| dw8083 |
Has anyone experimented with creating a JFET version of the BOSOZ? Seems like an interesting idea. Just wondering if this has any merit.
-David |
|
|
| Nelson Pass |
It has a tremendous amount of potential.
While simply hooking up a 2sk389 in that topology actually
works well, it is easy to envision a fully cascoded and
constant current sourced version, with output followers
if you want to "gild the lily"
If there is any enthusiasm, I'll go dig up some of my curves and
get back to you next week.
:cool: |
|
|
| apassgear |
| quote: | Originally posted by Nelson Pass
It has a tremendous amount of potential.
While simply hooking up a 2sk389 in that topology actually
works well, it is easy to envision a fully cascoded and
constant current sourced version, with output followers
if you want to "gild the lily"
If there is any enthusiasm, I'll go dig up some of my curves and
get back to you next week.
:cool: |
I for one would be very intresded in seeing such topology in action. Any posibility that you could post also a schematic? May be this is asking too much :xeye:
Thanks very much Nelson :cool: |
|
|
| Nelson Pass |
You can chew on this if you like...
:cool: |
|
|
| Vix |
Wow! Woo!
I started thinking about BOSOZ, Aleph P....and now this one!
Q3-Q6 could be replaced by mosfets CCS, or resistors...
What devices do you recommend for Q1 and Q2? Would LU1014D work in this circuit?
Thanks,
Vix:) |
|
|
| dw8083 |
| quote: | Originally posted by Nelson Pass
You can chew on this if you like...
:cool: |
Very chewy!
-David |
|
|
| cviller |
Nice!
Then we just need to adjust the gain to work nicely with F4... ;)
I think NP already suggested 2sk389 (or matched 2sk170).
|
|
|
| Zen Mod |
simplified enough?
even if I delete degenerative resistors randomly |
|
|
| Vix |
| quote: | Originally posted by Zen Mod
simplified enough?
even if I delete degenerative resistors randomly |
:) Almost. I wanted to use resistors instead of CCS. Just keep the gain Jfets. How could we call this circuit. Bride of F4? :) |
|
|
| Zen Mod |
| quote: | Originally posted by Vix
:) Almost. I wanted to use resistors instead of CCS. Just keep the gain Jfets. How could we call this circuit. Bride of F4? :) |
really depends on your power needs with F4.
if you need smaller swing,then resistors are good ;
for bigger swing you need CCSes and even higher PS voltages......and then even cascoding of jfets.
B200 still? |
|
|
| Vix |
| quote: | Originally posted by Zen Mod
B200 still? |
Yes, it's love and hate relationship. At the moment I put them in ML TL boxes previously used for Fostex FE206. That was Martin J.King project 5. http://www.quarter-wave.com/Project05/Project05.html
I used to fill the port with foam, (without it was sometimes boomy). So it worked more like sealed box, at least until I make my mind for an OB. On the other hand, Fostexes are waiting for a BIB. As things look now, I am going to have two setups, one with B200 's, OB in one room, and a BIB with FE206 in another. F2 is fine with B200, nice for easy evening listening, but not louder. ZV9 is better, overall. I am also using a pair of B&W's 601, with two sealed 12" sealed subs, just for reference, but I prefer the sound of fullrangers. B&w's are more calm, but sometimes boring.
Short explanation of the "items" on a picture: Top: ZEN V9 (used to be zen V3). CD -player: Rotel RCD-02. Below that is what looks like
JVC equalizer. It is not. It's a BOZ in a new enclosure. On top of it is a chocolate can that houses an active X-over for the two 12" sealed subwoofers. They are powered by a smallish amp at the bottom (looks like a satellite receiver, in fact is was a satellite receiver until I put an LM3886 in it).
Overall, I am getting good sound out of my system, but feel that with a little more work, it could be modified to sound better. Somehow I get an impression that, apart from the speakers that did not get appropriate "home" yet, BOZ is the "bottleneck". That's why I was considering building another preamp. I was wondering about BOSOZ, then almost decided on Aleph P 1.0 and then...this circuit:scratch: :confused:
So, most probably, this preamp would drive ZEN V9 with B200 on OB, but it would be nice to know that is has enough capacity to drive F4 in the future. Who knows....
Sorry for such a long post that appears to be OT, but I wanted to let you know what I use at the moment and what I'd like to do, so it could be easier for you to give a proper advice.
Big thanks to all of you, specially to Nelson. I could have never imagined that learning about amplifiers could be so motivating.
Regards,
Vix |
|
|
| Magura |
Well, what's keeping us from an adapted (read downscaled) balanced version of the ZV9 ?
Magura :) |
|
|
| adwsail |
| I didn't think the 389 was still available. Who has them and how much???? |
|
|
| jupiterjune |
Nelson-
I would love to see some measurements -- especially on a cascoded jfet BOSOZ. I fooled around with it for more than a year.....but gave up on it for various reasons (unbalanced to balanced conversion, implementing volume control for 5.2 channels)
To my ear, it sounded better using resistors for the CCS( 72v rails, 3000 ohms per CCS). No follower.
