NAD 1020 preamp upgrade to 3020 integrated

leesonic

Captured on security camera...
Subscriber
Ever since I worked on my first NAD 1020 preamp, I have always wondered if it was possible to build a 3020 integrated out of one? From initial appearances, the circuit board in the 1020 looks to be the same as the 3020, just without any of the power amp components fitted. But as I found out, not all 1020 preamps are suitable candidates for this modification. The 1020B model has all the holes drilled in the PCB, but turn it over and you'll see there are no copper tracks under the power amp section. The 1020A is the same. None of this applies to the "i" and "e" models, these are significantly different to the earlier ones.

So why modify a 1020 to make a 3020, why not just buy a 3020? Well, unless you can find a bargain at a Goodwill, Thrift Store or yard sale, most 3020 amps sell for close to $200, or more. For that money, you'll still be getting 35 year old electrolytic caps that weren't any fancy name brand to begin with. Contrast this to 1020 preamps, which at the time of writing, you could still find for less than $100, non-working units sometimes for less than $50. Since you'll be replacing a bunch of caps anyway, why not start afresh? I'm not saying this is the most economically viable way of getting into vintage NAD amplification, but if you like a challenge and you're itching to get soldering on your next project, this thread might be for you.

So for this recipe, you will need :

1) A 1020 preamp.
2) Components to build the power amp section.
3) A heatsink.
4) A power transformer.
5) Some other small items from parts units.

The last three items are the hard to find ones. It is possible to drill your own heatsink, I did it myself once before using nothing but a piece of angle aluminum, "clamped" in the grooves of a picnic table and drilled with a hand drill. The power transformer on the other hand is uniquely NAD. It features twin secondaries, one for the preamp stage (and parts of the power amp), and the other to feed the final couple of stages of the power amp. You could in theory just use a single secondary transformer, but you might lose some of the benefits of the dual secondary design.

But help is on hand in the form of the NAD 7020 receiver, which has the 3020 amp and 7020 tuner combined into one chassis. Like the 1020 preamp, parts 7020 receivers can be purchased for around $50. You might ask the next question, why not just rebuild the 7020 in the first place, you get a free tuner in with the deal? Why not indeed! Nothing is stopping you from rebuilding either the 1020 preamp as purely a preamp, or restoring the 7020 receiver you bought for parts. This thread is about using a parts 1020 and 7020, and using them to make a 3020.

Here goes...
 
Last edited:
Board preparation.

OK, so you have your 1020 on the workbench. The first step is to remove the board from the case so it looks like the picture below. You can see, I started with a real "parts" 1020, the board had already had the balance pot and muting circuit FETs (both circled in yellow) removed. The board at the bottom left is for the headphone amp circuit, and the smaller board at the top right is the switch to select the capacitance loading for the phono stage.

1020into3020-01.jpg

One nice thing about the NAD circuit board is that it's very well labelled, NAD might have cut corners elsewhere, but at least the board is nicely silk screened. Fair enough, it might not be that fancy green fiberglass board that other manufacturers use, but for this purpose, it's good enough. In using the 3020 board to make a 1020 pre, NAD had to get power and signal feeds to and fro across the board using the existing copper tracks. This means there are wire links where we do not want them. The wire links are all labelled "J" for jumper presumably. Anywhere you see a wire link that is not going across a "J" marking on the board, this needs to come out. I have circled the ones that need removing on the picture below. Any wires that go to the headphone amp board, cut these off at the main board as well, and move this board out of the way.

What you also might notice is the cracking around the mounting screw hole in the top center of the board. 99% of the 1020 and 3020 boards I've worked on have cracks here. Luckily for us, the cracks rarely go through and take out the copper tracks, and some of the tracks here are for the soft clipping circuitry which we won't be using anyway.

1020into3020-02.jpg

Here is what your board should look like with all the unwanted wire links removed. Notice the holes marked C, D, E, E1, F, F1 and P5 on the board, we will need these later.

1020into3020-03.jpg

Because the board was probably assembled using a flow-solder process, all the holes we need to use have been covered over with solder. You could either open these up by desoldering them, and on ones with a thick layer of solder that would be best. Alternatively, you could heat the solder pad up, and use a sharp pencil point to make yourself a new hole. You'll need to open all of the holes in the power amp section, but I would recommend doing it as you go along, rather than do them all at once.

1020into3020-04.jpg

Now, go ahead and insert the jumpers into position where marked on the board. I used some of the wire from the headphone amp board, making sure to sand each little piece down with some 400 grit sandpaper, so as to give some nice clean metal for the solder flux to do it's work. You should have 38 jumpers in the area shown in the picture below, some were there already, most are new.

1020into3020-05.jpg
 
Last edited:
Adding components.

