The HISSBUSTER - for sensitive headphones

[j-curve]

Modern headphones are too sensitive. How sensitive should they be? Some people find that previous favourites like the HD580 (at 102 dB/V) are hard to drive with portable players but I've had more frustration due to the opposite problem, i.e. background hiss from headphones which are too sensitive. Even the HD215 (which is only rated at 112 dB/V) throws out continuous noise from most soundcards and some headphone amplifiers too. For the sources I use, 95-105 dB/V seems to be a reasonable range of sensitivity and 100 dB/V would be optimum. [Note: Sensitivity (dB/V) shouldn't be confused with efficiency (dB/mW). This thread might help.]

Now look at some of the crazy specifications for headphones in the marketplace today. Full-size cans run up to 123 dB/V, earbuds and canal phones even higher. An inline volume control is one way to eliminate background hiss. Here's another, simpler method.

Recommended values depending on the reduction (insertion loss) desired, all resistors in Ohms:-
For -10 dB, R1=15, R2=22
For -15 dB, R1=6.8, R2=27
For -20 dB, R1=3.3, R2=27
For -25 dB, R1=1.8, R2=27
For -30 dB, R1=1.0, R2=33
NB: Consider using a 1/2 Watt resistor for R2 as it may dissipate some heat when the volume is cranked.
Input impedance is about 30 Ohms with headphones connected.

I built mine into an extension cable.
Sorry, no macro setting on the digital camera.


[Aside from busting hiss, sound quality will be improved in some cases. e.g. If you're using a portable player with a digital volume control which truncates bits, you'll hear how bad it sounds at low volume settings. That's not the fault of the headphones(!), however it is indirectly caused by the headphones because they are too sensitive.]

[j-curve]

Bumpety bump.

After a few weeks of use I can say this simple attenuator is a winner. All my hissy sources (soundcard, integrated amplifier etc.) are now hiss-free. They still have plenty of headroom too. My only concern is the packaging as I'm not sure it will withstand being squashed under foot.

I used R1=5.6, R2=27 which takes the HD215 down to 95dB/V.

[audiomagnate]

Doesn't anybody make one of these comercially? Built into a 1/4 to 1/8 in aadaptor would work for me.

[Bob_McBob]

How does this affect the sound of Westone UM2s? I have heard most attenutators affect it negatively, but the hiss from my iPod is really bothering me.

[kramer5150]

can you use SMT resistors? I find them a LOT smaller. Or does it have to be metal film variety?



Whats the difference between your solution and a resistor in line with one of the leads?.
I did something similar with my koss plugs, which are ~12 ohms by themselves.

[audiomagnate]

Originally Posted by kramer5150

can you use SMT resistors? I find them a LOT smaller. Or does it have to be metal film variety?



Whats the difference between your solution and a resistor in line with one of the leads?.
I did something similar with my koss plugs, which are ~12 ohms by themselves.

For one thing, he's using one resistor in parallel and one in series. If you do it right you get attenuation w/o an impedence change, if thats what you want.

[pikasu321]

wow, someone should totally build and sell these as adapters. I would love to have this for my E500s and UM2s.

[j-curve]

Bob_McBob: How does this affect the sound of Westone UM2s? I have heard most attenutators affect it negatively, but the hiss from my iPod is really bothering me.

Wow, a Westone UM2 owner. Did you know that in the spreadsheet (see link in first post) they came up as the most sensitive headphones? 135 dB/V(!!!) If there's any source which doesn't hiss with those then it's probably got a flat battery. All I can say is... you've come to the right place. Unfortunately I can't speak to the effect on the sound, except to say that with the hissbuster plugged in, everything should look about the same to both the source (which still sees a 30 ohm load) and the 'phones (which will still see a very low impedance source). I'd recommend the -30dB version for starters, although there is a bit of a risk as different manufacturers may use different measuring methods.

If most attenuators affect the sound in a negative way, it's probably because they don't provide a low impedance output to drive the 'phones, for example if they consist of a simple series resistor (R2 without R1 in the circuit above), as audiomagnate has pointed out. Technically this turns the circuit into a current source instead of a voltage source, causing the frequency response to be altered according to the varying impedance of the headphones and impairing the electrical damping factor to the diaphragm.

kramer5150, if you have SMD skills then go for it! I envy you. Metal film has a lower thermal coefficient, but I wonder if we have to worry about that. I used carbon film for R2 but that was only because I'm paranoid and wanted half a Watt of heat dissipation for the day someone plugs some old AKG cans into this thing. Overkill really, a 1/4 Watt would have been fine. If you do worry about thermal effects on the signal then just use the same type for both R1 and R2, since this is a divider circuit.

audiomagnate: Doesn't anybody make one of these comercially?
pikasu321: wow, someone should totally build and sell these as adapters.

I wish someone would. I believe there have been a couple including Shure's inline volume control for use on planes, however I can't vouch for those products because this kind of low impedance gear is really quite specialised. Try finding 50 or 100 ohm dual gang potentiometers, or metal film resistors with values below 10 ohms. Rare stuff.