Using Jfet CCS's sounded similar to the resistors, but the "classic" CCS using a small signal bipolar transistor and a mosfet sounded edgy. I would be curious to get see if the meter shows a difference.....
I just drilled a few holes in my AudioXpress boards and plopped them in. |
|
|
| moe29 |
May 2007 will be the 10 Year Anniversary of the BZLS!
(probably the best Zen article written to date)
((well, the ZenV9 is close... let's see how it looks 10 years from now!)) |
|
|
| GRollins |
I have done a bit of fiddling with current sources versus resistors. Yes, sometimes a resistor sounds better, specs be damned. As for why, I'm not certain. I have long held a suspicion that part of the problem is high frequency compliance problems in current sources. A resistor, at least a decent one, is going to give pretty much the same performance at any reasonable frequency.
In practical terms, a resistor of 5-10k is usually sufficient to get good results from something like a differential. If you want to use a higher resistance you will get better performance still, but at the expense (sometimes) of having to provide a higher voltage rail.
The main problem you run into is when you need a lot of current. By definition that means a low resistance, which tends to negate the things that a current source is supposed to do. You also start burning a lot of power in the resistor.
Like many things, you have to think it through. In some cases a CCS is the best solution. In others, a resistor does as well...sometimes better.
Grey |
|
|
| juma |
| quote: | Originally posted by Nelson Pass
it is easy to envision a fully cascoded and
constant current sourced version, with output followers
if you want to "gild the lily" :cool: |
Yes we want, and the lily is all for it ;)
Like this, perhaps ? |
|
|
| dw8083 |
If this Jfet BOSOZ was connected only to an Aleph30, are the output coupling caps necessary? It would be great to eliminate them if possible.
-David |
|
|
| flg |
| There is a DC level at the output of all these JFET BOSOZ variants. A cap will be needed either at the output of these pre's or the input of the A30. Does the A30 already have one??? |
|
|
| Nelson Pass |
| quote: | Originally posted by dw8083
If this Jfet BOSOZ was connected only to an Aleph30, are the output coupling caps necessary? It would be great to eliminate them if possible. |
The + input is direct coupled, and usually the - input has a
capacitor to give low offset. I would keep the coupling caps,
but you could then eliminate the cap on the A30
:cool: |
|
|
| jupiterjune |
| quote: | with output followers
if you want to "gild the lily" |
Were you thinking of something like seven watt output followers? |
|
|
| Nelson Pass |
I believe that would qualify.
:cool: |
|
|
| dw8083 |
Here's a first start at a design. Any thoughts? I'd like to get some feedback before building it. Source followers could be added.
-David |
|
|
| Zen Mod |
| quote: | Originally posted by dw8083
Here's a first start at a design. Any thoughts? I'd like to get some feedback before building it. Source followers could be added.
-David |
increase current (change J type,or use two or use another type of CCS);
after that -be careful with dissipation; decrease PS voltage ,if needed
edit:
R7 and R8 are in 10x magnitude bigger for my liking.........susy effect is smaller with them,degeneration is greater.........
for vinyl (next post,edit after):
Q2 and cousin are CCSes,cascodes,what you like?
:devilr: |
|
|
| Onvinyl |
OK folks,
I'll be the fool of the day and the one who asks the question that no one dares to ask...
What is the function of Q2 and how does it work?
Rüdiger |
|
|
| dw8083 |
Thanks Choky for the insight.
This version has a second J511, lower voltage, and adjustments to R5-8. I would appreciate help in verifying the design and specifically in selecting values for R5-8. I think the values maybe a bit low.
Thank you,
-David |
|
|
| Zen Mod |
| quote: | Originally posted by dw8083
Thanks Choky for the insight.
This version has a second J511, lower voltage, and adjustments to R5-8. I would appreciate help in verifying the design and specifically in selecting values for R5-8. I think the values maybe a bit low.
Thank you,
-David |
try like this
you must adjust R5 and R6 to catch J511's 4,7mA through each side
connection of J511 corrected and few other things
but,I usually make a mess............:devilr:
edit:
move both 100K resistors to 2k2/22k node (not before 2k2 but after,ditto on gate) |
|
|
| dw8083 |
Thanks Choky, I missed the wire between R7 & R8, it's now added. I'm still wondering if R5 & R6 are on the small side?
Here's the updated schematic.
-David
EDIT: Moved C1 & C2 back, outside of R2 & R4
|
|
|
| Zen Mod |
| quote: | Originally posted by dw8083
Thanks Choky, I missed the wire between R7 & R8, it's now added. I'm still wondering if R5 & R6 are on the small side?
Here's the updated schematic.
-David |
if your intention was to make CCS-es with two upper LSKs (hehehe) ,then you must look in data or test them ..........I don't know from top of my head what's Ugs for 4,7mA.............off course-you must match them,so you'll know then what resistors to choose
besides that-for later -you can add followers on output,naturally with overall feedback connected all around
edit:
if you are lazy for testing and matching,you can always send me handful of little critters and I'll do that for ya :devilr: |
|
|
| Zen Mod |
| quote: | Originally posted by Zen Mod
if your intention was to make CCS-es with two upper LSKs (hehehe) ,then you must look in data or test them ..........I don't know from top of my head what's Ugs for 4,7mA.............off course-you must match them,so you'll know then what resistors to choose
besides that-for later -you can add followers on output,naturally with overall feedback connected all around
edit:
if you are lazy for testing and matching,you can always send me handful of little critters and I'll do that for ya :devilr: |
editedit:
I just saw your edit............
put them back (caps)
this way you don't have CCSes anymore.......in fact -heaven knows what you have like this :clown: |
|
|
| Zen Mod |
I know;
besides-you wrote LSK on schematic?
hehe,regarding handful.........am not picky.........2SK or LSK........ :clown:
edit:
disregard my words about matching upper LSKs...........I just forgot that LSK is already pretty good matched:D .............late night..........or I'm just bussy with coffee ?
I told ya that I make a mess often :rolleyes: |
|
|
| dw8083 |
I re-edited the edited post, and replaced the schematic with C1 & C2 moved back outside of the resistors.
BTW, Linear Systems calls their jfets LSK rather than 2sk, so I just went with it. :)
It is nice to see someone making these jfets again. They also have a 2sk170 replacement, LSK170 in all three current levels.
Thanks again Choky! You da man!
-David |
|
|
| Zen Mod |
| quote: | Originally posted by dw8083
I re-edited the edited post, and replaced the schematic with C1 & C2 moved back outside of the resistors.
BTW, Linear Systems calls their jfets LSK rather than 2sk, so I just went with it. :)
It is nice to see someone making these jfets again. They also have a 2sk170 replacement, LSK170 in all three current levels.
....................
-David |
yup,I know that........but I'm too far from US of A ,and Europe distribution is shalala.........and Europe is still not part of Serbia:cool: ........I hope that these problems will be solved soon,at least for me ;) (Paypal,here I come!!):smash:
| quote: | | Thanks again Choky! You da man! |
you're wellcome......if you ever read Castaneda (old goat brujo) you know what's behind my left shoulder.........
but -I have Papa somewhere behind my right shoulder......mostly to remind me how little I know :clown: :devilr:
edit:
this days I must hack this site ,and switch >cool< to >devilr< ,so Papa will have adequate signature ..............
:cool: |
|
|
| Zen Mod |
btw- do you have them,and what's the price there? |
|
|
| dw8083 |
| Chocky, YGM! -David |
|
|
| Zen Mod |
| quote: | Originally posted by dw8083
Chocky, YGM! -David |
I don't know what this acronym is,but YT...........!
("you too" ,just in case :devilr: )
OK- to stay on topic (strange for me........I'll try to not embrace that nasty habit...stayin' OnT)........
in your boots-I'll change that sisy two J511 sink to ,say ,two BJT sink,or any other "led referenced" sink........why do not use parts easy to find in every corner tobacco store,especially when it is probably even better than some fancy schmancy part..........as J511 certainly is |
|
|
| jupiterjune |
Well, now you have to build it! It should not be too tough to wire point to point.
Did you run this design on a simulator?
Do you plan on feeding it single ended or balanced? It seemed to me higher source resistance caused my 'almost stock' circuit to perform not so good at converting an unbalanced input to a balanced output.
I believe the original BZLS article showed much lower distortion with lower gain (from more source resistance). It also had a pot in between r7 and r8. This pot doesn't really work well as volume control, unless, perhaps, you are driving a unity gain amp or follower.
Just curious about your plans for this circuit, and hoping to see learn the best way to do unbalanced to balanced conversion.
JJ |
|
|
| Zen Mod |
| quote: | Originally posted by jupiterjune
Well, now you have to build it! It should not be too tough to wire point to point.
Did you run this design on a simulator?
Do you plan on feeding it single ended or balanced? It seemed to me higher source resistance caused my 'almost stock' circuit to perform not so good at converting an unbalanced input to a balanced output.
I believe the original BZLS article showed much lower distortion with lower gain (from more source resistance). It also had a pot in between r7 and r8. This pot doesn't really work well as volume control, unless, perhaps, you are driving a unity gain amp or follower.
Just curious about your plans for this circuit, and hoping to see learn the best way to do unbalanced to balanced conversion.
JJ |
David or me?
me-not ..................hehe
David must do this ..............that's his homework |
|
|
| dw8083 |
Hi Choky, YGM... You got Mail. Look for a message from me to your DiyAudio email.
I'm planning on wiring it up and testing it. Just would like some more confirmation on the design. It would be great if someone had a spice sim.