Now we can start inserting the resistors. I didn't use anything "boutique" or fancy, I just searched Mouser for 1% metal film resistors, and came up with these ones by TE Connectivity, which are actually 3/5W. Curiously enough, what would seem a common value of 1k you can only buy in batches of 4000, so for this value I used some from Xicon, hence the slightly larger case you can see on the second resistors in from the sides on the bottom row.

1020into3020-06.jpg

NAD has some resistors that are mounted up on little ceramic legs to either keep the heat off the board, or to stop the board from getting burnt if they go pop (I've seen the same thing on Pioneer amps). The orange ones are Vishay/Dale 1% 2w.

1020into3020-07.jpg

Transistors in next...

1020into3020-08.jpg

Followed by the caps. The little blue ones are TDK multi-layer ceramic in place of the original disc type ceramic, and they're a LOT smaller. For the ones that were originally mylar film, I splashed out (not that they were that expensive) on some WIMA films, either MKS2 or FK2. There was one value I couldn't get, I had to use a Panasonic ECQ.

1020into3020-09.jpg

Jumping ahead a bit here, not sure how that happened, maybe I got caught up in a soldering frenzy and forgot to take pictures. OK, so I cheated a bit because I happened to have a spare 3020 heatsink, here are the new output transistors mounted to it. The four driver transistors just below it are BD237 and BD238, which are replacements for the original 2N6553 and 2N6556. They have a different pin-out than the originals, but it helps because there is an extra center hole that is drilled offset from the rest, meaning you don't have to bend the legs like crazy. The 2N6553 and 2N6556 are weird package that have an extra long metal tab on them, I'll be keeping an eye on how hot these drivers run, and adding some heatsinks if necessary. The good thing about these new transistors is the all plastic package, meaning you don't need to insulate them from the heatsink.

Further down the board, you will notice the blue Bournes trimmers. I used multi-turn for the bias adjust simply because there is very little room here, I used plain old single turn ones for the offset adjust. All the electrolytics have been fitted, Nichicon KW series except for ones in the signal path where I used KL low leakage. I tried removing as many electrolytics as I could, the four WIMA caps along the very bottom of the picture were electrolytics in the original design. But sometimes, the values are so high that film caps would have to be huge and/or expensive in order to use them.

The red wire is linking the two F points (F and F1), there is another wire linking the two E points (E and E1) which connect the speaker ground terminals to the power supply ground. Points C and D connect the input of the bridge rectifier to the two fuseholders at the far edge of the board (out of the frame unfortunately). The two inductors on the left of the picture came from a parts 7020, but you could wing your own. The row of empty spaces you can see at the bottom of the board are for the soft clipping circuit, I decided not to use that.

1020into3020-10.jpg

Sorry, but I couldn't resist attacking the power supply while I was there. All new caps, resistors, bridge rectifier and transistors fitted. There is nothing special about the two gold color caps, they are FW series as Mouser didn't have any KW series in stock at the time of ordering.

1020into3020-11.jpg

But we're not finished yet, this area of the board needs some attention. The area within the yellow rectangle has some missing links, those blue caps need to go, as do the resistors. There are also two fuseholders missing which can be sourced from a parts 7020. The area within the red rectangle is for the muting circuit, this actually mutes the output of the preamp, and has nothing to do with the power amp stage. I shall leave this as is for now, and see what happens. The two white wire links have to go though.

1020into3020-12.jpg

This is how that area should look. R657 and R658 are the resistors for the headphone output, and C633-636 and R655/656 filter off the high frequencies.

1020into3020-13.jpg

By this time, the board was getting awkward and heavy to keep turning over, so I put it back in the case. There are a few more jumpers just behind the source selector switches that need to be put in, otherwise the preamp won't put out any sound.

1020into3020-14.jpg

Here they are in place.

1020into3020-15.jpg
 
Last edited:
All together.

Here is a picture of it somewhat back together. The wires you can see dangling out the front are the test points for the bias adjustment. I bring them out with some scrap wire, rather than try and probe the board and adjust the bias pot at the same time.

1020into3020-16.jpg

Here is the back view. I was lucky enough to have a 3020 back panel with transformer attached, so all I had to do was make a hole for the phono cartridge loading switch. If I had used a 1020 panel, I would have had to drill new holes to mount the transformer, and also cut out holes for the speaker terminals, and enlarge the hole for the pre out/power in jacks.

1020into3020-17.jpg

After setting the bias and offsets, I hooked it up to some old Goodmans car speakers, and let it play. These are 4 ohms, so they helped get some heat into the new outputs. Obviously it's not 100% finished, I still need to recap the phono and line driver stages, but IT WORKS!

Lee.

1020into3020-18.jpg
 
Last edited:
Sorry I don't have any detailed expertise or comments to offer, but WOW - amazing what you did there. Wouldn't have thought it was possible.
 
That looks great Lee , lots of work but like I have said in the past you have a lot of patiance . I am impressed with the 1020 that we did our deal on , it sounds nice and will last a long time since you did a recap on it . I had never thought about NAD until Lee turned me on to this pre and must say if you are looking for affordable gear you should try NAD . The 1020 pre is a clean sounding unit and when Lee is finished with his project I bet it sounds exceptional .
 
