[adidex]

Hello there,
I searched for an headphone amplifier schematic for my AKG K240M headphones (600 ohms), and I am not satisfierd with what I have found till now. The CMOY amplifier will hapilly work I guess with 600 ohm headphoes but I think it doesn't have the current capacity to drive lower impedance headphones (32 ohm), for more than 9+9V power supply. The other amplifiers seem too complex for me (more than 5 transisotrs for the diamond buffers, FETs, etc.).
So, I was thinking about using a normal power amplifier IC (Like TA8205, or something similar) as an headphone amplifier. What do you think, what results should I expect ? Will the noise be a problem ? A specialised medium power IC amplifier at more 32 ohm might be better than an AO and a 2-transistor output stage ... It's not hard for me to build one, but I don't have another specialised amplifier to compare it with, so I was thinking that maybe somebody else here allready tried this ...

Thanks.

 

[Garbz]

There are plenty of opamps out there which are capable of driving low-z headphones with sufficient current. Look at the chip used in Grado's amp (someone else will need to tell you the number). The OPA627 has 45mA current output instead of the OPA134's 35mA, and arguably sounds better too. The OPA227 can also supply this.

/EDIT: Look also at the AD8022 which can supply 55mA output linear and 100mA in short circuit.

 

[adidex]

Well .. I allready got a few OPA2134, but then I read on the datasheet this: "OPA134 series op amps are capable of driving 600ohm loads with power supply voltage up to ±18V.". Ok, that's perfect for the 600ohm headphones, but for 32 ohm I guess it will be a problem (I will have to use the same voltage supply, so the power dissipation will increase dramatically). The only solution is to decrease the voltage supply for the low-impedance headphones ... That's not nice ...
Of course, it is also possible to use a AO with a higher power dissipation capability (maybe an AO that has 50-60mA output capacity would be good enough), but I was curious to find out if an medium-power amplifier is also up to the task (since it only needs to drive an higher impedance it should behave better - damping factor), but maybe the noise will be higher (and also the distortion ?). What do you think ?

 

[Tomo]

Hey,

You can try some ethernet cable drivers. ie THS4022 or THS6012. They are powerful. I think peak current is like 150mA.

Most common method is to upgrade CMOY to make Buffer-Opamp combinations. You should immediately see the design on the spec sheet for OPA132 or BUF634. This combination is very very common and more well known under different name, META. You might already heard about upgraded version of META with different feedback scheme.

Tomo

P.S. This is my personable opinion, but I think K240M sounds better with a single opamp as opposed to my BUF634-OPA627 amp.

P.P.S. Supply voltage will make no difference. Opamps don't burn up if you drive the sukka at +-18V or +-4.5 as long the voltages are within the tolerance range. Reducing the supply voltage will not suddenly make it better for low impedance cans.

 

[adidex]

Thanks, Tomo, for your answer ...
Yes, I heard about different other AO-based flavors of amplifiers, but right now I just want to build as soon as possible an amplifier, and don't want to wait another 10-15 days 'till I get some new components .. But I guess i will build a simple amplifier with just a single AO, and then see how it works with different impedances (anyway, I just want to use 32ohm hedphones ocasionally ..).
You said about the voltage supply .. I think you still have a problem, I know the AO would not be fried using it in various configurations, but I am more concerned about taking it to the limit (the performances decrease). For High-Impedance headphones i will need high voltage, and for LowImpedance High Current. So if i will supply the AO with with more than 12+12V let's say, it will work just fine for the 600ohm, but the power dissipated on a 32ohm load will increase dramatically, so there's the problem .. But enough talking, i'll just build it, and see how it behaves in various circumstances ..
Thanks for the replyes ...

 

[Tomo]

Hey,

What is an AO? I dont get it because I drive both my HD580 (300ohms) and PX100 (32ohms) without much trouble with a single opamp. My opamps dont go up in smoke even when I stress tested by running my opamps for like a whole month.

Only trouble in using an amp I built for 300-600 ohm cans to drive my 32 ohm can is that the GAIN is too high.

TA8205 (Toshiba?) is rather too powerful. But if you get it to work, you are set. I found a schematics for you. This schematics will tell you how to wire BTL amp to a can.

http://www.fulcrum.ru/Read/CDROMs/NS/LM4867.pdf

Check out the schematics on the second page. I dunno what you do with the Mono input though.

Good luck, and tell me how it comes out. Ciao.

Tomo

 

[rickcr42]

Those headphones were never meant to be driven by what we consider to be a headphone amp but instead directly from a studio monitor amp through a 120 ohm 1W resistor.
Common practice in studios is to use a dedicated 20-30 WPC "Pro" power amp with left/right volume controls (Crown D-60 was the standard for years) straight from the speaker posts with no intervening resistors.

This headphone is not suited to having a whimpy opamp driving it

 

[adidex]

Thanks, Tomo,
AO means opamp (I was using jargon, sorry

),
It is very unlikely that you will ever see an opamp in flames, even if you shortcircuit the output (every modern opamp that I know has output current limitation circuitry). The easiest way to burn an opamp is to supply it at a bigger voltage that it was built for, or reverse the voltage supply sense (or maybe ESD-it). The output limitation circuitry is in fact the problem, 'cause when the output current is getting closer to the maximum current that is specified in the datasheet the performances will degrade very fast. There are 3 types of limitations possible: voltage (high impedance load), current (low impedance load) and power (a combination of both), so that's why I was concerned, because I want the amplifier to work with both high and low impedance loads. But I guess you allready know this ...
Thanks for the link that you sent me, but I think I need a higher power amplifier, because I tried the AKGs with an 12W amplifier that I build a few years ago (it has an general purpose car amplifier in it - TDA1510) and it only delivers enough power for the headphones when it's at full power (I guess it cannot generate more power on 600ohm because of the single 12V supply). I cannot get more power for the 600ohm headphones without increasing the power supply (that's why I was planning to use minimum 9V differential for the opamp). I also used TA8215A (Toshiba) in another amplifier, so it won't be a problem to make another one for the headphones (I allready got one TA8215 and an TA8205), and that's what I am going to do in the following days (I don't have that much time, unfortunately), and i will post here the results (comparison between an simple OPA2313 opamp amplifier and an TA8215). The amplifier that I allready have is very noisy and it is distorting (it's just an low-end amplifier for the PC output ..).
I'll get back with the answer in a few days .

 

[rickcr42]

Guess I should add that I use this headphone as a recording monitor and have for about twnety years and have always driven it with a 30W amp.The reason for the 600 ohm impedance and low efficiency is an attempt to make this headphone "blowout" proof and AKG succeeeded there.
A tougher can has probably never been made but I notice looking at the site recently it has been discontinued in favor of the 240 "Studio" version which is an entirely different beast.
To me this is a huge mistake but most likely a decision based more on consumer ignorance (trying to use the cans where not suited such as "headphone amp" asamp

) causing its decline rather than sound reasons for the design itself.I expect a lot of 240 "S" cans will be showing up to be serviced when the voice coils "pop" because they are made more with an eye towards the audiophile consumer rather than the true pro who needs reliability and bullet proof more than any other parameter.

"hold that performance guys.I have to go get another set of headphones before we can continue" and all at $500/hr for even a low end basement studio

 

[adidex]

Yeah, rickcr42, those 600ohm headphones were made to last forever (I had an 20+ year old pair of K240Ms working perfect), but I think AKG is still making 600ohm phones (the M version), they just have more types now (impedances).

As far as i know, they made 600ohm phones for another reason also : to make it possible to hook-up more phones in parallel (with a "normal" 4ohm amplifier you can drive ... 150 pairs of headphones without any problem

), and that's good for the studio use.

And it's been an misunderstanding here, we can only talk about the power of
an aplifier only when we know the load value, an amplifier that generates, let's say, 30W per 4 ohm will only generate something like 0.2W on an 600ohm load (it will generate 30W if you put 150 headphones in parallel). So that's why an normal opamp can easily drive one pair of 600ohm headphones, but it will not have the output capacity of an 30W/4ohm amplifier. We only need an small output current for an 600ohm load.

Regards,


Adi.

 

[rickcr42]

On paper yes but they sound like crap without a real amp having a high damping factor,one that obviously reduces with the advancing load but still WAY higher than any opamp/buffer combo will provide.

The other thing is why and how of design.i wish more manufacturers would spec exactly what type of equipment they actually voiced/tested their headphones on.
The older AKG monitor series and some of the older Beyer and KOSS cans were meant from the start to be driven by an 8 ohm tap on a low power speaker amp so anything else means you are not hearing the cans as they are meant to be heard.
This goes to how many report vintage cans with vintage amps sound great even though those same amps or cans used with a modern counterpart usually sound like crapola.

all about the matchups but they need to give us more than equations and tell us exactly how they voiced the cans for actual use before we can truly evaluate them as intended