Fixing dead keys on my Nord Piano 2 HA88

[csciguy8]

Hey guys, I'm a new poster on this forum but a long time Nord user.

I've owned my Nord Piano 2 a couple years, recently noticed 5 keys on the lower register no longer producing any sounds. It happened recently after a moving company moved it, although I did have it shipped in the original packaging. Mechanically the keys felt fine, so I assumed it was just dirt / debris in the key action.

After taking everything apart and cleaning, I was right! I THINK it was really just excess grease that had made its way to some of the rubber contacts in the key action. After cleaning the contacts, removing / putting back the circuit board, everything seems to work perfectly, yay! It took me about 3-4 hours across 2 days, and I went really slow. I'm pretty comfortable with taking apart electronics, so if you aren't, this whole procedure may freak you out. But for me, it was worth it to save the money and time versus sending it to a repair shop.

This forum helped a lot, and I used these threads as reference...
http://www.norduserforum.com/nord-stage ... tml#p22246
http://www.norduserforum.com/nord-stage ... t7446.html

But since there isn't any thread about opening up the Nord Piano 2 specifically, I documented everything I did, and with pictures!

(here we go)

1) From the bottom of the keyboard, unscrew the 4 screws that connect to the keybed, and the 2 screws that attach to the top panel. I posed the keyboard upright for the pic, but you can do this while the keyboard is on a keyboard stand most likely.



2) From the back, remove the 7 screws securing the top panel to the bottom casing



3) From the wooden side panels, remove 4 of the 5 screws, and just loosen the far bottom one. This will act as a hinge when we open the top panel. Once the top panel data connector is disconnected, then we can remove this screw.



4) Facing the front of the piano, grab the wooden sides, and lift up, hinging on that last screw you didn’t remove completely.



5) Next we will start disconnect cables, but I recommend you find some good plastic prying tools. These are cheap, and won’t damage the circuit board if you accidentally slip or press too hard.



6) Disconnect the top panel data connector. You should be able to do this with your fingers by grabbing the side tabs and moving them away from each other. Once disconnected, you can return to those wooden panel last screws, disconnect them, and lift the top panel up and away.

You will also need to disconnect the 2 orange connectors + 1 white connector on the bottom circuit board. We need these disconnected to remove the keybed. Use your plastic pry tools and go slowly. Don’t yank anything or apply any significant amount of force here.



7) The top panel removed. Now is a good time to blow some canned air or wipe down the keys.



8) The bottom circuit board connectors disconnected. Make note of which connectors go where.



9) The connectors set to the side



10) Let’s get to removing the keybed now. Remove all the screws securing the top part of the keybed chassis to the outer casing.



11) Once the keybed screws are removed, we are ready to lift the keybed straight up and out of the casing. Two people help here, but I was able to do it by myself, just go slowly and be careful.

I found the best place to grab the keybed was on the sides, on the metal frame. It’s probably not a good idea to grab it by the green circuit board, it’s pretty heavy.



12) The case with the keybed removed..



13) Lift up and place on a table with some kind of padding so you don’t scratch your keys. In this picture, I’ve also flipped the keybed so the keys are facing down.



14) Remove the 3 data cables, then remove all metal struts. There are 9 struts, each attached with 3 screws.



15) Remove all screws that attach the circuit board to the frame. Keep a note of which screws go into which holes (take pics like I did).



16) The circuit board is now free to be removed. There are a lot of pieces and screws loose here. Lift straight up to remove the circuit board. It's actually made up of two pieces connected by a ribbon cable. I didn't bother disconnecting the ribbon, I just carefully lifted both pieces at the same time.



17) The circuit board removed and flipped over to reveal the rubber contacts for each key. For those that don't know, each key makes contact with the circuit board twice on a single press, each time completing a circuit and registering a signal. The difference in time between these signals is how velocity is calculated.



18) Now here's where everyone's experience may vary. What is causing your dead keys? In one post I read, it was underneath the rubber contacts that was dirty and preventing an electrical connection. I checked under a key I knew was dead, and the contacts looked clean.



19) I did notice that the top of some of my rubber covers and a lot of excess grease on them. I figured this could affect the mechanical point of contact at least, so I cleaned them by wiping off the excess with my plastic pry tools.



20) I also cleaned the part of the each key that makes contact with the rubber contact, there was excess grease here too



21) Here's an optional step. I'm a total nerd and like playing with multimeters and know how to check continuity with circuits. So if you are like this too, now is the time to check your dead keys before reversing all these steps and putting everything back together.

22) Put everything back together using these instructions in reverse. Follow any pictures here and any pictures you took to help you out.

If you've taken your time and fixed any issues properly, your Nord should play beautifully like mine did! Good luck!

[csciguy8]

(oops, posted again by accident)

[Spider]

GREAT tutorial, thank you!
Yes most of the "contact cleaning" topics are about semi-weighted actions, there is not much regarding the HA action... and surely not as detailed as yours!
This will serve well all piano, piano2, stage and stage2 users.

Of course I strongly hope that I'll never need to do this, but if it happens this will be the first topic I check!
(having the heart to actually open up my beloved Stage2 is another matter...)

[mi-ji]

Who is this mystery man who joins the forum and dumps this amazing photo journal post ON THE SAME DAY?!?!?

Pretty good job, I'd say. Stage model owners will need to additionally disconnect the aftertouch cable from the mainboard and ensure that the cables are freed from the lower chassis at the extreme left end before attempting to remove the entire keybed assembly.

I disagree with leaving the two contact PCBs connected to one another. If you are careful with things, this can work, but it is an invitation to having a accident/problem, a temptation to the Murphy's Law gods. Optimal practice will be to disconnect one side or the other.

With regard to:

Now here's where everyone's experience may vary. What is causing your dead keys? In one post I read, it was underneath the rubber contacts that was dirty and preventing an electrical connection. I checked under a key I knew was dead, and the contacts looked clean.

Irregular note behavior such as intermittent response or full velocity output will ALWAYS be caused by something beneath the rubber contact strip. It cannot possible have anything to do with anything happening on the outside of the contact strip.* I have walked people through this operation over the phone more than 300 times over the past 5+ years (and done it myself quite a few times) and on average there is visible evidence of contamination about half the time -- but the repair ALWAYS works. There are actually two switches there, with the descending carbon-infused black dots coming down to bridge the gap between corresponding pairs of black ovals on the circuit board (PCB) to close the two switches. 95% of all cases involve dusty linty contamination; occasionally it's a drink spill and therefore icky sticky. Compressed air cleans the linty ones, q-tip in isopropyl or acetone cleans the icky ones.

It may be nice practice to clean excess lube here and there, but these things have nothing to do with curing the defect. Notice in the photo between 18) and 19) showing the upturned rubber contact strip that there are tiny ventilation channels molded into the rubber, leading diagonally from the chamber to the perimeter of the strip. When a note is played and the chamber is smashed down, air is expelled. When that note is released, the dome springs back into shape and air is sucked in as a result. THERE IS ALWAYS A CHANCE THAT ANY DIRT IN THE AREA WILL BE SUCKED IN WITH IT.

* Note: it is possible that a diode is blown on the contact PCB. I have encountered this once in 400 cases. If a diode is blown there will be NO intermittent aspect to the note behavior. It will never work. Thus, if there is any intermittent aspect to the note behavior, the possibility of diode failure is ruled out.

Bless,

Pablo

[csciguy8]

Ha! Thanks. I joined and posted on the same day, because I needed to join to post, but I've been lurking for a while now. I've learned a lot from this forum, so hopefully this gives back a bit.

So for this...

It may be nice practice to clean excess lube here and there, but these things have nothing to do with curing the defect. Notice in the photo between 18) and 19) showing the upturned rubber contact strip that there are tiny ventilation channels molded into the rubber, leading diagonally from the chamber to the perimeter of the strip. When a note is played and the chamber is smashed down, air is expelled. When that note is released, the dome springs back into shape and air is sucked in as a result. THERE IS ALWAYS A CHANCE THAT ANY DIRT IN THE AREA WILL BE SUCKED IN WITH IT.