I run balanced all the way, both sides source and amp.
-David |
|
|
| Zen Mod |
| quote: | Originally posted by dw8083
Hi Choky, YGM... You got Mail. Look for a message from me to your DiyAudio email.
............
-David |
I know.......JJ
just jokin' ;)
YT |
|
|
| dw8083 |
| quote: | Originally posted by Zen Mod
I know.......JJ
just jokin' ;)
YT |
ARRGH! That's the second time you've pulled my leg! I forget your sense of humor.
You make this board fun.
:D
-David |
|
|
| wuffwaff |
Hi,
just a few remarks/ questions:
you have 20dB of gain (2k2 /22k). This is a bit high for a preamp (even if you want to mate it with an F4)
The value of R5 and R6 can be calculated so the output sits at aproximately 10-11V (half the supply voltage)
The 100k´s could be moved to the gate again
Why use 2sk389 for the current sources?
William |
|
|
| stefanobilliani |
| quote: | Originally posted by Zen Mod
btw- do you have them,and what's the price there? |
I remember Linear System was kind enough to send me 5 pieces of LSK389 as samples
:cool: |
|
|
| Zen Mod |
| quote: | Originally posted by stefanobilliani
I remember Linear System was kind enough to send me 5 pieces of LSK389 as samples
:cool: |
hehe
they don't think in stereo sets?
;)
they charge for shipping,naturally? |
|
|
| Zen Mod |
| quote: | Originally posted by wuffwaff
Hi,
just a few remarks/ questions:
**you have 20dB of gain (2k2 /22k). This is a bit high for a preamp (even if you want to mate it with an F4)
*******he value of R5 and R6 can be calculated so the output sits at aproximately 10-11V (half the supply voltage)
The 100k´s could be moved to the gate again
Why use 2sk389 for the current sources?
William |
** really depends ; more mater of taste and need -you're right-easy to change with just pair of resistors
***and where needed current for lower CCS will pass,if you choke upper 389 (upper CCSes)?
R5 and 6 must be set for current sum |
|
|
| trango |
This is very interesting!
I'm afraid I'm not into electronic design very much so cannot contribute with any comments about the schematics, but I'm thinking about trying to build one.
Which type of LSK389 would be suitable for the schematics posted, type A, B or C?
I hope you can excuse my ignorance.
/Fredrik |
|
|
| Zen Mod |
| quote: | Originally posted by trango
This is very interesting!
I'm afraid I'm not into electronic design very much so cannot contribute with any comments about the schematics, but I'm thinking about trying to build one.
Which type of LSK389 would be suitable for the schematics posted, type A, B or C?
I hope you can excuse my ignorance.
/Fredrik |
current is important.
B or ,even better,C
btw-nobody is born smart......
so-just chill :):D |
|
|
| stefanobilliani |
| quote: | Originally posted by Zen Mod
hehe
they don't think in stereo sets?
;)
they charge for shipping,naturally? |
heheh ;) no it was complitely free , shipping included |
|
|
| dw8083 |
| quote: | Originally posted by wuffwaff
Hi,
just a few remarks/ questions:
you have 20dB of gain (2k2 /22k). This is a bit high for a preamp (even if you want to mate it with an F4)
The value of R5 and R6 can be calculated so the output sits at aproximately 10-11V (half the supply voltage)
The 100k´s could be moved to the gate again
Why use 2sk389 for the current sources?
William |
Hi William,
What would be your suggestion for values R1-R4, If the gain is too high?
I used the LSK389 for the current source I had them already. But you are right a mosfet or something else could be used. If you would like to suggest a better part, I'm open to ideas.
Thank you!
-David |
|
|
| Zen Mod |
| quote: | Originally posted by dw8083
Hi William,
What would be your suggestion for values R1-R4, If the gain is too high?
I used the LSK389 for the current source I had them already. But you are right a mosfet or something else could be used. If you would like to suggest a better part, I'm open to ideas.
Thank you!
-David |
ratio 2k2/22k gives you gain
make it 4k7/22k to halve it
for CCSes (both upper and one lower) you can use whatever you want.........but-stay with LSKs if that suits you,or make two BJT CCSes .........as in Babelfish,for instance.
boyz at toob forum are frenzy with LED referenced CCSes ,but I have no any valid link handy in this moment.
anyway- I'll choose CCS a la Babelfish before LSK one ; both- cheaper and better
hehe-and just for fun-when you going to make final version,make shunt regs for both sides ;) |
|
|
| Babowana |
>>O<<
Hey schematic, post#29, forgive me for my question.
How are you going to work with no input coupling cap?
:darkside: |
|
|
| Zen Mod |
| quote: | Originally posted by Babowana
>>O<<
Hey schematic, post#29, forgive me for my question.
How are you going to work with no input coupling cap?
