Headphone amp from AN-272 Op Amp Booster Designs

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
B

BJosephs

Member
Joined 2011
Hi all,


Does anyone know if someone has tried leveraging figure 1 of this design note into a headphone amp? It looks like it could work will with a high quality op amp.

http://www.ti.com/lit/an/snoa600b/snoa600b.pdf

I was bored so I started up a simple PCB layout, using Express PCB and their 3.8X2.5 proto-board size. No traces around the OA yet, just working out the output stage. I doubled up on the output transistors and I think the 2.5 ohm ballasting resistors should drop to about 1 ohm to suit. The green plane will be for audio, red will be for power rails and ground plane.

An externally hosted image should be here but it was not working when we last tested it.
[/URL]

An externally hosted image should be here but it was not working when we last tested it.
[/URL]


So, bad idea? Errors in the layout? Order parts, build it and report back?

-Brian
 
dchisholm

dchisholm

Member
Joined 2011
That circuit looks to be well thought-out and actually designed - not just presented as a textbook example of a concept. It offers a lot of possibility for learning and experimentation. I suspect you can find the same basic architecture in quite a few projects published on the 'net and elsewhere. For example,

Using the LM334 current sources in the output stage bias network should improve the overall linearity compared to simple resistor bias but may compromise the usable output voltage swing. (I had to check whether the LM334 is still available - haven't seen it mentioned in quite a while.)

Here on the DIYAudio Forum you can probably collect a few dozen suggestions for more modern, higher-performance substitutes for the LF35x family opamps. I'm partial to the TL07x family over the LF35x.

The 2N2219/2N2905 devices may be difficult to obtain from reliable sources. There are several current-production (at least as of June, 2014) devices in TO-126 or TO-225 packages that should be suitable substitutes.

If you can't find the solid tantalum bypass capacitors, a good quality (low ESR) electrolytic - perhaps with a film capacitor in parallel - should do the job. In fact, I'd probably park a pair of film caps right at the opamp power pins even with the tantalums on the power input lines.

Dale
 
J

jcx

Member
Joined 2003
don't use Ta electros - expensive and Al have more than caught up in performance compared to 30+ yrs ago when solid Ta were the datasheet standard - but unexamined even then suggestion

if boredom is the design criteria then its something to do

if performance is the desire then embrace surface mount, use last decade or so better op amps, can still use Scott Wurcer/Walt Jung multiloop topologies but with DSL driver op amps as high current, low audio distortion output stage if your headphone specs are fine with op amp voltages
 
Last edited:
B

BJosephs

Member
Joined 2011
Sure!

A couple of those circuits are right out of Walt Jung's IC Op-Amp Cookbook
Click to expand...

OK, cool. Thanks, Fast Eddie D

That circuit looks to be well thought-out and actually designed - not just presented as a textbook example of a concept. It offers a lot of possibility for learning and experimentation. I suspect you can find the same basic architecture in quite a few projects published on the 'net and elsewhere. For example,
Click to expand...

Great, those examples show that there is some precedent. I think the 3rd one I stumbled upon while googling. I'm sure the use of the LM334 is superior to resistors. Any idea why they would use 2 diodes per base and raise the idle dissipation of the output transistors?

The LM334, 2N2219 and 2N2905 all seem to be available from Mouser. I'm open to suggestions for replacements on the "2Ns"... maybe something in a beefier package or that takes a more space efficient heat sink?
 
dchisholm

dchisholm

Member
Joined 2011
BJosephs said:
. . . why they would use 2 diodes per base and raise the idle dissipation of the output transistors?
Click to expand...
Lower crossover distortion? Avoiding reduced Hfe (or lower Ft) at low collector currents? Class-A operation into moderate-resistance loads?

. . . suggestions for replacements on the "2Ns"... maybe something in a beefier package or that takes a more space efficient heat sink?
Click to expand...
The old standbys would be MJE340/350, MJE172/182, and BD139/140. There are several threads on this Forum discussing possible, more modern, alternatives.

Dale
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom