I am well aware there are standards. What I asked was what are normal listening levels i.e what levels do people actually listen at rather that the levels they are supposed to listen at.
Cheers
Ian
ED 80 to 83 db’s
Ian,
I listen to music for therapy or pain management. In my world music has a positive emotional response. Last night I sat down in the chair in my listening room with db meter in hand. An effective dose of music was 80 to 83 db’s. The db meter does not fit inside the headphone but my guess is that an headphone ED is about the same or in the range of 80 to 83 db’s.
Yes the music is addicting.
Cheers
DT
All just for fun!
Ian,
I listen to music for therapy or pain management. In my world music has a positive emotional response. Last night I sat down in the chair in my listening room with db meter in hand. An effective dose of music was 80 to 83 db’s. The db meter does not fit inside the headphone but my guess is that an headphone ED is about the same or in the range of 80 to 83 db’s.
Yes the music is addicting.
Cheers
DT
All just for fun!
Can you say a bit more about exactly how you used the dB meter - I assume it's a sound level meter - to measure the SPL. Did you just stick the earphone onto the mic of the SPL meter? I ask because I have such an SPL meter and for some reason it never occurred to me to use it to measure headphones level - LOL.
Cheers
Ian
Ian,
This is how I did it. I cheated and extrapolated a SWAG at it. I turned on the tubes put on the vinyl and used MTM dipoles plus subwoofer in the near field. With a Radio Shack sound meter setting on the floor I adjusted the spl to where I liked it. I tried a little louder then a little less loud and then somewhere in between. When it was just right I picked up the hand held meter and watched it for a few minutes and made the extrapolation that headphones would be not much different. 80 to 83db’s with the meter set on the DBa range.
I digress briefly. After playing with this stuff, spending time and money (the time is well spent, the money was not wasted) for personal listening I have come to prefer the near field dipoles to any headphones.
In the past I have put a Behringer sampling microphone through holes drilled into a plywood box about the size of my headO on one side of the box I glued on skin and a fake ear made from closed cell neoprene rubber. Then I placed the headphone on the box head. This was to look at frequency response. This might be an approach to look at spl’s inside the headphone when it is on the dummy’s head. (no hearing loss)
I think that the microphone needs to be inside the headphone when it is on an ear and head (even fake head) to get a good reading of spl. Put the headphones on your head, adjust the spl then place the headphones on the test dummy.
DT
All just for fun!
This is how I did it. I cheated and extrapolated a SWAG at it. I turned on the tubes put on the vinyl and used MTM dipoles plus subwoofer in the near field. With a Radio Shack sound meter setting on the floor I adjusted the spl to where I liked it. I tried a little louder then a little less loud and then somewhere in between. When it was just right I picked up the hand held meter and watched it for a few minutes and made the extrapolation that headphones would be not much different. 80 to 83db’s with the meter set on the DBa range.
I digress briefly. After playing with this stuff, spending time and money (the time is well spent, the money was not wasted) for personal listening I have come to prefer the near field dipoles to any headphones.
In the past I have put a Behringer sampling microphone through holes drilled into a plywood box about the size of my headO on one side of the box I glued on skin and a fake ear made from closed cell neoprene rubber. Then I placed the headphone on the box head. This was to look at frequency response. This might be an approach to look at spl’s inside the headphone when it is on the dummy’s head. (no hearing loss)
I think that the microphone needs to be inside the headphone when it is on an ear and head (even fake head) to get a good reading of spl. Put the headphones on your head, adjust the spl then place the headphones on the test dummy.
DT
All just for fun!
Last edited:
Hi DT,
Thanks for the info. I just did a similar experiment in my little studio sitting about a metre from the speakers and got very similar results. However I did notice the meter was set to the A weighting setting. When I switched to the C weighting the levels peaked at 95dB. Even so, it seems to me that designing for a max level of 100dB SPL should be more than adequate.
Thanks for your help.
Cheers
Ian
Thanks for the info. I just did a similar experiment in my little studio sitting about a metre from the speakers and got very similar results. However I did notice the meter was set to the A weighting setting. When I switched to the C weighting the levels peaked at 95dB. Even so, it seems to me that designing for a max level of 100dB SPL should be more than adequate.
Thanks for your help.
Cheers
Ian
Ian,
I like your question or thought about the required design power or db output of a headphone amplifier should have. The same question could be asked of a power amplifier for loudspeakers. I am afraid that if we look closely there will be some disappointed audiophiles. I believe clipping is the rule. Put the question another way, how much headroom is needed above average listening spl’s to avoid clipping of the peaks. Somewhere here or in the other current headphone thread someone said that headphone amplifiers have power to burn.