:darkside: |
you're dumb almost as Serb .......... :devilr:
gates are referenced to gnd via 100K resistors
i REALY LOOK AT OTHER THINGS,PRdamn caps lock.....esuming that schematic is almost correct;
this way it will work ;)
R5 and R6 really needs trimming - to achieve 4,7 mA through each half of upper LSK, and this way- as in every proper susy ,output caps are included in fdbck loop........theoretically killing cap induced nasties (if any) through susy effect ;) |
|
|
| Babowana |
| quote: | Originally posted by Zen Mod
you're dumb
|
Heh . . . ^db^
Hey! Schematic above! Only two questions.
What is going to be your voltage gain? 100/33 < 22K/4.7K
You really going to waste potential thru D1 and D2?
:darkside: |
|
|
| Zen Mod |
| quote: | Originally posted by Babowana
Heh . . . ^db^
Hey! Schematic above! Only two questions.
What is going to be your voltage gain? 100/33 < 22K/4.7K
You really going to waste potential thru D1 and D2?
:darkside: |
re-read my previous post
100 will not be 100,they mmust be trimed for current of 4,7mA
D1 and D2 are CCSes-where is waste? |
|
|
| Babowana |
I have thought R5 and R6 as too low.
And, -22V minus few voltages will be used across D1 and D2.
How about introducing the input coupling caps and using
only +rail? If so, we could use lower V across D1 and D2,
and could have bigger R5/R6. Then, the voltage gain will
follow the value of 22K/4.7K . . .
Don't take this seriously. Just an opinion ^^.
:darkside: |
|
|
| Zen Mod |
| quote: | Originally posted by Babowana
I have thought R5 and R6 as too low.
And, -22V minus few voltages will be used across D1 and D2.
How about introducing the input coupling caps and using
only +rail? If so, we could use lower V across D1 and D2,
and could have bigger R5/R6. Then, the voltage gain will
follow the value of 22K/4.7K . . .
Don't take this seriously. Just an opinion ^^.
:darkside: |
I have no graphs for LSK and I can't know values of R5 and R6 for 4,7mA without graphs and without LSKs on my bench
they're there as CCSes,so I can't see connection between values of R5 and R6 and overall gain
degeneration resistors in sources of "active" (lower) LSK will have influence on OLG,certainly.
anyway-without graphs I can't tell most adequate voltage for plus and minus PSUs,but I presume that these voltages are good starting point.
you can always choose single sided supply ,as you say-implementing caps on inputs,moving 100K resistors away from gates......
why not-if that's your Zen.........
;)
you know that I don't take seriously anything.......... |
|
|
| dw8083 |
The input to the design is great!
I'm thinking of builting an integrated amp with my Aleph30. I thought I'd take advantage of the negative voltage, since it is there already.
The first batch of parts arrived today. Time to finish the design, and gas-up the soldering iron. ;)
-David |
|
|
| jupiterjune |
| quote: | | The input to the design is great! |
yeah, I don't think Choky and Bobo ever get to build anything themselves, their too busy helping everyone else out!
(no I didn't expect you to build it choky! Didn't I see you with some Krell clone PCB's?)
If you are going to use this with an Aleph 30, you might take a look at increasing the source resistance to reduce the gain. (I am assuming you are using this with a CD player not vinyl).
The BOSOZ has lower distortion with reduced gain (if you reduce the gain by adding source resistance, that is.) This circuit, I think, is more like an Aleph "J"P -- I am not sure if the aleph P also has lower distortion with more source resistance -- maybe one of the gurus out there can confirm.
Anyway, if your CD player is putting out 2 volts, I think you only need about 6dB of gain (if that) to get full output from the Aleph 30. I may have to dig up my tutorial sheet on dB calculations.
The fact that you have balanced sources is a benefit. I saw some problems with unbalanced to balanced conversion with the BOSOZ. But then, this is not really a BOSOZ.
JJ |
|
|
| Zen Mod |
| quote: | Originally posted by jupiterjune
................
(no I didn't expect you to build it choky! Didn't I see you with some Krell clone PCB's?)
JJ |
yup;
KSA100 pcbs-present from my friend AR.......I'll build it just because I can,not because I need it ;)
| quote: | | This circuit, I think, is more like an Aleph "J"P -- I am not sure if the aleph P also has lower distortion with more source resistance -- maybe one of the gurus out there can confirm. |
not that I'm guru........just not too much modest student .....with more source degeneration - local feedback is greater,so also distortion is decreased ; where is sweet spot ? that's really matter of compromise -what you need -greater OLG or lesser already small distortion....ratio between OLG and CLG or local feedback and overall feedback......for my ears -too much of both aren't good
| quote: | | If you are going to use this with an Aleph 30, you might take a look at increasing the source resistance to reduce the gain. (I am assuming you are using this with a CD player not vinyl). |
I already build X-CSCCSWHATEVER BOSOZ,in fact -Chokyfied salvaged pair of KK's pcbs far Balanced BOZ or what it really was.........that was a favor to little less xperienced friend who spent a little fortune on parts and thing just didn't work........
