Is the Dynalo still a good choice?
May 13, 2011 at 7:19 PM Post #16 of 45
I don't think the concern of 11x gain was noise floor, I think it was that there would be a very small usable range of the volume with sensitive cans. That being said, the output level of your source should be a key consideration when choosing gain as well. A source that puts out 1 Vrms with an amp with gain of 11x is about the same as a source that puts out 4 Vrms with an amp with gain of 3x.
 
May 13, 2011 at 7:20 PM Post #17 of 45


Quote:
I highly suggest reading the construction guild, and all the edits along with with the current guide.
 
Try investing in building a better PSU rather then a simple LM3xx bipolar supply, which allows it to be up and above the commercial counterparts. Or you could build it for the sake of DIY, whatever reason works best.
 
May I suggest you stick with the default gain of 11, I remember someone said that despite the high gain, it still was quite quiet. You could also use the default gain for the possibility of using a crossfeed.
 
Anyways,
Have fun!


Yeah, I just looked into it, and several grado users are using 5x or 8x gain.
 
I'll be using the o11 with this. Still in the preplanning phase, though.
 
 
May 13, 2011 at 7:24 PM Post #18 of 45


Quote:
I don't think the concern of 11x gain was noise floor, I think it was that there would be a very small usable range of the volume with sensitive cans. That being said, the output level of your source should be a key consideration when choosing gain as well. A source that puts out 1 Vrms with an amp with gain of 11x is about the same as a source that puts out 4 Vrms with an amp with gain of 3x.


My DAC is the gamma 2, which has an output of 1.4Vrms at 0dBFS.
 
EDIT: I'd like to be perfectly clear that I have very little knowledge in electricity, and the definition of Vrms is beyond me. The sound straight out of the y2 is extremely loud (my guess is 120dB+), so judging from this, I'd want a low gain, right?
 
May 13, 2011 at 7:46 PM Post #19 of 45
Cool, I use the gamma2 as my source as well. My beta 22 has a gain of 5, and it gets more than loud enough with senn hd600 and lcd2s. I wouldn't recommend doing any higher than 5x if you want to have reasonable volume control, and if you don't plan to get less efficient headphones you probably want to go for something around 2x.
 
BTW, Vrms is root mean square voltage. It is calculated by taking the integral of a signal squared over a certain period of time, dividing by the time interval, and taking the square root of that result. For practical purposes you can generally assume the rms value to be the maximum voltage level divided by the square root of 2. So from this you can see that the peak output of the gamma2 would be 2 volts, giving it a Vrms of 1.4 volts.
 
May 13, 2011 at 9:48 PM Post #21 of 45
What kind of horrible things will have to be endured with a gain of x11 and Grado's?
What happens if said user upgrades to a higher impedance headphone?
Why was the Dynalo designed with a 20K pot in mind, and a gain of x11 times?
Is 5x ok with music that's brick walled? or rather more music with dynamic range?
 
This has to be somewhat confusing to someone? Namely the 3rd question.
 
Cheers
 
May 13, 2011 at 9:55 PM Post #22 of 45
No idea why it was originally designed with an 11x gain. 5x should be fine with most music, though you may not get a huge volume travel with high sensitivity headphones. Nothing horrible would be endured, but you would only be able to turn the knob like 5% to get loud volume.
 
May 13, 2011 at 11:58 PM Post #23 of 45
OK, so I found this in the official thread as I was trying to figure out how to adjust the gain.
 
 
Quote Pars:
I would probably increase the 1K but I don't see that it would make much difference either way. BTW, the default gain on a Dynalo is for some reason not too loud with most phones as it would be in most amps. I know even with my SR60s, I still have my home dynalo turned up over 1/3 way, with CD as a source

 
This is blowing my extremely tired mind. Also, the way the pot is wired in this confuses me, but I'll figure it out tomorrow after sleep.
 
 
May 14, 2011 at 1:21 AM Post #25 of 45

Quote:
What kind of horrible things will have to be endured with a gain of x11 and Grado's?
What happens if said user upgrades to a higher impedance headphone?
Why was the Dynalo designed with a 20K pot in mind, and a gain of x11 times?
Is 5x ok with music that's brick walled? or rather more music with dynamic range?
 
This has to be somewhat confusing to someone? Namely the 3rd question.
 
Cheers


1: You (generally) attenuate a LOT which puts you in a poorly matched section of the pot, and IME screws with frequency response a bit. Higher gain results in more noise due to funky resistor effects, bummer. The only advantage is that the volume control is WAY down, like you are holding back a more powerful amp - purely psychological. 
 
2: The word of the lord passed down in the headphone chapter of the audio bible that you need gain of 11 to run anything more than 75ohms will have to be broken then. Obviously Im making a cheap joke. The lord did not come down and give us this rule, a human being (who like all human beings is prone to errors, and unlike many human beings listens loud as hell with 600ohm headphones) pawned it off on us - the rule may even predate people listening to headphones for hifi! gain of 10 was VERY popular for preamps, despite the fact that power amps have been running with 1v input sensativity since the 60's or earlier. Anyways, Despite what the easy to follow rule tells us, high impedance headphones generally run just fine on amps with voltage gain in the 1-4 ranges. Try it! Some people MAY top it out, but then gain of 11 is only a little more than 3 (if you max out an amp with gain of 3 easily, 11 is not even enough!). If the volume knob NEVER gets within 1/3 of the max you have more gain than you need - save yourself some resistor noise, and reduce the gain.
 
3: Why a 20K pot? If Im not mistaken Kevin Gilmore had a Penny + Gilles 20K pot "just sitting around" waiting for an application... Good enough reason to use a $250 pot for me. 
 
Many MFR's make 20-25K audio pots, its just that nobody uses them*. EVER. P+G is awesome so they can do anything they want.
 
* another rule given us by the good lord in the general section of the audio bible is that the only values of potentiometer to use as attenuators are 10K, 50K, and 100K. Values below 10K, or between those 3 values must never be used as an attenuator least you be smitten. In all seriousness, 20K is a sweet value: Its a LITTLE bit more than 10K (twice as much!) which buys you an extra octave of bass in cap coupled sources, but not so high impedance as to cause undue noise. 50&100K ohm pots have much more inherent noise which coupled with the trend for too much gain gets you nothing but too much noise * too much gain! values lower than 10K (1K,2K,5K are common) are SWEEEET because they have VERY low noise, but may prove too difficult for some sources to drive well, and forget it with cap coupled sources. 
 
Quote:
No idea why it was originally designed with an 11x gain. 5x should be fine with most music, though you may not get a huge volume travel with high sensitivity headphones. Nothing horrible would be endured, but you would only be able to turn the knob like 5% to get loud volume.


EVERYTHING was designed for 11X gain back then. 
 
Quote:
This is blowing my extremely tired mind. Also, the way the pot is wired in this confuses me, but I'll figure it out tomorrow after sleep.
 


poosticks... the quote in a quote thing didnt work. Just immagine the part you quoted just before the part I quoted is there. 
 
Everyone talks about gain without consideration of the source voltage level, ASSUMING that the user has a "standard" 2v source. Some people have high output 4 or 10v sources, others have a very low 0.25v. With a 4 or 10v source you need NO gain to run almost anything - just a good stiff buffer that can handle the heat. With a 0.25v source you actually need gain to run 300 ohm headphones!
 
May 14, 2011 at 4:13 AM Post #26 of 45


Quote:
 
I'll be using the o11 with this. Still in the preplanning phase, though.
 


I don't know if someone has already reacted to this but i'm going to answer anyway ; you can not use a sigma 11 with a dynalo unless you add a ground channel, which is complicated. Actually I am not sure if Kevin Gilmore ever considered a ground channel for any of his amp. In short you need to use a bipolar power supply such as the sigma22 with a dynalo.
 
To respond to the original question, I have a dynalo and it is indeed a great amp, you can not go wrong with it ! I lowered the gain too but I don't remember how much...
 
Again I think it has been said (that won't prevent me from repeating 
tongue.gif
) if you have a high gain, you need to use the potentiometer in a "region" where the two channels are not matched well at all (especialy with a cheap pot). A low gain allows you to turn it up a bit which "compensates" the differences of volume between the right and left channels.
Edit : Nikongod said : "You (generally) attenuate a LOT which puts you in a poorly matched section of the pot" It is exactly what I was trying to say.
 
 
May 14, 2011 at 8:58 AM Post #27 of 45


Quote:
Everyone talks about gain without consideration of the source voltage level, ASSUMING that the user has a "standard" 2v source. Some people have high output 4 or 10v sources, others have a very low 0.25v. With a 4 or 10v source you need NO gain to run almost anything - just a good stiff buffer that can handle the heat. With a 0.25v source you actually need gain to run 300 ohm headphones!



Actually if you go back a few posts I pointed out the same thing:
 
Quote:
I don't think the concern of 11x gain was noise floor, I think it was that there would be a very small usable range of the volume with sensitive cans. That being said, the output level of your source should be a key consideration when choosing gain as well. A source that puts out 1 Vrms with an amp with gain of 11x is about the same as a source that puts out 4 Vrms with an amp with gain of 3x.


The OP said he will be using a gamma2, aka a 1.4V source, so I suggested a gain somewhere between 2 and 5. 
 
The rest of your points are excellent.
 
 
May 14, 2011 at 12:16 PM Post #28 of 45
As someone who has had severe pot imbalance issues in the past, I can attest to Nikongod's statements about it putting you in a bad section of the pot. The solution to this would seem to be a stepper, which generally has no channel imbalance, but you're still attenuating a huge amount, and that can lead to two problems with steppers.

A resistor with a 5% tolerance will be much less accurate at higher resistance values than lower ones. (That said, if your stepper uses 5% resistors, you've got bigger problems in store...)
The bigger problem is that your steps will probably be HUGE! Splitting 11x gain into equal chunks means that you have very little precise control over volume.

The moral of the story is to not overdo the gain, in the end. I've built every single one of my amps with gains of 5 or less, and hardly ever had an issue with 2V or lower unbalanced sources. (Nearly all of my sources have been 2V/6dBV.)
 
May 14, 2011 at 1:28 PM Post #29 of 45
OK, so I'm awake and reading more about this amplifier.
 
Quote:
I don't know if someone has already reacted to this but i'm going to answer anyway ; you can not use a sigma 11 with a dynalo unless you add a ground channel, which is complicated. Actually I am not sure if Kevin Gilmore ever considered a ground channel for any of his amp. In short you need to use a bipolar power supply such as the sigma22 with a dynalo.
 
To respond to the original question, I have a dynalo and it is indeed a great amp, you can not go wrong with it ! I lowered the gain too but I don't remember how much...
 
Again I think it has been said (that won't prevent me from repeating 
tongue.gif
) if you have a high gain, you need to use the potentiometer in a "region" where the two channels are not matched well at all (especialy with a cheap pot). A low gain allows you to turn it up a bit which "compensates" the differences of volume between the right and left channels.
Edit : Nikongod said : "You (generally) attenuate a LOT which puts you in a poorly matched section of the pot" It is exactly what I was trying to say.
 

Yeah, I meant sigma 22
 
So, you guys are saying that the low end of pots aren't matched very well, and that I should lower the gain in order to increase the amount of travel so that I can avoid issues? Will do. 

 
Quote:
1: You (generally) attenuate a LOT which puts you in a poorly matched section of the pot, and IME screws with frequency response a bit. Higher gain results in more noise due to funky resistor effects, bummer. The only advantage is that the volume control is WAY down, like you are holding back a more powerful amp - purely psychological. 
 
2: The word of the lord passed down in the headphone chapter of the audio bible that you need gain of 11 to run anything more than 75ohms will have to be broken then. Obviously Im making a cheap joke. The lord did not come down and give us this rule, a human being (who like all human beings is prone to errors, and unlike many human beings listens loud as hell with 600ohm headphones) pawned it off on us - the rule may even predate people listening to headphones for hifi! gain of 10 was VERY popular for preamps, despite the fact that power amps have been running with 1v input sensativity since the 60's or earlier. Anyways, Despite what the easy to follow rule tells us, high impedance headphones generally run just fine on amps with voltage gain in the 1-4 ranges. Try it! Some people MAY top it out, but then gain of 11 is only a little more than 3 (if you max out an amp with gain of 3 easily, 11 is not even enough!). If the volume knob NEVER gets within 1/3 of the max you have more gain than you need - save yourself some resistor noise, and reduce the gain.
 
3: Why a 20K pot? If Im not mistaken Kevin Gilmore had a Penny + Gilles 20K pot "just sitting around" waiting for an application... Good enough reason to use a $250 pot for me. 
 
Many MFR's make 20-25K audio pots, its just that nobody uses them*. EVER. P+G is awesome so they can do anything they want.
 
* another rule given us by the good lord in the general section of the audio bible is that the only values of potentiometer to use as attenuators are 10K, 50K, and 100K. Values below 10K, or between those 3 values must never be used as an attenuator least you be smitten. In all seriousness, 20K is a sweet value: Its a LITTLE bit more than 10K (twice as much!) which buys you an extra octave of bass in cap coupled sources, but not so high impedance as to cause undue noise. 50&100K ohm pots have much more inherent noise which coupled with the trend for too much gain gets you nothing but too much noise * too much gain! values lower than 10K (1K,2K,5K are common) are SWEEEET because they have VERY low noise, but may prove too difficult for some sources to drive well, and forget it with cap coupled sources. 
 
Quote:

EVERYTHING was designed for 11X gain back then. 
 
poosticks... the quote in a quote thing didnt work. Just immagine the part you quoted just before the part I quoted is there. 
 
Everyone talks about gain without consideration of the source voltage level, ASSUMING that the user has a "standard" 2v source. Some people have high output 4 or 10v sources, others have a very low 0.25v. With a 4 or 10v source you need NO gain to run almost anything - just a good stiff buffer that can handle the heat. With a 0.25v source you actually need gain to run 300 ohm headphones!


OK, so I have several more questions about pots. I had almost no idea how they worked until I watched the intro on tangent's website. 
 
1. You connect potentiometers directly from the dc output, and then from the wiper you connect it to the board, right?
2. 20k pots are kind of hard to find. Does anyone have any suggestions? Alps is the recommended brand, right?
EDIT: 3. Is it even necessary to use a 20k ohm pot? From the bolded it appears that it's not critical.
 
 
So, I've decided on a gain of 4-5. The only issue is figuring out how to lower the gain. I'll look around for some info, since it is called DIY for a reason 
tongue.gif
.
 
 

Users who are viewing this thread

Back
Top