Hi,

Unfamiliar with electronics, so bear with me please:

I'm looking at picking up a presonus HP4, being sick of dealing with audiophile headphone amplifiers and finding they all pretty much sound the same when blind-tested. I'm selling off my inventory and looking at a decent, relatively inexpensive method to drive my headphones. I've pretty much landed on the Presonus HP4 - it's expensive, but I'm buying it mainly for build quality (steel and aluminum will probably hold up much better than plastic).

However, I have a question: will it provide enough power to drive my beyerdynamic DT880? The power supply indicates 16V, which should be more than enough. However, it's a 4 channel headphone amplifier, which leads me to believe that (even though I pretty much will never use all 4 channels, and 99.999% of the time use only one) each channel only gets 4 volts (which, if my admittedly meager knowledge of electronics serves me right, would be insufficient). Is this erroneous? Also, the output impedance is 51 ohms, and the impedance of the DT880's is 250 ohms. Is this enough of a mismatch to cause a problem?

Also, I've heard that the output impedance of a given amplifier should meet the headphone's impedance as closely as possible, however I'm not sure to what extent this matters insofar as how big of a gap is significant. If it turns out that the HP4 is indeed insufficient, can someone recommend a decent headphone amplifier? I'm basically only looking for something that can provide ample voltage/amperage (I assume given the impedance of the DT880's that current doesn't matter all too much, but I could be wrong), with decent build quality. I've seen mixers recommended, but I'm scared that given my inadequate knowledge of how to work them that I'll end up tampering with the sound in a suboptimal manner.

Thanks.

Two things. Output impedance needn't to match your headphones, but the lower the better (it's the inverse for input impedance). Input voltage doesn't tell you anything about the amps quality or ability to drive high impedance headphones as yours. By using large capacitors one could build the most audiophile headphone amp around a single 1,2 volt battery.

Read some trustable reviews instead. Trustable means that they judge by both subjective and objective data and also publish the latter.

To compare models use the following guide:

Input impedance: higher
Output impedance: lower
Rarely published (but important): Slew rate: higher
Frequency response from at least 20-20000 Hz: flat
SNR/Dynamic range: higher
THD: lower

If you buy a model not just by recommendation of its sound but also the above data, it will be a safe bet. The DT880's 250 ohms impedance is much, but not that much, so you shouldn't have to worry about an amps ability to drive them. All higher class amps (regarding above numbers) I know should be able to handle it easily.

Each channel will benefit from the 16 volts going in, but each will only be able to draw one fourth of the available current.

Looks cheap to me. ;-) My best headphone amp is a Rane whose street price is at least twice that. Not only that, the Rane is rack-mountable and has at least twice the maximum output voltage. It has so much output voltage that it's totally fried at least one pretty good pair of headphones (Grado) when I err, made a little mistake some years back. :-(

Seriously, the Presonus looks pretty good to me, But, I've got to point out that Behringer has a similar product for less than half the price. They might even be made in the same factory in China. ;-)



Yes, your presumption is very erroneous. A product that runs off of 16 volts AC from a wall wart can potentially have internal +/- 12-15 volt power supplies. In fact my Rane runs off of a wall wart that if memory serves, puts out 18 volts. Not much more.

At any rate, a product with +/- 12 volt internal DC power can put out up to 6-7 volts rms without clipping, from each output. That's because the power is distributed to the 4 headphone amp sections in parallel, not series as you seem to have presumed.



There's no mismatch. It is good practice to drive a higher impedance load with a lower impedance source. In fact this is how amplifiers and speakers work in general. Your typical stereo amplifier has an output impedance of maybe 0.1 ohm or less. The rated load impedance is different from the output impedance.



Somebody must be talking trash. One reason for driving higher impedance loads from lower impedance sources is to minimise the degree to which the load whose impedance often varies with frequency, changes the frequency response of the source. Some headphones have highly variable load impedances, just like loudspeakers. You can vary the timbre of a loudspeaker or headpones by raising the source impedance.

As far as your DT 880s goes, they are typical of high quality european headphones, like the Sennheiser HD580s that I loved dearly until this fall when some theives broke into my house and ripped me off for about $5k of computer, audio, bicycles, and jewelry. The 580s had a high impedance too, but they were also very sensitive. The high impedance meant that I could run them directly off of the line outputs of some audio gear, without a headphone amp of any kind.



This time you're right. High impedance headphones mean that you don't need to worry much about supplying them with a lot of current. However, look to the future, down the road you might buy some new headphones that are more typical of Japanese headphones, and are only 16 ohms.



The worst thing that could happen to you with a mixer is that you might get scared away by all the knobs, and maybe have to fiddle for a while to even get any output at all! ;-)

My mixer of choice has 56 channels and over 2,000 adjustments, many of which can render it perfectly silent. ;-)

Cheap mixers generally have very good headphone jacks because lots of musicans use them for personal monitor mixers. They are professional tools, and may well lead you into places you never expected to go, and have so much fun! ;-)

Thanks for the replies.

