advertisement


hackernap advice thread

John, thanks for your findings and post.
The chokes idea is very interesting but I don't understand couple of thins - please pardon my ignorance:
* how do you determine the required choke resistance (I am using 1500uF FE caps)?
* why bigger than 20mA current is required and does it need to match the one the chokes are specified at (156M - 3H @ 100ma)?
Thanks,
Ivo
 
John, thanks for your findings and post.
The chokes idea is very interesting but I don't understand couple of thins - please pardon my ignorance:
* how do you determine the required choke resistance (I am using 1500uF FE caps)?
* why bigger than 20mA current is required and does it need to match the one the chokes are specified at (156M - 3H @ 100ma)?
Thanks,
Ivo

All good questions Ivo, and I had to do my homework on this before the building as well:

Explanation
When you put the choke before the first capacitor you are creating an inductor capacitor series circuit. It has a natural frequency of resonance which needs to be damped to prevent it singing away like a tuning fork provoked by the regular kick from the incoming mains pulses. Damping is achieved by getting the critical amount of resistance into the circuit. Some of this is already there in the choke and a small amount of ESR in the cap. Thankfully there's a pretty easy calculation for this as follows:

Calculations
For critical damping i.e no ringing the series resistance R = 2 times the square root of L/C. So for the 1.5 Henry choke and 0.001 Farads of capacitance we get R = 2 times the square root of 1.5/0.001 which comes to 77.45 ohms. That's why I added 22 ohms to the choke which already had 56 ohms resistance (makes 78 ohms).

There's also a calculation that gives the minimum current draw needed to keep a choke supply working as it should. This minimum current is just dependant upon the value of the choke (bigger means a lower minimum current) and the mains frequency. For 50 and 60 Hz it approximates to the following:

The minimum (critical) current in milliamps = the RMS secondary voltage divided by the choke inductance in Henrys. For my 1.5 Henry choke and a 55 volt transformer secondary that came to 55/1.5 mA or 36.7 mA.

The 100ma quoted for the 3 Henry choke 156M is the maximum rated current for the choke. It is either the current limit bofore it overheats over time or before the core saturates. As long as you don't exceed it in the long term you'll be OK.

Your setup calcs
With your 1500uF caps and the 3H choke you are looking at a needing a minimum of 89.4 ohms of damping resistance and a minimum current draw (in milliamps) of whatever your transformer secondary RMS voltage is divided by 3. The damping resistance in the choke is specced as 86 ohms and probably close enough on its own. If you have one 40-0-40 transformer serving a pair hackernap front ends then the minimum current draw of 13mA is already exceeded by the 20mA or so the front end uses, so no bleeder resistors are needed. From a 40-0-40 transformer you will get a minumum of 36 volts to the front ends which will be enough for evaluation of the benefits.

I hope that helps get you there. I'm keen to share the better sound this has brought me.

John
 
Thirdly I don't believe that the bypassing of electrolytics with film caps is beneficial. On three separate occasions I have tried these bypass caps and then removed them. In my experience they add a fuzziness to the mid and bass. I was trying the values of 100nF in the positions of C10, C12 and C15, both on the NCC clone and the NAP clone. It was always a relief to remove them. There is a good reason why they can degrade the sound. They can create a resonant circuit with the inductance of the electrolytic capacitor they bypass. IMHO it is better to use the best possible quality electrolytic alone at these points in circuit, or even large value film caps in place of the electrolytics altogether (as suggested by others). I'm building with enough space around the board to allow me to fit the much larger film caps in place of C9 and C11 and omitting C6 and C7 altogether.

Thanks for the very detailed and considered post, John. I just wanted to address the above point by saying that I agree. It should be pointed out that the optimizations I'm suggesting for the hnap front end supply (the use of MMKs instead of electrolytics) is consistent with best practice here: we're not augmentig electrolytics, but ditching them from the circuit entirely. Well, almost entirely: the reservoir caps will still be electrolytic.

Cheers,
Carl
 
John,
Thanks for the good and well explained for novice like me reply - much appreciated!
So you put the choke before the first cap - do you keep the low value 3W resistors between the caps on HackerCap board then?

Cheers,
Ivo
 
Gentlemen,

May be a dumb question but I need to ask:
If the FrontEnd supply is so important but with lower and steady consumption why we don't use more complex regulators like the ones designed from Andy, Teddy, Russ, etc.
I suppose they can be tuned or re-designed to give higher output voltages or am I missing something?
Thanks,
Ivo
 
The front end needs a constant current low noise supply hence the need for a nice regulator.
Any noise injected into the front end gets amplified through the 2nd (VAS)and 3rd (Output) stages
The output stage needs a raw fast high current capability to add welly:)

Alan
 
There was a design for a teddy 'powerreg' on here a while ago, I am planning on building some on stripboard for my gainclones.. Perhaps this would be another good use for them?

Sam
 
That's a good question, I'd like to know as well. Maybe somthing to consider for my ncc200 amp front ends.

For my NCC 200 I am going to butcher 2 Hackernap board's just to use the front end supply as a reg ;)
You could also use a little girator board.

Alan
 
John thanks for the info, I am going to build a second pair of amps so will see if I can try this out.
Ivo from memory the reason the front ends don't have more complicated regs is Les from Avondale posted some of his experience in trying out different options in regulating the front end, his conclusion was that the VBE aproach was best.
Geoff
 
Alan why butcher the NCC200 just sell the boards and build the hackernaps, there are a few other tweaks on them as well that you will miss out on.
I can tell you they are a good step up on the already good NCC's expect a more powerfull dynamic and detailed sound thats what I got.
Geoff
 
This was covered a bit on page 5 by hacker - depends on your transformer V for the FE input (and actual mains voltage).

If your using the standard 40v for the FE, it might want nearer 2M than 1M - to drop less voltage and keep the FE higher than the output V.

Quick q

R30, 34. Says 2M in the BOM, has 1M on the link. Which should it be?
Thanks,

Andy
 
Apologies, my search skills are weak!

This was covered a bit on page 5 by hacker - depends on your transformer V for the FE input (and actual mains voltage).

If your using the standard 40v for the FE, it might want nearer 2M than 1M - to drop less voltage and keep the FE higher than the output V.
 
hacker: have you dusted off your soldering iron yet? I'd still love to get my hands on a stereo set of HNAP and HCAP boards! Please!

Chris
 
Ivo from memory the reason the front ends don't have more complicated regs is Les from Avondale posted some of his experience in trying out different options in regulating the front end, his conclusion was that the VBE aproach was best.
Geoff
Thanks Geoff for pointing me this out!
On the other hand Les does not use a separate transformer for the FE as the HackerNAP does so may be new tests should be run?
If John finds improvement with better chokes filtering why not to use complicated regs then?
 
Most of the complicated regs have parts in that limit the output voltages to 24V or so. If you want a regulated 50V line, there are many fewer off the shelf choices.

Something like a LM317 has lower output impedance (at low frequencies), but has performance that drops off radically as frequnecy goes up, while Hacker's solution is very constant and frequency independent.

Note that in a power amp, the signal levels are much bigger, so the noise floor matters less than it does in a pre-amp.
 
Pete, my mistake for the single transformer - apologies!

PD, thanks for the reply - I thought there must be a good reason for Carl to not utilize more complex FE regs...
 


advertisement


Back
Top