Using a fet as cascode device is not possible with these voltages. If you really want this (why?) the voltage should be lowered to 30-40 volts.
--------------------------------------------------------------------------------
I was thinking that using a less noisy and more linear device as the cascode device could make the result even better?
Yes I know, I need to reed some more about how a cascode actually works....
--------------------------------------------------------------------------------
I was thinking that using a less noisy and more linear device as the cascode device could make the result even better?
Yes I know, I need to reed some more about how a cascode actually works....
GRollins said:
Thanks, I`ll have to try that some day.
Just a wild guess,,, your upcoming headphone amplifier is going to have JFET as the cascode device???
BTW.. thank you for your great effort in the JFET group buy, but living in Norway I guess I`ll have to wait until dhole gets a shipment from you.
Regards,
Peter
Well, gee, I've got scads of the things lying about. Might as well do something with them.
Let's see, starting at the beginning of the system, there's the phono preamp (no, not the front end, silly...use 'em in the outputs), the preamp, active crossovers, amps, headphone amps, you know...the usual suspects.
Grey
Let's see, starting at the beginning of the system, there's the phono preamp (no, not the front end, silly...use 'em in the outputs), the preamp, active crossovers, amps, headphone amps, you know...the usual suspects.
Grey
Oooh! Oooh! I haven't built a phono pre yet! Do that one Grey! With balanced outputs, yeah yeah! (Picture the crazy weasel from the Looney Toons drooling prodigiously!)
Although ... the idea of an all JFET headphone amp ... scribble, scribble, sketch, scribble, scribble ... muttering about career getting in way of hobby ...
I might be able to put together an idea or two, but don't hold yer breath, I have about as much time available as Grey these days.
Terry
Although ... the idea of an all JFET headphone amp ... scribble, scribble, sketch, scribble, scribble ... muttering about career getting in way of hobby ...
I might be able to put together an idea or two, but don't hold yer breath, I have about as much time available as Grey these days.
Terry
I've been working at the discrete phono thing for what seems like centuries. It keeps getting shoved aside to run other projects through the mill. I was about 90-95% done when the Lovoltechs arrived on the scene last summer. Now I'm rethinking the output stage. Since more people would like--and be able to use--an amp, I'm probably going to push that through to completion. Once again, the phono stage gets the short end of the stick.
An all JFET headphone amp? Yeah, I think it's probably going to work out, but a scalable power amp will come first. After all, my speakers aren't nearly as efficient as Nelson's, and neither are most of yours. And possibly another amp, totally different from the first. I have a third idea that I may unleash if I haven't gotten rid of all the devices by then. And one so bizarre that I dismissed it entirely when it first came to me, but still it haunts me.
And we'll need a preamp, just to demonstrate that it can be done, so you guys keep steaming straight ahead. I've got a few ideas, but it will take a bit for me to get around to doing a preamp circuit. I'll leave it in your capable hands.
Grey
An all JFET headphone amp? Yeah, I think it's probably going to work out, but a scalable power amp will come first. After all, my speakers aren't nearly as efficient as Nelson's, and neither are most of yours. And possibly another amp, totally different from the first. I have a third idea that I may unleash if I haven't gotten rid of all the devices by then. And one so bizarre that I dismissed it entirely when it first came to me, but still it haunts me.
And we'll need a preamp, just to demonstrate that it can be done, so you guys keep steaming straight ahead. I've got a few ideas, but it will take a bit for me to get around to doing a preamp circuit. I'll leave it in your capable hands.
Grey
Hi, Nelson
I am intersted in JFET U310, and I want to use it for my BOSOZ (version: single line voltage and CCS tail).
Data sheet:
http://www.ee.washington.edu/stores/DataSheets/fet/u310.pdf
Can I use this? I am thinking of the cascode with IFR620.
Your advice will encourage me to order this right away.
Regards
jh
I am intersted in JFET U310, and I want to use it for my BOSOZ (version: single line voltage and CCS tail).
