Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards Old CVP105 Keys Sound problem

Hi, thank you everyone for such a wonderful forum. I've been reading for a while, and gained a lot of information about digital piano. This is my first post.

A few days ago I acquired a CVP-105. I know it's old, but I'm a totally beginner and this is quite cheap and I like the touch and sound. However, three black keys around D of the third octave have sound problem. Some time they have sound, some time the sound is low and some time there is no sound, especially after repeat hitting. After search on the internet, I suspect it is not the key itself but the things beneath the key, because the key’s action is normal. What’s beneath the keys? Spring? Sensor? Ruber? Is it hard to fix by myself? After search the internet, I’m thinking open up and do it myself. But want to cautiously ask the experts here first.

Please help!
Thanks

I repaired my CVP 96 some time ago.
Beneath the keys is a kontakt strip or kontakt bubbles.
These sit on a PCB that has printed carbon kontakts.
These carbon contacts lost connection with the underlying PCB traces.
It was not visible, but measurable.
Here is how I repaired it:




If you disassemble the keyboard then be sure to mark the parts, especially the rubber strips, because it is possible to reassemble them in the wrong orientation and then it would not work.
Best is, when you get the service manual somewhere, because there are some important precautions when the keyboard is taken out and the PCB removed it must be prevented that all keys fall out...

Peter

Thank you hpeterh. Thank you for the quick response.

Sorry I'm not tech savvy. Let me understand how did you do: you use the conductive pen glue back the carbon kontakt (contact?) to the PCB track. Is that right?

Thanks again!

Hello,

The contacts where not mechanically loose. Normally the resistance between the carbon and the copper is < 100 Ohms. It was some kOhms or more in some cases. For security I repaired them all. I scratched the traces blank and painted new connection with special silver color.
As said, these where not loose, possibly oxidation or something like that caused the problem.

It should be mentioned that I purchased this piano in a used and defective condition. A Yamaha technician tried to repair it and replaced the rubbers without success. Then they gave up, too expensive and sold it on ebay...

Another possibility for your keyboard could be broken soldering points. That happens easily with these one sided PCB's. Had to resolder the speaker connections on my Kawai Piano some days ago ....

If you are without tools and without experience (theory alone does not help) then you run into problems if the causes are others for you keyboard.

On the other side, if only some particular keys are affected then it is very sure that the fault is localized on the keyboard PCB. Normally it is trivial broken connections and else. Other stuff like digital semiconductors usually dont fail after such a long time.

Best,

Peter

Thanks Peter!

Even though I cannot compltely understand the tech talk, I got a good sense of it. I'll first get a service manual and open it up. and go from there.

Thanks again for the great help! I'll let you know the outcome.

Hi, Peter,

I think I’m really slow at learning the electronic stuff. After look carefully again at your picture and your post, I notice that in the top picture, at left side the traces is slim line (which is the original trace and you scratched out),while on the right side and also in the lower picture (the finish product?), the line became a silver spot. So the silver color stuff is the new connection? Is it the paint from the conductive pen shown on your lower picture?

Thanks

Exactly yes to all questions.

A little bit more explanation: All contacts are double contacts.
Each key has a pair of double contacts that means each key has 8 contact pads. that means 88 x 8 Pads = a lot of work to do.

Also there is a risk to damage the traces or pads or to make shorts.
This color doesnt flow nicely, this is why it looks so ugly. I had to scratch away some shorts I produced and had to test and unmount the keyboard twice until I discovered the last short....

So I would urge you, before you try this measure if this really is the problem. Possibly it is enough if you clean the contact board and the conductive rubber contacts, using alcohol.

wish you success,

Peter

Thank you very much!

Now I'm like a child (a 40 years old one) with a new toy. I'm not only going to play it, I'm also going to open it up to fix it. Isn't it exciting!

I cound not thank you enough!

Hi, Peter,

I opened the CVP105 on weekend, armed with the service manual and some video from YouTube. The whole procedure was much easier than I thought. I directly open up the circuit board (the PCB board shown on your picture). Sure enough, there is some dust or gray color fuss around the carbon contact and the traces. I think it is probably the oxidation. I used a cotton Q-tip cleaned it up. Put everything back, vola! Everything work now!
However, the new day, my wife notice another black key some time not work again. And I notice one white key some time low sound and some time high pitch sound. So I open up again last night, used the cotton Q-tip cleaned all the PCB board. Put them back. Everything seems to work again. But then I found a black key no sound. This key worked perfectly before. Since I have no time left, I didn’t open up and clean.
Now I think shall replace the whole circuit board (the PCB board)?

Can the carbon get loose or fell off the PCB board?

When the contact between the carbon and the copper became loose it will give low sound or no sound, right? How about the shorts you produced. Will it produce the loud sound?

Also how about the rubber contact? Could that be a problem? I cleaned a little the rubber to, it seems fine to me. But I’m not sure. It has two vein-like lines cross from left to right.

Thanks for the help!

One more thing. before you re-paint the traces, do you have to scratch the original traces. can you just paint over it? and will the new silver paint's resistance lower than the original trace and cause the sound volume different?

Thanks

Hi,

I think the rubber is ok when it is clean and looks ok.
This is solid conductive material.
I also first cleaned and that improved the problems but did not cure all problems.

I dont know if a replacement PCB is available from Yamaha....

The effects depend from the contact that is affected.
It can happen that a key gives no sound or the key could produce 100% loud sound always or inbetween.

A bad carbon contact cannot change the pitch. If the sound has wrong pitch then there must be a shortage or interruption somewhere else on the board.

If possible measure the resistance of the carbon-copper connection.
I dont know the value, but all connections must have the same resistance. This is not the case when they are defective.

But be sure to make a factory reset first to reset all custom tunings or transposition and switch off the autoaccompaniement.

There are also builtin service test routines. I dont know if these are the same as with my CVP 96 but I believe so. Look for this in the servicemanual..

Peter

Thanks Peter.

I found the test program on the service manual. but where is the factory reset? I can't find it on the service manual and the user's manual.

I have a meter, I'll check the resistance between carbon and copper. I guess I shall put one niddle on the carbon and the other on the copper line a little bit away from the carbon?

About the question of scratch the trace before paint it. Can I paint over the trace without scracth it first? or I'll mess up the connection?

Yamaha do have the circuit board, but only the right part, for $92, When I input the part no. for the left board, it said it's not orderable. of course I put the part no. for the cvp96, but probaly is the same. I'll call yamaha to check.
thanks again

Hello

Place the needle /flat/ on the carbon. This is hard as glass, dont break it! Both parts should be clean, no fingerprints. Place the other needle on any soldering point that is connected to the copper trace. So far I remember the resistance must be some 100 Ohm, but was > 1kOhm in many cases.
The resistance of the copper trace can be neglected it is virtually zero if the trace is not broken.

To be sure you might measure the carbon resistance alone for comparison. Place the neddles on both ends of a carbon stripe.

If you dont scratch the trace then you dont get contact. Usually there is green or red color. This must be removed.

With the PCB, be careful. There are different revisions for the keyboard. The PCB must fit to the rubber contacts. Some have bubble contacts, some have stripe contacts. I know this because I have seen it but I dont know partnumbers.
Best is, ask a Yamaha technichian. He will probably ask for the keyboard model or serial no. Look if there is a sticker somewhere or if the PCB has a number.

Also look here www.cvpug.com . This is the CVP user group. The members of this forum are very experienced with CVPs and where to get stuff. Many know more than I do ;-)

Good luck,

Peter

Many thanks Peter!

Now I'm ready to venture again! By the way I found the factory reset on yamaha's site. it's the rightmost key (hold down while turn on), same as the cvp96.

Thanks again! I'll report back.

I think the volume is controlled by the time between when one set of contacts is shorted to the time the second set are shorted. The piano deduced kay velocity be the time between two points.
The contacts themselves are digital, they are either open or closed so variation in resistance should not mater

I've not done this repair but I've worked on PCBs. That conductive silver pencil is a great invention. I've used solder and a jeweller's loupe. You should be able to paint over any metallic traces but I think you are best off polishing with #800 wet or dry sand paper wrapped over a pencil eraser.

Fixing a PCB is very detailed work. I'd use a good magnifier and test every connection more than once. Slightly flex the board and test again.

Alcohol is a decent cleaner but there are much better products with brand name "Deoxit" that will actually remove oxidation from contacts. I'd use that and an old tooth brush. You will likely need to vacuum and clean the entire inside of the piano it not dust and gunk will fall onto you newly clean PCB.

You can buy bottles of compressed air at the same place as the deoxit contact cleaner. I use the air and a small 1" paint brush for dusting parts like this.

@Chris:
That all is true but overkill in this case.
The copper is covered with protection color and if this is removed, then it is blank and fresh and without oxidation.

The carbon is chemically inert and doesnt oxidize. It is hard as glass and any scratching could damage it. It must be clean, thats all.
The tools shown in the image was all that I used. Experiments with other tools where less successful.

This is: a scalpel (sharp knife) a glassfiber eraser pencil and a conductive silver pen. And spirit. (Wet cleaning doesnt develop electrostatic)
Oh yes, and a strong magnifier google headset.
So scratch the color away and clean everything and use this conductive silver color. Soldering in this case is impossible because the carbon cant be soldered.

BTW, be very careful with vacuum cleaners and dust brushes.
These develop a considerable amout of electrostatics and there are reports on CVPUG.COM where people have damaged their piano with a vacuum cleaner. It can be used, but should always keep a security distance of 10cm from electronic parts.

Best,

Peter

Peter & Chris,

Thank you all for the advice, last night I opened it again, with good result.

Let me tell you first how strange the sound became before I fix it. One black key have no sound (it was good before), but when I hold down the black key next to it (either side will do) and push it down, it's normal. This is just getting stranger every day.

Anyway, Now I know what my problem is: I did not put the rod between the hammer and the keys bed!

The last two times I took the whole key board off, I didn’t put the rod in, so when I put the key board down on the floor to work on the PCB, it's upside down. The keys then were all in the press down positions. So when I put the rubber strip and the PCBs back, I have to forcefully push them down and screw them. Somehow I think because the force, the rubber strip was not perfectly in line or in contact with the PCB, which then causes the delicate system behave strangely.

Now after put it back carefully and perfectly as I can. Everything is normal.

Thank you Peter for your great help, I learn a lot during the process.

By the way, I did try to measure the resistance, but my meter is a cheap digital one and my hands and sight are not very good. But I did manage to measure the few that give me problem. Their resistances are between 100 ohm and 300 ohm. Again because my hands was a little shaky and the thing is so tiny. So I think they are in the acceptable range.

Thanks again!

Glad when it works.
The measurement values are ok and normal and typical.
The variation is because it is not easy to find reliable contact.

No expensive measuring instrument needed, that would not deliver other results in this case ;-)

Peter

"Protection color"? Do you mean "solder mask"?
See here for definition: Solder Mask

Mostly rather scape off the mask I'll run a wire to the nearest l=place the trace is exposed, maybe a via or a pad for a component pin

I do keep a bottle of alcohol on the work bench but recently I decided to try and see if the $15 per bottle stuff they sell in the electronics stores really is better than the $2 stuff I buy at the drug store. I'd say yes not not 7.5 times better. It is designed to clean "pots" which have a metal wiper that runs over a carbon track. I think if there is any oxidation it would be on the metal wiper.

I have to take back what I said about resistance not effecting the sound because the switch is digital. It might. It takes some finite time for a voltage to drop to zero and the time depends on the resistance to ground. I'm sure the controller does not consider the switch closed until after the voltage is reached some threshold and resistance determines the rate of change in voltage after the switch closes. This effect may be trivial or not depending on the details of the design.

Sorry, the german word is "Lötstoppmaske". Did not know the correct english word.
But I know what I know and hope with some goodwill it is understandable.

So, a little bit detail about resistance: Usually the pullups used are 10kOhm. If TTL levels are used then < 0.8V is "Low" and > 2.0V is "High".

So, if the switch resistance is below 1kOhm then "Low" = 5V/11 and that is about 0.45V. However that are normative values, practically the TTL transition is at 1.0V.

So all switch-resistance values below 1kOhm (if stable) are absolutely secure with a large security factor included.

And a 300 Ohm value is super-secure ;-)

But probably that will only confuse the OP. I can explain it even more acurately and also for CMOS levels but that is beyond the topic of this forum.

Peter