I am new to DIY and would like to build a desktop amp that will work well with my Denon D2000 headphones. They are low impedance (25 ohm) with 106dB/mW sensitivity. I am looking to build either a solid state amp or a hybrid amp that is suitable to drive them. I have heard that the Starving Student Hybrid is too noisy to work well with these phones.
Basically I want to know which desktop DIY amps have low output impedance, a black background, and are able to deliver a good amount of current into 25 ohms. Also thinking of getting a pair of orthos, if I can find something that delivers enough current that plays nice with my D2ks.
Any suggestions? I am working on an O2, but really want a nice big thing for my desktop for my next project. I am curious about CKKIII, but don't know the specs I am interested in.
Basically I want to know which desktop DIY amps have low output impedance, a black background, and are able to deliver a good amount of current into 25 ohms. Also thinking of getting a pair of orthos, if I can find something that delivers enough current that plays nice with my D2ks.
Any suggestions? I am working on an O2, but really want a nice big thing for my desktop for my next project. I am curious about CKKIII, but don't know the specs I am interested in.
you don't need V gain at all - buffer not amp - consumer Desktop source mostly 2 Vrms, most DAP 1 Vrms 0 db fs
even then a ~12 dB stepdown audio ouptut xfmr would be best to preserve S/N
even then a ~12 dB stepdown audio ouptut xfmr would be best to preserve S/N
That did explain it quite well. It also makes me want to try to build something similar to what was made in that link, but I don't have the knwoledge to put it together without pcbs, circuit diagram, and BOM. I might be able to just get it done with a circuit diagram, but I think I should get some more experience first.
I will have to check how loud my headphones get when plugged directly into my source (I think it is 1.7v RMS). I am concerned about the prospect of just using an attenuator, and I think this blurb from headphone.com sums it up: "Despite the AH-D2000's seemingly efficient impedance rating of 25 Ohms, it's overall sensitivity and heavy voice coil driver mass will still benefit from dedicated headphone amplification and control to hear exactly what it can sonically accomplish; otherwise, the sound may tend to be lightly 'flat' and un-dynamic, especially from portable audio sources."
I have had less than stellar results by plugging D2000s into my iPhone, namely a poor low frequency response. They get quite loud, but sound disappointing.
I will have to check how loud my headphones get when plugged directly into my source (I think it is 1.7v RMS). I am concerned about the prospect of just using an attenuator, and I think this blurb from headphone.com sums it up: "Despite the AH-D2000's seemingly efficient impedance rating of 25 Ohms, it's overall sensitivity and heavy voice coil driver mass will still benefit from dedicated headphone amplification and control to hear exactly what it can sonically accomplish; otherwise, the sound may tend to be lightly 'flat' and un-dynamic, especially from portable audio sources."
I have had less than stellar results by plugging D2000s into my iPhone, namely a poor low frequency response. They get quite loud, but sound disappointing.
a buffer provides Current amplication but only unity Voltage gain
for the D2000 Z, sensitivity # you should be able to caclulate SPL: ~120 dB for 1 Vrms drive
iPods and other DAP often have too small a output coupling C which rolls off the lower few octaves - worse with lower Z headphones
if you don't need to reach 120 dB SPL often then even DAP could be used with 25 Ohm load with bass boost EQ to compensate for the RC roll off
for the D2000 Z, sensitivity # you should be able to caclulate SPL: ~120 dB for 1 Vrms drive
iPods and other DAP often have too small a output coupling C which rolls off the lower few octaves - worse with lower Z headphones
if you don't need to reach 120 dB SPL often then even DAP could be used with 25 Ohm load with bass boost EQ to compensate for the RC roll off
I do BTW have a pair of Fostex T50rp which need 3-4 times more power than the Denons and if I plug them directly to the relatively week line out of my Ibasso D7 they are to loud. You will BTW find measurements of a lot of headphones at Headphone Data Sheet Downloads | InnerFidelity
Ok, I tried my headphones connected to my 1.7v RMS source, and it was about as loud as I would ever want it. You guys are really good, and I hope one day to be as smart and capable as you all.
burgunder: Is there a kit to make what you linked? That looks like it would be perfect for me, although I would prefer having a power supply rather than 4x 9v batteries. If there is no kit I will have to read through it a few more times and try to come up with some questions about what was used to make it and how I should start.
burgunder: Is there a kit to make what you linked? That looks like it would be perfect for me, although I would prefer having a power supply rather than 4x 9v batteries. If there is no kit I will have to read through it a few more times and try to come up with some questions about what was used to make it and how I should start.
I'm with @jcx on this one. I recommend you go after a CMOY with unity gain (i.e. a buffer), but configured with <1Z to get all the benefit possible from your 'phones.
I really DONT think jcx was thinking to the cmoy, nor any other IC-based output stage cause they dont have the required current to drive low Z 'phones.
some single op amps have plenty of current, but may not like the load Z
adding series R is OK for flat impedance curve headphones, can make a R divider "L-pad" but then need excess current capacity to "waste" in order to get lower drive Z to the headphone (== deliberate Z mismatching)
at the extreme it depends on what you consider low z, high sensitivity
JH Audio JH16 IEM:
118dB @ 1mW
Impedance: 18 Ohms
~200 mV, 10 mA peak to get to ~120 dB SPL
clearly the only rational way to drive these from 1-2 Vrms standard audio sources involves a step down transformer
adding series R is OK for flat impedance curve headphones, can make a R divider "L-pad" but then need excess current capacity to "waste" in order to get lower drive Z to the headphone (== deliberate Z mismatching)
at the extreme it depends on what you consider low z, high sensitivity
JH Audio JH16 IEM:
118dB @ 1mW
Impedance: 18 Ohms
~200 mV, 10 mA peak to get to ~120 dB SPL
clearly the only rational way to drive these from 1-2 Vrms standard audio sources involves a step down transformer
Last edited:
I have pretty much decided to build a 2 channel Beta22 with low gain. I may get other headphones in the future and have to change the gain settings, but a Beta22 should drive anything I throw at it admirably.
I had looked into making something that is just a current source + attenuator, but there don't seem to be any viable (well documented) designs.
In fact, I think it was a post on another forum by jcx that convinced me to go 2 channel on my Beta22 build instead of 3 channel.
I had looked into making something that is just a current source + attenuator, but there don't seem to be any viable (well documented) designs.
In fact, I think it was a post on another forum by jcx that convinced me to go 2 channel on my Beta22 build instead of 3 channel.
Last edited:
TPA6120 is designed for 25 Ohm load, good audio specs by itself - some headphone project amps use it
some may not know but CFA op amps can be used "unity gain" - you still need a feedback R between the output and -in
its value provides frequency response compensation, increasing it provides "overcompensation" - less peaking, better immunity to destabilizing C load
better (but probably inaudible) performance can be had by wrapping a "good", "audio" input op amp feedback loop around a CFA buffer op amp in a multiloop composite amplifier
some may not know but CFA op amps can be used "unity gain" - you still need a feedback R between the output and -in
its value provides frequency response compensation, increasing it provides "overcompensation" - less peaking, better immunity to destabilizing C load
better (but probably inaudible) performance can be had by wrapping a "good", "audio" input op amp feedback loop around a CFA buffer op amp in a multiloop composite amplifier
I've been liking the use of the NJM4556D that the Objective2 employs. That thing can handle 1 Ohm to 600 Ohm and puts out about 70mA... enough for every headphone out there (other than the K1000) to get over 105dB.
Anyway, I was thinking you would've liked to build something like this: http://nwavguy.blogspot.com/2011/05/cmoy-ebay-headphone-amp.html
I hope the Beta22 turns out nicely.
Anyway, I was thinking you would've liked to build something like this: http://nwavguy.blogspot.com/2011/05/cmoy-ebay-headphone-amp.html
I hope the Beta22 turns out nicely.
Last edited:
I did actually build an Objective2, and I think that it simply can't be beat for the price of parts, or its retail price for that matter. I had considered building another one and putting it in a larger enclosure for desktop use, but decided to try a Beta22. I might take a crack at some of the other amps mentioned in this thread as well, since I find building these things just as much fun as listening to them.
Other than the B22, AMB has a few other offerings such as the CKKIII, M3, or even the A20 that can be configured as unity gain buffer. Or you can make a simple tube voltage gain stage to drive the A20 buffer if you fancy a hybrid.
I have only build the B22 but all of them are supposed to be good. Glass jar audio offer kits for some of these and all you have to do is put them together.
They also have other kits such as the SOHA II, EHHA and the Bijou from Cavalli audio. FWIW I prefer the SOHA II over the B22.
I have only build the B22 but all of them are supposed to be good. Glass jar audio offer kits for some of these and all you have to do is put them together.
They also have other kits such as the SOHA II, EHHA and the Bijou from Cavalli audio. FWIW I prefer the SOHA II over the B22.
- Status
- Not open for further replies.