yairf
Trade: Teddy Pardo Audio
Hi all 
I have recently received a NAC-112 with a serious heart disease.
The symptoms were an unstable output signal, plus background noise, which would become intolerable whenever you cranked the volume up.
Eventually I decided to replace the entire circuit board with a conventional volume pot and fitted a pair of 821 boards from Avondale Audio. Here is the story behind it:
NAC-112 internals
The NAC-112 is basically a remake of the classic Naim preamp configuration, similar to the NAC-102:
Input switching --> Input stage (Time-Aligned) --> Volume Control --> Output (Gain) stage.
Supposedly this is a tried and tested configuration that can't go wrong, right...?
WRONG
They had to change something, so as to justify a new model number, so they:
1. Replaced the input relays with CMOS Switches
2. Replaced the good-old volume pot with a CPU-Controlled, CMOS-Switched Attenuator (see below).
3. Changed the classic gain stage to become a CMOS-Switched-Variable-Gain.
4. Changed the local regs to 15V, as required by the CMOS switches.
NAC-112 Volume Control
This system is based on a motorized analog-pot, acting as a DC voltage divider, with an A/D that samples the voltage into a digital reading, which is sent into the Micro-Controller, which sends a command to the CMOS switches, which switch resistors in the attenuator circuit. Theoretically this should have a much better accuracy and linearity than a conventional volume pot. How does it work in real world terms?
In the words of an esteemed colleague: "with the CMOS switching, the NAC-112 strangles everything it tries to play!".
A little "Going Over"
After hooking in my venerable Tektronix scope, I found out that the unstable output was caused by the volume control system. I'm not sure if the fault lies in the CMOS switches or the analog pot that triggers the system, but it was rather complicated to find out the root source, and seemed like a waste of energy.
I decided to apply the following changes:
1. Disconnect the input stage from the volume control system. I generally like the TA circuit, so I decided to leave it in and let it drive the new volume pot.
2. Remove the old volume pot from its position, but leave it powered on, so as to avoid CPU communication errors. Also, I needed the motor-control for the new pot.
3. Install a new stereo volume pot (ALPS RK-27 "Blue"). This is a motorised pot, which means I was able to maintain the remote-control functionality.
4. Disconnect the output stage and rewire the signal to the Avondale 821 boards, having the output signal wired directly to the output relay.
5. Power the 821 boards with an ALWSR regulator, which is one of my old-time favorites.
6. The NAC-112 is powered by the auxiliary power supply in its teammate - the NAP150. I have increased the auxiliary voltage to 30V, so as to leave enough headroom for the 112 internal regs.
Here is the new layout, with some captions:
The result? From the first turn-on, I got goosebumps ! It's stunning !
The 821's yield a gain of 3:1 (about 9.5dB - just about perfect!) and sound so alive, so effortless, so transparent, it makes you wanna get up and dance.
The NAC-112 retained all its original functionality, except one: I wasnt able to recreate the "Start-up Mute" function, as its applied through the switched attenuator. I'm still searching for a reasonable solution
Cheers,
Yair
I have recently received a NAC-112 with a serious heart disease.
The symptoms were an unstable output signal, plus background noise, which would become intolerable whenever you cranked the volume up.
Eventually I decided to replace the entire circuit board with a conventional volume pot and fitted a pair of 821 boards from Avondale Audio. Here is the story behind it:
NAC-112 internals
The NAC-112 is basically a remake of the classic Naim preamp configuration, similar to the NAC-102:
Input switching --> Input stage (Time-Aligned) --> Volume Control --> Output (Gain) stage.
Supposedly this is a tried and tested configuration that can't go wrong, right...?
WRONG
They had to change something, so as to justify a new model number, so they:
1. Replaced the input relays with CMOS Switches
2. Replaced the good-old volume pot with a CPU-Controlled, CMOS-Switched Attenuator (see below).
3. Changed the classic gain stage to become a CMOS-Switched-Variable-Gain.
4. Changed the local regs to 15V, as required by the CMOS switches.
NAC-112 Volume Control
This system is based on a motorized analog-pot, acting as a DC voltage divider, with an A/D that samples the voltage into a digital reading, which is sent into the Micro-Controller, which sends a command to the CMOS switches, which switch resistors in the attenuator circuit. Theoretically this should have a much better accuracy and linearity than a conventional volume pot. How does it work in real world terms?
In the words of an esteemed colleague: "with the CMOS switching, the NAC-112 strangles everything it tries to play!".
A little "Going Over"
After hooking in my venerable Tektronix scope, I found out that the unstable output was caused by the volume control system. I'm not sure if the fault lies in the CMOS switches or the analog pot that triggers the system, but it was rather complicated to find out the root source, and seemed like a waste of energy.
I decided to apply the following changes:
1. Disconnect the input stage from the volume control system. I generally like the TA circuit, so I decided to leave it in and let it drive the new volume pot.
2. Remove the old volume pot from its position, but leave it powered on, so as to avoid CPU communication errors. Also, I needed the motor-control for the new pot.
3. Install a new stereo volume pot (ALPS RK-27 "Blue"). This is a motorised pot, which means I was able to maintain the remote-control functionality.
4. Disconnect the output stage and rewire the signal to the Avondale 821 boards, having the output signal wired directly to the output relay.
5. Power the 821 boards with an ALWSR regulator, which is one of my old-time favorites.
6. The NAC-112 is powered by the auxiliary power supply in its teammate - the NAP150. I have increased the auxiliary voltage to 30V, so as to leave enough headroom for the 112 internal regs.
Here is the new layout, with some captions:
The result? From the first turn-on, I got goosebumps ! It's stunning !
The 821's yield a gain of 3:1 (about 9.5dB - just about perfect!) and sound so alive, so effortless, so transparent, it makes you wanna get up and dance.
The NAC-112 retained all its original functionality, except one: I wasnt able to recreate the "Start-up Mute" function, as its applied through the switched attenuator. I'm still searching for a reasonable solution
Cheers,
Yair
