Headphone Amp - AD797 compensation, distortion cancellation caps

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
F

FritzS

Member
Joined 2008
stockhammer.at/hifi/
Reason - past years ago, I build an headphone amp from a kit - the WNA MKII and testet some OPAs - IC1
LM6171 (designed for), LME49710, AD843, AD847, LT1056, LT1022, OPA602(AP), TLE2071, OPA627, Supreme Sound Opamp V5; and now AD797.
With AD797 I fear oscillation, but they work very well and with the best sound I heard with WNA MKII.
Now I will spend them the suggested 'Distortion Cancellation and Bandwidth Enhancement' caps, 47pF or 56pF (which are better?).

Unfortunately I don't have distortion measure equipment. :Ouch:
My measure only my ears. ;)

The WNA MKII Thread in head-fi:
New! WNA MKll Head-amp kit. | Headphone Reviews and Discussion - Head-Fi.org
New! WNA MKll Head-amp kit. | Page 51 | Headphone Reviews and Discussion - Head-Fi.org

The schema and modifications of my WNA MKII:
Audio Project Headphone Amp WNA MKII

I found this too:
distortion cancellation for beginners
 
traderbam

traderbam

Member
Joined 2002
rayma said:
The distortion reduction capacitor is really only useful for high gains >100 times.
With your low gain circuit, the distortion will be well below the noise level,
and the capacitor will not be helpful.
Click to expand...
Hmm, controversial.
Anyhow, the magic mathematical "cancellation" formula relies on both the external capacitor and internal capacitor being exactly equal. They can never be equal not least because one has a silicon dielectric.
However, the external cap does provide negative feedback around the output stage in, I assume, a tight electrical loop, and should be better in than out. Relying entirely on global NFB is usually inferior in my opinion.
 
Last edited:
scott wurcer

scott wurcer

R.I.P
Joined 2004
traderbam said:
Hmm, controversial.
Anyhow, the magic mathematical "cancellation" formula is a bit mythical and relies on both the external capacitor and internal capacitor being exactly equal. They can never be equal not least because one has a silicon dielectric.
Click to expand...

No actually the junction capacitances are balanced as much as possible the MOS caps are a grown oxide that is quite good they make 24 bit switched cap circuits after all. As I said if you trim the cap through 50pf (null) to 100pF the error flips sign that is the distortion is the same magnitude but reverse phase. If you can make a normal negative feedback loop behave this way, I would be interested.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom