advertisement


Hackernap, NCC2x0: omit input coupling cap for wire link?

C

cubeasic

pfm Member
Hello,

This is maybe a dumb question, but as I understood it once, with Hackernaps and NCC amps, the 2,7k and 24k resistor an the input are the counterpart of the 27k feedback resistor, presenting nearly equal resistance to the long-tailed pair inputs (from output and GND, respectively).

In case of DC-absence at the input, a wire link instead of the input coupling capacitor is often promoted here.

When using a wire link for the input coupling cap, the 100k resistor of the input RC highpass is in parallel to the 24k resistor, effectively reducing the 26,7k resistance of 24+2,7k to something around 22k.

Since the hackernap then is directly coupled to the source any source-side parallel resistance reduces this even more.

Does this matter in terms of stability an/or sound quality? Is there a best practice approach?

Thanks
Florian
 
It won't effect stability or sound quality and there is also nothing worth while to be gained from removing the input coupling cap. It will effect DC offset. You may end up with say 100mV offset, which is still harmless.

I wouldn't bother if I were you.
 
If you have a preamp with a relatively small value of output couping cap and you use the NCC input cap then you could end up with an audible effect at the lowest frequencies. Indeed if you have a capacitor coupled preamp there is just no need for the power amp to have an input cap.

If you have a DC coupled pre then your thinking is correct and it could have an effect on the DC offset of the power amp if you link the input cap. This is usually minimal and is easily measured - initially without speakers connected to be on the safe side.

Stability will not be affected.

In terms of sound - trust your ears. Many people think a link sounds better than a cap.
 
Thanks for the feedback. Much appreciated. I guess I‘ll try it soon, and check for DC offset in the process.

The problem in my case goes a little bit further.
I use a balanced output buffalo DAC and would like to use the unbalancing transformer at the hacker nap input (10k:10k Sowter 3575). This guarantees negligable DC offset, and just for DC protection, the input cap can go.

But the secondary winding DC resistance of the transformer is much lower than 24k, reducing the total resistance to something much lower than 20k...

I just will try it at the next opportunity and see how the DC offset behaves, and also how the frequency response in the bass department changes.
I assume the lone 100k input resistor can go as well in this case? It seems kind of pointless when omiting the cap while having a fixed ground reference by the transformer...

Thanks
Florian
 
It should be OK to remove the 100K and also to increase the 24K to keep the total the same when the transformer is in parallel but I would be wary of removing the 2K7 or the 220pF cap. Whilst mainly there to provide a low pass filter at the input, some amplifiers need them to prevent instability. It may well be fine to remove them but I wouldn't do so without testing their possible impact on stability.
 
Ok, that is reassuring. I will keep the 2k7 and 200pF. I figured that stability could be an issue, even though the transformer would provide low pass filtering by itself.
 
You must log in or register to reply here.


advertisement


Back
Top