High impedance doesn't equally not loud

[Ahoy Hoy]

Since there has been a few more then usually asking about headphone amps and whether this headphone or that headphone is going to be able to run on this or that. I thought Id make this quick explanation to why just because a headphone is 250 Ohms doesn't mean it wont be loud or is too difficult for something to produce a loud enough sound.


Everyone seems to understand Ohms/Impedance/Resistance means the difficulty for the electric signal to go through the headphones. 
But no one ever discusses the sensitivity of the headphone. This is how much power is needed to produce a volume. (It doesn't matter if the headphones are hard to give power if they only need a little to get loud)
With both of these bits of information you can find out very easily if your motherboard or your phone or what ever will be powerful enough for your headphones.


So Im going to take a few popular high impedance headphones and some use cases and work out if they are good enough to run the headphones.

                                                      Headphones

Model                                                          Impedance                                  Sensitivity 
Beyerdynamic DT990 Pro                            250 Ohms                                   96dB/mW

Beyerdynamic DT1990                                 250 Ohms                                  102dB/mW

Sennheiser hd600                                        300 Ohms                                   97dB/mW



                                                          Source
Name                                                 Impedance                                         Output power
Iphone                                                300 Ohms                                               3mW
Gigabyte X570 Aorus                         250 Ohms                                             4.4mW



So the two source a Iphone and a motherboard both at a high resistance produce enough power to not only run these "hard to drive" headphones but at loud enough volumes that if that was the noise level at work you'd legally be required to have protective equipment to stop hearing damage.


So regarding the question do I need a Amp to power these headphones the answer 99% of the time is no. 
Most modern motherboards which are not the bottom spec (even then they manage) have good enough amplifiers built in you that don't need a standalone amp. But if youve got money  to spend on a amp just spend it on a better motherboard. Not only will you get better audio parts built in you will also get better motherboards with extra features.

If the headphone is loud enough currently you don't need a amp.

[Senzelian]

Seems plausible. Thanks for the information!

 

With the numbers you mentioned, would it be possible to calculate the output volume?

And if so, can you give me an example?

[Jurrunio]

then what would happen if impedence is indeed too high (or if my board is just really old and no longer capable of doing what it is rated for)?

[Glenwing]

9 minutes ago, Jurrunio said:

then what would happen if impedence is indeed too high (or if my board is just really old and no longer capable of doing what it is rated for)?

Maximum power is achieved when the impedances are matched. If the impedance of the headphones is higher or lower than the source's output impedance, then the volume will be lower than if the headphones were the same impedance as the source device.

[Jurrunio]

13 minutes ago, Glenwing said:

Maximum power is achieved when the impedances are matched. If the impedance of the headphones is higher or lower than the source's output impedance, then the volume will be lower than if the headphones were the same impedance as the source device.

It doesnt affect quality right? volume is not my greatest concern, I rarely even go beyond 5% in Windows for volume unless I really need to figure out what I am hearing.

[Ahoy Hoy]

24 minutes ago, Senzelian said:

Seems plausible. Thanks for the information!

 

With the numbers you mentioned, would it be possible to calculate the output volume?

And if so, can you give me an example?

You can. I was taught it by my audio lecture but I will refrain from giving a example as I cant remember the equation that well and sound is logarithmic so its a annoying equation. But there is a rule of thumb which makes it actually really easy. To increase the volume by 3dB you have to double the power. So in the case of the Iphone and the Sennheiser headphones you can double the power 1 and a quarter times so thats a extra 3.75dB so max SPL should be around 100.75dB. Reason why its 1.25 times and not 1.5 is from 1mW to 2mW is double then the next double would take it to 4mW which is over the max.



 

 

23 minutes ago, Jurrunio said:

then what would happen if impedence is indeed too high (or if my board is just really old and no longer capable of doing what it is rated for)?

If the amp is not very powerful and the headphones are not sensitive then the volume would be low.

Headphone impedance isnt equal along the frequencies and normally the low end so you bass frequencies have a high impedance then the rest. So in a extremely underpowered headphone will produce quieter low end as well as being quite.

 

 

3 minutes ago, Jurrunio said:

It doesnt affect quality right? after volume is not my greatest concern, I rarely even go beyond 5% in Windows for volume unless I really need to figure out what I am hearing.

At 5% volume in windows your amp is powerful enough. Your running the amp at 1/20 of its max output and its producing a loud enough volume. If you had it at 100% volume and it wasnt producing a loud signal then you would have problems with quality where you would possibly losing low end. But you may also start hearing interface at 100% of a amp due to other factors such as power management and signal quality.

[akio123008]

24 minutes ago, Ahoy Hoy said:

You can. I was taught it by my audio lecture but I will refrain from giving a example as I cant remember the equation that well and sound is logarithmic so its a annoying equation.

 

47 minutes ago, Senzelian said:

would it be possible to calculate the output volume?

You can calculate intensity in a non logarithmic fashion with the inverse square law;     I = P / (4PI * r^2)     plug in the source power as P (watts) and the distance from the source as r (meters) and you get the intensity in W/m^2.

 

To get the intensity in dB, you can use        L = 10log(I / Iref)         where L is the intensity in dB and I the intensity in W/m^2 just found. Iref is some reference that depends on human hearing, some value you can find on the internet.

 

Problem with all this, the inverse square law assumes a point source and the sound propagating in all directions like a sphere (hence 4PI * r^2, the surface area of a sphere) with no obstacles. In this case, we're talking about headphones, where the sound's somewhat trapped in a small tunnel, meaning the intensity will probably turn out way higher than one calculated purely with the maths above. It's extremely hard to calculate an accurate value for this.

[rice guru]

58 minutes ago, Ahoy Hoy said:

f the headphone is loud enough currently you don't need a amp.

I repeat this all the time. but I will say 250 ohm open back beyers do need more power than it seems they dont seem to be as easy to drive as noted. on a weaker source. you get a looser bass response not as much impact. the mids are pulled back slightly . on a well paired source it becomes a more controlled headphone with more impact. on phones they dont get loud at all. on my portable dac amp they get loud but have the previously stated issues. on my motherboard they get to a decent listening level but they dont get loud with 0 headroom,  you can definitely deny me with math and I will admit this just purely anecdotal admittedly. there is a reason I tend to push sennheisers or the tygr instead of the 250 ohm 990 , 880 for people in the first place.

[The Flying Sloth]

11 minutes ago, rice guru said:

I repeat this all the time. but I will say 250 ohm open back beyers do need more power than it seems they dont seem to be as easy to drive as noted. on a weaker source. you get a looser bass response not as much impact. the mids are pulled back slightly . on a well paired source it becomes a more controlled headphone with more impact. on phones they dont get loud at all. on my portable dac amp they get loud but have the previously stated issues. on my motherboard they get to a decent listening level but they dont get loud with 0 headroom,  you can definitely deny me with math and I will admit this just purely anecdotal admittedly. there is a reason I tend to push sennheisers or the tygr instead of the 250 ohm 990 , 880 for people in the first place.

I completely agree and @Tigerleonwill back that up, a higher quality, higher power external DACAMP or interface vastly improves the quality of 250 ohm Beyer's along with many other high-impedence headphones.

 

Numbers are important but they aren't everything. If Henry Saylene (sorry if I misspelled that) or Derkoli want to come in and grace us with proper technical analysis I'm sure we'd all appreciate it.

[Tigerleon]

12 hours ago, The Flying Sloth said:

I completely agree and @Tigerleonwill back that up, a higher quality, higher power external DACAMP or interface vastly improves the quality of 250 ohm Beyer's along with many other high-impedence headphones.

 

Numbers are important but they aren't everything. If Henry Saylene (sorry if I misspelled that) or Derkoli want to come in and grace us with proper technical analysis I'm sure we'd all appreciate it.

That I will . If you get good listening levels doesn't mean you have got the full potential of them. The reason why you would need an amp, as rice guru said. You have to look at the headphone specs first and see if they have good efficiency to drive them off your mobo. If they have similar specs to the DT 990 pros for example, you wont be able to push them without a good amp. The sensitivity of the DT 990's are 96dB/mW which is much lower than for example 104dB of the HD58X's. They are still 150ohm but can be driven pretty good without an amp.

