I found a pair of SABA H200 (custom AKG K240 sextett) headphones at a thrift store yesterday. They are the best sounding headphones I've ever used (I have K142HD, K404, PXC250, Siberia V2) but they're a bitch to drive. They do basically nothing when connected to my smartphone and are still relatively quiet with my laptop. Only thing that seems happy driving them is a Sony MXP-290 mixing console. But the console takes up too much space so I can't keep it on my desk, let alone take it with me 
So I'm looking for something to drive them. I'd like to build a super simple CMoy type thing, would that work? I want to do it compact by using a single 9V battery with a voltage splitter and do pin-to-pin on a DIP8 socket.
http://www.mypage.tsn.cc/dakiller/cmoy.png
Would the above circuit do the job? I'll be using NE5532 at first but I'll try to get some OPA2134 soon
Cheers!
So I'm looking for something to drive them. I'd like to build a super simple CMoy type thing, would that work? I want to do it compact by using a single 9V battery with a voltage splitter and do pin-to-pin on a DIP8 socket.
http://www.mypage.tsn.cc/dakiller/cmoy.png
Would the above circuit do the job? I'll be using NE5532 at first but I'll try to get some OPA2134 soon
Cheers!
many DAP use single cell LiIon batteries and give ~ 1 Vrms out
splitting a "9 V" battery and accommodating typical op amp output stage saturation V may end up giving you less than 10 dB, possibly only 6 dB more
those # are not subjectively very much louder
http://www.innerfidelity.com/images/AKGK240Sextett.pdf give dB sensitivity and Ohms Z
if you want >110 dB clipping free peak capability you want more than 8.9 Vrms = >+/-12 V pk - more than even dual "9 V" supplied amp like the O2
you do need to consider safe all day background music average SPL vs critical listening to highly dynamic recorded Jazz and Orchestral music with >20 dB peaks above the average
http://headwize.com/?page_id=266
splitting a "9 V" battery and accommodating typical op amp output stage saturation V may end up giving you less than 10 dB, possibly only 6 dB more
those # are not subjectively very much louder
http://www.innerfidelity.com/images/AKGK240Sextett.pdf give dB sensitivity and Ohms Z
if you want >110 dB clipping free peak capability you want more than 8.9 Vrms = >+/-12 V pk - more than even dual "9 V" supplied amp like the O2
you do need to consider safe all day background music average SPL vs critical listening to highly dynamic recorded Jazz and Orchestral music with >20 dB peaks above the average
http://headwize.com/?page_id=266
Last edited:
Well, I could just build a dedicated power supply, I don't really need to carry these around, I can make do with the K404s on the road and K142HD without the amp. Really only need it when listening at home.
I have two transformers I could use.
One is 9-0-9 AC or about 13-0-13 rectified.
Second is 12-0-12 or about 17.3-0-17.3 rectified.
Which one should I use?
I have two transformers I could use.
One is 9-0-9 AC or about 13-0-13 rectified.
Second is 12-0-12 or about 17.3-0-17.3 rectified.
Which one should I use?
for 600 Ohms a carefully chosen dual op amp may suffice
I'd use the higher V supply and use 44 Vs_max LME49860 or the older OPA2604 which handles even higher Vsupply - 48 Vmax, set Av ~ +20 and use ~25 Ohm output series R after the feedback to protect from cable load, intermittent shorting when plugging TRS
the LME49860 has fine specs, input noise is well matched to a 10 k volume pot
for likely inaudible, pointless improvements buy a higher V xfmr to safely regulate down to +/-20 V
double up the op amps in the dual for each channel in parallel (independent but identical value feedback), added together with the ~25 Ohm series R
keeping it simple I wouldn't recommend the master/slave O2 circuit with these op amps - requires too much input common mode range, also runs the slave at unity gain which requires better implementation for stability against oscillation
for 600 Ohm headphones I wouldn't bother with series R||L, just the R alone doesn't require thinking about the TRS shorting problem
I'd use the higher V supply and use 44 Vs_max LME49860 or the older OPA2604 which handles even higher Vsupply - 48 Vmax, set Av ~ +20 and use ~25 Ohm output series R after the feedback to protect from cable load, intermittent shorting when plugging TRS
the LME49860 has fine specs, input noise is well matched to a 10 k volume pot
for likely inaudible, pointless improvements buy a higher V xfmr to safely regulate down to +/-20 V
double up the op amps in the dual for each channel in parallel (independent but identical value feedback), added together with the ~25 Ohm series R
keeping it simple I wouldn't recommend the master/slave O2 circuit with these op amps - requires too much input common mode range, also runs the slave at unity gain which requires better implementation for stability against oscillation
for 600 Ohm headphones I wouldn't bother with series R||L, just the R alone doesn't require thinking about the TRS shorting problem
Just for practice, I built the classic CMoy I did it with a +-12 regulated supply from my Rod Elliot P06 phono stage. Used a NE5532 for now. Sounds pretty good and is plenty loud. What size input coupling should I use? I used .47uF K73-16V NOS for now, but I also have 4.7. Would the amp benefit from that or is .47 fine?
