BUF634 Open Loop with Single Supply - diyAudio
Go Back   Home > Forums > Amplifiers > Headphone Systems
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Headphone Systems Everything to do with Headphones

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 2nd November 2015, 02:54 PM   #1
diyAudio Member
 
Join Date: Sep 2010
Default BUF634 Open Loop with Single Supply

Can I use a BUF634 with open loop configuration with a Single Supply?

I want to use V+ = 8V and V- = Ground.

Do I have to use a virtual ground at the BUF634 Input ?

Do you know some schematics with the BUF634 using single supply?

Regards.
Alfredo Mendiola
  Reply With Quote
Old 2nd November 2015, 03:11 PM   #2
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: The great city of Turnhout, BE
Blog Entries: 7
You can use a single supply, but check in the data sheet whether it will work with 8V total supply - IIRC 10V total is minimum for guaranteed performance.

You want to couple the signal to the input via a cap, and use a resistive divider to set the DC at the input to half supply. The output will then also be centered around half supply and you need to capacitively couple it to the load.

I assume you know it is a buffer with gain of 1 x? So 'virtual earth at input' cannot be used as this would set Vin at zero

Jan
__________________
Whether we like to think of it this way or not, an audio engineer shares the professional goal of a magician - Richard Heyser
Linear Audio Vol 11 is out!
  Reply With Quote
Old 2nd November 2015, 03:34 PM   #3
diyAudio Member
 
Join Date: Sep 2010
Quote:
Originally Posted by jan.didden View Post
You can use a single supply, but check in the data sheet whether it will work with 8V total supply - IIRC 10V total is minimum for guaranteed performance.

You want to couple the signal to the input via a cap, and use a resistive divider to set the DC at the input to half supply. The output will then also be centered around half supply and you need to capacitively couple it to the load.

I assume you know it is a buffer with gain of 1 x? So 'virtual earth at input' cannot be used as this would set Vin at zero

Jan
Yes I don't need Gain, just transform 200 ohm output impedance to 50 ohm impedance.

I Will use virtual ground from an opamp output (Vcc/2) with a 200 ohm resistor to have an input impedance of 200 Ohm. The virtual ground will be connected to the Buf634 input after the decoupling capacitor.

I will use a 50 ohm resistor at the BUF634 output to have an output impedance of 50 Ohm.

Is it Ok?

I don't understand this part:

So 'virtual earth at input' cannot be used as this would set Vin at zero

Regards.
Alfredo Mendiola Loyola
  Reply With Quote
Old 2nd November 2015, 06:02 PM   #4
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: The great city of Turnhout, BE
Blog Entries: 7
Quote:
Originally Posted by amendiola View Post
Yes I don't need Gain, just transform 200 ohm output impedance to 50 ohm impedance.

I Will use virtual ground from an opamp output (Vcc/2) with a 200 ohm resistor to have an input impedance of 200 Ohm. The virtual ground will be connected to the Buf634 input after the decoupling capacitor.

I will use a 50 ohm resistor at the BUF634 output to have an output impedance of 50 Ohm.

Is it Ok?

I don't understand this part:

So 'virtual earth at input' cannot be used as this would set Vin at zero

Regards.
Alfredo Mendiola Loyola
I think the problem is that you use the term 'virtual ground' in a way I am not familiar with.
Normally 'virtual ground' is a circuit node that is not connected to ground but has zero signal, like the -in terminal of an inverting opamp circuit. So I don't see how a virtual ground can feed a buffer. But you probably use the term in a new unknown way.

Can you post a schematic?

Jan
__________________
Whether we like to think of it this way or not, an audio engineer shares the professional goal of a magician - Richard Heyser
Linear Audio Vol 11 is out!

Last edited by jan.didden; 2nd November 2015 at 06:05 PM.
  Reply With Quote
Old 2nd November 2015, 10:39 PM   #5
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
Quote:
Originally Posted by amendiola View Post
Can I use a BUF634 with open loop configuration with a Single Supply?
Keep in mind that the BUF634 (and LME49600 for that matter) have a huge output DC offset voltage if not used within the loop of an op-amp. 30-100mV from the datasheet excerpt below. That is likely enough to damage headphones and IEMs over time.