The answer is; it depends on the source or program material. If the source is digital, say a CD that is uniformly compressed and uniformly loud where the maximum output voltage from the CD player is 2 peak volts you can follow that maximum peak voltage through the chain of components and you can know with certainty that there are no clipped transients coming out the other end. I saw this illustrated this way. I put the sound track CD from the first Pirates of the Caribbean movie on the CD player and attached the oscilloscope probes to the output of the headphone amplifier. Wow it was a surprise to me for a given setting of the volume knob there was an absolute maximum voltage output. It was like a stone ceiling.
The short version is that the amplifier headphone combination needs to produce the program peaks without clipping.
DT
All just for fun!
I like your question or thought about the required design power or db output of a headphone amplifier should have. The same question could be asked of a power amplifier for loudspeakers. I am afraid that if we look closely there will be some disappointed audiophiles. I believe clipping is the rule. Put the question another way, how much headroom is needed above average listening spl’s to avoid clipping of the peaks. Somewhere here or in the other current headphone thread someone said that headphone amplifiers have power to burn.
The answer is; it depends on the source or program material. If the source is digital, say a CD that is uniformly compressed and uniformly loud where the maximum output voltage from the CD player is 2 peak volts you can follow that maximum peak voltage through the chain of components and you can know with certainty that there are no clipped transients coming out the other end. I saw this illustrated this way. I put the sound track CD from the first Pirates of the Caribbean movie on the CD player and attached the oscilloscope probes to the output of the headphone amplifier. Wow it was a surprise to me for a given setting of the volume knob there was an absolute maximum voltage output. It was like a stone ceiling.
The short version is that the amplifier headphone combination needs to produce the program peaks without clipping.
DT
All just for fun!
Headroom is indeed the big issue and you are quite right that so much of today's music has the life compressed out of it such that the level is almost constant. But for recording, for example, where there may well be a lot more dynamic range to the signal, there will often be peaks well in excess of the average level and the headphones amp needs to be able to handle them cleanly. I think the ability of my current design to handle over 110dB SPL with any headphones gives it a more than adequate margin.
Cheers
Ian
Cheers
Ian
Last edited:
I would like to thank Ian for all of his help in finishing the design of his Improved Headphone Amplifier, getting a small pcb run made, and then answering all of my stupid questions that I asked when assembling his amplifier. A birthday present for my brother turned into an inspiring learning experience and an excellent final product.
It took me a couple months to assemble all of the parts that I needed to build his amplifier as it was the first time that I built an amplifier from scratch. Fortunately, everything came along well and I got enough of the design assembled to test his amplifier for about an hour today.
I would like to mention that a couple modifications were made to reduce costs and adapt Ian's headphone amp to 120v power here in the US. Following his recommendations, I elected to go with two Edcor XSE10-50-8k output transformers since these are each only $18. Although these only have one secondary winding, their impedance is well-suited to the 38 ohm headphones my brother is using. $18 sure beats the cost of two Sowter 8665 output transformers which cost more than $200 for both.
For a mains transformer, I purchased an Antek AN-05T240 on Ian's recommendation. Due to these savings, the amplifier cost less than $300 (including the case).
Anyway, while I don't have an Oscilloscope to empirically verify the quality of the audio on this amplifier, I can say that the sound I am hearing on a cheap pair of test headphones is excellent! Even at very high volumes, my music sounds very clean but warm. Ian's implementation of the NFB circuit must be great because distortion is much lower than it is on most less expensive tube amplifiers that I have heard, however the characteristic "tube" sound is still there.
I would like to independently give this design an A+ for its simple but incredibly effective layout and usage of parts. I will take and post some pictures once I get the case assembled.
It took me a couple months to assemble all of the parts that I needed to build his amplifier as it was the first time that I built an amplifier from scratch. Fortunately, everything came along well and I got enough of the design assembled to test his amplifier for about an hour today.
I would like to mention that a couple modifications were made to reduce costs and adapt Ian's headphone amp to 120v power here in the US. Following his recommendations, I elected to go with two Edcor XSE10-50-8k output transformers since these are each only $18. Although these only have one secondary winding, their impedance is well-suited to the 38 ohm headphones my brother is using. $18 sure beats the cost of two Sowter 8665 output transformers which cost more than $200 for both.
For a mains transformer, I purchased an Antek AN-05T240 on Ian's recommendation. Due to these savings, the amplifier cost less than $300 (including the case).
Anyway, while I don't have an Oscilloscope to empirically verify the quality of the audio on this amplifier, I can say that the sound I am hearing on a cheap pair of test headphones is excellent! Even at very high volumes, my music sounds very clean but warm. Ian's implementation of the NFB circuit must be great because distortion is much lower than it is on most less expensive tube amplifiers that I have heard, however the characteristic "tube" sound is still there.
I would like to independently give this design an A+ for its simple but incredibly effective layout and usage of parts. I will take and post some pictures once I get the case assembled.
Last edited:
- Status
- Not open for further replies.