Data sheet:
http://www.ee.washington.edu/stores/DataSheets/fet/u310.pdf
Can I use this? I am thinking of the cascode with IFR620.
Your advice will encourage me to order this right away.
Regards
jh
The J310 (aka U310) is a nice device. The problem you run into is the power dissipation. You can't run both current and voltage through the thing at the same time or you'll roast the poor thing.
Ideal for cascode, you say?
Yep.
Run it with about Vds=15V or so and stick something else on top of it that can take a little bit of voltage and you're set to go.
Mouser has both Fairchild and Vishay/Siliconix for decent prices.
Grey
Ideal for cascode, you say?
Yep.
Run it with about Vds=15V or so and stick something else on top of it that can take a little bit of voltage and you're set to go.
Mouser has both Fairchild and Vishay/Siliconix for decent prices.
Grey
Hi jh6you,
just a few remarks
1. this is not an X, you forgot the input resistors.
2. you don´t need input caps (see point 6)
3. you don´t need source resistors
4. you don´t need a compensation cap
5. what are the resistors between gate and source of the jfets?
6. you will need a negative power supply voltage
William
just a few remarks
1. this is not an X, you forgot the input resistors.
2. you don´t need input caps (see point 6)
3. you don´t need source resistors
4. you don´t need a compensation cap
5. what are the resistors between gate and source of the jfets?
6. you will need a negative power supply voltage
William
Hi, William
I always respect your advices.
1. Right, no X-feedback. I am trying to have no NFB.
2-4. I will take, thanks, into consideration.
5. It is to provide JFET with a negative Vgs.
6. I am trying to make this possible with no -V supply.
Regards
jh
PS. Is there any moderator . . . ?
I feel I am disturbing the thread.
If you agree, you could send post#30 through this to a new thread titled "Jfet-cascoded BOSOZ."
I always respect your advices.
1. Right, no X-feedback. I am trying to have no NFB.
2-4. I will take, thanks, into consideration.
5. It is to provide JFET with a negative Vgs.
6. I am trying to make this possible with no -V supply.
Regards
jh
PS. Is there any moderator . . . ?
I feel I am disturbing the thread.
If you agree, you could send post#30 through this to a new thread titled "Jfet-cascoded BOSOZ."
Short on time...no surprise there. This is not to be construed as a complete list of comments...just a starter.
--The Source resistors will need to be there. Assuming that we are talking about depletion mode devices, this will bias the JFETs (considering that--as drawn--the Gates return to the node where the Sources connect to the resistor/CCS).
--With the previous point in mind, the input caps will need to be there as the Gates will be a volt or two above ground.
--I am not sure how well running the Gates back to the node between the Sources will work in the real world. On paper, it will DC bias properly. In the real world, there is some variation at that node (look at it on a scope and you will see signal), hence this will constitute feedback. Approach this with caution. If feedback is your goal, then you'll still have problems, as device variations will vary the amount of signal you see at the Sources to some degree.
--If the caps from the Source of the JFET to the Gate of the cascode MOSFET are intended as "modulation" you'll do better to use a variable resistor to drive the cap; full signal from the Source is probably not the best setting.
--I'm not sure I see the purpose of the resistors in parallel with the CCS (the ones going to ground).
--I agree with William, although you don't need a negative rail, it would simplify things tremendously. You could then do away with the Source resistors (unless you intentionally want to degenerate the JFETs). You could also do away with the input caps. You could ground reference the Gates instead of pointing them back towards the node with the Sources. Keep in mind that a negative rail need not be the same voltage as the positive rail. Ten to 15V will do just fine, depending on what you're using for a current source. Small transformers like that are cheap. I've easily got a dozen of them in my junkbox and quite a few Tamura transformers on hand that cost less than $5 apiece.
All said, I think that the general idea has promise. It's just a few details.
Grey
--The Source resistors will need to be there. Assuming that we are talking about depletion mode devices, this will bias the JFETs (considering that--as drawn--the Gates return to the node where the Sources connect to the resistor/CCS).
