NAD 106 Preamp Restoration/Upgrade Project

Leestereo

Super Member
Here are some pictures from my circa 1990s NAD 106 preamp project. As can be seen, the 106's very plain looking fascia belies its designation in NAD's 1992 product line up as their "top of the line analogue preamp".



But there are some hints of its ambitions to be seen on the outside: the nearly ubiquitous "loudness" control is no where to be found and around the back, there are gold plated dual RCA outputs as well as XLR connectors.



Inside, there is a very clean layout on a single glass epoxy pcb. Immediately noticeable, is the sealed ALPS "Blue Velvet" volume control, the use of 1% metal film resistors and numerous encapsulated polystyrene capacitors. The topology appears quite nice with two regulated power supplies, and the line and phono stages are fully discrete transistor designs.



In stock condition, the preamp sounds very good, but I think that the potential of the design is let down by the use of TEAPO branded electrolytic capacitors. Accordingly, in this restoration/upgrade project, all of the non-signal path electrolytic capacitors will be replaced with low ESR, 105°C types and the signal path electrolytics will be replaced with "Audio Grade" electrolytics (e.g., Nichicon MUSE) or film caps (whenever possible).

Restore/Upgrade (Part 1):
In the main regulated power supply, the replacement capacitors used were all Nichicon PW types. There was no opportunity to increase the capacitance values due to space constraints on the pcb (the replacement capacitors are essentially the same size as the original ones). Nevertheless, the power supply is quite robust for a preamp with over 6300µF of filtering capacitance. I added heatsinks to the TO-126 regulator transistors since these run quite hot and the top cover is not ventilated.



Part 2 of the project will cover the phono stage.
 
Last edited:
Restore/Upgrade (Part 2):



In the stock phono stage, the input high-pass filter uses a 47µF electrolytic capacitor and a 0.1µF polyester by-pass. These were replaced with a 2.2µF WIMA MKS2 polysester film capacitor and a Kemet PHE426 polypropylene 47nF capacitor, respectively. The replacement film capacitors are large enough to provide a low enough cut off below the audible range (i.e., F3 = <1.5Hz) and have the advantages of lower leakage, lower ESR, and low %DA.



The stock DC filtering/decoupling capacitors were 220µF/25V and these were replaced with 470µF/35V United Chemicon KZE low ESR capacitors.



The stock RIAA low-end cut-off capacitors were 2200µF/6.3V and these were replaced with Nichicon PW capacitors of the same capacity but with a higher voltage rating (16V). Whenever possible, it is best to avoid low voltage (e.g., 6.3V and 10V) rated capacitors as their performance and longevity are typically poorer than their higher voltage counterparts. I usually try to use capacitors rated at a minimum of 25V, but in this case there wasn't sufficient space on the board.

The stock RIAA equalization capacitors are very high quality polystyrene types and hence there is no need to replace them.

The stock 10µF output electrolytic capacitors and the 0.1µF polyester by-passes were replaced with 22µF Nichicon FG (Fine Gold) audio grade capacitors and Kemet PHE426 polypropylene 47nF capacitors, respectively.



The phono stage output also includes a low pass filter with a 2.2nF polyester film capacitor. This capacitor was replaced with a better quality polypropylene of lower value (470pF) to essentially eliminate any high frequency phase shift. With the 2.2nF capacitor, the phase shift starts within the critical midrange, around 2kHz. With the replacement 470pF capacitor, the start of the phase shift is now above 10kHz.
 
Last edited:
Restore/Upgrade (Part 3):

There are three electrolytic capacitors per channel in the signal path of the NAD 106 line stage. These were replaced with high quality Nichicon KZ MUSE capacitors. The stock 100nF MKT film bypass was replaced with a better quality 47nF PHE426 polypropylene one.





The remaining signal path capacitors in the line stage are all film types. Except for one, they are high quality polystyrene capacitors; there is no need or advantage to be gained by replacing the polystyrenes. The stock capacitor which immediately follows the sealed ALPS volume control is a 470nF MKT (polyester) film. It would have been good to replace it with a polypropylene (or similar) type, but unfortunately one that has a 5mm lead pitch and would fit the allotted board space was not to be found. The use of a larger film capacitor and installing it off the board is not generally recommended since it may add noise to the signal (due to the long unshielded capacitor leads and the capacitor itself).
 
Last edited:
Restore/Upgrade (Part 4):

The NAD 106 has a separate headphone circuit which has its own regulated power supply. The replacement capacitors in the regulated supply were United Chemicon KZE types and there was enough space on the pcb to increase the capacity over the stock values. The original 47µF 25V capacitor at the base of the pass transistor was replaced with a 100µF 25V one. The original capacitor for the final DC filtering was a 100µF 25V and was replaced with a 220µF 35V one.



In the stock headphone stage, the input high-pass filter uses a 1µF electrolytic capacitor. The replacement was a 3.3µF WIMA MKS2 polysester film capacitor. Interestingly, in this case the film capacitor is larger than the original electrolytic. The output capacitor was a 47µF electrolytic and this was replaced with a 100µF Nichicon FW audio grade capacitor.
 
Last edited:
awesome stuff....would rebuilding a 1020 look anything like this...?

Although a much earlier NAD design, the 1020 actually contains a nearly identical phono stage topology. Its line stage also appears decent and it also uses a well designed regulated power supply. The build quality of the 1020 isn't anything to write home about, with ceramic caps and 5% carbon film resistors; but the design is very good and a rebuild would likely be worth the effort as it would yield many sonic improvements.
 
Some Listening Impressions:

The NAD 106, even in stock form, is a very good/excellent sounding preamp with the all of the hallmarks of the NAD house sound. Those who like the "NAD sound" often describe it as: neutral, smooth, musical, laid back, non-fatiguing. Conversely, those who don't particularly like the "NAD sound" have described it as: too polite, un-dynamic, uninvoling, dull. I am obviously in the former camp, but I do think that the other side's point of view does have some merit.

In the ~3 weeks since the final upgrades to the NAD 106, its been played every evening for at least 2-3 hours. The restoration/upgrade has changed the sound for the better (did anyone think I would say otherwise?). The differences in sound are not "night and day", but are readily noticeable with careful listening. One of the first differences I noticed was that the bass was subjectively deeper and "tighter"; bass transients have more impact. Previous to the restoration/upgrade, the treble range could be characterized as "soft"; its now subjectively, more extended and detailed. The 106 was always a very quiet preamp, but that too seems improved; the music seems to emerge from a "blacker" background silence. Playing CD and LP recordings that I'm very familiar with (e.g., Steely Dan, Aja; Miles Davis, Kind of Blue; Dire Straits, Love Over Gold; Beatles, Abbey Road; Karajan, 1962 Beethoven Symphony cycle; Duke Ellington Orch., Digital Duke), I was hearing the little details that much more clearly. The Steely Dan recording in particular seemed to gain clarity in the bass range. And for most recordings, the imaging/soundstage seems improved: wider and more 3D-like.

The Nichicon KZ MUSE capacitors that were used upgrade the final line stage, are sometimes characterized as "dark sounding", but that isn't my impression in the NAD 106. If anything, the overall sound is now subjectively more "open" and the overall presentation more lively/dynamic. The use of the Nichicon KZ capacitors was not due to any preference over the equally lauded Elna Silmic II capacitors; the KZs are just what I had on hand. And, I'm not convinced that there is that much difference between the various "audio grade" capacitors of the same rank. Similarly, I consider most low ESR, 105°C capacitors of the same rank to be essentially equivalent.

Final thought: The restored/upgraded NAD 106 is definitely a "keeper"; its still second to my Conrad Johnson PV-8, but the gap between them has now narrowed.
 
Lee this is beautiful and so well thought out work! I have done a couple of 3020's and have picked up some nice tips here.

On Audio Grade e-caps:

For signal path duties I really like:

Elna Silmic II
Wima MKS2
Wima MKS4 (if they will fit)
Elna Silmic bipolar
Nichicon ES

For filtering (local)

Silmic II
Nichicon KZ
Elna Cerafine
not audio but Nichicon PW and HE

Where voltage and or size are importiant:

Nichicon KA
Nichicon KL

Power Supply Filtering

CDE 381LX
Silmic II
Nichicon PW

As you have run into I really wish that there were more options for the 5mm lead spacing, alot of times I may use a Silmic II instead of a film.

As you say what cap to use depends on what you are trying to acheive.

-If the sound of the pre is slow and or soft sounding you may want to add Elna Cerafine to the amp section.
-If you are looking for more dimensionality and air-Silmic II
-If you are looking to put more body into the vocals-Silmic bipolar.

So listening to your finished product after burn in since you say you can hear the difference, you may say I wish the vocals had a little more body to them, you may want to add the Silmic Bipolar to the preamp area.

just an example

love your work, thanks for sharing!
 
Last edited:
Thanks for the compliments Pat; its always good to hear from fellow enthusiasts that one's efforts are appreciated.

I am familiar with, and have used, most of the capacitor types you have listed in various restorations/upgrades. The only exceptions being the Elna Silmic bipolar and the Nichicon KA. It is interesting that you don't recommend using the Nichicon KZ Muse in the signal path, they are purportedly Nichicon's best "audio grade" capacitor. I think they are really good for signal path use, provided that there is enough space for them; they are quite large for any given capacitance/voltage (see the picture above of the KZs next to the ALPs Blue Velvet pot).

IMO, I think that most of the "sound" of the NAD 106, or any given unit, is from the circuit design and that the quality of the components used in the build maximize/enhance the sound potential of the design. By corollary, installing the most expensive components will not transform a poor design into a great sounding one. And, due to the time and effort needed to do a full restoration/upgrade, I think that it is good practice to only consider such for a component that one already likes the sound of (assuming that it is in good working order). Of course, YMMV.
 
Thanks for the compliments Pat; its always good to hear from fellow enthusiasts that one's efforts are appreciated.