I have experience with not properly driving the DT 990's. When I first got them. I did not have an amp. So to get my DT 990's sounding great to me. I had to put the EQ all the way up and same on windows. Without them sounding boring and dying. So when I got my DAC/amp sound card I realized how much I was under powering them because I even had to turn down some EQ because the treble got so much fuller same with the bass, and the mids didn't hide as much as well.

[Ahoy Hoy]

8 hours ago, Tigerleon said:

That I will . If you get good listening levels doesn't mean you have got the full potential of them. The reason why you would need an amp, as rice guru said. You have to look at the headphone specs first and see if they have good efficiency to drive them off your mobo. If they have similar specs to the DT 990 pros for example, you wont be able to push them without a good amp. The sensitivity of the DT 990's are 96dB/mW which is much lower than for example 104dB of the HD58X's. They are still 150ohm but can be driven pretty good without an amp.

I have experience with not properly driving the DT 990's. When I first got them. I did not have an amp. So to get my DT 990's sounding great to me. I had to put the EQ all the way up and same on windows. Without them sounding boring and dying. So when I got my DAC/amp sound card I realized how much I was under powering them because I even had to turn down some EQ because the treble got so much fuller same with the bass, and the mids didn't hide as much as well.

This is the frequncy response Dt990 600 ohm (old model which is actually hard to run at 95dB mW) being tested on five different amps. Can you even tell a difference. Where is this clarity in the mids and fuller bass your getting as that graph has a difference of less then 1dB.


Heres another one again DT990 this time only two amplifiers. Again only a minor difference of around 1dB at the low end.

I have done a amplifier difference test at university. One where we used different amps with the same power outputs. That gave small difference in frequncy response as you expect with different designs. Then we used two amps one with a higher output then the other both were the same brand. Set gains so the SPL out of the speaker was the same. The difference in frequncy response again was small.


Any amplifier with the same input impedance same output impedance and has a flat response will sound the same. But there are amplifier out there especially in the headphone market which are not flat responses. People buy headphone amplifiers with tube amps and other features which change how the signal sounds because they don't amplify equally with a flat response. If you think a headphones needs a different sound response to be more enjoyable. I'm not going to argue that as that's personal opinions.

Now if your amplifier really is terrible you will have problems like you said where the low end sounds bad. The difference between a great amp and a okay amp is how flat of response they create. BUT nearly every electronic device in the last decade has been built with a good enough amplifier that the difference in response is so minor that its not noticeable to most people and the power output is good enough for majority of headphones.

If your comments about the dt990 are from two decades ago then your probably correct as amps back in those days were terrible inside electronics. But if your comment is about a modern decent spec device then I don't understand how you can say it.

Balls in your court I am ready to have my mind changed on this. If you can give me frequncy responses or blind hearing tests which prove your theory I am ready to go buy a amplifier and become a believer.

[Tigerleon]

3 minutes ago, Ahoy Hoy said:

This is the frequncy response Dt990 600 ohm (old model which is actually hard to run at 95dB mW) being tested on five different amps. Can you even tell a difference. Where is this clarity in the mids and fuller bass your getting as that graph has a difference of less then 1dB.


Heres another one again DT990 this time only two amplifiers. Again only a minor difference of around 1dB at the low end.

I have done a amplifier difference test at university. One where we used different amps with the same power outputs. That gave small difference in frequncy response as you expect with different designs. Then we used two amps one with a higher output then the other both were the same brand. Set gains so the SPL out of the speaker was the same. The difference in frequncy response again was small.


Any amplifier with the same input impedance same output impedance and has a flat response will sound the same. But there are amplifier out there especially in the headphone market which are not flat responses. People buy headphone amplifiers with tube amps and other features which change how the signal sounds because they don't amplify equally with a flat response. If you think a headphones needs a different sound response to be more enjoyable. I'm not going to argue that as that's personal opinions.

Now if your amplifier really is terrible you will have problems like you said where the low end sounds bad. The difference between a great amp and a okay amp is how flat of response they create. BUT nearly every electronic device in the last decade has been built with a good enough amplifier that the difference in response is so minor that its not noticeable to most people and the power output is good enough for majority of headphones.

If your comments about the dt990 are from two decades ago then your probably correct as amps back in those days were terrible inside electronics. But if your comment is about a modern decent spec device then I don't understand how you can say it.

