• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Why is headphone amp input sensitivity so high?

nothingman

Active Member
Joined
Sep 15, 2020
Messages
233
Likes
261
Why is the input sensitivity so high on headphone amplifiers? What am I missing? On my loudspeaker power amp, the input sensitivity is 1.43V for the SE inputs. That makes total sense to me, since any home audio preamp is going to be in the meaty section of the power band and not clipping at that level.

I went and bought the HD6XX and the Sabaj A10H to get started. The problem is the input sensitivity on the A10H is apparently quite high. I’m running into problems getting enough voltage out of my miniDSP SHD, especially when running EQ, even with the A10H in high gain mode. I went looking at the Topping alternatives like the L30, and it’s the same story: 3V input sensitivity even on high gain mode.

Who in home audio has all these DACs and preamps outputting a solid 3V, especially when so many are using EQ on headphones? The counterpart to the L30 is the E30 which has 2V max output. Why is the headphone amp world totally different in this regard? Is this all part of measurement-chasing by manufacturers, or is there a good engineering reason why power amps have input sensitivity accessible to any DAC/preamp and headphone amps do not?

Lastly, paging @JohnYang1997 here, would I be better served by a DAC/amp combo, say the Topping DX3 Pro+? I can output coax from my miniDSP SHD and then run volume via remote on the Topping, but I don’t fully understand if a DAC-amp combo will intrinsically fix the underlying issue of getting enough power to the headphones (e.g. is the built-in DAC output a better match for the headphone amp sensitivity?).

Two things are most important to me beyond raw amp performance, 1) good channel matching, and 2) remote functionality since I use all of my 10 foot cable to sit on my couch while listening. Hence I thought I scored a home run with the Sabaj A10H, but it seems I missed the mark purely based on headphone amp input sensitivity. It doesn’t seem like it should be this hard. Thanks, all.
 
Last edited:

TurtlePaul

Active Member
Forum Donor
Joined
May 19, 2020
Messages
272
Likes
389
So you want the inputs to clip sooner? Amir tests the amps with the normal 2V nominal input. This amp performs fine with this level, which is the consumer standard reference level. Having an amp that doesn't clip the input until 3V or 4V just gives you 3-6 dB of headroom above the standard, which isn't a bad thing. Are you complaining that the amp doesn't have enough gain (it is too quiet with the pre and Sabaj at max)?
 
OP
N

nothingman

Active Member
Joined
Sep 15, 2020
Messages
233
Likes
261
I’m saying that this amp and others like it don’t follow what I consider to be a fundamental, commonsense rule from the loudspeaker world: that in any matching preamp and power amp combo, the preamp’s output surpasses the power amp’s input sensitivity spec. Name me a single outlier to this, please, because I’ve never seen it.

When a preamp cannot reach the input sensitivity of an amp, then the amp is leaving power on the table. The amp’s max power rating becomes rather meaningless in a situation where the preamp isn’t even capable of driving it to max power. Can you imagine if… I dunno… Cambridge Audio said, “our latest and greatest amp is capable of 300W/8ohm except if you use our matching preamp because we didn’t make that preamp strong enough to drive the amp to that power (i.e. reach the input sensitivity). Sorry, but you only get 200W in that case. But please know the amp is capable of more on paper and please publish the power spec everywhere like we made a preamp that can take advantage of that power, but we didn’t. It’s up to you to find one that does.”

If my headphone amp has 3V sensitivity, and my preamp is only capable of 2V, then I’ve already lost power that the amp is capable of providing, and then if I’m doing the common thing of running EQ on my headphones, then I’m so far below the amp input sensitivity that I’m just wasting my time.

How is any of that better than an amp with 1.5ishV input sensitivity, where any standard preamp can drive the amp to full power output without the preamp itself constantly at its absolute limit?

Yes, this is closely intertwined with gain, but I‘m not technical enough to write a white paper on it. It‘s the same effect — if at most I’m going to get 9db gain and my preamp can’t drive the amp to full power, I’m left without enough power. In the case of the DX3 Pro+, I don’t have the input sensitivity spec because it’s not published, but I do have +19db gain which means what my preamp can provide is enough to get proper output from the amp.
 

TurtlePaul

Active Member
Forum Donor
Joined
May 19, 2020
Messages
272
Likes
389
As long as you have enough gain to get to your desired output volume, that is all that matters. A lot of headamps are capable of 1 watt+ output ( and why not, that isn’t much power) but if you drive a 300 ohm headphone to 1 watt you would instantly go deaf. Also, power isnt a useful metric because these devices are voltage multipliers and the power output will differ by an order of magnitude between 30 ohm IEMs and 300 ohm over-ears. The gain on headamps is set so the volume adjustments will work in a reasonable range. It assumes that it will be driven direct from source or from a preamp which is capable of at least unity gain. The input capabilities of the Sabaj ensure that a +4 dB pro source or a preamp with positive gain wont clip the input. Very few headphones need more than 5V output pole-to-pole to get insanely loud.
 

restorer-john

Master Contributor
Joined
Mar 1, 2018
Messages
8,735
Likes
24,259
Location
Gold Coast, Queensland, Australia
There's a whole bunch of poor misguided folks who think they need to 'stage' their gain so their preamp or D/A converter output matches the rated input sensitivity for the power amplifier stage.