Cool project! I was actually thinking of an opposite one, I have a 3220PE with blown outputs and some other issues i might transform into a preamp.
 
I have to say that is an amazing (re) build, done simply because you wanted to do it, essentially to prove a point to yourself. Great attention to detail and Well done!
 
Love it! I dig "can I do that?" projects. My boombox that I use almost daily was a similar endeavor. Great mad scientist job, Lee!
 
I did the phono preamp and line driver stages this afternoon...

I have it so the first cap the signal sees from the phono input is a nice, juicy WIMA film. Oh yeah... Can you feel it? Do you like it like that? :banana:

The output cap for the phono pre is also a WIMA film, I managed to squeeze in a 10uF one by moving a resistor to the underside of the board.

For the line driver stages, the values of the electrolytics were too high to use a film, it would be huge, so I stuck with Nichicon KL low leakage ones. Keen eyed observers will notice the green mylar films and disc ceramics still in these sections. I suppose I could swap them out to WIMA films and TDK ceramics like I used for the power amp stage.

I'm wondering how much gain the line driver stage has, if any? Looking at the circuit diagram, I wonder what would happen if I connect the wiper of the volume control to the output jacks of the preamp? Would this give the same effect as the "CD direct" button on some amps? Would the phono pre have enough power to drive the power amp directly?

The picture below shows the before (left) and after (right) shots.

Lee.

1020into3020-19.jpg
 
Last edited:
I am indeed feeling the WIMA magic :)

As we like to say around these parts, wicked awsome project!!!
 
Since the 1020 has it's own headphone amplifier board which we no longer need anymore, you'll need to use the pads on the PCB to connect your headphone jack. The picture below shows these pads and what they do. It should be fairly self explanatory, "left input" means the feed coming from the output transistors, "left speaker" means the track going to the speaker terminals, "left headphone" means the output that goes via the 220 ohm resistors.

1020into3020-20.jpg

There is no reason why you could not desolder the headphone jack from the 1020 headphone amp board and use that. If, like me, you are stuck without a headphone jack to use, help is on hand in the form of Mouser part number 161-0087-E. The picture below shows the output part of the 3020 schematic, with the numbers labelled which correspond to the 161-0087-E jack.

1020into3020-21.jpg

And here it is installed. There is clearance between the headphone jack and the metal front panel, but to be on the safe side I put a thin rubber washer here. It looks better with a "proper" headphone jack with a metal insert, rather than the all plastic one NAD uses, don't you think?

1020into3020-22.jpg
 
Last edited:
This is what I was talking about when I said certain 1020 preamps are not suitable for upgrading into a 3020. The picture below shows the copper side of the board from a 1020A preamp. Notice how the holes are drilled in the center of the board for the power amp components, but there are no copper tracks there. Instead there are the half a dozen or so tracks that take power and signals from one side of the board to the other. Interesting that they chose to run the power tracks from the fuse holders (center right of the board) to the bridge rectifier, rather than connect the two with wires on the other side of the board like the 3020 amp.

1020into3020-23.jpg
 
Last edited:
As the proud owner of this FrankenNAD 1020/3020, I can say that it is amazing. I have it hooked up to my ADS L810s in the living room and it makes a glorious sound. Wish I could leave the cover off so I can admire Lee's beautiful work, but the kids would destroy it. (I do allow them to press the buttons and enjoy the warm glow of that 5mm LED.)
 
Thank you for your kind compliments, glad you are pleased with it.

Maybe I should build some more? :naughty:

Lee.
 
Am I interpreting this to show that the headphone amp section of the 1020 is different than the 3020?

I have heard in the past that the 3020 is great for headphones....and then also heard that the 1020 was just a 3020 w/out the speaker amplifier....which of course I extrapolated to mean that the 1020 would also be great for headphones! But what I'm reading/seeing in this thread and admittedly barely understanding seems to indicated that was all a false presumption.
 
Am I interpreting this to show that the headphone amp section of the 1020 is different than the 3020?

I have heard in the past that the 3020 is great for headphones....and then also heard that the 1020 was just a 3020 w/out the speaker amplifier....which of course I extrapolated to mean that the 1020 would also be great for headphones! But what I'm reading/seeing in this thread and admittedly barely understanding seems to indicated that was all a false presumption.

Yes. The "headphone amp" of the 3020 is nothing more than two 220 ohm resistors in series with the speaker outputs. That means all the musicality of the 3020 will go into your headphones. The headphone amp of the 1020 is a different design, using a quasi-complimentary pair of low power BC556 transistors, which also means there is a 100uF capacitor on the output. How different do they sound? Only some extended listening can tell you that.

Hope this helps,

Lee.
 
Back
Top Bottom