Arnold: is 6V output really enough to drive high impedance headphones? I remember I saw some scope tests done by wmax (http://www.hydrogenaudio.org/forums/index.php?showtopic=39106&st=25&p=354399&#entry354399) that seemed to indicate that 9V was insufficient. Granted I don't listen to such dynamic music, but I do have a fair share of dynamic music (mainly classical) that has some pretty large peaks. Is the 6V output going to be a problem? I figure I'm being overly paranoid about this, but I don't want to make a purchase I will regret.

Thanks.

It certainly is for the high impedance headphones I owned, namely the Sennheiser HD 580s.

I was delighted with a headphone amp that ran off of 5 VDC, and was thus limited to under 2 volts.




I see that the DT 880s may be less efficient than my HD 580s. You might want to look at the compartive specs, and reach your own conclusion. Please ask any questions that may arise in that investigation.

Also, I am far and away *not* the conoisseur of loud music that some are. That's one reason why I can still effectively mix live sound at age 62. ;-)

Some of the musos I work with *need* what I consider to be ear-splitting volumes. Consider clipping out one channel of a 240 wpc (25 dB above 1 watt) power amp into a stage monitor that is about 4 feet away from the ears, and has about 98 dB/watt sensitivity (123 dB calculated and within the dynamic range of the speaker). We had a drummer (electronic drums) in his early 40s that neeed those kinds of levels.

Well, if you want to go further, then you might look at one of the *pro* headphone amps like my Rane that can probably put out about 10 volts RMS. Note the giant price jump, too.

I read the article you reference as suggesting that a 15 volt supply is about enough, but 9 volts isn't.

I myself use a Samson S-amp. It's a 4-channel stereo headphone amp with really rugged aluminum casing with rubber feet. It has no audible hiss, at a normal listening level. It comes with some spec sheet about different headphones. It uses internally max 16 volt system (checked inside, all caps max 16 V). It drives my HD595 (50 Ohms) at reasonable listening level about 1/8 to 1/4 sector of the volume knob. I'm quite sure that if you're not too nitpicky person about this stuff, it could drive more demanding load too. I tried it with my cheap philips headphones with inline attenuator: the amp gave clean, undistorted sound near the max volume position too.

If you haven't bought your amp of choice yet, take a look. I'm about to experiment with my own by changing some components inside.

More info here: http://www.samsontech.com/products/product...cfm?prodID=1698

EDIT: Oh yeah, the amp is priced from 60 € to 90 € depending on the store. Little expensive when compared to other similar products but I'm happy with it.

Just a minor point of pedantry...

Opening up a product and simply reading the voltage rating of the smoothing capacitors in the power supply section is not a reliable way of determining what voltage the unit is running at internally. 16Volts is the lowest voltage rating commonly available in many ranges of electrolytic capacitors so is highly likely to have been used whether necessary or not. The internal power rail voltage could be anything from a theoretical zero all the way up to and including 16Volts.

Cheers, Slipstreem.

Basically, it's not the power section, power section is rated at 35 V, but the rest of the circuit is 16 V. This is why I said 16 V MAX The opamps can take a few volts more and the power supply even more as said.

EDIT: Typos

EDIT2: Checked with a voltage meter, the internal voltage is indeed 15 V

Good. It's the only way to know for sure. I only bought it up as the first post starts with the words, "Unfamiliar with electronics...".

Cheers, Slipstreem.

Thanks for all of the info and advice.

I think I'll go with the Presonus. I don't really listen to headphones at a loud level (granted, this is compared to people around me who just blast their headphones) but on my ipod with stock earbuds (not that I use them, but for comparison), the volume never goes above 1/8th of the bar. That being said, I looked over the scope tests again and it appears that vRMS is only 3.90 with the 16.7V power supply (5.51 peak), so I think the presonus should be adequate in that regard. I was considering the s-amp, but decided against it because I couldn't find it in a good online store that also had the other stuff I wanted at a good price. I could've gotten it at sweetwater but I would've ended up paying more because the other things I needed were more expensive. Also, the presonus has a monitor out - a small feature, but one that would come in handy seeing as I use speakers on the same setup. Thanks for all of your help and dealing with my OCD-ish tendencies; I tend to over-research things so I don't end up disappointed down the road.

Also Arnold, offhand, how did you figure 5-6vRMS from a 16V power supply? I just ask because I'm curious if that stems from practical experience, or if there's some mathematical way to determine it.

Thanks.

Ok, I got it just because had some great deal from a shop here in my country.

EDIT: For the Vrms values, if not provided by the manufacturer, always measure, there's no rule (that I know of) that applies in every single occasion

take 16 volts, divide in half because we want full waves, take off a volt for losses in the output stage, and divide by 1.414 convert to RMS.

Letsee:

16/2 = 8

8 - 1 = 7

7/1.414 = just about 5 volts RMS.

According to my amp's opamp specs, it can output Vmax - 1.2 V

The power rail voltage used is 15 V
15 V- 1.2 V = 13.8 V
13.8 V / 2 = 6.9 V
divide by square root of two

we have 4,8 Vrms

the older opamp had max output of Vmax -1.5 V and therefore 4,77 Vrms

The difference isn't big and as Arnold said, take off the volt if you want to have an estimate.

