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

Headphone Systems Everything to do with Headphones

The fetHAv4 amplifier
The fetHAv4 amplifier
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 19th August 2019, 07:58 PM   #1
jstott is offline jstott  United States
diyAudio Member
 
Join Date: Feb 2016
Location: Connecticut
Default The fetHAv4 amplifier

I know this isn't earth-shatteringly original, but it's a very simple and very nice sounding class-A headphone amplifier that only needs a 9V single-ended supply.

To make it a little more interesting, I'll walk through how the design came to be (jump down a few posts if you just want the schematic).

My goal was to come up with a headphone amplifier that takes phone/laptop audio input and gives me a more robust output. I wanted a JFET input stage for two reasons: first, the high input impedance makes it easy to work with and second, the gentler clipping of a JFET means it shouldn't sound quite as harsh as a would BJT when the peaks start to clip. I just have ordinary headphones (16-32 ohms), so that's my design output impedance.

I've attached the first amplifier I came up with. J1 supplies the input voltage gain, Q2 is a follower that supplies the output power gain. It's a pretty classic single-ended tube design actually (except I have a capacitor instead of an output transformer).

Yeah it works, but frankly, it sucks.
  • R6 parallels the load, so you have to make R6 small to get enough current. When you do that, you end up dissipating as much power in R6 as you do in the output.
  • Large output currents mean large base currents into Q2, so Q2 ends up
    loading the JFET
  • The JFET gain is low - you're lucky to get 2:1 before clipping even when it isn't loaded by Q2.
  • The distortion is too high.

Time to start working on the design.
Attached Images
File Type: png fetHAv1.png (143.0 KB, 168 views)
  Reply With Quote
Old 19th August 2019, 08:05 PM   #2
jstott is offline jstott  United States
diyAudio Member
 
Join Date: Feb 2016
Location: Connecticut
Default fetHAv2

There are three quick changes I can make to clean up this amplifier.
  1. 1 To improve the power efficiency, I need to parallel the output load with something that has a high input impedance but a low voltage drop (i.e., not a resistor). A current source will do the job nicely.
  2. 2 To keep the output from loading the input stage, I need to decrease the current draw. Changing Q1 from a BJT to a Darlington pair increases the input impedance 100-fold [or equivalently, decreases the base current 100-fold]. That solves the loading issue.
  3. 3 To get more gain out of the JFET, I again need something with a large impedance but a small voltage drop - another current source.

With these changes, this is starting to become a decent amplifier. The distortion is still high, and it has way too much gain now, but those are solvable problems.
Attached Images
File Type: png fetHAv2.png (168.9 KB, 169 views)
  Reply With Quote
Old 19th August 2019, 08:13 PM   #3
jstott is offline jstott  United States
diyAudio Member
 
Join Date: Feb 2016
Location: Connecticut
Default fetHAv3

Now that I've got a reasonable amplifier, let's flesh it about a bit.

For the JFET's drain, I've implemented the current source as a bootstrapped resistor. While I could have done transistors, the advantage of the bootstrap is that it lets the JFET's drain voltage go all the way up to the rail voltage, increasing the power output.

For the output current source, I've replaced it with a classic discrete version. The LED sets a constant voltage level and R7 sets the current.

Finally, I've added negative feedback (C2 and R8). This trades in some of the excess gain I had and uses it to reduce the distortion at the output. Another way I could have done this is to eliminate C2 and connect the gate to the headphone output directly (the output side of C6). There's no good reason to prefer one over the other.

If you put in some component values and run this through Spice you'll find it's starting to become a pretty good little amplifier. If you build it, though, you'll find there's a problem. Single-ended amplifiers like this one are notorious for their lack of supply rejection and this is no exception: if your supply hums, so will your output. So that's the last thing I need to do is clean up the power supply.
Attached Images
File Type: png fetHAv3.png (198.6 KB, 167 views)
  Reply With Quote
Old 19th August 2019, 08:27 PM   #4
jstott is offline jstott  United States
diyAudio Member
 
Join Date: Feb 2016
Location: Connecticut
Default fetHAv4

This is my final version (and this one has component values).

