This is my first post, so let me introduce myself. I'm Tom from Poland. I started building DIY amps in 2008. For the first few years I worked with tubes exclusively (guitar amps, RIAA preamps, SET stereo amps), but recently I've "converted" to solid state (Pass Zen amp with IRFP9240 and 2SK1058 rules 😀). I've made a few Le Pacific based passive RIAA preamps (schematic), with both 2SK117 and 2SK170 and now I'm thinking about improving the design a little. I'd like to increase the output level and, if possible, improve the linearity of Jfet gain stages a little. I consider the following changes:
1. Increase R3 and R5 to 10k
2. Increase the supply voltage to 70-80 Volts (2SK117 have 50V of breakdown voltage, so they shouldn't bang. They would run at about 3-4mA, so they wouldn't overheat either).
3. Decrease R2 to 47k
4. Decrease R1 to 3-5k (with suitable changes to the rest of RIAA circuit of course).
What do you think about it? Did anyone try to make such modifications? Or maybe there are some other ways to improve this preamp?
One Love
1. Increase R3 and R5 to 10k
2. Increase the supply voltage to 70-80 Volts (2SK117 have 50V of breakdown voltage, so they shouldn't bang. They would run at about 3-4mA, so they wouldn't overheat either).
3. Decrease R2 to 47k
4. Decrease R1 to 3-5k (with suitable changes to the rest of RIAA circuit of course).
What do you think about it? Did anyone try to make such modifications? Or maybe there are some other ways to improve this preamp?
One Love
There's one little gotcha: R3 and R5 determine output impedance of the respective stage, and 10k is a bit too much for a line-level output. You'll probably want to add a buffer when enlarging R5.
The first stage may also be somewhat unhappy having to drive a much heavier load if you significantly reduce R1 et.al.. Again, a buffer could prove useful. It would also permit increasing R3 even further.
And while Crss would be expected to be appreciably reduced at higher Vds (from about 3.5 pF to ~2 pF or so), the much higher levels could in fact increase high-frequency distortion a fair bit. Consider cascoding, possibly even the bootstrapped variety.
The first stage may also be somewhat unhappy having to drive a much heavier load if you significantly reduce R1 et.al.. Again, a buffer could prove useful. It would also permit increasing R3 even further.
And while Crss would be expected to be appreciably reduced at higher Vds (from about 3.5 pF to ~2 pF or so), the much higher levels could in fact increase high-frequency distortion a fair bit. Consider cascoding, possibly even the bootstrapped variety.
2SK170 were pretty unique transistors and 2SK117 are not equivalent types. They may "work" but won't give the best results. If you are looking to improve the performance of the original design but you are buying copies of those Toshiba parts from Ebay stores, your chances of optimum performance are even less.
You can purchase 2SK170 equivalents, type LSK170 from our Diyaudio store, if you want the circuit to work as intended with the sort of RIAA equalization this very simple design permits. It may be a little more expensive, but at least you can make fair evaluations.
You can purchase 2SK170 equivalents, type LSK170 from our Diyaudio store, if you want the circuit to work as intended with the sort of RIAA equalization this very simple design permits. It may be a little more expensive, but at least you can make fair evaluations.
Take a look at the Salas phono stage for inspiration..he has taken this topology about as far as anyone:
http://www.diyaudio.com/forums/analogue-source/129126-simplistic-njfet-riaa.html
It'a a long read but worth it.
http://www.diyaudio.com/forums/analogue-source/129126-simplistic-njfet-riaa.html
It'a a long read but worth it.
For BJTs, significantly exceeding the breakdown voltage results in failure short. With J-fets, exceeding Vgd max causes the electric field to be distorted in shape resulting in a very non-linear transfer. i e, distortion.
There is a limit of course, but exceeding the Vdg by 20% or more will probably not be an optimum amplifier. Similar to triode behavior. That being said, I did use a 'comp' pair of 40V J-fets as the VAS in the previous version of my mosfet amp with peak Vgd of each Jfet at 55V, but they were operating as push pull common gate amplifier and this arrangement nullifies the effect. Great sounding amp.🙂😎
There is a limit of course, but exceeding the Vdg by 20% or more will probably not be an optimum amplifier. Similar to triode behavior. That being said, I did use a 'comp' pair of 40V J-fets as the VAS in the previous version of my mosfet amp with peak Vgd of each Jfet at 55V, but they were operating as push pull common gate amplifier and this arrangement nullifies the effect. Great sounding amp.🙂😎Thanks for the answers. I've noticed the difference between 117's and 170's and I agree that the 170's give a little better sound (at least to my taste). I usually use 117's for experimenting due to higher Vgds and lower price (so I won't worry so much if I burn one 😛).
I was thinking about a simplistic phono preamp for a ZEN/ZCA type amp (2SK1058 with IRFP9240 current source, biasing circuit as in the ZCA, but with lower resistance values and 1uF cap). It doesn't have to drive it to max power, but it'd be good if I didn't have to turn the vol all the way up...
I was thinking about a simplistic phono preamp for a ZEN/ZCA type amp (2SK1058 with IRFP9240 current source, biasing circuit as in the ZCA, but with lower resistance values and 1uF cap). It doesn't have to drive it to max power, but it'd be good if I didn't have to turn the vol all the way up...
I made one using 2SK117's with 10k drain resistors powered from 75v supply. It didn't work out - the distortion is clearly audible and the overall sound is crappy :<. But i was thinking about using this distortion to my advantage, so I removed RIAA filter and I'm experimenting with using this circuit as a guitar preamp 😛.
Well, there's no reason not to try. In the worst case the jfets will pop and I'll feel like Jimi Hendrix blowing his amps up ^^. It works pretty nice as a guitar preamp. I decreased the interstage cap to 1n and put another 1n in front of first jfet. It overdrives pretty hard, so it's important not to "clog" it with too much low end. I also used a low pass filter on the output so it won't sound too "gritty". It sounds somewhat similar to a tube preamp, I'll post a sample soon.
- Status
- Not open for further replies.