Gain in Dynahi
Jan 4, 2005 at 3:10 PM Thread Starter Post #1 of 6

chillysalsa

Headphoneus Supremus
Joined
Sep 26, 2002
Posts
1,987
Likes
186
What is the best way to lower the gain in a Dynahi? I am using it with Grados and a stepped attenuator, and I'm only using the first 4 steps before it gets too loud.

The first 4 steps on the SA give me:

-60.35
-52.11
-46.38
-42.52

of attenuation. I guess the steps are just too big... but I selected values that were finer closer to the middle of the attenuator!

Some solutions I am considering:
*My gain resistor in the amp (the 2000R location beside the 33pf cap) is 1.2K, what should I lower this to make the amp usable with both Grados and Senn's?

*I may put in a 75 ohm resistor at the RCA inputs to attenuate there

*I may put in a 10K input resistor together with the attenuator to change the log curve of it

What would be the best sounding solution??? Suggestions are appreciated...
 
Jan 4, 2005 at 10:58 PM Post #2 of 6

amb

Member of the Trade: AMB Laboratories
Joined
Apr 1, 2004
Posts
4,933
Likes
41
Quote:

Originally Posted by chillysalsa
What is the best way to lower the gain in a Dynahi? I am using it with Grados and a stepped attenuator, and I'm only using the first 4 steps before it gets too loud.


If I recall, the feedback resistors on the dynahi is 2K ohm around the amp and 100 ohm to ground. Since you have already reduced the 2K to 1.2K, I wouldn't recommend reducing that much further. However you could increase the 100 ohm resistor value. The gain of the amp is computed by: G = (R1 / R2) + 1, where R1 is the "2K ohm" resistor and the R2 is the "100 ohm". With a 1.2K/100 resistors you still have a gain of 13 which is too much for Grados. Try changing the 100 ohm resistor to something like 560 ohms, and it will give you a gain of around 3.
 
Jan 4, 2005 at 11:59 PM Post #3 of 6

chillysalsa

Headphoneus Supremus
Joined
Sep 26, 2002
Posts
1,987
Likes
186
Thanks AMB!

I put in a 300R for R2, so the gain is about 5. Much better.

My RA-1 had a gain of 5 as well and it drove sennheisers just fine, so I think I will stick with this setting. And still, I can use a good 6 or 8 positions of the attenuator before it gets too loud, that is plenty of range for me (I'm not super picky about the volume!).
 
Jan 5, 2005 at 5:51 AM Post #4 of 6

louissit

New Head-Fier
Joined
May 18, 2004
Posts
1
Likes
0
eek.gif

Is there any sonic different between changing 100 ohm resistor 2000 ohm feedback resistor?

No sure the 100 ohm is optimum value or not.... can kelvin say something about this? What is the best way to adjust gain?

I'm planning to make a balanced dynahi (dynomite?). It seems like I have to reduce the gain anyway.
 
Jan 5, 2005 at 7:07 AM Post #5 of 6

amb

Member of the Trade: AMB Laboratories
Joined
Apr 1, 2004
Posts
4,933
Likes
41
Quote:

Originally Posted by louissit
eek.gif

Is there any sonic different between changing 100 ohm resistor 2000 ohm feedback resistor?

No sure the 100 ohm is optimum value or not.... can kelvin say something about this? What is the best way to adjust gain?



There is no single "optimum" value for these. The ratio of the two resistors determine the gain, and there is a wide range of values you could select to achieve the desired gain. The only considerations are the following:
1. Keep the resistance low to prevent resistor noise from becoming a factor.
2. However, it must not be too low such that the resistor around the amp becomes a significant load to the amp's own output. In such a case power dissipation on the resistor also becomes an issue.
3. For minimum DC offset the differential input transistors should "see" approximately the same equivalent resistance to ground on both sides. This however is difficult to achieve due to conflicting requirements on the inverting side vs. the non-inverting side, and the fact that a volume control pot changes this depending on its wiper position. In the case of the dynahi (and any FET input amp), this is pretty much a non-issue due to very low input bias current. Also, the dynahi has a DC servo to keep things in check.
 

Users who are viewing this thread

Top