IPB

Welcome Guest ( Log In | Register )

4 Pages V  < 1 2 3 4 >  
Reply to this topicStart new topic
Is a headphone amp needed?
fewtch
post Nov 29 2005, 10:12
Post #26





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



QUOTE (mirrorsawlljk @ Nov 24 2005, 10:01 PM)
So, is it worth dropping $80 on a dedicated headphone amp to sit between my cans and my soundcard?
*

Why not drop $69 on a go-vibe (www.go-vibe.com) and see if you get "better" sound directly from the card to a standalone amp. Not very expensive experiment. I've owned a go-vibe a long time, and it drives HD580s, DT-880 and whatever else you throw at it to ear-splitting volumes with low output impedance, plenty of dynamics and low distortion. It's a hand-built amp with good quality parts, so the price is justified IMO.

I'd say try the experiment, if it doesn't work you can sell it for close to what you paid on eBay or head-fi. Subjective matters call for subjective listening.

P.S. I have no affiliation with the aforementioned seller, except I own an earlier version of this amp and enjoy it a lot. The only reason I recommended it is that the price is reasonable and I have personal experience with it.

This post has been edited by fewtch: Nov 29 2005, 10:32


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post
WmAx
post Nov 29 2005, 18:31
Post #27





Group: Members
Posts: 541
Joined: 22-May 04
Member No.: 14243



QUOTE (fewtch @ Nov 29 2005, 05:12 AM)
Why not drop $69 on a go-vibe (www.go-vibe.com) and see if you get "better" sound directly from the card to a standalone amp.  Not very expensive experiment.  I've owned a go-vibe a long time, and it drives HD580s, DT-880 and whatever else you throw at it to ear-splitting volumes with low output impedance, plenty of dynamics and low distortion.  It's a hand-built amp with good quality parts, so the price is justified IMO.


DT880 on a G0-VIBE? While I'm sure that you can achieve annoyling loud SPLs, a standard 400 ohm DT880(as opposed to one of the new high sensitive 32 ohm versions) is probably going to distort on dynamic recordings with only a 9VDC supply due to voltage clipping. 18VDC seems to a safe supply voltage when listening to my most dynamic music program examples on a standard 400ohm DT880. A 9VDC supply results in a static-like harsh distortion on high amplitude peaks in these dynamic recordings when coupled with a DT880. However, if one does not use very dynamic program material(uncompressed dynamically music, such as classical, opera, etc.), it is not likely to be noticed or become an issue. Also, glancing at the Go-Vibe description, it appears to be directly coupling a AD8066AR with the output load(no mention of a current buffer). I am not an electronics technician(so keep this is mind ), but looking over the technical document for the AD8066AR, appears to be a poor choice for directly coupling to low impedance(<100 ohm) loads. I would recommend a design that uses a more suitable operational amplifier for low impedance loads(unless only high impedance loads are used) or get an amp design that used a current buffer so that the op-amp loading current is irrelevant.

-Chris

This post has been edited by WmAx: Nov 29 2005, 18:40
Go to the top of the page
+Quote Post
Melomane
post Nov 30 2005, 15:50
Post #28





Group: Members
Posts: 643
Joined: 13-January 03
From: Marseille
Member No.: 4552



mirrorsawlljk, sorry for my poor english

my humble opinion:
source is the most important thing because amplifier cannot recover lost signal.

have you good soundcard or external DA converter?

if not, buy good soundcard with headphone integrated amplifier , for exemple revolution 5.1 ( i am very happy with!) or Audiotrak Prodigy 7.1 LT ( 80 -> 100 $ )
and never mind!

or buy very high quality EXTERNAL DA converter and very high quality headphone amplifier ( 1000 -> 500000000 $ ...)
and make me a gift for coaching!!!

if you have good soundcard, for exemple revolution 7.1 or 5.1, terratec dmx 6fire or santacruz the question is :
have it good headphone amplifier?

ansver for revolution 5.1 or santacruz ( tanks KikeG!) is yes!
ansver for 7.1 and dmx 6fire is no!
and you have need amplifier.

which amplifier?
because soundcards use OP amplifier and cannot recover lost signal ...
Do-It-Yourself (20 $) and never mind..., for exemple

Pocket Headphone Amplifier by Chu Moy

if you cannot DIY sell your soundcard
and buy revo 5.1

melomane


--------------------
Music is my first love.
Go to the top of the page
+Quote Post
fewtch
post Dec 6 2005, 07:08
Post #29





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



QUOTE (WmAx @ Nov 29 2005, 10:31 AM)
DT880 on a G0-VIBE? While I'm sure that you can achieve annoyling loud SPLs, a standard 400 ohm DT880(as opposed to one of the new high sensitive 32 ohm versions) is probably going to distort on dynamic recordings with only a 9VDC supply due to voltage clipping.

My DT-880 is driven just fine (earsplitting levels, no audible distortion) with my Go-Vibe V3. BTW they are all 250 ohms now, I don't know where you got the 400-ohm figure.

QUOTE
18VDC seems to a safe supply voltage when listening to my most dynamic music program examples on a standard 400ohm DT880. A 9VDC supply results in a static-like harsh distortion on high amplitude peaks in these dynamic recordings when coupled with a DT880.

You seem to be generalizing this voltage issue to include all amps. As long as the op-amp[/buffers] in the amp put out enough current (voltage is not the issue at all) you will not get distortion. Voltage swing relates only to loudness, not distortion.

QUOTE
Also, glancing at the Go-Vibe description, it appears to be directly coupling a AD8066AR with the output load(no mention of a current buffer). I am not an electronics technician(so keep this is mind ), but looking over the technical document for the AD8066AR, appears to be a poor choice for directly coupling to low impedance(<100 ohm) loads.

You would think so, but in practice it is not. Or if a person believes so, another opamp can be used; the go-vibe is set up to allow changing out opamps easily.

I don't know where people get the idea that an expensive amp is needed to drive cans like the DT880 or HD600... electrically speaking, it's just not so. In fact, they are easier to drive than low-impedance Grados, in every way except loudness. People who spend hundreds of dollars on headphone amps are either looking for "audiophile jewelry" or are ignorant of basic electronics (or both).

As a person with vocational school certificate in electronics (fwiw) I can tell you the Headsave Go-Vibe is an excellent amp, particularly for higher impedance cans. Low output impedance, no coupling caps, nice RMAA results, long battery life, no "audiophile" overpricing.

This post has been edited by fewtch: Dec 6 2005, 07:21


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post
WmAx
post Dec 6 2005, 10:01
Post #30





Group: Members
Posts: 541
Joined: 22-May 04
Member No.: 14243



QUOTE (fewtch @ Dec 6 2005, 02:08 AM)
BTW they are all 250 ohms now, I don't know where you got the 400-ohm figure.
Maybe I am thinking of my other headhones impedance... sorry about the confusion.

QUOTE
You seem to be generalizing this voltage issue to include all amps.  As long as the op-amp[/buffers] in the amp put out enough current (voltage is not the issue at all) you will not get distortion.  Voltage swing relates only to loudness, not distortion.


But voltge is a real issue. When the output signal nails the rail limits, the result is hard clipping. I have analyzed the waveform input signals to the amp, the gain and the output signal, it can not achieve satisfactory voltage output with my very dynamic material using a 9VDC supply. The result is very offensive distortion. However, merely increasing the supply voltage renders this a non issue(and is exactly as calculations would predict). Or I can simply not use my very dynamic selections and it's also okay. Though, I have another headphone that is substantially less efficient than the DT880, and 18VDC supply is not enough for them on very dynamic material...

QUOTE
You would think so, but in practice it is not.  Or if a person believes so, another opamp can be used; the go-vibe is set up to allow changing out opamps easily.


Well, maybe AD chose to be conservative in their specs? A good set of measurements under the respective loads is all that is needed to demonstrate suitability, of course. But the data sheet of the output device used did not look favorable for low impedance loads.

QUOTE
I don't know where people get the idea that an expensive amp is needed to drive cans like the DT880 or HD600... electrically speaking, it's just not so.


For the record, I am the last person to suggest an expensive amp(unless it has some special feature). But other portable amps besides the Go-Vibe in the same approximate price class exist that have current buffers on the outputs, that was my point.

-Chris

This post has been edited by WmAx: Dec 6 2005, 10:03
Go to the top of the page
+Quote Post
fewtch
post Dec 7 2005, 08:22
Post #31





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



QUOTE (WmAx @ Dec 6 2005, 02:01 AM)
But voltge is a real issue. When the output signal nails the rail limits, the result is hard clipping. I have analyzed the waveform input signals to the amp, the gain and the output signal, it can not achieve satisfactory voltage output with my very dynamic material using a 9VDC supply. The result is very offensive distortion. However, merely increasing the supply voltage renders this a non issue(and is exactly as calculations would predict). Or I can simply not use my very dynamic selections and it's also okay. Though, I have another headphone that is substantially less efficient than the DT880, and 18VDC supply is not enough for them on very dynamic material...

For what it's worth, I own a Go-Vibe v2 (9v, AD8620) and a Headsave Home-Vibe (24v, OPA627) and notice no difference whatever in dynamics between the two amps using my DT880. They are audibly identical in the dynamics category. Maybe I have tin ears or something, but I do listen to very dynamic classical music with both amps and have never noticed any clipping with the go-vibe at 9v.

Moreover, I don't understand that there's anything magical about 9v, or 18v or (insert voltage figure here). As long as you aren't exceeding Vmin with the particular opamp involved, you are fine. See this page for a list of voltage specs (ignore the subjective judgments on sound quality, unless inclined to believe such stuff):

http://www.tangentsoft.net/audio/opamps.html

As far as your experience goes, I can only suspect that you were either using a voltage-hungry opamp, or the battery was run down. There's no reason why 9 volts is too little for dynamic music, or a particular headphone, or whatever. Many headphone amps in portable CD players drive a wide variety of headphones just fine (albeit lacking a bit in volume swing) powered by 3VDC. In fact, some portable radios and MP3 players use a single 1.5v "AAA" battery and don't get returned in droves for nasty clipping with dynamic music.

QUOTE
But other portable amps besides the Go-Vibe in the same approximate price class exist that have current buffers on the outputs, that was my point.

Fair enough, what ~$69 amp comes to mind? I've had no problems at all myself going unbuffered, but some people may want the extra insurance of current buffers, particularly with low impedance cans. As for the go-vibe specifically, the reason I recommended it is that I own one, and am happy with it.

P.S. don't forget about the power supply capacitors, when envisioning power delivered to the opamp. The go-vibe has a couple of nice, beefy ones that seem to handle dynamic swings just fine. In fact, these capacitors probably matter more (in terms of dynamics) than the amount of voltage supplied by the battery.

This post has been edited by fewtch: Dec 7 2005, 09:48


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post
Shade[ST]
post Dec 7 2005, 08:30
Post #32





Group: Members
Posts: 1189
Joined: 19-May 05
From: Montreal, Canada
Member No.: 22144



I personally made myself a Cmoy amplifier, and I can confirm that having too low a voltage is an issue; However, in this case, I really ran down my batteries to an insane level, and using two batteries would only allow me to run them down more. The distortion will be a corellate of both voltage and current, so you're actually both right. However, more importantly, clipping is due to insufficient voltage.
Go to the top of the page
+Quote Post
WmAx
post Dec 7 2005, 19:31
Post #33





Group: Members
Posts: 541
Joined: 22-May 04
Member No.: 14243



QUOTE (fewtch @ Dec 7 2005, 03:22 AM)
For what it's worth, I own a Go-Vibe v2 (9v, AD8620) and a Headsave Home-Vibe (24v, OPA627) and notice no difference whatever in dynamics between the two amps using my DT880.  They are audibly identical in the dynamics category.  Maybe I have tin ears or something, but I do listen to very dynamic classical music with both amps and have never noticed any clipping with the go-vibe at 9v. 


Refer to [A].

QUOTE
Moreover, I don't understand that there's anything magical about 9v, or 18v or (insert voltage figure here).


[A] 9V simply is not enough for my combination of music and listening level when using the Beyer DT880. 18V is 2x9V batteries(so the jump is logical) and provides 6dB additional amplitude potential vs. 9V. 18V is not sufficient for my AKG K340.

QUOTE
As long as you aren't exceeding Vmin with the particular opamp involved, you are fine.  See this page for a list of voltage specs (ignore the subjective judgments on sound quality, unless inclined to believe such stuff):


I will put together a measured analysis for you within the next couple of weeks.

QUOTE
As far as your experience goes, I can only suspect that you were either using a voltage-hungry opamp, or the battery was run down.


The batteries new. The opamps are not demanding as used in the evaluatory device. The distortion is logical, as it occurs as it should accordig to my calculations based on the gain setting I use with particular music and the amplifier voltage output limitations.

QUOTE
There's no reason why 9 volts is too little for dynamic music, or a particular headphone, or whatever.  Many headphone amps in portable CD players drive a wide variety of headphones just fine (albeit lacking a bit in volume swing) powered by 3VDC.  In fact, some portable radios and MP3 players use a single 1.5v "AAA" battery and don't get returned in droves for nasty clipping with dynamic music.


It would be poor engineering for a totally enclosed system, such as a PDCP, to distort. It is easy to limit the gain factor of the internal amplifications so that 0dBFs of the input signal after max, gain never exceeds the voltage available at the rails. This is not a controlled situation, however, when external devices are combined(external portable amp, etc.).

QUOTE
Fair enough, what ~$69 amp comes to mind?  I've had no problems at all myself going unbuffered, but some people may want the extra insurance of current buffers, particularly with low impedance cans. 


Well, at a little bit higher price, the Shellbrook MiniMoy comes to mind.

-Chris

This post has been edited by WmAx: Dec 7 2005, 19:32
Go to the top of the page
+Quote Post
Axon
post Dec 7 2005, 20:23
Post #34





Group: Members (Donating)
Posts: 1984
Joined: 4-January 04
From: Austin, TX
Member No.: 10933



9V batteries are usually rated to only 25mA of continuous current, and they have a rather large source impedance IIRC. (I've pulled 9Vs out of their package and have not been able to get 20mA out of them in a short!) This suggests that the maximum continuous DC power output is somewhere in the vicinity of 225mW. Given the source impedance, the short-term voltage loss at such currents, the efficiency of the DT880s and the efficiency of opamp designs I'd say it's not outside the realm of possibility that a 9V battery may not have enough juice. Additionally, opamps do have measurably increased distortion getting closer to the rails, even though clipping per se may not occur. 2 9's are going to suffer from source impedance-induced voltage drops and rail-related distortion to a much lesser degree, although the current capacity is unchanged.

That said, if you replace the 9V with a large block of D cells you definitely will not have this problem. If you use a DC-DC converter you will probably not have this problem, as long as it's not current related. And if you actually drive a DT880 to even 5 volts you are probably going to fry the coils.

I think this discussion has nothing to do with voltage.

This post has been edited by Axon: Dec 7 2005, 20:23
Go to the top of the page
+Quote Post
MillerLowlife
post Dec 12 2005, 05:28
Post #35





Group: Members
Posts: 2
Joined: 18-November 05
From: Alpharetta, GA
Member No.: 25822



QUOTE
You must take notice of the fact that a dynamic headphone, no matter how expensive, needs very little power, so the requirements for the amp are quite low. Only a bit of care is needed to build a good amp



You wouldn't think so, but there can be a substantial difference in the current/power requirements for a given set of headphones. The driving source just sees the headphones as an impedance load. Although there is not a direct connection between the price of the headphones and the power needed to drive them, some of the more expensive headphones just happen to have high impedance values relative to some less costly models. For example, look at the impedance values of 3 different headphones.

Sony MDR-XD400= 24 ohms
Beyerdynamic DT880= 250 ohms
Sennheiser HD650= 300 ohms

P=Power
E=Voltage
R=Resistance (Impedance)
Power= (E squared / R)

As you can see, the DT880 cans will need over 10 times more current for a given input voltage (rms) and the Senns over 12.
Just like people, the money hungry headphones are also the power hungry ones , as well. :-)


--------------------
I can't live up to my name so I drink a lot... of tequila, vodka, etc.
Go to the top of the page
+Quote Post
KikeG
post Dec 12 2005, 09:17
Post #36


WinABX developer


Group: Developer
Posts: 1578
Joined: 1-October 01
Member No.: 137



QUOTE (MillerLowlife @ Dec 12 2005, 05:28 AM)
As you can see, the DT880 cans will need over 10 times more current for a given input voltage (rms) and the Senns over 12.
*

Well, Ohm's law says actually the opposite. P=(V^2)/R, or P=(I^2)*R. So, in fact, the greater the impedance (R ), the lower the current (I) required for same power (P). But the greater the voltage (V) needed. That's the main problem with high impedance headphones. They usually need a higher voltage to sound at same loudness. I say "usually" because it all depends also on the headphone efficiency. If a headphone was very efficient, it could sound even louder despite having high impedance. However, it seems that high impedance headphones are usually not among the most efficient ones.

As to DT880 clipping... if the measurable distortion al high levels goes away when doubling the voltage of the supply, it seems like the cause was clearly the insufficient voltage of the supply. Putting two batteries in series increases the voltage, but also increases the source impedance, so distortion should not be supply source-impedance related. But, doing the maths, I find strange that 9V would not be enough to drive a 250 ohm or even a 400 ohm headphone a high enough SPLs. Anyway, actual measurements are the final proof of performance.