now he is a happy camper ;
this (jfet) iteration I'll not use for anything ;
I'll make toob susy to compare with my-for now - reference WOT preamp........I have some other intention with tiny jfets,besides Babelfish thingies
we'll see for vinyl; I have plenty of TTs,so I'll certainly make not just one new phono preamp;LR EQ tube one is a must,all other can be just for fun
| quote: | | The BOSOZ has lower distortion with reduced gain (if you reduce the gain by adding source resistance, that is.) |
I wrote already.......depending on taste ; no free lunch
| quote: | | Anyway, if your CD player is putting out 2 volts, I think you only need about 6dB of gain (if that) to get full output from the Aleph 30. I may have to dig up my tutorial sheet on dB calculations. |
depending what speakers are,..........and-what if someone want to use line stage like this for FW4? :devilr:
my spks are almost sensitive as good mic........:clown:
| quote: | | The fact that you have balanced sources is a benefit. I saw some problems with unbalanced to balanced conversion with the BOSOZ. But then, this is not really a BOSOZ. |
problems can be only if you didn't match impedances ......... |
|
|
| Zen Mod |
that's for babo
pretty tired now-I'll double check it tomorrow
if someone see something funny-just say so......... |
|
|
| Babowana |
| quote: | Originally posted by Zen Mod
pretty tired now
|
I know you are tired.
Babo singles (2SK370) not really to be matched as there are
the caps blocking dc to ouside world. The matching is always
preferable tho . . .
I'm sending you some of 2SK370 as a small gift pakage. Try them.
JJ is complaining you not building any.
JJ, I have learned many different things in this place,
I'm just trying to return part of them.
I always have weekend, and I'm going out for beers, cigars and
live music -- Blues, Jazz, Salsa . . . when Papa spends time in
the kitchen ^^.
Papa! life is short, enjoy too! ^^.
:darkside: |
|
|
| apassgear |
Hey Choky,
Why would you need 4K7 resistors at the input?
Toob guys usualy don't use them if they can be avoided :D
Cute circuit BTW :up: |
|
|
| Zen Mod |
| quote: | Originally posted by apassgear
Hey Choky,
Why would you need 4K7 resistors at the input?
............. |
ya must be joking me...................;) |
|
|
| Zen Mod |
for babo-checked
if someone see something unusual,I'm dumb :devilr: |
|
|
| apassgear |
| Now it makes sense... ;) |
|
|
| apassgear |
| Those 1u caps seems rather big for the purpose to me, I would test something smaller like 0.1 if you are looking for noise suppresion. |
|
|
| Zen Mod |
| quote: | Originally posted by apassgear
Now it makes sense... ;) |
:clown: |
|
|
| Zen Mod |
| quote: | Originally posted by apassgear
Those 1u caps seems rather big for the purpose to me, I would test something smaller like 0.1 if you are looking for noise suppresion. |
you're right.....I have u22 in Babelfish,along with 33K;
but-it's pretty irrelevant-1u or u1 if cap is of good quality |
|
|
| apassgear |
| babo is :drink: or :spin: or :sleep: |
|
|
| apassgear |
| quote: | Originally posted by Zen Mod
but-it's pretty irrelevant-1u or u1 if cap is of good quality |
not the same for the purpose... but I'm sure we could not hear the difference for that matter... so be it...
we could go with the digital guys for oppinions... :clown:
BIB? :D |
|
|
| Zen Mod |
| quote: | Originally posted by apassgear
....................
we could go with the digital guys for oppinions... :clown:................
:D |
why?
I'm still listenin' with my ears ,not with theirs ;)
almost always ;) |
|
|
| Vix |
I was thinking about deleting input caps and makin' it with negative rail. One more change, can CCS be LED referenced? By the way, I'd increase the input resistor to 10-20 K and feedback one to 47K
Choky, can you help with a schematic, I am not good at drawing... |
|
|
| Zen Mod |
| quote: | Originally posted by Vix
I was thinking about deleting input caps and makin' it with negative rail. One more change, can CCS be LED referenced? By the way, I'd increase the input resistor to 10-20 K and feedback one to 47K
Choky, can you help with a schematic, I am not good at drawing... |
you mean -babo- dedicated with dual rails? |
|
|
| Babowana |
Babo should be with single rail ^^.
No time for beer tonight, and no sleep either, because of the
reason given in another thread ^^.
:darkside: |
|
|
| Vix |
| quote: | Originally posted by Zen Mod
you mean -babo- dedicated with dual rails? |
Exactly. Dual Babo, with (red) LED referenced CCS. :clown:
thanks :) |
|
|
| Zen Mod |
| quote: | Originally posted by Vix
Exactly. Dual Babo, with (red) LED referenced CCS. :clown:
thanks :) |
I like this type of ccs ;)
I can do LED variant later tonight |
|
|
| Vix |
| quote: | Originally posted by Zen Mod
I like this type of ccs ;)
|
Yeah, I see :) I guess it works fine.