Since you will have to use a resistive divider anyway at the input to set the DC input at Vcc/2 for single supply usage, like Jan Didden said, you might be able to do something really slick here with a DC servo as an alternative to looping with an op-amp. You may be able to replace the bottom (midpoint to ground) resistor of the voltage divider with a transistor set up an a DC servo circuit, or a DC precision op amp (look at the OPA140) set up as the same. That would adjust the BUF634 input to not only be Vcc/2, but also would take that 30-100mV of DC output offset into account. With an OPA140 you should be able to adjust than down to less than 200 microvolts.

The BUF634 will fine on 8V. Its good down to +/-2.25Vdc = 4.5V when used as a single supply.
Attached Images
File Type: jpg buf634_offset_1.jpg (85.3 KB, 206 views)

Last edited by agdr; 2nd November 2015 at 10:41 PM.
  Reply With Quote
Old 3rd November 2015, 12:17 AM   #6
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
Whoops - I forgot that with a single supply you will have a coupling capacitor on the output, so that 30-100mV DC offset of the buffer won't make it out to the real world.

What will be a problem though (specific to these single supply designs) is the leakage current of the output capacitor. It can be substantial during the first minute or so after the DC voltage hits (there is a specification in the data sheet) and the electrolyte barrier forms, enough to make a "pop" in the headphones initially. You can minimize that by trying to find "low leakage" electrolytics at Mouser or Digikey. The pop shouldn't be nearly as damaging to headphones though as the 30-100mV of continuous DC offset out of the buffer would have been. It is a good idea to put some large value minimum load resistor on the amp output (parallel to the headphone), like 2K or so, to allow the output capacitor to charge and the barrier to form even if the headphones are not plugged in.

Here is a schematic I just whipped up, below.

The 20K resistors set the DC voltage at Vcc/2 like Jan said, 4V in this case. They are essentially in parallel as far as the input signal is concerned (the Vcc power rail is effectively an AC short) so that gives you a 10K input impedance. 10K plus the 10uF capacitor gives you an input corner frequency of around 1.5Hz.

The 270R resistor is in there for two reasons. It forms an RF filter with the 620pF capacitor and also reduces the bandwidth to the audio range. If you run this sim with no 270R and 620pF the amplifier corner frequency is around 11MHz - the Buf634 is a wideband part. Amplifying anything above 100Khz or so can only lead to trouble for audio work.

The other purpose of the 270R is short-circuit current limiting. In the datasheet for the 634 they explain that the part has input protection diodes that get forward biased if the output is shorted. That happens for an instant when you plug a TRS headphone plug in the headphone output jack. Here it happens for an instant when that output capacitor is initially a short at power-up. In this circuit it really doesn't matter, but if you had an op amp powering the buf634 input it might.

The 2200uF output cap plus a 32R load gives a matching corner frequency of around 1.5Hz on the output.

In the transient plot note that the input (blue) and output (green) are both centered around zero, thanks to the coupling capacitors on the input and output. But the output of the buf634 itself (red) is centered around that Vcc/2 point, as it should be.
Attached Images
File Type: jpg buf634 no servo schematic.jpg (101.6 KB, 192 views)
File Type: jpg buf 634 no servo transient plot.jpg (51.2 KB, 182 views)
File Type: jpg buf 634 no servo ac plot.jpg (39.1 KB, 185 views)

Last edited by agdr; 3rd November 2015 at 12:38 AM.
  Reply With Quote
Old 3rd November 2015, 12:50 AM   #7
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
Default 200R input Z

I just remembered you wanted to translate impedance from 200 ohms. In the schematic above you could use 390R virtual ground resistors, instead of 20K, along with a 560uF input cap instead of 10uF to keep the corner frequency around 1.5Hz.
  Reply With Quote
Old 3rd November 2015, 06:38 AM   #8
diyAudio Member
 
Join Date: May 2010
Location: Skokie Il
With one more capacitor and one more resistor, you can decouple the buffer's input from the power supply. You can easily set the input impedance with the decoupling resistor.

I am curious how well this circuit will work. I recently built a single ended unity gain headphone amp using a BUF634 and a 5532. It is hardly more complicated than your circuit. It definitely sounds better with more voltage; 25 volts seems good.

The single ended LME49600 circuit I built has a voltage gain of 2, and it works great from a 12-15 volt supply; higher voltage does not improve the sound into low impedance, high efficiency headphones, although it is designed to work on a supply as high as 30 volts.

The datasheet circuit along with good decoupling and a good ground layout is all you need.
  Reply With Quote
Old 3rd November 2015, 10:03 AM   #9
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: The great city of Turnhout, BE
Blog Entries: 7
Quote:
Originally Posted by agdr View Post
What will be a problem though (specific to these single supply designs) is the leakage current of the output capacitor. It can be substantial during the first minute or so after the DC voltage hits (there is a specification in the data sheet) and the electrolyte barrier forms, enough to make a "pop" in the headphones initially. You can minimize that by trying to find "low leakage" electrolytics at Mouser or Digikey. The pop shouldn't be nearly as damaging to headphones though as the 30-100mV of continuous DC offset out of the buffer would have been. It is a good idea to put some large value minimum load resistor on the amp output (parallel to the headphone), like 2K or so, to allow the output capacitor to charge and the barrier to form even if the headphones are not plugged in.
Not to pick nits but this has nothing to do with cap leakage. At switch-on, the capacitor has to be charged to half supply and that is what causes the plop. This will happen with ANY capacitor here, low leakage, high leakage, film, whatever type.

With a 300 or less load headphone the leakage even at a few 10's of uA will at any rate not be a problem.

Jan
__________________
Whether we like to think of it this way or not, an audio engineer shares the professional goal of a magician - Richard Heyser
Linear Audio Vol 11 is out!
  Reply With Quote
Old 6th November 2015, 04:17 AM   #10
agdr is offline agdr  United States
diyAudio Member
 
agdr's Avatar
 
Join Date: Sep 2010
Quote:
Originally Posted by jan.didden View Post
Not to pick nits but this has nothing to do with cap leakage.
I respectfully disagree.

About this time last year I posted this single supply inverting CMOY project:

Inverting CMOY with inverting DC servo for 5V USB

A fellow had posted a thread asking for help with a CMOY. For fun I split off this thread where I whipped up a board, had it fabricated, built and tested it. Worked great (lol, I even sent one to the op of the first thread for putting up with me hijacking it). The board worked great, except for this same hiccup of leakage current through the output capacitor.

Initiallly in the BOM I had specified an organic solid polymer 470uF 6.3 volt capacitor. I hadn't checked the leakage specs, just built it, and was expecting only a small pop from that intial charging current as you've noted. Intead I had a massive pop that caused me to jerk the headphones off my head. Tracking that one down the leakage current turned out to be the culprit.

Below I've included what turned out to be the leakage specs of that capacitor, vs. the revised low-leakage aluminum electro I went to in the final BOM. The original had 592uA = 0.59mA of leakage, and that's after 2 minutes. During the first minute or so it was likely more in the neighborhood of a couple of mA. The low leakage cap spec was 6uA after 1 minute. So that is a 592/6 = 100 times greater leakage current.

Going to the low leakage cap resulted in a roughly 80% (subjective) reduction in turn-on pop. Nothing else in the circuit changed, just the cap.

In fairness this is a bit of a special case. These organic solid polymer's are turning out to be especially leaky. I believe that opc ran into that a couple of months ago in a thread on one of his "Wire" projects. Someone had noticed an exceptionally high voltage initially in one place and I believe opc tracked that down to this same problem of high leakage in these type of caps.

At any rate, that is my 2 cents, as we say here.
Attached Images
File Type: jpg solid_polymer 1.JPG (133.6 KB, 111 views)
File Type: jpg solid_polymer 2.JPG (46.7 KB, 8 views)
File Type: jpg low_leakage.JPG (178.8 KB, 10 views)

Last edited by agdr; 6th November 2015 at 04:19 AM.
  Reply With Quote

Reply


Hide 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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Is there a better way to implement buf634 other than place in a a loop? proid Headphone Systems 2 21st October 2013 11:19 PM
How to plot open-loop gain for single-transistor-input feedback amplifier? steveh49 Solid State 0 14th March 2012 02:59 AM
OPA637 - BUF634 power supply? AudioGeek Solid State 5 21st April 2010 08:14 PM
BUF04 Open Loop Bonsai Solid State 0 7th August 2006 10:09 AM
open loop gain,how many is enough? Leolabs Solid State 28 17th May 2006 09:00 PM


New To Site? Need Help?

All times are GMT. The time now is 09:05 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Copyright 1999-2016 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki