Casio Electronic piano AP-24 ''loud keys"

[MARSHALL BUSTER]

Hey Chaps, how's it going??

Its been a while since iv posted a thread so I hope your all keeping well

Ok someone passed on a Casio piano to me today and to be honest its not my sort of thing and I cant even play one tune on it haha,, but I liked it so I thought what the hell I'd like to see what's its like inside one of these anyhow so here I now have a piano!!

Ok to start with, its a Casio AP-24 digital piano and it all works fine apart from 8 of the keys sound alot louder than the rest of the keys. I notice if I turn off the touch select then the problem goes away and all keys play the same, so its as if the soft touch for them 8 keys is not working regardless how soft I press the keys they sound loud like i'm hitting them hard.

Iv had the usual look inside and it all looks to be as it should.

Maybe this is a common sort of problem for pianos?? I really wouldn't know. Any ideas would be greatly appreciated

[Jazz P Bass]

I have no idea how the Casio gets the 'pressure' reading.
I have seen on other synths the use of two 'switches' under the key.
Actually they are contacts.
When a key is pressed the first contact closes & then the second one does.
The amount of time between the two closures is the 'pressure'
Anyway, if the first contact is not working, all you get is a loud note.
Which 8 keys?
Are they all together (something got spilled into the keys?)

[MARSHALL BUSTER]

Hi Jazz thanks for the reply.
"Two contacts" right I see!
Well the guilty keys are all together they start at 34-38 on the white and the three black keys above them. No spillage that I can see.

[Jazz P Bass]

The keys are set up in a matrix.
A failed board after the keyboard does not make sense. (because of the matrix)
I still suspect the actual contact pad/ circuit board underneath the keys.
The small black 'carbon dots' under the contact pad should not be touched at all.
If they are contaminated or worn out then replace them.
The circuit board contact area can be cleaned without damaging anything.
Here is a SM for sale: Casio AP-25/28 Service Manual

[Mick Bailey]

It's certainly common on older Roland pianos that get a lot of use. Jazz P is right, though it's velocity and not pressure sensitivity. Commonly the centre of the keyboard gets the problem. Whilst a spillage will cause this type of problem, it would be unusual to only affect the velocity and not the pitch.

If the Casio is similar to the Roland it may use a silicone moulding with two carbon pads moulded in to form each contact pair. The PCB is similarly coated with 4 contacts arranged in two pairs for each key. These are then bridged by the moulded contacts. One moulded contact is slightly higher than the second so will always make contact with the PCB contact first.

The usual problem is wear to the edge of the carbon pads on the silicone moulding - they get rounded off initially, then fail to make proper contact. When they lose contact altogether the note will not sound. The difference between the first pair contacting and the second pair is (usually) converted to a midi value, but this can get screwed up when the first contact gets worn more than the second and many keyboards then output the maximum midi value (127)

When I pull the strips off they're nearly always showing excessive wear on the upper contact. It's difficult for liquids or dust to get in them by accident.

If you have this system on your keyboard you'll need to remove the contact strip PCB and remove a strip - usually 12 pairs of contacts in one moulding. The contact pads need to be crisp looking disks with no chamfering or rounding. Many that I see are showing the base silicone through the edges of the little pads.

I never clean and replace them. I've found this to be a waste of time and always replace the entire contact set for the keyboard. The PCB pads are usually fine and I use 100% isopropyl alcohol to clean them.

You need to be meticulous when reassembling. An eyelash or greasy fingerprint can undo all your work.

An interesting thing with Roland pianos is the newer turquoise strips have thicker contact pads than the old grey ones.

Then again, the Casio may be totally different system......

[Jazz P Bass]

Casio AP Service Manual

Here is the SM .
There are some pointers on page 5 about the key pad.

[Bob W.]

If it is the key contact pads you can clean them with alcohol, or if really stubborn I have used lacquer thinner, seems to soften the contacts too. But most likely because the 8 are together the feed line to the start pads, or first contact, is open somewhere between the pads and the plug to the cpu board. Quite often the start pads are hooked together to cover each octave of the board. Bob

[Jazz P Bass]

Casio Matrix

You may be on to something there.
Here is the matrix layout & 8 keys are indeed tied together in the left column.

[Enzo]

Each key has two contacts. When you press down a key, one closes before the other. The computer tells how fast the key was moving (ie how hard you played it) by the time it takes between the two contacts closing. The longer the time, the softer it plays. Now what if the second contact is already closed? Then as soon as the first one closes, the computer sees the other one closed as well, and assume maximum velocity, and so maximum loudness.

