I redid the bias supply and made some other changes. Got the 60 Hz ripple down too. It is sounding quite good now. The imaging is other worldly. I get why Mike named it LuminAria (lit from within). Will build a buffer next.
In the process of tweaking, I put the drain voltage on the gate, and one of the devices is mostly toast. It produces sound, but very weak. 100V between G-S is a bit much even for these stealthy devices. I am absolutely hating myself for that mistake. I wonder if it is possible to salvage the part. I made some resistance measurements: gate to drain or source is open, but D-S measures <1 ohm, whereas the good ones measure >1 ohm.
While I had a quad of 2sk82, I am now down to just one extra. I have a few of the Tokin devices, but I am looking to buy a few more of these preamp-suitable devices. Is the KP926 the best option right now from a price and availability POV?
In the process of tweaking, I put the drain voltage on the gate, and one of the devices is mostly toast. It produces sound, but very weak. 100V between G-S is a bit much even for these stealthy devices. I am absolutely hating myself for that mistake. I wonder if it is possible to salvage the part. I made some resistance measurements: gate to drain or source is open, but D-S measures <1 ohm, whereas the good ones measure >1 ohm.
While I had a quad of 2sk82, I am now down to just one extra. I have a few of the Tokin devices, but I am looking to buy a few more of these preamp-suitable devices. Is the KP926 the best option right now from a price and availability POV?
Sorry to hear about your wiring issue. I had one recently with my THF-51S but it was a reversal of voltage to drain and source, and the device survived. Unfortunately high voltage Vgs is a much worse situation.
I started experiments with a KP926B common source choke loaded voltage gain stage recently and it seems that the KP926B is not a drop-in for a Sony VFET. First I tried high voltage low current with fixed bias and mu was around 30. I switched to source bias and mu drop to around 20. Output impedance was very high at around 4000 Ohms. I then changed to lower voltage and higher current and mu went back up but output impedance dropped to around 400 Ohms.
Then my lab power supply malfunctioned so now I am waiting for parts so that I can attempt to fix it.
I have four KP926Bs and their Vgs are varied. My experiments so far have been with the one with the lowest Vgs. I may have to do some manual curve tracing and learn how to create a Spice model and do some LTSpice simulations.
I started experiments with a KP926B common source choke loaded voltage gain stage recently and it seems that the KP926B is not a drop-in for a Sony VFET. First I tried high voltage low current with fixed bias and mu was around 30. I switched to source bias and mu drop to around 20. Output impedance was very high at around 4000 Ohms. I then changed to lower voltage and higher current and mu went back up but output impedance dropped to around 400 Ohms.
Then my lab power supply malfunctioned so now I am waiting for parts so that I can attempt to fix it.
I have four KP926Bs and their Vgs are varied. My experiments so far have been with the one with the lowest Vgs. I may have to do some manual curve tracing and learn how to create a Spice model and do some LTSpice simulations.
Sorry to hear about your wiring issue. I had one recently with my THF-51S but it was a reversal of voltage to drain and source, and the device survived. Unfortunately high voltage Vgs is a much worse situation.
I started experiments with a KP926B common source choke loaded voltage gain stage recently and it seems that the KP926B is not a drop-in for a Sony VFET. First I tried high voltage low current with fixed bias and mu was around 30. I switched to source bias and mu drop to around 20. Output impedance was very high at around 4000 Ohms. I then changed to lower voltage and higher current and mu went back up but output impedance dropped to around 400 Ohms.
Then my lab power supply malfunctioned so now I am waiting for parts so that I can attempt to fix it.
I have four KP926Bs and their Vgs are varied. My experiments so far have been with the one with the lowest Vgs. I may have to do some manual curve tracing and learn how to create a Spice model and do some LTSpice simulations.
