An idea to solve the problem on the RH3

[afr]

A possible solution to the problem of mechanical RH3

Hi everyone, I state that using a Kronos 61
Fortunately I do not have the problem occurred with some models of 73/88,however, I tried to read and understand what is happening on these models,also comparing ideas of user on differents forum, I tought to express an idea as possible solution to the problem


He seemed to understand that the problem occurs when the same key
as a result of a rebound, produces a ghost note stopping the sound of the main note

I was wondering if is enough to introduce a filter that eliminates
ghost note if it is played in a short period of time
(2 or 3 ms?) and if the level of dynamics is very different

Do you think that this might be an acceptable and functional solution?

[michelkeijzers]

Something like this is also the possibility that Korg will analyze probably.

[petem]

There is a well established technique called "debouncing", which is applied to many mechanical switches, often as a software algorithm. When a mechanical switch is closed the contacts often bounce apart and then close again one or more times before settling, which won't matter if it's a light switch but if it's driving something electronic the bounces may be seen as the button having been pressed more than once.

For example if it's a button on a telephone it might send the tone for that button more than once when you only hit the button once. Debouncing routines look at things like the time between the events and if it's too short to be real separate button presses then the later one is ignored.

This issue is very well known to electronic engineers so I've no doubt that Korg will have considered it if the problem with the RH3 is actually some form of bounce. In fact I'd guess there will already be some sort of debouncing being done on the keys, maybe the routine just needs some fine tuning. If the key is bouncing slowly it does become a problem as the routine needs to decide what is a bounce and what is another very fast keypress by the player, and if the bounce is very slow there might be no way to be sure it's not another intentional note. In that case the only good solution might be to change the mechanics so that the keys don't bounce this way (obviously not as easy to do as a software upgrade to the debounce routine). As afr suggested, they may be able to make use of the difference in velocity if it's detectable that the bounce is always much more gentle than the original hit.

Hopefully Korg will have a solution soon and it will be interesting to see what it is.

[Sharp]

http://www.korgforums.com/forum/phpBB2/ ... hp?t=67537