Thanks heaven! I've done volume matched tests like this myself and come to the same conclusion. After all the tests I did, I ended up plugging my headphones (which everybody on head-fi says HAS to have a expensive amplifier) straight into my motherboards onboard sound and it sounds great.

Just a point of clarification. The HP4 says the Balanced TRS output is 51 ohms. I don't think that means the headphone jack impedance is 51 ohms. I think that is for the jacks at the back, although that doesn't make much sense to me. If the HP4 is using a typical opamp circuit, the output impedance is probably much closer to 0 than 51 ohms.

If the output impedance really is 51 ohms, it probably won't be a major issue for your DT880s. However, if you have a headphone with a much lower impedance that can affect the frequency response. There is a short article about it Meier-Audio

Honestly, why don't pro audio manufacturers spec their stuff properly? What is the output impedance of the headphone jack? I'd expect such shoddy specifications from audiophile stuff.

I seriously question the credibility of the meier source. If your headphones sound better with higher impedance outputs their design is flawed. Any output impedance adds serious non-linearities to your frequency response and 120 ohm is not an international standard for headphone outputs as claimed. Any decent amp will virtually have 0 ohm output impedance, that's the target good headphones are designed for. Meier's proposition to put a resistor in parallel into the signal path for the reduction of output impedance would be the first time I have heard this. Can an output stage, that is too weak to drive low impedance headphones be tamed by lowering the total impedance of its load?

Actually, Meier Audio's website is probably referring to the headphone jacks of Integrated amps. Many of them (I actually asked the customer service of many manufacturers) use resistors off the main amp circuit to power the headphone jack. You end up with an output impedance far more than 0 ohms. Some people have also made headphone adapters using the speaker outs of their integrated amp.

That being said, my own experience is this method does not make a huge difference for high impedance headphones like the DT880. It's a different issue when you have something that is 25 ohms like the Denon DX000 series.

As for adding resistors to "adjust" the output impedance of something like an Integrated amp headphone out with a high (i.e. 50 ohms) impedance, I don't support such a method. I posted the link to the page just to illustrate the effects of a high impedance output jack.

You bring up an interesting question, so I decided to follow up on it. I e-mailed Presonus last night, they still haven't responded so I called them up. The tech support guy said that the 51 ohm output only applied to the TRS jacks (presumably on the back of the unit, the monitor out), but couldn't get an engineer on the line to get the output impedance of the actual headphone out. He said that as far as he remembered it was 60 ohms, and that sounds odd. I'm hoping that my e-mail will be responded to by an engineer or someone who knows for sure, because that sounds kind of puzzling.

I recall a thread on Head-Fi where people were discussing the headphone jack of the Central Station and HP4. People claimed there was a audible difference (they always do on that site). Somebody contacted Presonus and an engineer said the Central Station and HP4 use the same headphone amp circuit. Some people still insisted there was a difference

The Headphone Jack specs for the Central Station.

Type ¼" TRS Active Stereo
Maximum Output 150mW/channel @ 60 Ohm load
THD+N 0.015% (150mW/channel @ 60 Ohm load)
Frequency Response 10Hz-50KHz, +1dB

Now, the spec above doesn't automatically mean the jack impedance is 60 ohms. It simply means it can put 150mW into a 60 ohm load.

If you ever do buy it, please post your opinions here. If you can run a loaded RMAA test, even better.

I do own it now. It sounds... transparent. Exactly the same as all of the "audiophile" amplifiers once tested level-matched and blind. I don't know how to do a loaded RMAA test - do I run the outputs to the headphone amp, and then from the amp straight back to the soundcard input? That wouldn't be loaded probably, and even then I don't think I have the necessary cables. It all depends, I guess, on whether or not I can obtain the cables from somebody (my roommate does campus radio, he may be able to hook me up with another 1/4" TRS cable, but even then I'd need a splitter to make it 1/4" mono). It all depends on the equipment I need versus the equipment at my disposal, I suppose.

To do a loaded RMAA test you need a stereo splitter with one male end and 2 female ends. The input end (male) of the splitter goes to the headphone jack of the HP4. You plug headphones into one output (female) of the splitter and run a wire from the other output of the splitter to your sound card. Be careful when you are turning up the volume is such a situation. I imagine you can easily damage a headphones by turning it up too loud. Also be prepared to hear your headphones squeal with test tones.

From what I've seen, RMAA tests loaded with high impedance headphones rarely shows anything interesting (i.e. other than a flat frequency response) unless there is something drastically different about the amp like a intentional bass boost or a defective design. It's only with low impedance (~32 ohms or lower) that you start to see bass roll off and things like that.

It's good to know the HP4 sounds good. I tried a pro-audio headphone amp/splitter once (borrowed from somebody who had drawers full of electronic junk) and there was a lot of noise. From your review, the HP4 should be a good recommendation for people who want freedom from the audiophile headphone amp world.