much lower internal resistance (rp) than 6SN7, which working into the choke's limited inductance will yield better bass response. 6N6P are still readily available, even NOS in box. Or you could use JJ ECC99, which I think are available as new production.
6SN7 rp with 7mA Ip will be about 9k ohms or thereabouts.
6N6P rp with 15mA Ip would be about 2k ohms or thereabouts.
ECC99 rp will be similar to 6N6P.
6SN7 rp with 7mA Ip will be about 9k ohms or thereabouts.
6N6P rp with 15mA Ip would be about 2k ohms or thereabouts.
ECC99 rp will be similar to 6N6P.
My suggestion :- differential 6SN7 with CCS at the cathode. LL1660 PP inter-stage. This can be easily switched from balance or SE for input and output. The distortion is very minimal.
Johnny
Johnny
Thanks all for the ideas.
@rongon - I have a few dozen 6N6P tubes, but I've been disappointed with the quality control. These were supposedly NOS tubes, but the section matching is terrible, which makes them less suitable for a LTP. This is one of the reasons why I am down on NOS tubes. But I may give them a try with plate chokes to see if they sound ok.
@jcalvarez - yes, the two windings will be magnetically coupled.
@kmtang - interesting idea. I don't have any experience with interstage transformers, but this might be a good way to learn a bit about how they work. I'll run some simulations to see if I can improve my understanding. It's possible the output impedance would be low enough directly from the transformer which would eliminate the need for another stage.
@gerrittube - I'm building a stereo line stage, but each channel is completely separate with its own power supply.
@rongon - I have a few dozen 6N6P tubes, but I've been disappointed with the quality control. These were supposedly NOS tubes, but the section matching is terrible, which makes them less suitable for a LTP. This is one of the reasons why I am down on NOS tubes. But I may give them a try with plate chokes to see if they sound ok.
@jcalvarez - yes, the two windings will be magnetically coupled.
@kmtang - interesting idea. I don't have any experience with interstage transformers, but this might be a good way to learn a bit about how they work. I'll run some simulations to see if I can improve my understanding. It's possible the output impedance would be low enough directly from the transformer which would eliminate the need for another stage.
@gerrittube - I'm building a stereo line stage, but each channel is completely separate with its own power supply.
regarding NOS tubes, I do not know if all, but a lot I used need to be operated for a while to recover their full emission. Morgan Jones proposed baking (actually putting them in a hot oven for some hours), but operating the filament for some hours also work. I am actually surprised this is barely mentioned, as I think a lot of valves may be deemed bad because of this lack of burnin after decades of storage.
I would indeed look into a pair of LL1667 gapped for about 2mA, that should give much more inductance and still allow for some disbalance.
In a recent amp I made I used a much cheaper PP choke for the driver, the FP230-25 from Triad. It worked well from 20Hz to at least 30kHz loading an ECC81, but I am convinced the Lundahl is better 🙂 https://www.diyaudio.com/community/threads/dc-coupled-4p1l-ecc81.403347/
I would indeed look into a pair of LL1667 gapped for about 2mA, that should give much more inductance and still allow for some disbalance.
In a recent amp I made I used a much cheaper PP choke for the driver, the FP230-25 from Triad. It worked well from 20Hz to at least 30kHz loading an ECC81, but I am convinced the Lundahl is better 🙂 https://www.diyaudio.com/community/threads/dc-coupled-4p1l-ecc81.403347/
@ErikdeBest - thanks for the tip on running the filament for a while. I'll give that a try.
@kmtang - I've run some simulations using the LL1660 with the four secondary windings in parallel driving the load directly (with no source followers) and this looks like it might work really well. -3db at approximately 20Hz to 35Khz driving a 50K 500pf load, with a gain of about 13db (plenty for my purposes). As I mentioned, I don't really need the low frequency response since I have a separate SS buffer driving my woofers. I'm going to order a couple and try them out.
@kmtang - I've run some simulations using the LL1660 with the four secondary windings in parallel driving the load directly (with no source followers) and this looks like it might work really well. -3db at approximately 20Hz to 35Khz driving a 50K 500pf load, with a gain of about 13db (plenty for my purposes). As I mentioned, I don't really need the low frequency response since I have a separate SS buffer driving my woofers. I'm going to order a couple and try them out.
Hello Jaytor,
Please order the LL1660 with PP version. This would have much higher induction and will provide better low frequency response.
Please order the LL1660 with PP version. This would have much higher induction and will provide better low frequency response.