I am familiar with, and have used, most of the capacitor types you have listed in various restorations/upgrades. The only exceptions being the Elna Silmic bipolar and the Nichicon KA. It is interesting that you don't recommend using the Nichicon KZ Muse in the signal path, they are purportedly Nichicon's best "audio grade" capacitor. I think they are really good for signal path use, provided that there is enough space for them; they are quite large for any given capacitance/voltage (see the picture above of the KZs next to the ALPs Blue Velvet pot).

IMO, I think that most of the "sound" of the NAD 106, or any given unit, is from the circuit design and that the quality of the components used in the build maximize/enhance the sound potential of the design. By corollary, installing the most expensive components will not transform a poor design into a great sounding one. And, due to the time and effort needed to do a full restoration/upgrade, I think that it is good practice to only consider such for a component that one already likes the sound of (assuming that it is in good working order). Of course, YMMV.

If I have a choice of a signal path cap between the Silmic II and the Nichicon KZ, it will usually be the Silmic II getting the nod. Again it depends on the unit you are restoring, I do not have a templete of just using the same caps on every unit. Like you said the circuit will dictate the sonic signature, but it can be slightly tweeked. I agree 100% with you on the KZ, they tend to be larger and they have a very robust build with larger diameter leads that sometimes do not fit in the circuit board. I have a 1/16 hand tap drill for that but would really only do it if I have to.

Here are my listening impressions of the KZ from one of my other threads:

Please take with a grain of salt:

Initial Impressions: The KZ/FG are very analytical and clean sounding. They are probably the most accurate and natural yet across the board. Unlike the FC's they image well and go all the way down in the bass and have no high frequency roll off. There is a small problem in the lower mids they sound a bit scooped out here. I do want to break them in well before I commit to this though. They are not in the "tube like" catagory like the Silmics, and they do not have the detail and low bass articulation of the Cerafines.
 
Last edited:
Interestingly, the NAD 106 pcb layout appears to have been designed for the use of over sized "audio grade" capacitors in the signal path. In the attached picture, there are two 47µF 25V capacitors: the "audio grade" Nichicon KZ MUSE is approximately 4X the size of the 105°C, low ESR United Chemicon KY capacitor. Also, notice how there is much more space on the pcb allotted for the signal path capacitors.



As for the sound of the Nichicon KZ, I agree with your assessment, they are essentially "neutral", which is my preference in any passive component. The Elna Silmic II capacitors have been subjectively described as "smooth" or as you state, more "tube-like"; I don't disagree, but, these sound differences between "audio grade" capacitors are relatively subtle, less than the differences between audio grade capacitors and "regular" low ESR capacitors. Perhaps the perceived differences between "audio grade" capacitor types are by design and are the result of deliberate manipulation of capacitor ESR and/or ESL.
 
Last edited:
:lurk: Looking good. Keep us posted if there is actually any audio improvement. :)

Not sure if you are referring to the NAD 106 or jpb6793's NAD 1600. But I did post my listening impressions of the restored/upgraded 106 in post #9 above; in short, yes, there was a noticeable improvement in sound.
 
The stock capacitor which immediately follows the sealed ALPS volume control is a 470nF MKT (polyester) film. It would have been good to replace it with a polypropylene (or similar) type, but unfortunately one that has a 5mm lead pitch and would fit the allotted board space was not to be found. The use of a larger film capacitor and installing it off the board is not generally recommended since it may add noise to the signal (due to the long unshielded capacitor leads and the capacitor itself).



Update:
Although I have yet to locate a 0.47µF polypropylene film capacitor which has a small enough footprint and a 5mm lead spacing to directly replace the stock polyester film type at C303 and C304, there is an alternative way to upgrade to polyporopylene. On the NAD 106 main board, there is a jumper in series with each of the C303, C304 capacitors, and by swapping the position of the jumper with that of the capacitor, one can install a polypropylene film capacitor with a lead spacing of ~10mm and still have the capacitor close to the ground plane (to minimize noise/hum).



At the same time that the covers were off, I used the opportunity to increase the decoupling capacitors (C311/C312, C327/C228) from 0.1µF to 0.47µF (the replacement film capacitor is just slightly larger than the stock one). Similarly, all of the small feedback (Miller capacitors) were replaced with same value C0G types.



As I have stated earlier, this NAD 106 is a definitely a "keeper"; it is a great sounding preamplifier design and I continue to be impressed by it. Interestingly, its topology is essentially identical to that of the much lauded 1020 preamplifier, which in turn is the same as the preamp stage in the iconic 3020.
 
I have owned a NAD 106 pre amp. since 1992. The upgrade project mentioned here inspired me to do the same.
Recently i discovered looking at the diagram in the riaa section that the polarity of the capacitor C215/216 (2200 uf) is wrong. The positive lead should be attached to ground. Due to the bias current there is a small voltage drop over the resistor r207/208 about 75-100 mv meaning that on the two base in the differential coupling voltage will be negative compared to ground and it stays negative even when the amplifier is turned on or off.
The next thing i will mention is the phono stage input capacitor C205/206 (47 uf). The reason NAD has chosen such a large value is to avoid electronically resonans due to the inductance in the phono pickup.
 
Last edited:
Back
Top Bottom