This is utter nonsense and has never been the case in the HiFi world since it started.

Rated SE (single ended) sensitivities were 150mV on line inputs to deliver typical preamp outputs of 1,1.5 or 2.0V back in the day. Power amplifier inputs were the same, say 1.5V for full rated power. Throw a CD player (2.0V) into a 150mV line input and you have a whole lot more than 1.5V on the output of your preamplifier.

Balanced XLRs were never double the voltage until recently, when the pro levels started infiltrating and rearing their ugly heads in the HiFi world.

Decent preamplifiers will output ~8-10V on their line outputs before clipping, more than sufficient for any power amp. Recordings vary massively in level and also need more preamp gain at times. But gain is not evil- you need it someplace and the best place to have it is in a preamplifier where the signal levels are low, the currents are low and the noise is minimal.
 
Joined
Jun 24, 2020
Messages
16
Likes
18
Why is the input sensitivity so high on headphone amplifiers? What am I missing? On my loudspeaker power amp, the input sensitivity is 1.43V for the SE inputs. That makes total sense to me, since any home audio preamp is going to be in the meaty section of the power band and not clipping at that level.

I went and bought the HD6XX and the Sabaj A10H to get started. The problem is the input sensitivity on the A10H is apparently quite high. I’m running into problems getting enough voltage out of my miniDSP SHD, especially when running EQ, even with the A10H in high gain mode. I went looking at the Topping alternatives like the L30, and it’s the same story: 3V input sensitivity even on high gain mode.

Who in home audio has all these DACs and preamps outputting a solid 3V, especially when so many are using EQ on headphones? The counterpart to the L30 is the E30 which has 2V max output. Why is the headphone amp world totally different in this regard? Is this all part of measurement-chasing by manufacturers, or is there a good engineering reason why power amps have input sensitivity accessible to any DAC/preamp and headphone amps do not?

Lastly, paging @JohnYang1997 here, would I be better served by a DAC/amp combo, say the Topping DX3 Pro+? I can output coax from my miniDSP SHD and then run volume via remote on the Topping, but I don’t fully understand if a DAC-amp combo will intrinsically fix the underlying issue of getting enough power to the headphones (e.g. is the built-in DAC output a better match for the headphone amp sensitivity?).

Two things are most important to me beyond raw amp performance, 1) good channel matching, and 2) remote functionality since I use all of my 10 foot cable to sit on my couch while listening. Hence I thought I scored a home run with the Sabaj A10H, but it seems I missed the mark purely based on headphone amp input sensitivity. It doesn’t seem like it should be this hard. Thanks, all.
First off your mini dsp is not going to work for your setup. The L30 needs a d50 or d90 along with it. you’ve been misguided. second the mini dsp is not a pre amp. the input sensitivity is not what you think it is. You need a d90se or something similar if you want pre amp with XLR 4v out or RCA with 2volt out. simple as that you need a volume wheel on the pre amplifier. Don’t overthink these things. the mini dsp does not have a volume output. it’s basically something you would put in between two devices to add digital signal processing to an existing setup you cant just plug it into a power amp and think you can control it from the power amps volume knob. Also the headphones you picked are not something I would recommend and they are pretty much low end headphones in my book. get a topping combo like a90/d90 or even A lower model combo like the 30 or 50 series. or just add an old stereo or an reciever you have lying around with preamp outputs. another word of advice is to go with everything XLR you can. forget RCA interconnections on the amps/pre section. and then get yourself a balanced 4.4mm or 2.5mm Or 4pin XLR connector cable for the headphone itself.….any of the 3 balanced headphone cables will work . An all balanced system sounds like what you think you want. You just didn’t know what you were really talking about.
 
Last edited:
OP
N

nothingman

Active Member
Joined
Sep 15, 2020
Messages
233
Likes
261
What you’ve said is 99% straight-up gibberish, but I encourage you to look at the miniDSP SHD. It is a preamp.

As for the rest I can’t even start. The proof that you’re wrong is that the HD6XX driven by the SHD (serving as source and PEQ) into the DX3 Pro+ with a single coax cable is marvelous. I don’t need balanced anything or a $1400 A90+D90 stack in the least.

To @restorer-john and @TurtlePaul i appreciate the help distinguishing between input sensitivity and gain. Not sure I still 100% get it, but at the end of the day the +19db gain of the DX3 Pro+ works when the other lower gain headphone amps didn’t, and so that’s what I’ll credit for the good result. Thanks.
 
OP
N

nothingman

Active Member
Joined
Sep 15, 2020
Messages
233
Likes
261
You come across level headed, well informed, and exactly the sort of person I should be listening to. Do you by chance have a newsletter I can subscribe to, or a phone number I can call to receive these beatdowns verbally?
 
