Joachim, I see you have been using Tina. If I may suggest ltspice, which in my opinion is really much better, and also free. Once you get used to it it's very easy to use. Many people in this forum (and elsewhere) use it as their primary simulator, so you'd be able to exchange simulation files easily. If you ever decide to give it a try, I can help you with a huge library of models. Just saying, no pressure. In the end, use what you like using.
I tested the buffer on the scope and gave it a soft lowpass filter somewhere around 5MHz or it makes a slight overshot with a 100kHz squarewave.
Without the filter the buffer surpasses my 20MHz measurement bandwidth. The buffer is builld on a double metalised board with groundplane construction. I supply it from buffered batteries, then a 100mH common mode choke plus 10.000uF and 1000uF plus 2.2uF plus 0.1uF MKP locally. On the scope it looks clean. Maybe i should put the switches and the Duts in a metal box. Good idea to watch EMF. Here is the circuit. The bias Leds in the cascode somewhat limit the voltage that i can squeze out clean so more can be used if i need more dynamic range.
Without the filter the buffer surpasses my 20MHz measurement bandwidth. The buffer is builld on a double metalised board with groundplane construction. I supply it from buffered batteries, then a 100mH common mode choke plus 10.000uF and 1000uF plus 2.2uF plus 0.1uF MKP locally. On the scope it looks clean. Maybe i should put the switches and the Duts in a metal box. Good idea to watch EMF. Here is the circuit. The bias Leds in the cascode somewhat limit the voltage that i can squeze out clean so more can be used if i need more dynamic range.
Hi Ikoflexer !
I use Tina only for drawing circuit diagrams. I think circuits in LT spice look terible.
I have an engineer friend that does my simulations in Microcap and he is so good that i do not even try to compete.
Anyway, i heard good things about LT spice concerning simulation and thanks that you offer me advice.
I use Tina only for drawing circuit diagrams. I think circuits in LT spice look terible.
I have an engineer friend that does my simulations in Microcap and he is so good that i do not even try to compete.
Anyway, i heard good things about LT spice concerning simulation and thanks that you offer me advice.
Joachim,
I see that you are using 3x 246Y in parallel, and 1x 103BL.
Could it be that you are also getting them from Mr. Meurer ? 😉
The Sziklais run at about 50mA if my sums are not wrong.
Why not just use one stage with 3 pairs 2SK170V/2SJ74V in parallel ?
For more Class AB current out, or lower input capacitance ?
Schönen Gruß,
Patrick
I see that you are using 3x 246Y in parallel, and 1x 103BL.
Could it be that you are also getting them from Mr. Meurer ? 😉
The Sziklais run at about 50mA if my sums are not wrong.
Why not just use one stage with 3 pairs 2SK170V/2SJ74V in parallel ?
For more Class AB current out, or lower input capacitance ?
Schönen Gruß,
Patrick
Hy Patrick !
i bought them from Reichelt and to my surprise i got 245Y and 103BL so i need 3 N-channel.
Other wise the green bias Led whould not ignite and i had that problem in the beginning. Even consulting with Jan Didden and Sigurd Ruschkovki ( the best of the best in my book ) did not bring out what was wrong until i found that out. In general buying from Reichelt usually brings you low Idss and low Hfe but i can use that to atvantage sometimes especially in my advanced current mirrors. I wish we could meet some day and i can show you more. Reichelt seems to buy from the lower gain devices and get a discount that way. when you know that it is not a real problem. They have incredible low prices and a phantastic service so i would not like to critisise them much here.
I made a buffer with 2SK170 and 2SJ74, so do not worry, i am up to date. I simply hate to waiste too much 2SJ74 because even contacting Kirkwood of Linear Systems did not get me any further to persuade me that they are not lost forever. John got some so they seem to safe them for the american market but that is fine for me. See my schematic that also allows the Hawksford bootstrap. How that works is a bit compicated. It gives another distortion profile because the Fets have to drive the bases of the cascodes too but it gives some advantage in dynamic range.
i bought them from Reichelt and to my surprise i got 245Y and 103BL so i need 3 N-channel.
