Unresponsive keys of a Kork minilogue xd

[aiyiadam]

I am trying to repair a Korg minilogue, which has 7 unresponsive keys. This happened on a brand new device. The owner used it for a while, put it away for a couple of months. And when he turned it on again, he realized that those keys weren't working.

What I have done so far?

- Cleaned the carbon surfaces of the switches and the board.
- Swapped the rubber strips
- Swapped the UMG9N digital transistor of a working key with the transistor of a non working key.

Those didn't work.

If you press a working key the both output pins of the relevant transistor reach 2,78V. This is different with unresponsive keys.. With those, I measure 2,78 at pin 4, but the pin 5 reaches only 0,8V.

Any ideas maybe?

 

[Khron]

What about the input pins, with a "bad" key pressed or not?

 

[ElSmurf]

If those green dots represent the unresponsive keys my first guess would be there's a line disconnected between the keyboard matrix and the controller.

 

[aiyiadam]

What about the input pins, with a "bad" key pressed or not?

0,156V for all keys. Pressing the key doesn't change the value..

 

[aiyiadam]

If those green dots represent the unresponsive keys my first guess would be there's a line disconnected between the keyboard matrix and the controller.

Yes.. Those are the unresponsive keys.. There is a connection with 16 wires between the keyboard and the mainboard. The connections on the keyboard side look okay to me.. I will also check the connections on the mainboard. But before taking everything apart I just want to be sure that this has nothing to do with the keyboard.

There are also two CD74HC238PWR on the keyboard PCB, which could also be suspected I guess.

 

[Khron]

0,156V for all keys. Pressing the key doesn't change the value..

And compared to the "good" keys?

 

[aiyiadam]

And compared to the "good" keys?

They are the same..

 

[NoisyIndividual]

Find the trace that is common to all those keys, and then follow it back to the connector/connector on mainboard.
Somewhere it will be broken/corroded.

 

[zamproject]

If those green dots represent the unresponsive keys my first guess would be there's a line disconnected between the keyboard matrix and the controller.

My guess is the same at first look, but then I'm puzzled with the pattern ?

The pragmatism would call for a 8x5 matrix
But we have a break of 5 keys between Bb and Eb where all other are spaced by 4 keys.
We have 7 dead key not 8 or 5.
We have 9 consecutive keys working.

Make non sense to me ?

Or maybe it's a 7x6 matrix with non consecutive key routing for some obscure design reason ?
But pcb clearly show parallel bus lines of 5 and 8 traces

 

[NoisyIndividual]

The break may be between key 7 & key 8, with only key 8 working.

 

[zamproject]

The break may be between key 7 & key 8, with only key 8 working.

Not sure to get your logic ? but I find a possible working pattern !

Row 2 and 6 broken but only after first column, If you count from the upper C key.
It imply one exception with C1 being attached to the fifth column with the four lower key (C0 to Eb0), C1 being the fifth key in this row ?

So my guess is to find what happen on the second and sixth horizontal and paralleled traces we see on the pcb that should be the 8 rows. Location should be between upper E and F.

Hope this help

 

[NoisyIndividual]

Easiest thing to do is find the trace that is common to 2 of the dead notes. A meter will then show the same pads on the other dead notes are also connected to this line. That line needs to go back to the mainboard, it probably doesn't.

 

[NoisyIndividual]

Well, that's how things normally are in keyboards. I notice a 16 pin SMD on the right of the photo, this may be some multiplexer or something which these common lines connect to. So one of these outputs may be dead instead.

 

[zamproject]

 

[zamproject]

So one of these outputs may be dead instead

Not sure ? we have a culum of the keyboard where all rows work ?

(note that in the above illustration it can also be row 3 an 7)

 

[aiyiadam]

Thank you all guys.. I still don't know how this thing works, but I think I solved the problem..

There are a few key chains. In my case the chain is 3,7,11,16,20,24,28 (non working) and two working keys after 28 (32,36).. This damaged pad is at the key 32.



The pad is almost completely gone and the circuit is open. I tried to repair it using the remaining piece of the pad and the keys seem to be working now..

 

[zamproject]

So you had 9 keys not working ? not 7 ? and a single wire solve all ?

I was totally wrong then... and have no idea how the matrix and scanning is done

 

[aiyiadam]

So you had 9 keys not working ? not 7 ? and a single wire solve all ?

I was totally wrong then... and have no idea how the matrix and scanning is done

Yes.. A single wire solved the problem..

7 of a total of 9 connected keys weren't working.. The keys 32 and 36 were also in the same key "chain". The damaged pad was at the key 32 and that is where the chain broke..

 

[zamproject]

Fine you get it work again !

I miss the obvious.
We get two pad per key (velocity ?) so this request more than a 5x8 standard matrix for 37 keys.
And then there is two demultiplexer...