Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Analog Line Level Preamplifiers , Passive Pre-amps, Crossovers, etc.

BJD BUFFER stage for passive preamps
BJD BUFFER stage for passive preamps
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 5th February 2018, 12:59 PM   #1
kinku is offline kinku  United States
diyAudio Member
 
Join Date: Feb 2013
Location: Maryland,USA
BJD BUFFER stage for passive preamps
Default BJD BUFFER stage for passive preamps

Hi Pals,
Few months ago I started this journey with B1 buffer stage. A simple and successful design by Mr.Pass.
Further research lead me to Feucht buffer.
Then I got tips from EUVL's design which basically a version of Feucht buffer without tracking capabilities.(My understanding,but may not be true!)
So after few weekends here I am with this design.
So the disadvantage of B1: running the jfet's at saturation, there is not much voltage swing to play with before distortion products goes up .
EUVL design (post #5 ) runs them at lower ID gives a little more headroom.
B1-Turbo on a Chip


Now being running at or near saturation B1 has an inherent advantage of PSRR. But when jfets working below Idss means they need very good power supply else power supply ripple will be transferred to audio signal.


Now being said all these, I took an excellent power supply design by Jan Didden (with his permission) and made this buffer stage that basically is EUVL's design from Feucht's.
Since this design uses BF862 and running them near 80%-90% of Idss means - cooling. It is challenging to cool this rice grain sized fets. They are thermally coupled through their gate pin. But to dissipate near 250mW(18V supply at 13ma Id) the area of 1oz copper pad needed is few square centimeters. You can glue a small heatsink on to them but those glues are much less efficient in heat transfer.
So here I am with this design ,I am planning to use small copper squares and fix them to board using nylon screws with heatsink paste between the copper squares.Nylon is not going to melt before 200C and before that jfets will go into smoke.
But to thermally track coupled JFET devices I will still use a small copper wire glued to them on top( or rather bolt them if I can work around HOW to do it,provided M1 screw hole for it).Well all this is to reduce Dc offset which is blocked by a nice capacitor at output.
On board there is a trim pot to use as a single Feucht buffer which will dial down both source resistors to match. 0806 and 0603 options covers together a lot of 0.1% smd thin film combinations.
If you want to increase current for driving longer cables etc use the EUVL route by parallel use of 2 matched pairs of JFETS (on both sides of PCB).But SKIP the trimmer. Assuming you will use a nice cap at output to block any DC offset.
The output impedance should be in 20ohm range. Simulation shows predominant second harmonics with tad bit lower distortion figures than B1 design. But magnitude is in 10x-7so does not matter.
Let me know your thoughts.....
Attached Images
File Type: jpg back.jpg (118.4 KB, 358 views)
File Type: jpg front.jpg (240.6 KB, 359 views)
File Type: jpg pcb layout.jpg (411.8 KB, 352 views)
File Type: jpg SCH.jpg (432.0 KB, 358 views)
File Type: gif tran18.gif (14.2 KB, 348 views)

Last edited by kinku; 5th February 2018 at 05:50 PM.
  Reply With Quote
Old 6th February 2018, 05:28 AM   #2
ammel68 is offline ammel68  United States
Banned
 
Join Date: Sep 2009
Location: Southeast, U.S.
BJD BUFFER stage for passive preamps
Quote:
Originally Posted by kinku View Post
Hi Pals,
Few months ago I started this journey with B1 buffer stage. A simple and successful design by Mr.Pass.

Since this design uses BF862 and running them near 80%-90% of Idss means - cooling.

0806 and 0603 options covers together a lot of 0.1% smd thin film combinations.
But SKIP the trimmer. Assuming you will use a nice cap at output to block any DC offset.

Let me know your thoughts.....
Hmm...my thoughts.

First, as you stated the B1 may be a successful design by Mr. Pass, but all my examples both AC and DC-coupled have resulted in less-than-stellar SQ.
Adjectives such as "sterile" and "lifeless" come to mind.
ANY design that uses only 2 devices(JFET or BJT) per channel is going to result in $h_tty SQ.
Unfortunately, the old design philosophy that "simpler is better" just doesn't seem to apply to any 2-4 transistor designs that I've tried.