I have some red LEDs in my junkbox, and every time I digged there I imagined those LEDs doing some useful work (what better than referencing CCS).
Thanks ;) |
|
|
| Zen Mod |
| quote: | Originally posted by Vix
Yeah, I see :) I guess it works fine.
I have some red LEDs in my junkbox, and every time I digged there I imagined those LEDs doing some useful work (what better than referencing CCS).
Thanks ;) |
what about xmass tree?
I mean-out of any amp?
John Cleese: "what's wrong with ol' plain kiss?" |
|
|
| Vix |
| quote: | Originally posted by Zen Mod
what about xmass tree?
I mean-out of any amp?
|
Xmas tree? Or a mid-fi amp, with many glowing red LEDs, just below a tree? When you look at such an amp, don’t you think “they’d better put them inside”.
| quote: | John Cleese: "what's wrong with ol' plain kiss?"
|
..Yes sir. Sorry sir. :D |
|
|
| dw8083 |
Here's a possible circuit based on Nelson's chewy hints from the top of the thread.
Opinions?
-David |
|
|
| dw8083 |
WARNING! I just started using LTSpice. I have only a particial clue as to what I'm doing with with this very powerful tool. Hey I'm here to learn. What a great hobby.
I hope I attached the right signal sources to the inputs.
-David |
|
|
| Zen Mod |
| quote: | Originally posted by dw8083
WARNING! I just started using LTSpice. I have only a particial clue as to what I'm doing with with this very powerful tool. Hey I'm here to learn. What a great hobby.
I hope I attached the right signal sources to the inputs.
-David |
I'm dumb for proggies like that;
OK-I'm more lazy........used to ruler and calculator
keep it simple,but not simpler than needed.
you forgot input caps...........
anyway-this circ is more than simple enough ,so you can build it and try without sims;
better put some energy in constructing good enough shunt reg.
and-stay with bipolar supply (+/-) ,just because then you can omit input caps.
all in all-we already have starting schematic-now you can try to implement folowers |
|
|
| Vix |
Hi,
While looking at the schematic that Choky provided in post #78, I was thinking how to modify this circuit to suit the parts that I have in my drawer. And this is the result. The circuit is almost identical to Choky's, I just cascoded a Jfet. While this may seem unnecesary, the reasons were:
I already have a salvaged amp that has a working PSU, giving around +-50 V, so I wanted to use it. I have more Bipolars than mosfets in my drawer, that's why I prefer Choky's CCS. I have LU1014D, so while that my not be the preffered part for a preamp, it should give satisfactory performance in the cascode, as in ZV9.
I do need your help in determining the correct values for R1, R2 and R3 and I was not sure about the value of the Zener diodes, I guess they should be around 6.2 V?
Thank you very much,
Vix |
|
|
| Zen Mod |
| quote: | Originally posted by Vix
Hi,
While looking at the schematic ........................
Vix |
tonight.............
in a meantime this can be handy |
|
|
| Steve Eddy |
| quote: | Originally posted by dw8083
Here's a possible circuit based on Nelson's chewy hints from the top of the thread.
Opinions? |
Why not use LSK389s instead of separate 170s that would have to be matched?
se |
|
|
| Zen Mod |
| quote: | Originally posted by Steve Eddy
Why not use LSK389s instead of separate 170s that would have to be matched?
se |
dual or separates........that really depends only of what you are able to have/find
in fact-also p fets vs. n fets.........just rotate everything (flip vertically) and other type can be used |
|
|
| jupiterjune |
ZM| quote: | | with more source degeneration - local feedback is greater,so also distortion is decreased ; where is sweet spot ? that's really matter of compromise -what you need -greater OLG or lesser already small distortion....ratio between OLG and CLG or local feedback and overall feedback......for my ears -too much of both aren't good |
This is a very interesting thought, I had read in EB's article that the source resistor increased local feedback, but I didn't really understand the mechanism and I never considered using this circuit at very low gain might bring with it the negative attributes of mucho negative feedback......
Babo--sorry, I didn't mean to complain, you guys are aces in my book!
Vix| quote: | | and I was not sure about the value of the Zener diodes, I guess they should be around 6.2 V? |
The best method I found is to substitute a small pot in place of that zener. Then tweak it until input sine waves are not flat on top or bottom, but not too peaky either. Usually it is only one or two volts more and the circuit starts oscillating. |
|
|
| dw8083 |
| quote: | Originally posted by Steve Eddy
Why not use LSK389s instead of separate 170s that would have to be matched?
se |
I'm actually going to use LKS389's, I just have not taken the time to build the part in my schematic capture program, so I used LSK170's instead.
BTW, I'm very impressed with SWCad III (LT Spice) from Linear.
http://www.linear.com/designtools/s...egistration.jsp
I'm happy to share my schematic with anyone who would like to use SwitcherCad to modify the circuit. The sowftare is free.