Dutchgfx, do you find yours uncomfortable? Here's a tip for anyone who has them and hates the plastic against their ears: put a very thin layer of low density damping wool (like MDM3) inside the cup between the plastic grill and your ear. A very thin and low density layer will be soft on the ears, but basically acoustically transparent. I did notice a slight boost in bass, but not sure. Overall, no change in sound but a huge upgrade in comfort.
When I'm still here, I thought I'd ask whether any of the higher supply voltage op-amps are pin compatible with NE5532? I don't want to use up more board to build another one
I did it with a +-12 regulated supply from my Rod Elliot P06 phono stage. Used a NE5532 for now. Sounds pretty good and is plenty loud. What size input coupling should I use? I used .47uF K73-16V NOS for now, but I also have 4.7. Would the amp benefit from that or is .47 fine? Dutchgfx, do you find yours uncomfortable? Here's a tip for anyone who has them and hates the plastic against their ears: put a very thin layer of low density damping wool (like MDM3) inside the cup between the plastic grill and your ear. A very thin and low density layer will be soft on the ears, but basically acoustically transparent. I did notice a slight boost in bass, but not sure. Overall, no change in sound but a huge upgrade in comfort.
When I'm still here, I thought I'd ask whether any of the higher supply voltage op-amps are pin compatible with NE5532? I don't want to use up more board to build another one
The headband adjustment on mine broke so I used a hairtie (like for a girl's ponytail) to keep it in place. I don't find them uncomfortable. I don't listen to them often , I mainly use them for testing an amplifier so I don't break anything. They are like a wall of sound, but they are pretty neutral
I love mine. They are, as I said, slightly different from the originals. I also noticed that the driver itself is slightly different, having more vent holes on the back. I added a bit of damping, thick wool pads on the inner walls, while not blocking the semi-open back ports. This helps with damping but does not block the ventilation, having basically no drawbacks whatsoever. Also the insulation in the pads seemed to give a slight bass boost. I absolutely love these cans, they have a huge, open sound stage, deep and fast bass, very detailed sound, clear but not harsh highs. One of the best AKG headphones I've heard, literally nothing bad I can say about their sound. I guess the pleather pads make my ears a bit warm, but I'm getting some DT-990 pads to replace them.
600R headphones prefer voltage over current. Using I=E/R and P=(E**2)/R you'll find that for, say for 200mw of power, voltage is the limiting factor. However, the reality is that most headphones are very efficient, DT-880s need only 250-300mv AC to blow your socks off. I'd think that 2ea 9V batteries and a single rail-rail opamp (configured like a CMOY) would work just fine.
What you're saying is that I could get away with using 9V rails? What about two 4.5V rails? I see that the JDS Labs Cmoy uses that and they claim to support headphones up to 600ohm. A 9v battery with a voltage splitter is what they have. I could use resistors, I heard that for higher impedance cans, you can use a resistor splitter.
You see, I really want to make this a compact build now because at home, I can just plug the cans into my mixing console and be done with it. But it's not all well and good with my phone, that thing has a seriously lousy amp. It even has trouble driving a pair of Siberia V2s, a pair of headphones that get quite loid with pretty much any source.
Take a look a my thread:
http://www.diyaudio.com/forums/head...e-data-sheet-table-power-voltage-current.html
It seems your headphones are bitch to drive regarding voltage:
for 115dB SPL you'll need 17.86 Vrms and 28mA
Take all voltage you can take with you
http://www.diyaudio.com/forums/head...e-data-sheet-table-power-voltage-current.html
It seems your headphones are bitch to drive regarding voltage:
for 115dB SPL you'll need 17.86 Vrms and 28mA
Take all voltage you can take with you
Yep, at 17.86 volts, mine are one of the hardest to drive.Take a look a my thread:
http://www.diyaudio.com/forums/head...e-data-sheet-table-power-voltage-current.html
It seems your headphones are bitch to drive regarding voltage:
for 115dB SPL you'll need 17.86 Vrms and 28mA
Take all voltage you can take with you
The problem is that for now, I have to resort to a simple and cheap amp due to lack of funds, hence the Cmoy build.
So how much voltage do I need again to get a clean 18 volts out? Dual 9V battery obviously isn't going to suffice. I'm currently running it off of a dual 12 volt regulated power supply I built a while ago.
Current performance is quite good. Very low noise floor with no chassis or any kind of shielding. No distortion whatsoever on normal listening levels. In fact, I could hardly get it to distort at all. Using Benny Benassi's Satisfaction as a bassy test track, I could easily drive it to a point of uncomfortably loud before any audible low frequency distortion.
My Cmoy config: 10K feedback with 1K shunt, 100K input load, 470nF input coupling, 100nF film across supply rails right next to op-amp, 22uF filtering per rail on board plus a total of 1470uF per rail on regulator board. Using L7x12 regulators. Neither op-amp nor regulators get even warm. Transformer heats up slightly, tho. It's rated for about 40VA but it's a cheap piece of junk from a HT subwoofer. This single transformer was powering a total of 6 (!!!!) TDA2030A chips! Absolutely idiotic.
My current op-amp is the NE5532P. It was either that or a 4558, so.... yeah.... What would be an upgrade to this that I could get for less than like 5 bucks? I know the OPA2134PA is good and it's locally available, but what other good op-amps are there in that price range?
A big thank you to everyone for your contributions!
Last edited:
Consider the OPA551
Along with all the good suggestions already made, take a look at the OPA551 which is good for +/-30V rails. The opa551 is the chip used in the output stage of the Violectric V90 headphone amp. Here is the datasheet:
http://www.ti.com/lit/ds/sbos100a/sbos100a.pdf (opens PDF)
It comes in a regular 8 pin DIP or the DDPAK version for higher power dissipation. Farnell shows stock, but it is US stock with a big delivery charge unfortunately. Looks like RS components has stock. They are $5 here in the states at Mouser.
The OPA551s are single op amps. You would need one for each channel. Don't use the OPA552 version for your purposes. It is the uncompensated version for higher gains (above 5) and would require some external compensation.
As for power supply you could use a voltage quadrupler with a 12VAC transformer to get +/-28dc (or so, depending upon load) power supply rails: See here
(schematic at the bottom of the post on the left):
http://www.diyaudio.com/forums/head...rc-diode-cap-heatsink-mods-3.html#post3281967
Possible point of confusion - a single rail voltage quadrupler is the same as a dual rail voltage doubler. It just depends on which point you define as ground.
Along with all the good suggestions already made, take a look at the OPA551 which is good for +/-30V rails. The opa551 is the chip used in the output stage of the Violectric V90 headphone amp. Here is the datasheet:
http://www.ti.com/lit/ds/sbos100a/sbos100a.pdf (opens PDF)
It comes in a regular 8 pin DIP or the DDPAK version for higher power dissipation. Farnell shows stock, but it is US stock with a big delivery charge unfortunately. Looks like RS components has stock. They are $5 here in the states at Mouser.
The OPA551s are single op amps. You would need one for each channel. Don't use the OPA552 version for your purposes. It is the uncompensated version for higher gains (above 5) and would require some external compensation.
As for power supply you could use a voltage quadrupler with a 12VAC transformer to get +/-28dc (or so, depending upon load) power supply rails: See here
(schematic at the bottom of the post on the left):
http://www.diyaudio.com/forums/head...rc-diode-cap-heatsink-mods-3.html#post3281967
Possible point of confusion - a single rail voltage quadrupler is the same as a dual rail voltage doubler. It just depends on which point you define as ground.
Last edited:
Beyerdynamidc DT 880 600 ohm
10.28 Vrms for 115dBSPL
All I can say is at 250mv AC true RMS peak, measured (not calculated) with a pair of DT-880 and a Fluke 45 in min/max mode, my ears were ringing. I was afraid to go much higher. If you own a pair yourself, I'd be curious what your measurements show.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.