Second, I wouldn't run those cheesy little BF862s at 80-90% of their Idss rating unless you're desiring to shorten their lives and have one or more of them to self destruct.

Third, if you enjoy soldering 0806 and 0603 SM resistors, I say "Go for it Dude".
I won't attempt to solder anything smaller than 1206 and there are plenty of .1%(not that it really matters for this design) thin film resistors in that larger size.

Fourth, rather than use a "nice cap" at the output, why don't you figure out how to better match the devices so that NO cap is needed at the output.
Remember, coupling caps blow and should always be avoided if possible.

Now...let's bust out the moonshine and soldering iron and build something!
  Reply With Quote
Old 6th February 2018, 10:37 AM   #3
AndrewT is offline AndrewT  Scotland
R.I.P.
 
Join Date: Jul 2004
Location: Scottish Borders
Cascode will remove the high dissipation concern of the very small 862 and will help with PSRR.

Each device (4 in total for a paralleled CCS+Follower) needs it's own selected source resistor to try to get Id for each close to identical. Close enough may be trimmers of the lower Rs on both halves. When output offset becomes zero for one half, then you know that Id for that half has been set up accurately.

My eyes need reading glasses, and I have soldered 805/603/402 with optical aids without any real problem.
__________________
regards Andrew T.

Last edited by AndrewT; 6th February 2018 at 10:40 AM.
  Reply With Quote
Old 6th February 2018, 10:41 AM   #4
AndrewT is offline AndrewT  Scotland
R.I.P.
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally Posted by ammel68 View Post
..............
Remember, coupling caps blow and should always be avoided if possible....................
Could you explain?
__________________
regards Andrew T.
  Reply With Quote
Old 6th February 2018, 11:55 AM   #5
kinku is offline kinku  United States
diyAudio Member
 
Join Date: Feb 2013
Location: Maryland,USA
BJD BUFFER stage for passive preamps
Ammel, valid points. But remember this is just a buffer and so is B1. There are lot of satisfied people who use them. What was the potentiometer type you used in your design? What cables you used ?
Running them at 90% of Id may not be an issue of they are properly heatsinked. They are rated for 300mW and even at full Idss they only reach 270
-290 mW if you chose Less than 16 Idss devices. Not sure running a transistor hot necessarily means a short life span, it is a relative concept. Unless you have ran them in both ways with enough sample size it is just an observation from you.
Soldering 0603 may be a challenge but 0805 is not for most people. Well we are taking about using soldering paste and SMD soldering station. You can buy them for less than 60$ from Amazon . Most of us will spend double the money buying boutique components for our carnal pleasures. If you have not it is kind of late, don't you think?
No caps is still possible if you have perfectly matched Jfets not just IDSs and Vp . Do you know how to use a curve tracer ? Most people here don't . And doing that with SMD should be challenging too. A cap in signal path is not a dreaded thing to have.
We should stop thinking soldering SMD parts using regular iron , yes it is a mess. I don't think even soldering those Jfets with a regular iron is easy.1206 package is better but also expensive( again relative word). But defenitely a good suggestion. When I searched it was easier and cheaper to find 0805 and 0603 at mouser with lot more options.
Andrew thanks for suggestions. Yes all the source resistors are independent. I thought about using a trimmer for both sides. But practically it is tough to access both sides of the board and also may make it challenging to fine tune. Also Feucht paper recommend using only 1% matched resistors and with SMD resistors we have 0.1% available.
thank you guys.