My plan is to use a wall-wart, which come in two polarities. Because they're so easy to plug in backwards, I added an input diode to protect the circuit. After that is a bypass capacitor and a capacitance multiplier that smooths out most of the input ripple. The output follower doesn't care what the supply voltage is, so that's connected to the mains, but everything else has to be filtered.

I've put in some components, but feel free to substitute, there's nothing special about any of these parts. As long as your small-signal parts stay small-signal and your power devices stay power devices, you'll be fine (although it runs cool even without a heatsink using TIP31/32's, so maybe I've over-built it).

If you do build it, my only caution is R3,R4 and R5. JFET's vary significantly in transconductance, and the exact values you'll use depend on both the model and the individual devices. Plan on experimenting on a protoboard to get them right before you solder. The general rule of thumb, though, is R3=R4=R5 (give or take 20%). When it's right, the DC bias [no input signal] should be at about 4.5V or a bit higher.

You might also consider adding a pull-down resistor before R1. As it is, the amplifier can hum a bit with nothing plugged in.

On the whole, though, I'm very pleased with this amplifier. It sounds very good, it can do do better than 150 mW into headphones (which is uncomfortably loud). If you really insist, you could push the power up to 12 V and increase the bootstrap resistors to get 250 mW output (which is way beyond any ear-safe listening levels). The bass response is very good, probably because the phone is happier driving the 10k input resistance of my amplifier than it is driving the 16-ohm load of my headphones.
Attached Images
File Type: png fetHAv4.png (209.2 KB, 170 views)
  Reply With Quote
Old 20th August 2019, 05:47 AM   #5
PRR is offline PRR  United States
diyAudio Member
 
PRR's Avatar
 
Join Date: Jun 2003
Location: Maine USA
Via a valid path, you use the JFET for high input impedance, then evolve a plan with just 10k input impedance. (You say this is higher, I say it is lower, all relative.) I agree that 10k is acceptable in modern systems (and sure beats 16r).

Your C2 provokes a subsonic rise of gain. Gosh, 100uFd with 22k? Well, sub-sub-sonic. Perhaps innocuous. But can be avoided. (Also extends response through the 2,200uFd cap, not that it needs it.)

33uFd bootstrap against 1k pull-ups may be "weak" around 20Hz. For really solid precise bottom end the bootstrap cap should be "very large" relative to resistors. I'd take the 100uFd that I just drew-out and put it here.

I wonder about DC bias stability with various JFETs of the same type. I'd expect to do some selecting.
Attached Images
File Type: gif fetHAv4-PRR.gif (11.5 KB, 147 views)

Last edited by PRR; 20th August 2019 at 05:49 AM.
  Reply With Quote
Old 20th August 2019, 02:06 PM   #6
jstott is offline jstott  United States
diyAudio Member
 
Join Date: Feb 2016
Location: Connecticut
The problem with going to a massive bootstrap is peaking. Both the feedback and the bootstrap have a DC blocking capacitor and if they have similar frequency cut-offs, you get a bass bump. It doesn't matter which, but one of them has to be much bigger than the other so that a only single capacitor sets the lower cut-off. I went with the smaller bootstrap.

As I said before, DC bias varies with transconductance. Perfect matching isn't required, but for a stereo build you will need a pair of JFETs that are "reasonably close" to get similar operating points.
  Reply With Quote

Reply


The fetHAv4 amplifierHide 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
Amplifier kit HOOD JLH 1969 Class A Power amp kit 2 channel amplifier 10W+10W hifisuran Solid State 2 21st July 2019 06:57 AM
Hypex SMPS400 + 2x UCD180 wiring (DC error, Auto Amplifier Enable, Amplifier Standby) maxw Class D 48 26th May 2017 06:47 PM
Boss NXD3500 car amplifier hooked to an Onkyo 7.2ch home amplifier Evolver Class D 0 20th August 2016 12:57 AM
zenith console amplifier (a stereo mono amplifier :) ) dr_skoobie Tubes / Valves 1 14th February 2015 03:42 PM
Motional feedback amplifier based on Philips 40 wats amplifier 1970s sanbadgujar Solid State 14 13th October 2013 01:35 PM


New To Site? Need Help?

All times are GMT. The time now is 09:21 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki