NwAvGuy: O2 Details
or just buy it:
JDS Labs - Objective2 Headphone Amplifier
many more hobby project amps have been presented in the past:
or just buy it:
JDS Labs - Objective2 Headphone Amplifier
many more hobby project amps have been presented in the past:
how can these threads get to over a dozen posts without a pointer to the Headwize archive?
HeadWize Library - Projects
site is dead, but the archives are required reading for aspiring headphone amp diyers
head-fi.org has a diy forum - beware the relative lack of engineering knowledge/discussion, excess fanboyism compared to here - but some info
List of DIY headphone amplifiers
okay, the o2 is not a discrete design, so will be less good right?If you go cheap on the casework, you can build most DISCRETE designs, such as Marsh, CK2III, etc.
what is the cheaper to build, the CK2III or Marsh?
while "discrete is superior" is an entrenched Audiophile meme - they never placed a "expires by date" on it
O2 is likely indistinguishable in DBT against any other objectively good, flat frequency response, low output Z, low distortion design regardless of technology, within its I,V output capability
but similar chip count composite/multiloop amplifiers, using "better" speced op amps released in the past decade or so are really untouchable at line level audio today - if accuracy is any part of your hifi amplifier definition
isolated bipolar, SiGe, GHz semi processes, advanced internal circuit designs by big companies competing for modern telcom, imaging market applications have taken a big step forward from the era of 4556, 5534
even $5-10 for op amps shouldn't blow a $200 budget
I quite understand as a circuit designer the greater enjoyment I derive from discrete circuit design - I just don't confuse that with being able to reach the performance specs of properly applied best in class modern monolithic op amps anywhere their input noise, output I,V capabilities meet the application's demands
O2 is likely indistinguishable in DBT against any other objectively good, flat frequency response, low output Z, low distortion design regardless of technology, within its I,V output capability
but similar chip count composite/multiloop amplifiers, using "better" speced op amps released in the past decade or so are really untouchable at line level audio today - if accuracy is any part of your hifi amplifier definition
isolated bipolar, SiGe, GHz semi processes, advanced internal circuit designs by big companies competing for modern telcom, imaging market applications have taken a big step forward from the era of 4556, 5534
even $5-10 for op amps shouldn't blow a $200 budget
I quite understand as a circuit designer the greater enjoyment I derive from discrete circuit design - I just don't confuse that with being able to reach the performance specs of properly applied best in class modern monolithic op amps anywhere their input noise, output I,V capabilities meet the application's demands
Last edited:
The main issue of opamp based headamps is the lack of current, which makes them not very suited for low Z 'phones.
Now, the hd650 is NOT low impedance and the o2 would drive it fine.
About how it sounds, I don't really know. I don't have any friend with one to do an AB test.
My suggestion to go discrete was dictated by the budget of the OP and by the fact that this is a DIY forum.
Btw he can also buy a shiit M-something (forgot the name) for 99$.
Now, the hd650 is NOT low impedance and the o2 would drive it fine.
About how it sounds, I don't really know. I don't have any friend with one to do an AB test.
My suggestion to go discrete was dictated by the budget of the OP and by the fact that this is a DIY forum.
Btw he can also buy a shiit M-something (forgot the name) for 99$.
ah - you can choose op amps to suit the applicaiton - you have to with discrete Q too
the O2 choice of 4556 is driven by the +/- 70 mA output spec, and that is doubled by paralleling the 2 op amps in each chip, per channel - so +/- 140 mA peak should be available
M50 38 Ohms absorbs >300 mWrms @ 140 mA peak drive
unfortunately their site doesn't qualify their 99 dB sensitivity with the reference level, the old standard of 1 mW or 1 Vrms as is becoming more common
99 dB +25 dB re 1 mW is ridiculous @ 124 dB SPL
but even if we go with +17 dB re 1V, for 116 dB SPL clipping level with O2 driving the "low impedance" - thats not so shabby loudness wise either
but my default choice of a modern output drive op amp is the TPA6120 (rebadged THS6012 DSL driver chip) 100 MHz CFA op amp with 400 mA output current and ppm distortion, designed to drive DSL transformer's 25 Ohm impedance - and I have paralleled 2 of them...
the O2 choice of 4556 is driven by the +/- 70 mA output spec, and that is doubled by paralleling the 2 op amps in each chip, per channel - so +/- 140 mA peak should be available
M50 38 Ohms absorbs >300 mWrms @ 140 mA peak drive
unfortunately their site doesn't qualify their 99 dB sensitivity with the reference level, the old standard of 1 mW or 1 Vrms as is becoming more common
99 dB +25 dB re 1 mW is ridiculous @ 124 dB SPL
but even if we go with +17 dB re 1V, for 116 dB SPL clipping level with O2 driving the "low impedance" - thats not so shabby loudness wise either
but my default choice of a modern output drive op amp is the TPA6120 (rebadged THS6012 DSL driver chip) 100 MHz CFA op amp with 400 mA output current and ppm distortion, designed to drive DSL transformer's 25 Ohm impedance - and I have paralleled 2 of them...
Last edited:
CMOY not very good, and if the rA-1 is the Grado one its stupidly costly. I've had a dozenish headphone amps and messed about with mods, psus, valves, kits, off the shelf, essentially I've tried most sensible approaches except high voltage tubes, and my conclusion from 14 years of this is: if you have a stepped attenuator and a PSU per channel, any decent solid state amp will sound very good and blow away affordable shop stuff. People that are invested in one design or other will claim superiority or that practical perfection has been reached, they aren't right but aren't that wrong either.
If you want to build a kit look for one that lends itself to seperating the left and right channel power. If you want a ready built amp see if the case can accomodate a stepped attenuator or other fancy volume control, the latest version of the Neco MOSFET has a psu per channel but could do with modding for a stepped attenuator.
The Panda kit on ebay is very good with HD600, so should go well with HD650. Best budget for a box for the transformers and one for the amp, will come in at under a bit $200 unless you go for boutique upgrade parts in which case it your own fault.
The O2 is highly rated but is smd and if pre-made may be difficult to change to a larger volume control. An alternative would be Low distortion ( 0.0001%!) preamplifier headphone amp ! | eBay
Get free sample chips from TI and the thing could easily come in for a $100.
If you want to build a kit look for one that lends itself to seperating the left and right channel power. If you want a ready built amp see if the case can accomodate a stepped attenuator or other fancy volume control, the latest version of the Neco MOSFET has a psu per channel but could do with modding for a stepped attenuator.
The Panda kit on ebay is very good with HD600, so should go well with HD650. Best budget for a box for the transformers and one for the amp, will come in at under a bit $200 unless you go for boutique upgrade parts in which case it your own fault.
The O2 is highly rated but is smd and if pre-made may be difficult to change to a larger volume control. An alternative would be Low distortion ( 0.0001%!) preamplifier headphone amp ! | eBay
Get free sample chips from TI and the thing could easily come in for a $100.
but my default choice of a modern output drive op amp is the TPA6120 (rebadged THS6012 DSL driver chip) 100 MHz CFA op amp with 400 mA output current and ppm distortion, designed to drive DSL transformer's 25 Ohm impedance - and I have paralleled 2 of them...
Now this is interesting
My HP peaks about 440mA
You can build the CK2III for that price. It is a discrete no-global feedback design, with a laid back sound. I've built an LM4562->LME49600 headphone amp design similar to the O2, and it was just too bright and steely for my ears. Extremely fatiguing, although the distortion measured extremely low.
buying/building different amps because your headphones would benefit from EQ is a losing proposition
see the Carver Stereophile Challenge - Stereophile's professional reviewers couldn't tell a $600 Carver SS amp from their choice of "SOTA" tube amp, in their room, their choice of speakers, music - after Carver nulled his amp against their tube amp, padded his amp's output impedance to match
Headroom high end amps came with a output Z switch - trivial for a diyer to add
Linkwitz shows passive discrete "power" EQ networks for iem...
and you could give a digital EQ a fair trial
see the Carver Stereophile Challenge - Stereophile's professional reviewers couldn't tell a $600 Carver SS amp from their choice of "SOTA" tube amp, in their room, their choice of speakers, music - after Carver nulled his amp against their tube amp, padded his amp's output impedance to match
Headroom high end amps came with a output Z switch - trivial for a diyer to add
Linkwitz shows passive discrete "power" EQ networks for iem...
and you could give a digital EQ a fair trial
The O2 is highly rated but is smd and if pre-made may be difficult to change to a larger volume control.
No, O2 uses through hole components.
For an inexpensive, simple build consider the glass sand hybrid, low cost and good sounding, can drive 24ohm cans. I'm just using a 24V wallwart power supply now, I have a slow start +-12 volt PCB in design to see if it makes any difference. I currently have about $50 in this project.
http://www.diyaudio.com/forums/head...-mosfet-hybrid-headphone-amp.html#post3415913
http://www.diyaudio.com/forums/head...-mosfet-hybrid-headphone-amp.html#post3415913
Attachments
Last edited:
while "discrete is superior" is an entrenched Audiophile meme - they never placed a "expires by date" on it
O2 is likely indistinguishable in DBT against any other objectively good, flat frequency response, low output Z, low distortion design regardless of technology, within its I,V output capability
but similar chip count composite/multiloop amplifiers, using "better" speced op amps released in the past decade or so are really untouchable at line level audio today - if accuracy is any part of your hifi amplifier definition
isolated bipolar, SiGe, GHz semi processes, advanced internal circuit designs by big companies competing for modern telcom, imaging market applications have taken a big step forward from the era of 4556, 5534
I agree wholeheartedly.
OP: I strongly recommend The Wire, if you can find boards anywhere.
...it was just too bright and steely for my ears. Extremely fatiguing, although the distortion measured extremely low.
the distortion measurement was most likely not done right... if it sounds bright and steely, than measurements must confirm it...have you done IM distortion measurements? have you looked at 13+14kHz twin tone spectra?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Headphone Systems
- suggestion for a 200$ headamp