Did you see the kp926 preamp thread? Seems like lower Vgs and Vds relative to the 2sk82 but iverall topology is the same.
tons
with some wit, one can use plethora of types in plethora of configurations
what I like most (lazy bstrd, choosing easiest) is 6DJ8 (7DJ8) and similar ones - having moderate Mu, lowish Rp and, most important, decent xconductance
you can do tons with part having amplification in range of 30
but, really, getting lazy with age ...... most of my fun with tubes nowadays is just to laugh at Kosher constructions ........ and
when I see thinking out of boxGood to know! I might also experiment with this part. You can throw an FET source follower on the output, similar to the input buffer on the LuminAria. The buffer will provide high input impedance and low output impedance. Works like a charm and doesn’t add any of its own sound.
tons
with some wit, one can use plethora of types in plethora of configurations
what I like most (lazy bstrd, choosing easiest) is 6DJ8 (7DJ8) and similar ones - having moderate Mu, lowish Rp and, most important, decent xconductance
you can do tons with part having amplification in range of 30
but, really, getting lazy with age ...... most of my fun with tubes nowadays is just to laugh at Kosher constructions ........ and
Oh, you meant tubes. Yes, for sure. Lots of triodes there. Here’s one, though it is pretty obscure
Attachments
ra7, I am trying to avoid a buffer at the output, but it is an option.
The big issue with me is that I want a voltage gain stage between my Luminaria and my follower amps, and the mu of the KP926B is much more than I want or need. I don't want to replace the Luminaria with a low distortion high gain preamp. The Luminaria has enough voltage output for my usual listening sound levels but it does not have enough voltage output to clip my THF-51S mu follower amp. I don't really need another voltage gain stage but this is a hobby so it is something to do that occupies my time and brain.
Still more experimenting to do.
I have tubes too, but that is the past.
The big issue with me is that I want a voltage gain stage between my Luminaria and my follower amps, and the mu of the KP926B is much more than I want or need. I don't want to replace the Luminaria with a low distortion high gain preamp. The Luminaria has enough voltage output for my usual listening sound levels but it does not have enough voltage output to clip my THF-51S mu follower amp. I don't really need another voltage gain stage but this is a hobby so it is something to do that occupies my time and brain.
Still more experimenting to do.
I have tubes too, but that is the past.
Originally I tried fixed bias with a negative power supply and the gain was high, around 30, so I switched to a source resistor for less gain.
When I switched from high voltage low current to a lower voltage higher current operating point, I changed the load choke to accommodate the higher current. The new choke had lower DCR but with the higher current, the source resistor value went down. It ended up with higher gain at this operating point.
The particular KP that I was using needed a low Vgs so the resistor value wasn't very big. I will try a different KP that requires higher Vgs and larger value source resistor.
Also the more I think about it, the less that I like the idea of making an output buffer that can handle a 25Vrms signal to deal with the high output impedance.
Perhaps the smartest thing for me to do is to choose a device that is more suited to my needs rather than trying to force it. But at least I learned a few things about the KP926B.
When I switched from high voltage low current to a lower voltage higher current operating point, I changed the load choke to accommodate the higher current. The new choke had lower DCR but with the higher current, the source resistor value went down. It ended up with higher gain at this operating point.
The particular KP that I was using needed a low Vgs so the resistor value wasn't very big. I will try a different KP that requires higher Vgs and larger value source resistor.
Also the more I think about it, the less that I like the idea of making an output buffer that can handle a 25Vrms signal to deal with the high output impedance.
Perhaps the smartest thing for me to do is to choose a device that is more suited to my needs rather than trying to force it. But at least I learned a few things about the KP926B.
Just wondering if this has made it into the oven yet Michael?
As ZM would say - Greedy Boyz!
Cheers
Sadly, my day jobs leaves me only an hour or two a week to tinker and I don't think I'll get to this one before the holidays.
Having said that, it's just a choke-loaded CS stage, with perhaps a 120V B+, self-biased around 20-30mA.
Thanks Michael, I understand that, rather surprisingly, there are things more important than audio projects, even if it's just The Dude's 10 pin bowling...
I'm just about read your orginal LuminAria article.
Cheers and take care.