Other wise the green bias Led whould not ignite and i had that problem in the beginning. Even consulting with Jan Didden and Sigurd Ruschkovki ( the best of the best in my book ) did not bring out what was wrong until i found that out. In general buying from Reichelt usually brings you low Idss and low Hfe but i can use that to atvantage sometimes especially in my advanced current mirrors. I wish we could meet some day and i can show you more. Reichelt seems to buy from the lower gain devices and get a discount that way. when you know that it is not a real problem. They have incredible low prices and a phantastic service so i would not like to critisise them much here.
I made a buffer with 2SK170 and 2SJ74, so do not worry, i am up to date. I simply hate to waiste too much 2SJ74 because even contacting Kirkwood of Linear Systems did not get me any further to persuade me that they are not lost forever. John got some so they seem to safe them for the american market but that is fine for me. See my schematic that also allows the Hawksford bootstrap. How that works is a bit compicated. It gives another distortion profile because the Fets have to drive the bases of the cascodes too but it gives some advantage in dynamic range.
I am in almost daily correspondence with Kirkwood. Yes, they are still having delays, but they are NOT discriminating against you.
Hi Joachim,
Just to echo what ikoflexer suggested, I prefer LTspice and I have an audio-amplifier-centric tutorial chapter on it in my book, Designing Audio Power Amplifiers. You can see what is in this chapter 19 by looking at the book's table of contents, which is posted on my site at CordellAudio.com - Home. Moreover, Chapter 20 is a complete chapter on tweaking and creating transistor models.
Cheers,
Bob
John, it is good to hear that the LSJ74 will be done. Sorry about my somewhat disappointed tone but i got no reply to my e-mail to Kirkwood. Maybe he gets too much requests and he simpy missed it.
Bob, as i said, i use Tina ONLY for drawing schematics. They look better to me then the oes that come out of LT Spice. I have LT Spice running on my computer but usually i work with another engineer that makes the simulation on Microcap.
I have your book and will look up chapter 19 and 20 again. I look into the book from time to time but i did not find the time to go into the details yet.
Bob, as i said, i use Tina ONLY for drawing schematics. They look better to me then the oes that come out of LT Spice. I have LT Spice running on my computer but usually i work with another engineer that makes the simulation on Microcap.
I have your book and will look up chapter 19 and 20 again. I look into the book from time to time but i did not find the time to go into the details yet.
Kirkwood is only a consultant for LS, he doesn't work there on a daily basis. You should talk to Jan about LS, as they were at Burning Amp.
I have received early samples, last year, and my company Constellation Audio, has received a shipment of J74 specials. However, if you were just looking for samples, delays are probable.
I have received early samples, last year, and my company Constellation Audio, has received a shipment of J74 specials. However, if you were just looking for samples, delays are probable.
Thanks for this information. I will contact Jan.
Besides i am working on the little cap switchbox and this is how far i got.
Let me state right away that my goal is not to find an absolue ranking and the results are only valid in the context of my own system.
I will ony measure the value of the components and will try to match them to better then 5%. So no measurements of distortion, DA or anything else will be made before the listening test. After the subjective and informal listening test it will be interesting to measure for example the "worsed" against the "best" if that brings out any differences on the bench.
Besides i am working on the little cap switchbox and this is how far i got.
Let me state right away that my goal is not to find an absolue ranking and the results are only valid in the context of my own system.
I will ony measure the value of the components and will try to match them to better then 5%. So no measurements of distortion, DA or anything else will be made before the listening test. After the subjective and informal listening test it will be interesting to measure for example the "worsed" against the "best" if that brings out any differences on the bench.
Attachments
I've moved your book to the top of my Christmas Wish List after Rene Yeager (who had just finished your book), gave it a very high recommendation last Saturday at an informal gathering.
High Praise indeed!
Best Regards,
TerryO
Joachim,
I am most likely to see Jan Wednesday next week.
Probably at his place but not sure yet because of work schedule.
Please send me a email via the forum.
I'll tell you the rest then.
Gruß,
Patrick
I am most likely to see Jan Wednesday next week.
Probably at his place but not sure yet because of work schedule.
Please send me a email via the forum.
I'll tell you the rest then.
Gruß,
Patrick
I've built something similar, the difference being that the first two switch positions are known (position 1, V-Cap teflon; position 2, Radio Shack electrolytic) and the remaining 12 are randomized. All silver wiring, Cinema Engineering switch, gold-plated nylon dielectric RCA jacks. Hours of fun for your friends.
- Status
- Not open for further replies.