OP
N

nothingman

Active Member
Joined
Sep 15, 2020
Messages
233
Likes
261
I sort of wish @sanctum’s (he’s not coming up as a member anymore… suspended? banned?) complete meltdown could be bleeped out and presented for posterity. If only so my last reply made sense.

To clarify just a minute: ignoring EQ and associated headroom considerations, there’s zero reason why anyone couldn’t use the analog pre-outs from an SHD into a headphone amp. My thing is that with PEQ headroom and with a headphone (impedance, sensitivity) like the HD6XX, I wasn’t getting enough power to the headphones with just 9db of gain. It was close but not quite where I want it.

So I looked for something with more gain, increased my budget a little (got the Pro+ for $180) and now everything’s peachy. Signal is clean and strong, gain staging is low maintenance (set and forget SHD to -9db and the Pro+ volume takes it from there) and I’ve got everything I wanted out of a budget setup to sit alongside my considerably more expensive loudspeaker system which can’t always be played loud due to sleeping children and/or neighbors.
 

Sal1950

Grand Contributor
The Chicago Crusher
Forum Donor
Joined
Mar 1, 2016
Messages
10,073
Likes
10,385
Location
Central Fl

BDWoody

Chief Cat Herder
Moderator
Forum Donor
Joined
Jan 9, 2019
Messages
5,184
Likes
14,622
Location
Mid-Atlantic, USA. (Maryland)
you come Across like a *naughty word* idiot. i don’t even know why I waste my time with morons like you on this stupid *more pottymouth* website. just keep buying shit till something works for your dumb ass

Ok, I deleted several of your way out of line posts.

Not sure why you feel that's appropriate, but it's not how we communicate here.

You have a couple weeks off.
 

xnor

Member
Joined
Jan 12, 2022
Messages
61
Likes
34
2 Vrms standard?
Consumer/semi-pro audio standard nominal level is -10 dBV which is 0.316 Vrms, and professional standard nominal level in the US is +4 dBu = 1.228 Vrms.
Note that these are nominal levels, but "nominal" means different things in consumer vs. pro.

In consumer/semi-pro equipment the nominal level commonly corresponds to -7 to -9 dBFS, while in pro equipment it is -18 dBFS because in pro applications extra headroom is needed. This means that in the US, pro equipment should start clipping at +22 dBu or almost 10 Vrms.

In consumer/semi-pro electronics the clipping point would be closer to 0.7 to 0.9 Vrms. RCA connectors once were an indicator for consumer -10 dBV...

So where do 2 Vrms come into play? Dedicated D/A converters typically output 2 Vrms at 0 dBFS on their unbalanced outputs.
Why? As far as I know this is coming from SCART. Yeah, that old A/V standard with the ugly 21-pin connectors... When SCART was standardized they raised audio levels to increase dynamic range. They ended up at 2 Vrms at 0 dBFS.

So everything is good, right? No. Audio(phile) equipment manufacturers also have products that output 2.1... 2.8 or higher Vrms on unbalanced outputs. Why? Obviously to make more money because louder sounds better, and clueless consumers don't know how to level match properly, right? :p

So that leaves one question for manufacturers of both D/A converters and headphone/power amplifiers: why does your converter output 2.x Vrms but your amp's input sensitivity even at highest gain is 3.x or even 4.x Vrms?
Some possibilities:
  • they simply don't care, they designed the (high) gain not to maximize output levels but to match some common headphones with their converters
  • they are ignorant about consumer audio standard levels
  • higher gain would result in clipping with "hot" input levels which the manufacturer wants to prevent at all cost
  • higher gain would result in lower performance

Also, at the op: I think you confused high and low.
High sensitivity means that small input voltages result in full output. Low sensitivity means you need high input voltages for the same full output.
 
OP
N

nothingman

Active Member
Joined
Sep 15, 2020
Messages
233
Likes
261
This is all good insight, @xnor - thanks. Just on the high/low sensitivity thing, I meant it as “why is the input sensitivity spec such a high voltage?” but in retrospect my language wasn’t precise.

Yep, personally I still don’t quite get why it wouldn’t be common practice to have a headphone amp input sensitivity spec in the normal loudspeaker power amp range, but at the end of the day I’m glad I found a product with an amount of gain that allows me to use my DSP and headphones as I wish.
 

xnor

Member
Joined
Jan 12, 2022
Messages
61
Likes
34
One design where clipping is a real issue is when there's a combined input+gain stage and the volume control comes after that.
Because in such a design high gain combined with a hot input will cause clipping regardless of volume control position.

Btw, 2V/3V is "only" a 2 dB difference but if your source outputs 1V (which would be roughly what's to be expected of consumer -10 dBV equipment) then this grows to -9.5 dB (perceptually half as loud) which is significant.

My observation of headphone amps is that power usually is not an issue. Gain is.
Nobody is listening to full-scale sine waves. Real music has much lower (average) level. Ever listened to uncompressed classical recordings? Added EQ on top of that? :D
I can see how people think they require a really powerful amp when in reality they just need a small fraction of the power that those amps could deliver... their setup just lacks another 10 dB or 20 dB of gain.
 
Top Bottom