who knows.. it can happen, maby I will use it in may badroom... :green:
simple headphone amp
- Thread starter mik
- Start date
Help Support GroupDIY Audio Forum:
> make it easy for you? tpa302
Good but not stellar audio specs.
Probably sounds better than LM386, a classic 50-cent power amp.
Teeny surface-mount, too small for my eyes.
5V maximum power (6V if you want to live dangerously for little benefit).
4V p-p 1.4V rms maximum output.
OK for low-Z phones at sane levels.
Won't make crazy-loud levels in most 32Ω phones.
May not make musical level in hi-Z phones.
Mik's 24V BD139 plan will make 18V p-p or 6V rms, at least 250mA peak. This will put 100mW in 300Ω phones, and about 1 Watt in 32Ω phones. That would be my main "objection": careless mis-use could lead to burned-out phones. (And burned-out ears, if anybody is stupid enough to stick his head between two 1-Watt screamers.) About 22Ω or 27Ω in series with the output will reduce the maximum power available in low-Z phones without any harm to hi-Z phone level.
Good but not stellar audio specs.
Probably sounds better than LM386, a classic 50-cent power amp.
Teeny surface-mount, too small for my eyes.
5V maximum power (6V if you want to live dangerously for little benefit).
4V p-p 1.4V rms maximum output.
OK for low-Z phones at sane levels.
Won't make crazy-loud levels in most 32Ω phones.
May not make musical level in hi-Z phones.
Mik's 24V BD139 plan will make 18V p-p or 6V rms, at least 250mA peak. This will put 100mW in 300Ω phones, and about 1 Watt in 32Ω phones. That would be my main "objection": careless mis-use could lead to burned-out phones. (And burned-out ears, if anybody is stupid enough to stick his head between two 1-Watt screamers.) About 22Ω or 27Ω in series with the output will reduce the maximum power available in low-Z phones without any harm to hi-Z phone level.
Kev
Well-known member
A local mag did another headphone amp article
and
it's available as a kit from a few places here
has a slightly different approach to bias control
opamp followed by a couple of drivers BC557/547 and then BD139/140
right click my re-draw pdf
http://www.diyfactory.com/projects/headphoneamp/HPamp_01.pdf
do look for mistakes
I did it in a rush
here is a few of the listed specs
Frequency .......................................................... flat from 10Hx to 20kHz
Rated output power ............................................ 200mW into 8ohms and 32ohms, 85mW into 800ohms
Max output (current or voltage limited) ................ 575mW into 8ohms, 700mW into 32ohms, 130mW into600ohms
Harmonic distortion ............................................ typically .0005%(600ohm load), .001%(300ohm load) and .005%(8ohm load)
I've not confirmed any of that yet but have made one board ready to test
and
it's available as a kit from a few places here
has a slightly different approach to bias control
opamp followed by a couple of drivers BC557/547 and then BD139/140
right click my re-draw pdf
http://www.diyfactory.com/projects/headphoneamp/HPamp_01.pdf
do look for mistakes
I did it in a rush
here is a few of the listed specs
Frequency .......................................................... flat from 10Hx to 20kHz
Rated output power ............................................ 200mW into 8ohms and 32ohms, 85mW into 800ohms
Max output (current or voltage limited) ................ 575mW into 8ohms, 700mW into 32ohms, 130mW into600ohms
Harmonic distortion ............................................ typically .0005%(600ohm load), .001%(300ohm load) and .005%(8ohm load)
I've not confirmed any of that yet but have made one board ready to test
Samuel Groner
Well-known member
* Q1 is upside-down
* Q2/Q4 are interchanged
* there is an error in the bias network for LED2 (one 4k7 wrong connected)
* R12 looks pointless
* the 47 pF RFI protection cap could/should be much larger, say 1 nF
Looks like a well-thought design though the current sources for the bias diodes are perhaps overkill. I've never tested the drive capability of an OPA2134, so I might be wrong.
Samuel
Svart
Well-known member
Hey PRR, I used the 302s because I had them handy and they do sound fine with normal studio headphones. they do get loud enough to hurt! I only mentioned them because they are cheap and reliable and do sound ok but as I mentioned before they aren't for critical listening, but if you are tracking drums, etc. you aren't going to hear every little detail anyway.
Samuel Groner
Well-known member
I used OPA551's successfully--cheap as long as you get samples...
Samuel
Samuel
Kev
Well-known member
[quote author="Samuel Groner"]* Q1 is upside-down,* Q2/Q4 are interchanged
* there is an error in the bias network for LED2 (one 4k7 wrong connected)
* R12 looks pointless
* the 47 pF RFI protection cap could/should be much larger, say 1 nF
Looks like a well-thought design though the current sources for the bias diodes are perhaps overkill. I've never tested the drive capability of an OPA2134, so I might be wrong.
Samuel[/quote]
yep to Q1 Q2 and Q2 and yep to the 4k7
looks like I was having an upside down sort of day
The pointless resistor of zero ohms ... (I used 0.1 cos the software doesn't like 0)
is for a gain adjustment if needed ...
this may require other component choices
and the design also includes an 8 ohm option where extra diodes are used to help current limit.
I think I have show a bit of a hybrid of high and low headphone options
sorry ... I was rushed
1 nF for RFI protection
ok I'll keep that in mind :thumb:
" ... the current sources for the bias diodes are perhaps overkill. "
Yes I think so too but it is that feature that prompted me to look at it and included it here and at the Factory.
It was just another and different slant on a fairly typical idea.
thanks heaps for you time
I'll fix asap
I've had an edit
http://www.diyfactory.com/projects/headphoneamp/HPamp01.pdf
the one above is still there (/HPamp_01.pdf)
and the full page will be at
http://www.diyfactory.com/projects/headphoneamp/headphoneamp_2.htm
when I'm done
DON'T look at it yet cos I'm not finished and it's all crap so far
* there is an error in the bias network for LED2 (one 4k7 wrong connected)
* R12 looks pointless
* the 47 pF RFI protection cap could/should be much larger, say 1 nF
Looks like a well-thought design though the current sources for the bias diodes are perhaps overkill. I've never tested the drive capability of an OPA2134, so I might be wrong.
Samuel[/quote]
yep to Q1 Q2 and Q2 and yep to the 4k7
looks like I was having an upside down sort of day
The pointless resistor of zero ohms ... (I used 0.1 cos the software doesn't like 0)
is for a gain adjustment if needed ...
this may require other component choices
and the design also includes an 8 ohm option where extra diodes are used to help current limit.
I think I have show a bit of a hybrid of high and low headphone options
sorry ... I was rushed
1 nF for RFI protection
ok I'll keep that in mind :thumb:
" ... the current sources for the bias diodes are perhaps overkill. "
Yes I think so too but it is that feature that prompted me to look at it and included it here and at the Factory.
It was just another and different slant on a fairly typical idea.
thanks heaps for you time
I'll fix asap
I've had an edit
http://www.diyfactory.com/projects/headphoneamp/HPamp01.pdf
the one above is still there (/HPamp_01.pdf)
and the full page will be at
http://www.diyfactory.com/projects/headphoneamp/headphoneamp_2.htm
when I'm done
DON'T look at it yet cos I'm not finished and it's all crap so far
Samuel Groner
Well-known member
OK I see. But then you need to move the connection of the feedback resistor--right in front of the opamp input doesn't change gain.
Samuel
Kev
Well-known member
as shown I think it has a gain of 2 (6dB)
to get it up to around a gain of 7 you would need to change all three resistors and the cap in that neg input/feedback area
to get it up to around a gain of 7 you would need to change all three resistors and the cap in that neg input/feedback area
Samuel Groner
Well-known member
Not shure we understand each other--as shown, R3 does not influence gain at all. It can be 0r0 or 1 meg, gain stays at +6 dB.
As shown, the value of R3 might have a slight influence on offset, distortion and noise but probably nothing to worry about. That's why I would skip it.
To get higher gain, you simply lower the value of the 1k going to ground, i.e. to 110 for about +20 dB. There's no need to change any other part.
I thougth that you wanted to get variable gain; this is possible if you move R3 between the 1k shunt resistor and ground and use a rev. log pot.
Samuel
As shown, the value of R3 might have a slight influence on offset, distortion and noise but probably nothing to worry about. That's why I would skip it.
To get higher gain, you simply lower the value of the 1k going to ground, i.e. to 110 for about +20 dB. There's no need to change any other part.
I thougth that you wanted to get variable gain; this is possible if you move R3 between the 1k shunt resistor and ground and use a rev. log pot.
Samuel
Kev
Well-known member
[quote author="Samuel Groner"]As shown, the value of R3 might have a slight influence on offset, distortion and noise but probably nothing to worry about. That's why I would skip it.[/quote]
I agree
the article is long winded here and they do site the possible increase in disortion as a reason for this resistor ... at higher gains they suggest as high as 2kohms for R3
R1 at 7.5k R2 at 1.2k and C1 at 100pF
they also talk about the source impedance varying when a 50k pot is used as an input atten/volume control ... inducing distortion and again refer to the above as a possible solution.
As I've said I don't agree with all of it and you do need to see the article for a complete picture of what they are on about.
Could be a little early for me to re-print great chunks of the text here or on the site as it is only a month or two old. I push things as it is and don't need any more ... mmmm? ... issues.
:green:
No harm in having an extra resistor. A bit of trial and error and experimenting and people can make there own mind up. You can always use a link ... I did.
I agree
the article is long winded here and they do site the possible increase in disortion as a reason for this resistor ... at higher gains they suggest as high as 2kohms for R3
R1 at 7.5k R2 at 1.2k and C1 at 100pF
they also talk about the source impedance varying when a 50k pot is used as an input atten/volume control ... inducing distortion and again refer to the above as a possible solution.
As I've said I don't agree with all of it and you do need to see the article for a complete picture of what they are on about.
Could be a little early for me to re-print great chunks of the text here or on the site as it is only a month or two old. I push things as it is and don't need any more ... mmmm? ... issues.
:green:
No harm in having an extra resistor. A bit of trial and error and experimenting and people can make there own mind up. You can always use a link ... I did.
Samuel Groner
Well-known member
OK--happy now! :thumb:
Samuel
Samuel
Similar threads
