Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Gain setting of B22 for DT880/600ohm
New Posts  All Forums:Forum Nav:

Gain setting of B22 for DT880/600ohm

post #1 of 17
Thread Starter 
I have passed by several B22 with gain setting in the range of (SE/balanced) 1/2, 3/6, and SE:8 since I was told DT880/600 takes gains about 11/22 to be good choice.

So I would appreciate any B22+600ohm beyer users to share your experience.

Thanks!
post #2 of 17
No answer here, but I see you've 250R DT880s... I've those in pre-2005 model & the 5x gain budget B22 in my sig drives them fine BUT they are my only cans that require the zobel. They'd squeal on very low vol settings. Linuxworks, amb & others posted recently about zobels for B22 & seems like a good, harmless thing to do.

Think you'd want the lowest gain that did the job well. I've read about people having to lower gain more than raising it & others may know better but I can't imagine needing > 8x for dynamics (esp. like those in my sig).
post #3 of 17
Why does it matter what amplifier the individual has? last I checked gain of 11 into 600 ohms from a transformer coupled tube amp is the same as gain of 11 from a cmoy, M3, or a B22. whats the difference?

Now then, I dont own a DT880/600 either

I do however own a AKG K240sextet, and sennheiser HD414 (the real ones, the third best dynamic headphone senn has ever made, 2000ohms) They both work fine from buffers and amps with gain of 2 for me. I dont listen as well as some other people, only 65db not 90.

The real question is not how much gain do you need, but how loud do you listen? What is the voltage output of your source? Have you ever measured the voltage output into your cans at your desired listening level?

If you already have the DT880/600 why dont you hook them up to your amp and estimate based on where you park the volume knob? For every third of the knobs spin, you can safely say you need 16-20 db less gain. So if you have the knob at 9'ockock (about 1/3 of the way up) you have 2/3 of the spin left and 32-40db too much gain in your system. Consider what the gain of your amp is now, and what 32db less is. If the knob is set to "noon" you are much better off than most people.

Very few people with dynamic cans and reasonable sources need gain of more than 3 in their headphone amps. If you have never put the pot at the far end of its travel to listen where you want to you have too much gain.

Think about that with the volume knob that you have never spun all the way up. You payed for that resistive attenuator and yet you have never used 2/3 of it. Interesting perspective on that, eeh?
post #4 of 17
one thing to keep in mind is that sources sometimes are weak in their voltage out. bantam dac and y1 dac are less than 2v out (much less). when I switch to one of my other dacs (midiman super2496) that one has more like 4v out and I can tell the diff quite easily (never have to run my gains at max). on my hd650 and 3ch b22 at a gain of 5, I DO run my vol at 100% max quite a lot. gain of 5 with a less-than-2v dac is not recommended, in my experience. you need gain somewhere, either before the amp or AT the amp.

just know what input level you'll use before you figure out your gains. input level does matter, so don't ignore it.
post #5 of 17
I plan to make an order to GJA for a 3 channel B22 + Sigma22 later on today. Is 2x gain ok for Denon D-7000's or should I go more for future headphones? How easy is it to have a hi/low "gain switch" installed? Does active ground make a difference with ordering the gain setting? Since I'm kind of a noob how do I know or test the "input level" on a Stello DA100 dac? I'll wait a few hours for replies before I "wing it" and make my order with 2x gain. Thx.
post #6 of 17
Quote:
Originally Posted by linuxworks View Post
one thing to keep in mind is that sources sometimes are weak in their voltage out. bantam dac and y1 dac are less than 2v out (much less). when I switch to one of my other dacs (midiman super2496) that one has more like 4v out and I can tell the diff quite easily (never have to run my gains at max). on my hd650 and 3ch b22 at a gain of 5, I DO run my vol at 100% max quite a lot. gain of 5 with a less-than-2v dac is not recommended, in my experience. you need gain somewhere, either before the amp or AT the amp.

just know what input level you'll use before you figure out your gains. input level does matter, so don't ignore it.
yes, for use with all sources I think a gain of 10 is minimum for 600 ohm phones. I had an amp with a gain of 8 and it wasn't enough for my 600 ohm sextetts with my source (1.7V RMS.)
post #7 of 17
[QUOTE=Oberst Oswald;6413602]I plan to make an order to GJA for a 3 channel B22 + Sigma22 later on today. Is 2x gain ok for Denon D-7000's or should I go more for future headphones? How easy is it to have a hi/low "gain switch" installed? Does active ground make a difference with ordering the gain setting? Since I'm kind of a noob how do I know or test the "input level" on a Stello DA100 dac? I'll wait a few hours for replies before I "wing it" and make my order with 2x gain. Thx.[/QUOTE

I personally would go either 4 or 5 on gain with the Denon's. When I first built my 3 channel B22, I set gain at the default of 8. On my D5000's, I got a fair amount of amp hiss between songs. Per Ti, I changed it to gain 5, because I still wanted to drive my HD650's (which I have since sold). On gain 5, you can barely hear any hiss on Denon's (you really sort of sense it vs hearing it), and it still drove my HD650's just fine. It also does a great job on driving my new K701's. I'm sure gain 2 would be fine for Denon's, but if you're going to add other known harder to drive cans, you might want to set at 4 or 5. If you go 4, then PM or email Ti, and he can give you the value of caps and resistors for 4 gain, which are not listed on the AMB site. Hope this helps
post #8 of 17
Thread Starter 
Thank you for the help. Just to get my bearing right. So the "gain 11" for B22 means 11dB gain? I thought that was a non-universal spec of its own, and hence my questions for you all.

If that is the case, then it will be easier to figure out (but not completely, see below). Certainly source level is relevant. Both my DLIII and Ref-1 (2.5V/5V with SE/balan) are hotter than norm (2 V), and I always running balanced if possible.

For my Phoenix (gain 9dB, balanced), I only needs 10~14 out of 70 on volume setting with balanced DT880/600 depending on recording level. My listening level is at 65~75dB (may 80dB for short duration in symphony) measured by SPL meter.

For HP100 (gain 11.3/low or 16.3/high), I needs about 10 o'clock for SE DT880/600.

With my hotter source, I think gain 8(dB) may do for me with SE B22. As for balanced B22, gain 6 may be OK. But this is all my guess. So some practical experience will be nice.

One B22 owner told me that with gain 3/6 (SE/bal.), he need about 1 o'clock to make his HD600 sing at his satisfactory level. And his another B22 with gain 1/2 even sound louder to him. Apparently, there is something strange there. I don't know if his report came from SE or balan. setup, but that makes me worry about short of headroom there (5 o'clock max).
post #9 of 17
Quote:
Originally Posted by linuxworks View Post
one thing to keep in mind is that sources sometimes are weak in their voltage out. bantam dac and y1 dac are less than 2v out (much less). when I switch to one of my other dacs (midiman super2496) that one has more like 4v out and I can tell the diff quite easily (never have to run my gains at max).
That doesn't sound right..... 'standard' CD players are most commonly 2VMRS at 0dBfs (5.6V peak to peak), and the y1 is 1.6VRMS at 0dBfs (4.5V peak to peak).

Quote:
Originally Posted by regal View Post
yes, for use with all sources I think a gain of 10 is minimum for 600 ohm phones. I had an amp with a gain of 8 and it wasn't enough for my 600 ohm sextetts with my source (1.7V RMS.)
I would be very keen to know what the amp was; 1.7VRMS is 4.8V peak to peak, and at a gain of 8 and maximum volume you would be theoretically be trying to push 38.4V up to peak to peak into the phones - and almost certainly clipping, and/or deaf during 0dBfs transients.

The only exception would be very quietly recorded classical music.
post #10 of 17
Quote:
Originally Posted by Beefy View Post
That doesn't sound right..... 'standard' CD players are most commonly 2VMRS at 0dBfs (5.6V peak to peak), and the y1 is 1.6VRMS at 0dBfs (4.5V peak to peak).

...

The only exception would be very quietly recorded classical music.
or, a poorly implemented movie set-top box streamer (popcorn hour) that is always a good 10db lower, on movies, than music. if I want to hear the dialog, my vol knob is often at 100% and I still could use more gain.

plug in the other dac (4v out) and I can hear the movie. so yes, there is a real world diff. perhaps its the way dts and dd5.1 are rendered on that PCH box.

for music, its not a big deal but for movie watching, it really is.
post #11 of 17
Quote:
Originally Posted by linuxworks View Post
plug in the other dac (4v out) and I can hear the movie. so yes, there is a real world diff. perhaps its the way dts and dd5.1 are rendered on that PCH box.
I think movies soundtracks are often recorded with a much wider dynamic range than your average CD. You might need it louder for dialogue, but then the action movie explosions will knock your teeth out.

I'm not saying it is the right advice for everybody, but the current trend towards everything needing super-high gain is mostly pretty silly. Especially if you are limiting the range of your volume knob with stepped attenuators or carbon pots.
post #12 of 17
in the case of the PCH box, its a known bug, actually. not sure if sigma (chip maker) is at fault (I doubt this, actually) or syabas, the pch company. but its been that way for a few years now and they have not fixed it yet ;(

I guess my main point is: don't forget the input source when you are figuring your gain. its NOT just about the load
post #13 of 17
Take into consideration the volume of the recording too, especially when you test your levels to determine your gain setting.

For example, don't use Metallica to determine your gain and be surprised when a live jazz recording isn't loud enough. I have a few recordings that need to be turned up pretty high, so it's not impossible to choose insufficient gain.
post #14 of 17
Quote:
Originally Posted by linuxworks View Post
bantam dac and y1 dac are less than 2v out (much less).
Actually, at 0dBFS the output voltages are:

bantam and alien DACs: 1Vrms
γ1 (with WM8501): 1.6Vrms
γ1 (with WM8759): 1Vrms
γ2 (anti-clipping off): 1.4Vrms

Punnisher brings up an important point. Some recordings are highly compressed and hug the 0dBFS line. They are much "louder" than others that have much lower average levels. Many of the best, dynamically uncompressed recordings tend to have low average levels so the peaks don't clip. The amplification chain should thus have sufficient gain to accommodate these recordings. The output level of the source as well as the impedance and sensitivity of the headphones should of course also be taken into consideration.

One other factor to consider is the taper characteristic of the volume pot itself. While we speak of "log" pots as if they are all the same, the truth is that most of them are not truly logarithmic. Thus, depending on which pot or attenuator you use, the playback level may be different at particular knob positions when compared to another amp with the same voltage gain, using the same source, driving the same headphones and playing the same recording.

On many good recordings, my β22 with 8x gain (18dB), delivers a loud but not deafening playback level with the Alps RK40 pot near 12 o'clock, using a γ2 DAC as source and driving the HD800. The HD580 and HD600 are a little bit louder.

Quote:
Originally Posted by dukja
Just to get my bearing right. So the "gain 11" for B22 means 11dB gain?
No. To convert voltage gain to dB, use the following formula:

dB = 20 * log(Vgain)

By the way, an unbalanced β22 (with +/-30V rails) can swing over 40Vp-p into the load before clipping. A balanced β22 doubles that. Under virtually no circumstance should you encounter amplifier clipping driving dynamic headphones to extraordinary levels, regardless of gain.
post #15 of 17
i have 11x gain on my B22 driving DT990/600ohm with a y-2 upstream. pot travel is 10-2 o'clock depending on the recording, i.e. perfect.
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Gain setting of B22 for DT880/600ohm