--With the previous point in mind, the input caps will need to be there as the Gates will be a volt or two above ground.
--I am not sure how well running the Gates back to the node between the Sources will work in the real world. On paper, it will DC bias properly. In the real world, there is some variation at that node (look at it on a scope and you will see signal), hence this will constitute feedback. Approach this with caution. If feedback is your goal, then you'll still have problems, as device variations will vary the amount of signal you see at the Sources to some degree.
--If the caps from the Source of the JFET to the Gate of the cascode MOSFET are intended as "modulation" you'll do better to use a variable resistor to drive the cap; full signal from the Source is probably not the best setting.
--I'm not sure I see the purpose of the resistors in parallel with the CCS (the ones going to ground).
--I agree with William, although you don't need a negative rail, it would simplify things tremendously. You could then do away with the Source resistors (unless you intentionally want to degenerate the JFETs). You could also do away with the input caps. You could ground reference the Gates instead of pointing them back towards the node with the Sources. Keep in mind that a negative rail need not be the same voltage as the positive rail. Ten to 15V will do just fine, depending on what you're using for a current source. Small transformers like that are cheap. I've easily got a dozen of them in my junkbox and quite a few Tamura transformers on hand that cost less than $5 apiece.
All said, I think that the general idea has promise. It's just a few details.
Grey
GRollins said:
I would love to see a Cascoded jfet X-BOSOZ with a phono stage! That would be awesome.
And while I'm wishing... an inductively loaded X-SOZV8 (ZV9-T?)would be incredible. But that's a different thread, isn't it.
I'm way too unknowledgeable to design either one, but I can be patient
Maybe I'm gonna need more Lovoltechs.
You want 'em?
I got 'em.
You know where to find me.
And thanks again for your support in this Lovoltech thing, guys. The response has been wonderful.
Grey
P.S.: I hope no one will be offended, but I've spent the last couple of hours working on the amp. Things are coming along nicely. I'll get back in harness on the JFET buy tomorrow.
I got 'em.
You know where to find me.
And thanks again for your support in this Lovoltech thing, guys. The response has been wonderful.
Grey
P.S.: I hope no one will be offended, but I've spent the last couple of hours working on the amp. Things are coming along nicely. I'll get back in harness on the JFET buy tomorrow.
Hi, Grey
Source resistor: I want to have it, thinking of two things - for the bias and to give up some voltage gain (a game of compromise).
Cap between MOSFET Gate and JFET Source: I might most probably remove it. I might not need to secure low Vds of JFET with respect to heat dissipation.
The resistor from the input to the ground: It is not in parallel with CCS.
Negative voltage supply: Understood. Nevertheless, I would like to try without it. I know that many diYers like no input coupling cap, tho.
Expecting further comments.
Regards
jh
Source resistor: I want to have it, thinking of two things - for the bias and to give up some voltage gain (a game of compromise).
Cap between MOSFET Gate and JFET Source: I might most probably remove it. I might not need to secure low Vds of JFET with respect to heat dissipation.
The resistor from the input to the ground: It is not in parallel with CCS.
Negative voltage supply: Understood. Nevertheless, I would like to try without it. I know that many diYers like no input coupling cap, tho.
Expecting further comments.
Regards
jh
Removed the cap going from the mosfet gate to the jfet source.
Acually I need max.8 pieces, but ordered 130 pieces of J310, considering matching, maybe paralleling, poor soldering skill, and confusion of D/G/S legs, . . . I am not efficient guy often.
Hey, out there . . . Lunar New Year guys!
Happy New Year!!!
Regards
jh
Acually I need max.8 pieces, but ordered 130 pieces of J310, considering matching, maybe paralleling, poor soldering skill, and confusion of D/G/S legs, . . . I am not efficient guy often.
Hey, out there . . . Lunar New Year guys!
Happy New Year!!!
Regards
jh
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.