Last edited by kinku; 6th February 2018 at 12:06 PM.
  Reply With Quote
Old 6th February 2018, 11:58 AM   #6
stellavox is offline stellavox  United States
diyAudio Member
 
Join Date: Jan 2007
Location: Oneonta, NY; USA
Excuse me Ammel - but what is "SQ"?
  Reply With Quote
Old 6th February 2018, 02:49 PM   #7
kinku is offline kinku  United States
diyAudio Member
 
Join Date: Feb 2013
Location: Maryland,USA
BJD BUFFER stage for passive preamps
Quote:
Originally Posted by stellavox View Post
Excuse me Ammel - but what is "SQ"?
I believe it is Sound Quality!
  Reply With Quote
Old 6th February 2018, 03:43 PM   #8
Calvin is offline Calvin  Germany
diyAudio Member
 
Calvin's Avatar
 
Join Date: Nov 2004
Location: close to Basel
Hi,

soldering the BFs SOT23 casing should be no more difficult than soldering a TO92 throughhole transistor .... they share almost same pad sizes and distances.
The problem with temperature and JFETs is gate leakage.
Unfortunately gate leakage is often not or grossly unpractically specified in Datasheets.
Running JFETs at high Id and high temperature condition lets the leakage rise over due.
If the gate resistor is chosen large (quite often 1M and more) the rising leakage can drive the JFET into a latch up, or in other words destruction.
As a rule of thumb the drain-source voltage should not exceed 1/4 of the specced max value when You intend to run the JFET 'hot'.
If You can't reduce the supply rails then cascoding is a option -also reducing THD considerably- or adding a power booster (Darlington or Sziklai) and reducing the JFETs Id.
The latter case is what I did in my Calvin Buffer .... allowing to bias the JFET lower, close to its 0-tempco point, and still benefitting from the current drive capabilities of a bipolar booster transistor.
No need for a parallel configuration and the hassle with trimming 4 or more bias currents to match.

jauu
Calvin

ps. never seen a coupling cap blow up ..... must be another one of those highend fairytales
  Reply With Quote
Old 6th February 2018, 04:37 PM   #9
kinku is offline kinku  United States
diyAudio Member
 
Join Date: Feb 2013
Location: Maryland,USA
BJD BUFFER stage for passive preamps
Thank Calvin. Appreciate your comments. Always nice to hear from you.
I will take my comments back on soldering those jfets,I should have rather said handling than soldering.
I have chosen 200R for gate resistance.
Now about running hot, So these jfets pump out heat through their gate pins than over there slender bodies, is that correct?
I found this calculator
The CircuitCalculator.com Blog >> PCB Thermal Copper Area


Attached screen shot of my calculation. So it needs about 500 square mm of 1.4mils(0.036mm) copper pad to keep junction temp at 60degrees.
My pad area is about 40 mm square, but heat sinking fortunately is a volume issue. So assuming 18 cubic mm of copper pad (500 x 0.036) can keep the junction at 60degree ,I can use a 0.5 mm copper plate and get more volume for cooling (40mmx0.5=20 cubic mm). Ok will use two 0.5 mm copper plates making it double the volume(40) to cool!


I definitely has plans to make your buffer Calvin. But this design only uses local feed back and all others including yours uses global feedback, am I right?
Not sure how to cascode Jfets into a buffer. If anyone has a circuit with bf862 happy to adapt it.
  Reply With Quote
Old 6th February 2018, 05:12 PM   #10
kinku is offline kinku  United States
diyAudio Member
 
Join Date: Feb 2013
Location: Maryland,USA
BJD BUFFER stage for passive preamps
Sorry here is the screen shot.
Attached Images
File Type: jpg thermal.jpg (137.0 KB, 88 views)
  Reply With Quote

Reply


BJD BUFFER stage for passive preampsHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Tube and Passive Preamps-Auto Transformers for Attenuators/Volume Control - Overview tiefbassuebertr Parts 9 10th June 2018 11:42 AM
The THD+N of passive preamps mtoc Analog Line Level 6 16th June 2016 07:12 AM
WTB: Buffer passive preamplifier ad010685 Swap Meet 5 2nd January 2016 03:40 PM
Passive Pre + F5 + Sub-woofer need buffer? CLiu Pass Labs 28 18th August 2015 02:56 AM
What causes more disagreement? Religion? Politics? Passive preamps? franjy Parts 2 4th March 2005 07:06 PM


New To Site? Need Help?

All times are GMT. The time now is 08:55 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki