[odigg]

Quote:

Audio-gd DAC19, I have no idea what it outputs.

You could measure the output voltage.  All you need to do is play a 1khz tone file through Foobar at full volume and measure the output voltage of the DAC.  You also need to make sure it's not clipping - I'll let somebody else tell you how to do that as I don't want to give you the wrong information.
 
 
 

 

[deadlylover]

Quote:

See edit, seems like you may have pushed the gain too hard (your Audio-GD outputs 2.5V).

 
Pushing the gain too hard on the second lowest option? =P.
 
I'll adjust it and report back later, I hate the **** thing now.
 
Quote:

You could measure the output voltage.  All you need to do is play a 1khz tone file through Foobar at full volume and measure the output voltage of the DAC.  You also need to make sure it's not clipping - I'll let somebody else tell you how to do that as I don't want to give you the wrong information.

 
Yeah I get something like 2.7v, thanks.
 
I didn't think it would be a problem as the stupid schematic said input 4V max, that's mighty misleading isn't it?

 

[Shike]

Quote:

 
Pushing the gain too hard on the second lowest option? =P.
 
I'll adjust it and report back later, I hate the **** thing now.

1) your source is hotter than almost any DAC out there in terms of voltage.
2) the amp swings a lot of voltage, you're swinging the maximum amount 100% of the time.  Drop the gain down to 2x and it should be alright.
 
 
Quote:

Yeah I get something like 2.7v, thanks.  
I didn't think it would be a problem as the stupid schematic said input 4V max, that's mighty misleading isn't it?

 
Read the associated article with it - he stated at 2v that 3.1x gain would be the max roughly.  You're at 2.7v.  The 4v max is in relation to damaging the amp.

 

[evanft]

Quote:

1) your source is hotter than almost any DAC out there in terms of voltage.

Really? My E-MU 0204 outputs 6.7 dBV, which is about 2.2 Vrms or 3.06 Vpeak.
 
Calculator: http://designtools.analog.com/dt/dbconvert/dbconvert.html
 

 

[Shike]

Quote:

Really? My E-MU 0204 outputs 6.7 dBV, which is about 2.2 Vrms or 3.06 Vpeak.
 
Calculator: http://designtools.analog.com/dt/dbconvert/dbconvert.html

That .3v-.5v is substantial.  NwAvGuy said that in the case of 2.25v, 3.1x gain was about max - anything more and you're pushing it.  If the gain was dropped to say 2.5 it would be about 6.75v max, currently it's demanding 7.75v minimum causing clipping (7v max AC).

 

[b0ck3n]

Anyone who's likely to use unusually hot DIY DACs with the O2 is hopefully knowledgable enough to read the design guidelines and adjust the gain accordingly. I don't necessarily think that the O2 has to play to the esoteric, freakish power output equipment crowd - the O2 is about real world performance in real world situations. Ti Kan's example of running the 114 dB/V, 32 Ohm PX100 headphones with 5x gain is anything but "real world".

 

[Steve Eddy]

 
Here's the way I look at it.
 
If your design goal is primarily pure performance, you end up with this:
 

 
Not this:
 

 
se
 
 

 

[upstateguy]

Quote:

your gain setting is limited though.  if using a bog standard 2Vrms source, you need to keep the gain at 3.1x or lower or risk clipping at the input.  as stated before, go above this gain, and the signal will be clipping at the input no matter where the pot is sitting.  if using a lower V source, you have more flexibility with the gain.  do you need a gain higher than 3.1x with a 2Vrms source for the K601 - probably not.
 
 

If you're using a standard AC dac source, do you need a higher gain than 3 to listen to T-1s, LCDs, or 800s at decent volume levels?
 

 

[kwkarth]

Quote:

If you're using a standard AC dac source, do you need a higher gain than 3 to listen to T-1s, LCDs, or 800s at decent volume levels?

By spec, it doesn't have adequate output to properly drive those cans regardless of gain structure.
 
 

 

[nikongod]

 
Quote:

Anyone who's likely to use unusually hot DIY DACs with the O2 is hopefully knowledgable enough to read the design guidelines and adjust the gain accordingly.

 
Your key assumption that the amp will only be used by tech-savvy people is hopelessly invalid. 
It is infinitely more reasonable to assume that this amp WILL be sold to non-techy people second hand when DIYers tire of its sound, and new by DIY-MOT builders. It is also reasonable to assume that many people who are building the amp will lack the knowledge to set the gain appropriately - the default gain is seldom "right" for any particular end user, yet you still see people building are amps with the same gain as on the schematics. Surely you will agree with me when you reflect on your previous experience following Cmoy, M3, B22, ckkiii threads.
 
Quote:

I don't necessarily think that the O2 has to play to the esoteric, freakish power output equipment crowd - the O2 is about real world performance in real world situations. Ti Kan's example of running the 114 dB/V, 32 Ohm PX100 headphones with 5x gain is anything but "real world".
 

 
The O2 is not about real world performance, it is about catering to the esoteric "freakishly good measurements" crowd. 
 
I think the mention of the PX100 was meant to illustrate a situation where something that has always worked well for AMB and countless other people will not work at all here. 
On that note, how is using the PX100 with gain of 5 any different than using the HD600 (as shown on the designers webpage, sitting next to the amp and test gear) with gain of 5? It has been clearly illustrated, and puppet-admitted on behalf of the designer that using high gain with a high output source (lol, high output=2v in this case, lol) causes clipping in the gain stages. If people are given the opportunity to screw something up they will. All it takes is a flip of a switch and kabooooom. 
 
 
Quote:

If you're using a standard AC dac source, do you need a higher gain than 3 to listen to T-1s, LCDs, or 800s at decent volume levels?
 

Way to open the worm can....
 
You need to define decent volume levels and standard DAC levels  
Less than 100 db SPL and a 2v source: gain of 3 leaves more than ample volume-control headroom (for quiet recordings) for both Hd800 and T1. Just spin the volume knob way up high.
 
The problem is that people dont like to use their gear like that. I think there is some deep-rooted psychological thing going on here. Between mom yelling when we were kids to "never pass half way up" on the TV/radio volume control, to the feeling of power from only just barely pressing the gas pedal in a powerful car and taking off like a rocketship or maybe something totally different we get conditioned to like to use controls a certain way. 
 
Looking around at meets, and a quick survey on teh forums will show you that many people only use the section of the potentiometer just above mute (the -60 to -40db range). There is so much fail in this, but it is a case where it takes a deliberate and conscious effort to break the mold. What REALLY sucks is that when you do, people get visibly nervous about spinning the knob up past half way and then complain it lacks power despite not topping it out.
 
There are 2 kinds of need going on here... What you need for functionality, and what you need to get most people to give your gear 2 listens. 

 

[khaos974]

]

By spec, it doesn't have adequate output to properly drive those cans regardless of gain structure.
 
 

At how many dB SPL do you consider a pair of headphones properly driven?

 

[b0ck3n]

My assumption was that people who use DIY DACs are likely more tech-savvy than the agerage user. The average user is more likely to use a 1v iPod. Besides that, the O2 isn't a commercial product and requires some interest and research to find out about it. For example it's far less likely to harm the end user than a Schiit amp is, the latter also being marketed to a much wider spectrum of consumers.

Edit: the O2 won't harm the end user ever, but if your preferred source is ridiculously hot then you might want to adjust the gain accordingly.

 

[Maxvla]

@ Nikongod

"Looking around at meets, and a quick survey on teh forums will show you that many people only use the section of the potentiometer just above mute (the -60 to -40db range). There is so much fail in this, but it is a case where it takes a deliberate and conscious effort to break the mold. What REALLY sucks is that when you do, people get visibly nervous about spinning the knob up past half way and then complain it lacks power despite not topping it out."

Absolutely true. I see this all the time. I for one don't subscribe to this and put the volume knob wherever the hell I please. If I need it past halfway that's where it goes.

 

[kwkarth]

Quote:

At how many dB SPL do you consider a pair of headphones properly driven?

Well, I know this will bother many of you, but even though a couple of the cans will make noise, will the quality of that noise properly showcase what those cans are capable of?
 
No.  It all depends upon one's level of experience and expectation, quality of source, quality of program material, etc.  SPL alone does not even come close to being the whole story.
 
 

 

[khaos974]

Well, I know this will bother many of you, but even though a couple of the cans will make noise, will the quality of that noise properly showcase what those cans are capable of?
 
No.  It all depends upon one's level of experience and expectation, quality of source, quality of program material, etc.  SPL alone does not even come close to being the whole story.
 
 

I've run a few calculations using
- Tyll's measurements, specifically the "Vrms needed to reach 90 dB SPL" figure, which is the reference voltage
- the 1% THD clipping limit voltage at 33 , 150 and 600 ohms (the figure at 150 and 600 ohm is the same at 7.3 V)

Since we are not taking into account any possible problem about gain structure, those figures are fine, with a possible 1-2 dB margin of error.

So 20 log (clipping voltage/ Vref) give:
- 24.5 dB gain for the T1
- 26 dB gain for the LCD-2
- 25 dB gain for the HD 800

Multiplied by a 0.9 factor, all tho voltages fall under the 0.01% THD+N range, which is still above 110 dB SPL for any on the concerned headphones.

EDIT: Of course that's only taking into account distortion and not audible factors that give the perception of headphone being well driven.