The opposite is when one of the contacts is dirty and cannot close at all. If it is the second contact, the system times out and defaults to the lowest velocity. If it is the first contact, sometimes you get no sound, other times it too defaults to minimum. I don't think you have dirty contacts.

I don't see which keys it is, you mention 38 or something, but the schematic doesn't call them by number. Note the matrix chart. Each key has a common - KC for key common. Then FI and SI, for first input and second input. Figure out which keys are stuck loud, and see what they have in common. I'd wager you will find one of those lines, probably the SI is stuck. if you don't know what it ought to look like, just look at the same signal on a working key.

Now look at first page of schematic, lower right. That is where all the KC, FI and SI leave the board for the keybed. Make sure all the ribbons are intact. But those signals all come from the large IC lower left, LSI1, the key assigner. It could have a dead I/O pin, but I'd be more suspicious of the R-packs just below. Those groups of resistors in a single package can open, meaning a line expecting a 56k pullup resistor now has none, so it lies there at ground potential... "stuck" low.

Looking at the keybed schematic, the switches are all normally open, so I tend to think we don't have a shorted diode or anything like that.

[doctor]

Each key has two contacts. When you press down a key, one closes before the other. The computer tells how fast the key was moving (ie how hard you played it) by the time it takes between the two contacts closing. The longer the time, the softer it plays. Now what if the second contact is already closed? Then as soon as the first one closes, the computer sees the other one closed as well, and assume maximum velocity, and so maximum loudness.

The opposite is when one of the contacts is dirty and cannot close at all. If it is the second contact, the system times out and defaults to the lowest velocity. If it is the first contact, sometimes you get no sound, other times it too defaults to minimum. I don't think you have dirty contacts.

I don't see which keys it is, you mention 38 or something, but the schematic doesn't call them by number. Note the matrix chart. Each key has a common - KC for key common. Then FI and SI, for first input and second input. Figure out which keys are stuck loud, and see what they have in common. I'd wager you will find one of those lines, probably the SI is stuck. if you don't know what it ought to look like, just look at the same signal on a working key.

Now look at first page of schematic, lower right. That is where all the KC, FI and SI leave the board for the keybed. Make sure all the ribbons are intact. But those signals all come from the large IC lower left, LSI1, the key assigner. It could have a dead I/O pin, but I'd be more suspicious of the R-packs just below. Those groups of resistors in a single package can open, meaning a line expecting a 56k pullup resistor now has none, so it lies there at ground potential... "stuck" low.

Looking at the keybed schematic, the switches are all normally open, so I tend to think we don't have a shorted diode or anything like that.

If one or few arbitrary keys are not working properly, there is a problem with related ribbon contacts.
But there are two specific problems related to the matrix structure of the keys assigning system: either one (or more) complete octaves do not working properly, or each 8th key does not working properly. The source of these problems is incorrect signal in one (or more) “column” or “row” trace of aforementioned matrix and ribbon contacts are not blame.
The reason need to be searched in other place.

Possible sources of such problem are:

1. Micro crack of the keyboard print, which leads to opened trace (fit a jumper).
2. Trace corrosion (the same solution).
3. Contact problem in the harness between keyboard print and print with assembled key assigner chip (very seldom) or in corresponding connectors.
4. Problem with pull-up or pull down resistor (if any) corresponding to this trace– opened or bad solder.
5. Key assigner chip.

In two last cases the schematic may be helpful.

[MARSHALL BUSTER]

Hi guys thanks for all the interesting reply's to this problem.

OK here goes...

I removed the keys and the corresponding rubber strips, I cleaned and checked them very carefully and I even changed the strip around with the next strip along on the PCB to see if the problem would move along the key board to the next set of keys. The problem stayed the same! same problem same set of keys.

I came back to the piano yesterday after moving it around the room and the problem had gone and the piano was playing fine as if there was never anything wrong with it

I left it switched on for a few hours occasionally coming back to check the keys and all was good so I turned it on and off a few times and it still worked well. Then out of the blue the problem just occurred again all by itself while I was sitting at the piano!!

So I'm thinking that the Docktor may have a point in his list of possible causes somewhere.

I'll be getting back to it again this evening to start checking the wiring connections/solders and looms.

I'll be back with my findings.

Thanks again.

[Enzo]

Do you have a scope?

Look at the contact surfaces under the rubbers for the affected keys. See if any of the contacts are dead. I know you don't know what to look for, so just examine the working key next door as a reference. I bet even a volt meter would work. If there is a low duty cycle signal on a working strobe or return line it will indicate SOMETHING on a meter. But a line that is stuck low or stuck high will just sit there at zero volts or +5v.


FInd that key assigner chip and look for the R-packs involved with the strobes and returns to the keybed. Wiggle the R-packs, any effect?