[audiomagnate]

I'm doing the 10dB mod on Mitsubishi MA-01 tommorow. I hope that's enough. I'm just going to do it permanently to the jack instead of building a cord. I new this was possible, but now I don't have to look up the values. Thanks!
BTW, my UM2's are dead silent out of my lowly MPIO FL 100, and it sounds great too!

[j-curve]

Hi audiomagnate. Quick question - what's the power output of that Mitsubishi? I can't find any info on it, other than that it's a well regarded amp from circa 1980. The reason I ask is that you have to think about power issues when you build something like this into a device which is capable of driving speakers. Usually, power amps with headphone sockets have output resistors to protect the headphone socket from sparking, melted terminals, etc, but which can also adversely affect the sound. In that case an external IMPEDER (see link in sig) could be the way to go. Sorry to be a bore, just concerned that 1/4 Watt resistors might get rather hot in there.
It's still a good mod to do though. You'll probably find R2 is already built into the amp - then it's just a matter of choosing a suitable R1.

[lipidicman]

Originally Posted by audiomagnate

For one thing, he's using one resistor in parallel and one in series. If you do it right you get attenuation w/o an impedence change, if thats what you want.

Is this correct? (I haven't figured it out, so this is a genuine question). I have wanted to build one for ages but have found the info out there confusing.

Meier audio's tips and tricks confused me. I thought a resistor in line would increase the impedance, but on the site it talks about this increasing the amp's impedance (which in my mind further increases the imbalance. I suspect my Xcan has a high impedance for the 580 etc and hisses with low impedance phones, again I may be way off here)
http://www.meier-audio.homepage.t-online.de/ see tips and tricks

That aside the idea is exactly what I need. I use the Shuffle real low with my CX300 and the hiss remains the same across the volume scale (digital bit dropping?) This should be SAFER for my ears and drop the hiss. Anyone have any R values for this combo, or better still has actually tried it?

My final question regards the effect on battery life

[audiomagnate]

Originally Posted by lipidicman

Is this correct? (I haven't figured it out, so this is a genuine question). I have wanted to build one for ages but have found the info out there confusing.

Meier audio's tips and tricks confused me. I thought a resistor in line would increase the impedance, but on the site it talks about this increasing the amp's impedance (which in my mind further increases the imbalance. I suspect my Xcan has a high impedance for the 580 etc and hisses with low impedance phones, again I may be way off here)
http://www.meier-audio.homepage.t-online.de/ see tips and tricks

That aside the idea is exactly what I need. I use the Shuffle real low with my CX300 and the hiss remains the same across the volume scale (digital bit dropping?) This should be SAFER for my ears and drop the hiss. Anyone have any R values for this combo, or better still has actually tried it?

My final question regards the effect on battery life

Yes it's correct. A series resistor increases resitance/impedance (I know they're not the same thing, but they're closely related) and a parallel resistor reduces both. So if you use the correct values you get attenuation (through heat in the resistors) without an overall change in the resistance/impedance the amp sees. The concept is used in crossover design for speakers all the time, which I did in a past life.

[audiomagnate]

Originally Posted by j-curve

Hi audiomagnate. Quick question - what's the power output of that Mitsubishi? I can't find any info on it, other than that it's a well regarded amp from circa 1980. The reason I ask is that you have to think about power issues when you build something like this into a device which is capable of driving speakers. Usually, power amps with headphone sockets have output resistors to protect the headphone socket from sparking, melted terminals, etc, but which can also adversely affect the sound. In that case an external IMPEDER (see link in sig) could be the way to go. Sorry to be a bore, just concerned that 1/4 Watt resistors might get rather hot in there.
It's still a good mod to do though. You'll probably find R2 is already built into the amp - then it's just a matter of choosing a suitable R1.

The MA-01 is rated at 100 watts per channel into 8 ohms. It is a real gem, I have three. I was head of warranty for Mitsubishi in the late 80's, and gobbled up as many as I could at employee sales. Audio magazine was nothing short of astounded when they reviewed it. I have'nt checked out your link yet, but it's on my list, so I'll delay my mod.

[TimmyMac]

If I'm looking at this right... most of the current goes through the little cross resistor then? And the phones only see a small amount of current?

R1 / impedance = % amount of current that goes through the phones?

Not terribly familiar with audio electronics but I understand a little of regular electronics.

[lipidicman]

This thing is a voltage or potential divider.
http://www.electronics2000.co.uk/dat...smr/potdiv.htm
http://en.wikipedia.org/wiki/Voltage_divider
The voltage your headphones see is input voltage x R1/(R1+R2) (unloaded) see wikipedia link

So the signal and the noise floor is attenuated. Then you can up the volume thus reducing the prominence of the hiss

Right so I have 16ohm CX300 on a shuffle. Resistance figures so the impedance is unchanged?