This post has been edited by KikeG: Dec 12 2005, 09:24
Go to the top of the page
+Quote Post
antz
post Dec 12 2005, 14:39
Post #37





Group: Members
Posts: 225
Joined: 26-March 05
From: Wolverhampton UK
Member No.: 20961



QUOTE (KikeG @ Dec 12 2005, 09:17 AM)
QUOTE (MillerLowlife @ Dec 12 2005, 05:28 AM)
As you can see, the DT880 cans will need over 10 times more current for a given input voltage (rms) and the Senns over 12.
*

Well, Ohm's law says actually the opposite. P=(V^2)/R, or P=(I^2)*R. So, in fact, the greater the impedance (R ), the lower the current (I) required for same power (P). But the greater the voltage (V) needed. That's the main problem with high impedance headphones. They usually need a higher voltage to sound at same loudness. I say "usually" because it all depends also on the headphone efficiency. If a headphone was very efficient, it could sound even louder despite having high impedance. However, it seems that high impedance headphones are usually not among the most efficient ones.

As to DT880 clipping... if the measurable distortion al high levels goes away when doubling the voltage of the supply, it seems like the cause was clearly the insufficient voltage of the supply. Putting two batteries in series increases the voltage, but also increases the source impedance, so distortion should not be supply source-impedance related. But, doing the maths, I find strange that 9V would not be enough to drive a 250 ohm or even a 400 ohm headphone a high enough SPLs. Anyway, actual measurements are the final proof of performance.
*


I'll second that and add that higher-cost headphones are generally lower efficiency. Higher quality tends to yield compromises on efficiency, which doesn't usually matter too much at the power levels of headphones.

Doing the maths says that a 300 Ohm load driven by a conventional amp running off 9V would yield about 35mW RMS. 35mW, even on low efficiency headphones, ought to be painfully loud! That calculation assumes a genuine 9V supply, which a 9V (nominal) battery normally reaches on-load only when new.

Doubling the voltage by adding a second battery may actually cause more problems than it solves. Voltage is doubled but so is the current requirement, since the impedance is unaltered. A small battery is unlikely to be capable of supplying this extra current, the voltage on-load falls significantly and the amp clips badly. Theoretically there should be about 4 times the power output but in reality it's probably not going to rise much by adding a second battery.
Go to the top of the page
+Quote Post
MillerLowlife
post Dec 14 2005, 04:49
Post #38





Group: Members
Posts: 2
Joined: 18-November 05
From: Alpharetta, GA
Member No.: 25822



QUOTE
Well, Ohm's law says actually the opposite. P=(V^2)/R, or P=(I^2)*R. So, in fact, the greater the impedance (R ), the lower the current (I) required for same power (P). But the greater the voltage (V) needed.



I'm not sure where we are in disagreement as I did state that P=(E squared/R). I would like to point out, however, if the impedance is raised in a circuit, as in this case by plugging in a different set of headphones, the power transferred to the output load (headphones) will decrease. For example, if you have a signal source coming from a headphone jack that is 6v rms and you are using phones with 24 Ohms of impedance there will be a current of 250mA which will deliver 1.5W of power to the load. If you then replaced the 24 Ohm headphones with a set of 300 Ohm headphones then your 6v signal source will be dropping accross 300 Ohms, therefore causing the current to drop to only 20mA and now only 120mW of power is transferred to the headphone load. In your example you chose to increase the voltage coming from the headphone output jack to maintain the same current level. Power=V x I so in this case, that would require your output voltage to increase from 6v to now 75v which is a huge increase and definitely isn't going to happen unless you have something like a dedicated headphone amp to drive this thing. In any event, there is a greater power demand for more expensive (higher impedance) headphones, which I believe was my point to begin with.


--------------------
I can't live up to my name so I drink a lot... of tequila, vodka, etc.
Go to the top of the page
+Quote Post
KikeG
post Dec 14 2005, 09:08
Post #39


WinABX developer


Group: Developer
Posts: 1578
Joined: 1-October 01
Member No.: 137



??? You claimed that if impedance was increased, current requirements were increased, given same voltage:
QUOTE
the DT880 cans will need over 10 times more current for a given input voltage (rms)

I said it was the opposite, and your own example proves this. See, in an amp+headphone (or speaker) system, current flows according to output voltage and impedance (not the opposite) because amps are ideally constant voltage sources. And headphones, being passive devices, have constant impedance, of course.

QUOTE
In your example you chose to increase the voltage coming from the headphone output jack to maintain the same current level.

I did not provide any example, I did not say anywhere anything about keeping same current level. What I said is that, keeping power constant, if impedance increases, current requirements are lower. But *voltage* requirements are higher, which is the same thing you say in your example.

This post has been edited by KikeG: Dec 14 2005, 09:09
Go to the top of the page
+Quote Post
chelgrian
post Dec 24 2005, 14:51
Post #40





Group: Members
Posts: 316
Joined: 27-April 03
Member No.: 6228



QUOTE (WmAx @ Nov 25 2005, 08:27 PM)
An 'optimal' headphone amp need not be costly.


What do you think sonifex professional boxes, specifically

http://www.sonifex.co.uk/redbox/rbhd1_ld.shtml
Go to the top of the page
+Quote Post
WmAx
post Dec 24 2005, 19:34
Post #41





Group: Members
Posts: 541
Joined: 22-May 04
Member No.: 14243



QUOTE (chelgrian @ Dec 24 2005, 09:51 AM)
QUOTE (WmAx @ Nov 25 2005, 08:27 PM)
An 'optimal' headphone amp need not be costly.


What do you think sonifex professional boxes, specifically

http://www.sonifex.co.uk/redbox/rbhd1_ld.shtml
*



I have never used one, nor have I seen 3rd party measurements. However, I would be suprised if it had any audible problems such as noise. It should perform perfectly. As a precaution, I suggest that you buy from an outlet that gives a sufficient return period so that you can be sure it has no design/feature issues that make is non suitable for your purposes. At this price, however, if you don't require a rack mount device, I would recommend other more versatile products. The Headroom Micro is one such device in the price range of the Sonifix, and it has variable gain settings(to ensure optimal SNR regardless of headphone sensitivity), is able to be powered by 2x 9VDC batteries for portable applications, and has built in crossfeed filter function(which is important to some individuals to reduce listening fatigue on headphones). If you only need a transparent amplification for headphones(and do not need a specific form factor or other special features), then there are other much lower cost options available.

-Chris

This post has been edited by WmAx: Dec 24 2005, 19:39
Go to the top of the page
+Quote Post
fewtch
post Jan 3 2006, 04:35
Post #42





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



I performed an RMAA test on my Headsave Go-Vibe v2, driving 300 ohm Senn HD580s via a splitter and using a partially run down 9v lithium battery.

Although I've seen amps that measure better (amp is unshielded and sitting right next to the PC fwiw), there's nothing there that looks like clipping... it would be dead obvious. It's driving the headphones at ear-splitting levels, BTW... any louder and the drivers would be at risk of damage.

RMAA Test Results

Sorry for not including a soundcard baseline, but the performance of M-Audio AP 2496 is well known around here.

Conclusion, the Go-Vibe at 9v drives Senn HD580 (and by extension the lower impedance DT880) just fine. The drivers themselves are probably going to clip before the amp does.

So much wasted time talking, when measurements answer all questions... rolleyes.gif

This post has been edited by fewtch: Jan 3 2006, 08:18


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post
Leto Atreides II
post Jan 3 2006, 08:05
Post #43





Group: Members
Posts: 163
Joined: 13-January 02
From: Eugene, OR
Member No.: 1009



Am I the only one that can't load the images off of the page fewtch linked to?

Thanks for the analysis, fewtch!
Go to the top of the page
+Quote Post
WmAx
post Jan 4 2006, 04:54
Post #44





Group: Members
Posts: 541
Joined: 22-May 04
Member No.: 14243



QUOTE (fewtch @ Jan 2 2006, 11:35 PM)
Although I've seen amps that measure better (amp is unshielded and sitting right next to the PC fwiw), there's nothing there that looks like clipping... it would be dead obvious.  It's driving the headphones at ear-splitting levels, BTW... any louder and the drivers would be at risk of damage.


I don't know exactly what your reference point is supposed to be. I can get nice numbers as seen above, but not so under very dynamic music conditions. BTW, drivers easily take several times their continous power in short bursts(as is found in real music) without any damage. The short bursts are not sufficient in duration to heat the coils to dangerous temperatures.

I have made the measurements I promised a few weeks ago. I will post them next.

-Chris
Go to the top of the page
+Quote Post
WmAx
post Jan 4 2006, 05:02
Post #45





Group: Members
Posts: 541
Joined: 22-May 04
Member No.: 14243



Here are the test results, as promised. Values have been rounded.

Test equipment: Nicolet Digital Storage Oscilliscope, Micronta RMS bench top multimeter, Zen Extra 30GB MP3 Player, Shellbrook MaxiMoy Portable Amplifier, 9VDC Alkaline Battery Cells, Beyerdynamic DT880, Behringer UB1202 mixer(used as mic preamp), Behringer ECM8000 Measurement Microphone, DIY Headphone Jig, Radio Shack SPL Meter

DT880 *Measured Efficiency: 96dB/0.01W[apparent] @ 1kHz

*Efficiency measurement assumes a 250 Ohm impedance and non-reactive load. Real transducers produce reactive loads, so in reality, this is only an apparent value. I would preferably measure the actual impedance and phase of impedance for this device, but my computer is not currently set up to do these measurements.

1.72VRMS/250ohms=0.00688amps*1.72=0.01 W(apparent power)

I meausured dB value on the portable system of DT880 subject using a measurement microphone jig(that headphone mounts like a head, and contains a measurement microphone). SPL meter was used beside jig before headphones were mounted in order to establish an accurate dB reference point/calibration between the two. A voltage dB meter was used off of the measurement microphone's preamplifier to measure dB value vs. SPL meter reference. Volume set so that 1.72 VRMS was achieved. Db value recorded while driving DT880 with 1.72 VRMS with a 1 kHz sine wave test tone file that is recorded at -1 dBFs.


The following scope screens were made under volume setting relevant to (1)0.01 Watt DT880 efficiency [and] (2)(3)Minimal acceptable volume[as determined subjectively] setting for Douglas Allenbrook's Ethan Frome, Track 5, Disc 1, Mapleshade Records. Reminder: the Ethan Frome CD's average RMS values are at around -30dB, and the peaks are approaching 0dBFs. Most music would be listened to with the volume control set to a considerably lower gain, but this is a very highly dynamic music example.



At 1.72 VRMS output is clean with a 8.20VDC supply. 1.72 volts represents 96dB SPL as measured in the first part of this article. Distortion sets in soon after this point. Setting the volume to the reference volume positioin for the Ethan Frome opera work, 2.43VRMS is achieved, but the waveform is severely distorted at this gain setting at about 0dBFs, as was expected by the calculations I produced before testing. But is this really voltage limited distortion? It's easy enough to test by simply adding a 2nd battery cell in series with the first one. Figure (3) is with 16.70 VDC supply. Distortion is completely removed with the higher suppply voltage and the voltage output is now raised to the needed amplitude for the peaks in the Ethan Frome opera recording. The increments of voltage are not coincedence, they are the result of using 1 or 2 9VDC cells. The cells are about 1 or 2 hours used in the amp prior. They are still very good integrity, and this battery state was chosen to more closely represent slightly used(or rechargable with lower voltage) batteries.

((8.2VDC/2)-0.6)*0.707 = 2.47 VRMS theoretical maximum into a resistive load. However, 2.43 VRMS maximum was achieved. The actual load is not resistive, and the actual losses due to amplifier topology(0.6) may be slightly more than estimated.

5.48 VRMS is theotrically possible with the 16.70 VDC supply. 3.90 VRMS is what was measured at peak conditions under the preferred volume setting for the CD at subject in this analysis. Some headroom is leftover with the 16.70VDC supply. I did not measure the peak VRMS capability of the 16.70 VDC supply.

For effects of using a 9VDC supply on dynamic music as I specified eariler, it's easy to determine this(and it is blatantly audible as severe distortion in peaks) by looking at the waveform of the music, in relation to the dBFs relative measurements provided above. Between 1:40 and 2:00, several instances that are peaking at maximum possible value have occured, and severe distortion is heard. Note the average RMS values of this music. I have seperately highlited the first part and second part and provided an analysis for each, as well as the entire track, because of the large degree of variation of dynamics within the total track that would otherwise be masked by a single analysis. Note the histogram of relative dB distribution.



For most music(since most is not very dynamic) no distortion would be a problem. But as I asserted earlier, it is a real problem using a single 9VDC supply amp on the DT880 with highly dynamic music.

-Chris

This post has been edited by WmAx: Jan 4 2006, 08:48
Go to the top of the page
+Quote Post
fewtch
post Jan 4 2006, 06:13
Post #46





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



QUOTE
=WmAx]For most music(since most is not very dynamic) no distortion would be a problem. But as I asserted earlier, it is a real problem using a single 9VDC supply amp on the DT880 with highly dynamic music.

Would Beethoven's 5th and 9th symphonies suit you as dynamic enough? Or any of the other orchestral classical music I listen to entirely distortion-free with Go-Vibe/DT880 daily. Perhaps you're referring to a more dynamic musical genre than symphonic classical, but I can't think of one.