Balls in your court I am ready to have my mind changed on this. If you can give me frequncy responses or blind hearing tests which prove your theory I am ready to go buy a amplifier and become a believer.

It definitely depends on how good your mobo amp was in the first place. Me having experience with my mobo and then upgrading to a soundcard made a huge difference to me anyways. Could be a different story for someone else. I definitely heard a difference anyways.

[Ahoy Hoy]

Just now, Tigerleon said:

It definitely depends on how good your mobo amp was in the first place. Me having experience with my mobo and then upgrading to a soundcard made a huge difference to me anyways. Could be a different story for someone else. I definitely heard a difference anyways.

Blind test your self. If you ever stumble upon the audiophile forums (sorry if you do) once in a blue moon someone says I blind tested my self on my dacs and amps and couldn't tell the a difference

[Tigerleon]

Just now, Ahoy Hoy said:

Blind test your self. If you ever stumble upon the audiophile forums (sorry if you do) once in a blue moon someone says I blind tested my self on my dacs and amps and couldn't tell the a difference

So you are telling me there is no difference between a mobo amp and a desktop/soundcard amp? 

[Ahoy Hoy]

Just now, Tigerleon said:

So you are telling me there is no difference between a mobo amp and a desktop/soundcard amp? 

This is the whole point of this thread. IF its a decent spec modern motherboard the onboard audio is normally very similar to those of a external. Especially on the low end external (sub 150). The main difference is power and noise. Which is another point of thread that again most modern stuff has enough power for most headphones. I also dont remember a time in the last decade I've even thought about noise or gain structure because most stuff these days is good enough that I cant hear it.

If your building a computer and youve got the choice between a £150 motherboard with a 150 DAC Amp or just spending £250 on a motherboard and get the same result just spend it on the motherboard because you also get extra features on the board.

[Derkoli]

21 hours ago, akio123008 said:

In this case, we're talking about headphones, where the sound's somewhat trapped in a small tunnel, meaning the intensity will probably turn out way higher than one calculated purely with the maths above. It's extremely hard to calculate an accurate value for this.

It gets harder again once you factor in different types of headphones (Closed & Open back, etc.)

 

13 minutes ago, Tigerleon said:

So you are telling me there is no difference between a mobo amp and a desktop/soundcard amp? 

There is a fairly large difference. Really, for cheaper external stuff vs a motherboard it's mostly power output and output impedance differences.

 

For the higher end gear you will get some more differences. Mostly crosstalk, noise rejection benefits, near immeasurable distortion etc.

 

Typically, the output of a motherboard (even a high end one, I'll get back to this) is much below the nominal 2 volts we expect to see when measuring a DAC. There is often also channel imbalances and channels just not measuring the same. A motherboard that is very guilty of this is the Gigabyte Z390 Aorus. Channel 2 has always measured worse than Channel 1.

 

Funnily enough that Gigabyte Z390 also has a habit of having a huge bandwidth related noise issue at 65kHz, not an audible issue but it still makes me giggle.

 

Another huge, huge, huge issue with motherboards is having a very high output impedance. This will change how your headphones sound which often makes them sound bad.

 

The Gigabyte Z390 Aorus has an output impedance of 78 Ohms. Ideally, a Headphone Amplifier should have an output impedance of 0. The best amplifiers typically have an output of just under 1 Ohm.

 

Usually, the upgrade people hear when going from a motherboard to a standalone Headphone Amplifier is the result of having a low output impedance so your headphones tonality is unchanged. Usually the power output isn't a huge issue.

 

 

 

Personally, I've always found a huge difference between most amplifiers, but I have been trained to pick out the audible differences between different bits of gear.

 

I am firmly in the camp that if two bits of gear measure the same, they will sound the same.

[Tigerleon]

50 minutes ago, Ahoy Hoy said:

If your building a computer and youve got the choice between a £150 motherboard with a 150 DAC Amp or just spending £250 on a motherboard and get the same result just spend it on the motherboard because you also get extra features on the board.

Eerm. No. Amplifying a DAC on an already bad one will increase the interference even more if you have interference. The best way to go is an amp/DAC combo or soundcard. Because then you have more flexibility and less risk of interference.

Also, ALL motherboards I know can't output the same as a 100-150$ amp can.