-David |
|
|
| Zen Mod |
| quote: | Originally posted by Vix
Hi,
While looking at the schematic that Choky provided in post #78, I was thinking how to modify this circuit to suit the parts that I have in my drawer. And this is the result. The circuit is almost identical to Choky's, I just cascoded a Jfet. While this may seem unnecesary, the reasons were:
I already have a salvaged amp that has a working PSU, giving around +-50 V, so I wanted to use it. I have more Bipolars than mosfets in my drawer, that's why I prefer Choky's CCS. I have LU1014D, so while that my not be the preffered part for a preamp, it should give satisfactory performance in the cascode, as in ZV9.
I do need your help in determining the correct values for R1, R2 and R3 and I was not sure about the value of the Zener diodes, I guess they should be around 6.2 V?
Thank you very much,
Vix |
vix
looking at pdf I attached (sorry- I forgot who is here responsible for measuring and drawing that.........Respect!!),seems that LU is good for nothing bellow 300mA........that's 300mA per side,600mA through lower CCS ............meaning that R3=Ube/I = 0V6/0,6 =1E.......plenty for veeeny BD140
even with 300mA (where R1 and R2 are 2E) BD140 will be fried........
you can stay with tiny Jfets for this
.........
but-saythat you can live with some (!!?) NONLINEARITY........CAPSLOCK,then you can try with 100mA per side....R1 3E,R2 and R3 6E.........
use bc556C ,use bd140-16.....and zenners..........(0,1Ax33 +1V4 + 4V) = 8V7..............what is nearest standard value?
anyway-try.........but check dissipations for those bds and trim PS voltages accordingly |
|
|
| Zen Mod |
to vix again.........with these currents-time for reworking entire schematic......other (more robust) CCS........
anyway-just as illustration-my variant of famous (hehe) xccsblswhatever preamp for my friend Igor.........singing nice tunes along with Haf9505 on Fried Coffins etc.
in your boots-I'll make something leaned on this schematic......LUs,also CCS-ed above,300mA per side,minimum needed PS voltages |
|
|
| Vix |
| quote: | Originally posted by Zen Mod
in your boots-I'll make something leaned on this schematic......LUs,also CCS-ed above,300mA per side,minimum needed PS voltages |
Thanks!!
In that case it would start looking more like micro-X-ZV9 with dual rails. :) :cool:
If I got it right, there will have to be around 600 mA from the lower CCS. (So there would be 1 ohm sensing resistor to set a current of 660 mA)
If the upper CCS-es are used (like those in Aleph P 1.7) instead of 680 R resistors, how much current shall flow through each of them?
The last schematic looks interesting. It strikes me that it doesn't use source resistors, so you have direct X. I don't want to go OT, but it would be nice to know how it compared against plain BOSOZ, or CCS-X-BOSOZ, (with no cascode and source resistors). Maybe that's covered in some other thread, but I couldn't find it...
Again, thanks a lot!
BTW, I think its time that Nelson writes a nice paper about the preamplifier as at the beginning of this thread. It could be added to a family of firstwatt amplifiers. First milliwatt is also important:)
Regards,
Vix |
|
|
| Tom2 |
| quote: | originally posted by Zen Mod
...seems that LU is good for nothing bellow 300mA........ |
I built this test circuit a few months ago.
It's just an experiment.
The lu1014 seems to work down in the mA range.
Tom |
|
|
| Zen Mod |
| quote: | Originally posted by Tom2
I built this test circuit a few months ago.
It's just an experiment.
The lu1014 seems to work down in the mA range.
Tom |
at the botom of transfer curve?
yup
good for milivolts of input voltage
make B class mini-F4-like this way ;) |
|
|
| dw8083 |
Attached is a revised schematic with some simulation numbers. I'm experiencing issues with the simulation. At the output I'm only getting +-8mv.
Any ideas or comments would be helpful.
-David |
|
|
| Tom2 |
| quote: | Zen Mod said:
good for milivolts of input voltage
|
Thinking about mc gain stage. The input capacitance might affect
the loading of the cartridge. Easier to use 170's 389's though.
Also the pdf file of the transfer function is at a single value of
Vds. It would be interesting to see a family of transfer functions
as a function of Vds. This would show the "triodeness" of the
device.
Yes your right, nonlinear section. Maybe thats good?
Maybe the fets might make a good gravity wave detector
amplifier put into a liquid helium bath to lower the 1/f noise.
Tom |
|
|
| dw8083 |
Attached are simulation graphs for the schematic in post #96. Part 1 of 3 attached.
Thank you for any comments and guidance.
-David |
|
|
| dw8083 |
Attached are simulation graphs for the schematic in post #96. Part 2 of 3 attached.
Thank you for any comments and guidance.
-David |
|
|
| dw8083 |
Attached are simulation graphs for the schematic in post #96. Part 3 of 3 attached.
Thank you for any comments and guidance.
-David |
|
|
|