QUOTE (WmAx @ Jan 3 2006, 08:54 PM)
QUOTE (fewtch @ Jan 2 2006, 11:35 PM)

Although I've seen amps that measure better (amp is unshielded and sitting right next to the PC fwiw), there's nothing there that looks like clipping... it would be dead obvious.  It's driving the headphones at ear-splitting levels, BTW... any louder and the drivers would be at risk of damage.


I don't know exactly what your reference point is supposed to be.

Reference point for what, clipping? It would show up on the RMAA test as severe distortion. I listened myself, and the test tones played through the amp at the proper volume for the RMAA software are *extremely* loud. No dynamic peak is going to be hitting that kind of volume, unless you want your eardrums blown out.

I don't know why you're so persistent in insinuating that the *go-vibe* in particular has trouble with high impedance cans. You do know that it's a matter of amp design and opamp choice as well as voltage, right? Well, I've both made a subjective claim that I never hear any distortion with the amp no matter how dynamic the material, and posted a test result that indicates strongly that there's no distortion at high volumes. If you care to buy a go-vibe and post contradictory results, there's a basis for further discussion... otherwise, I don't see any.

This post has been edited by fewtch: Jan 4 2006, 06:21


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post
WmAx
post Jan 4 2006, 06:17
Post #47





Group: Members
Posts: 541
Joined: 22-May 04
Member No.: 14243



QUOTE (fewtch @ Jan 4 2006, 01:13 AM)
Reference point for what, clipping?  It would show up on the RMAA test as severe distortion.  I listened myself, and the test tones played through the amp at the proper volume for the RMAA software are *extremely* loud.  No dynamic peak is going to be hitting that kind of volume, unless you want your eardrums blown out.


I just provided one of my favorite albums as a case in point. -30dB average RMS values, with +30dB peaks.

QUOTE
I don't know why you're so persistent in insinuating that the *go-vibe* in particular has trouble with high impedance cans.  You do know that it's a matter of design as well as voltage, right?  Well, I've both made a subjective claim that I never hear any distortion with the amp, and posted a test result that indicates strongly that there's no distortion.  If you care to buy a go-vibe and post contradictory results, there's a basis for discussion... otherwise, I don't see any.


Your measurements did not correlate with any known levels. I provided correlation to actual measured SPL and to a specific dynamic musical track, as well as showing the results of voltage limited distortion in reference to both. It does not matter what design your Go-Vibe uses, it can not possible reproduce very dynamic music(such as Ethan Frome) on the DT880 without distorting due to it's limited rail voltage supply.

-Chris

This post has been edited by WmAx: Jan 4 2006, 06:20
Go to the top of the page
+Quote Post
fewtch
post Jan 4 2006, 06:23
Post #48





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



QUOTE (WmAx @ Jan 3 2006, 10:17 PM)
Your measurements did not correlate with any known levels. I provided correlation to actual measured SPL and to a specific dynamic musical track, as well as showing the results of voltage limited distortion in reference to both. It does not matter what design your Go-Vibe uses, it can not possible reproduce very dynamic music(such as Ethan Frome) on the DT880 without distorting due to it's limited rail voltage supply.

False. Sorry about your incapable POS headphone amplifier, I recommend you get yourself a better one that doesn't distort with a 9v supply.

BTW, I've heard bad things about Shellbrook and your results don't surprise me. It's probably a matter of inadequate PS capacitance not being able to respond properly to dynamic peaks. If you think voltage is all that matters, you need some electronics courses.

This post has been edited by fewtch: Jan 4 2006, 06:34


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post
WmAx
post Jan 4 2006, 06:26
Post #49





Group: Members
Posts: 541
Joined: 22-May 04
Member No.: 14243



QUOTE (fewtch @ Jan 4 2006, 01:23 AM)
False.  Sorry about your incapable POS headphone amplifier, I recommend you get yourself a better one that doesn't distort with a 9v supply.
*


Since you have resorted to this level of response[ a random and baseless profane comment], I will no longer reply to you within this thread.

-Chris
Go to the top of the page
+Quote Post
fewtch
post Jan 4 2006, 06:36
Post #50





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



Heh, you would have to have expanded an acronym, which happens in your mind and not here on the forum. Anyway, you continue to neglect the importance of power supply capacitance in responding to dynamic peaks, so I just assume you need more education in basic electronics... no reply necessary.

P.S. I suggest you look up what the go-vibe website means when it lists:

"2 x Panasonic 1800uF power supply caps"

Hint: What are these parts used for?

This post has been edited by fewtch: Jan 4 2006, 07:10


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post

4 Pages V  < 1 2 3 4 >
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 23rd September 2014 - 23:43