I was expecting to see dirt / contamination under the rubber too, but I didn't. In fact, I didn't even blow canned air on the circuit board for fear I would just move existing dirt around. I didn't perform any kind of cleaning under the rubber contacts or on the circuit board at all. Just the wiping up of some excess grease from the areas I mentioned. I didn't even detach the rubber contacts from the circuit board, other than the photos I posted.

So yeah, it's a mystery to me why all 5 of my dead keys (across the bottom 3 octaves) worked after I was done. I assumed it was the grease cleanup, or at least just the physical removing / putting back of the circuit board? Maybe dirt under the rubber contacts fell out during this process?

Everything plays great now, but I hope the problem notes don't come back.

[pablomastodon]

5 dead notes? randomly spread about the place? what was the interval between them? learning this detail puts a whole different spin on this particular case...

[csciguy8]

My five dead notes were Bb0, Gb1, D2, Bb2, and Gb3.

(using the scheme where white keys from left to right are: A0, B0, C1, D1, E1, ..., C2, D2, ...., C3, D3, ... etc)

[pablomastodon]

those failing notes occurred in intervals of 8 semitones. This is a fairly clear indication that the problem never was dirt at all, but something in a data line (or the data line's path) from the keys to/on the mainboard. Is all now good? Did this behavior occur immediately out of the box, or was it working normally for a time and then suddenly went bad?

There is a possibility that there may have been an irregularity in the ribbon cable connection for the lower contact PCB and that unseating and reseating it has corrected this. There is also a possibility that this problem will revisit. Perhaps an intermittent open in one of the wires on the ribbon cable. This can happen sometimes if a particular IC on the mainboard fails, but in that case the problem is NOT likely to simply disappear just as suddenly, so this possibility can perhaps be disregarded.

WIthout going too deeply into the science behind it all (perhaps I'll dig out a long-winded description of this science in my email archives and post it), all (or most all) modern keyboards are designed around an 8x8 matrix. This fact results in failure modes which involve either 8 (or multiples of 8) consecutive notes going out, or notes failing in intervals of 8 semitones, as in your case.

Play the instrument and keep an eye (ear) on it and cross your fingers that it does not revisit you. If it does come back, please contact me at my regular work contacts shown below.

Bless,

Pablo
pablo.mastodon@americanmusicandsound.com
800-994-4984 x3030

edit:

Philosophically speaking, there is little difference between those key contact switches and the wall switch on your kitchen light. An ordinary switch needs two wires. When ganging multiple switches, a common rail can be used, but with 88 notes and two switches per note, that would require a cable with 177 independent wires (2x88+1). Building keyboards in this manner would be stupidly expensive and bulky/cumbersome, so some clever engineer types figured out a better way years ago and essentially all keyboard instruments these days use the same basic design principal: they're done on 8x8 matrices with diodes. For this discussion we can ignore the deeper aspects of that theory, but the 8x8 matrix design yields some very specific failure modes when failures occur.

[csciguy8]

Did this behavior occur immediately out of the box, or was it working normally for a time and then suddenly went bad? There is a possibility that there may have been an irregularity in the ribbon cable connection for the lower contact PCB and that unseating and reseating it has corrected this.

It actually happened after I moved from my apartment. I packed up the Nord in its original packaging and put it on a big moving van. It was a same city move, but I'm sure the ride was bumpy, and it could have been jostled by the movers during loading / unloading. So yeah, I'll go along with the unseated ribbon cable theory, it certainly sounds more probable than 5 notes all suddenly getting dirt under them. Especially since my fix didn't actually involve cleaning any dirt.

Very insightful, thanks!

(And btw, I'm completely nerdy, and would probably take the time to read a description of the 8x8 matrix design I totally noticed the circuit board key contacts sharing data lines across different keys and wondered how that all worked out in the end.)

[pablomastodon]

My knowledge level is sufficient to study the diode matrix arrangement and understand how it works, but not sufficiently sturdy for me to feel comfortable in trying to teach it to someone else. Do a websearch on "diode matrix keyboard" and you will get LOTS of results. Note also that half (or more) of these results will involve computer QWERTY keyboards, but the general concepts will be the same except that computer keyboards do not expect more than one key to be pressed at a time. This is cool for monophonic instruments, but some folks do like to play chords from time to time.

Bless,

Pablo