Make Like A Banana - And Keyboard Split

Let's Split

 There are two voice groups on the SS-30 - Cellos and Violins. There are 49 keys with corresponding keying circuits which provide attack and sustain envelopes. However, it's possible for more than 49 notes to be generated in the SS-30. There are a maximum of 25 notes of the Cello voice and a maximum of 49 notes of Violin. Those of you who have studied mathematics at the very highest levels may have started to see that there is a problem with this arrangement. It has crossed my mind that building another two K boards would solve this problem, but the eminent designers of Yamaha Nippon Gakki are not totally wasting notes. So, what is going on?

The keyboard split rotary switch selector on the SS-30 is numbered. Each split point being given an integer corresponding to the markers just above the keyboard. But what do they represent? Well, as the User Manual,explains:

"Switching this section changes the position of tones of Cello and Violin systems.
There are marks on the panel above the keyboard, so that each point of 1, 2, and 3 may be noticeable during playing"

Actually it's got a huge typo in and repeats some of that which makes it hard to read. It 's also says the marks are on "the keyboard (the panel)".  But that's what they meant to say.


Things become clearer with the following description.

 Right. And so...

"The KBD.SPLIT enables you to produce
chords of the Cello and melodies of Violin by your left and right hands respectively."

 Exactly! It does seem a shame that I couldn't just get all 75 notes at one and the same time, but it's not that they actually all exist at once. Why not?

Gating Away With It

This diagram shows how each K and G board is used for each split.

Except it isn't even as simple as that. The G boards have 4 outputs each which are mixed or rearranged depending on the voice selected.


The key to all this is the split gate.

This is the schematic, but it's not terribly helpful.

What's happening is the outputs of the G boards are being switched on or off. Which sounds simple, but there is a lot going on here. There are 6 inputs but 9 outputs. Two of the inputs are from G1 and the others from G2, where most of the switching action occurs.

This, from the  Block Diagram, is a bit easier to follow.

But combining them both with additional information about the way the voices are put together yields this:

As you can see, you never have all the the outputs from both G boards on at the same time.

It's still a bit hard to follow though, so here it is with the voices grouped together.

Now it's clear!

Cello and Viola use the exact same G1 and G2 outputs as each other, but not at the same time.  The split gate sends those outputs to different filters to convert the raw output to become either Cello or Viola. There's nothing fancy about these filters! But that's for another post.
The Violin is ON at the same time as the Viola, obviously, and it uses the other outputs from G1 and G2.
This whole thing could be simplfied if the split was reduced to either Cello on or Cello off, of course. However, it's a versatile solution and I've no desire to change it.

Splitting My Sides

As I thought more about the split control and where to place it on a the SS-30M panel, I came up with this rearranged front-panel for the SS-30. It places the split in-between the two sets of voices.

The main problem with this design is that the Keyboard Split is going the wrong way! All the way to the left means no Cello and all to the right is all the Cello on. Can that be improved?

The switches positions can be marked up in any way you like, as long as they are in the same rotational order. This version inverts the control so it balances the other way.

This fails in two ways though, because having the knob upside down is unnatural - our hands not naturally designed for this kind of 'twiddling'. The leverage is also naturally top-side. Mostly though, it makes less sense in aligning with the keyboard.

This is the current alignment

 And this is inverted

OK. But that is a nonsense because I simply rotated the drawing! What needs to happen is that the switch is rotated but the markers are kept at the top and reordered. Except that comes to exactly the. same thing as before but the switches action is upside down! Oh.

I have another solution for this though. There is a long tailed knob from the B-75 donor that is like a one-sided chicken-head knob, but with a very long tail. These were designed so that you could easily move them without changing your hand from a fingering position to a grip. Using this control serves two purposes. It makes the rather stiff switch easier to rotate, due to the extra leverage. It also means that the pointer can still be aligned to the split points on the keyboard but the arm points to the balance of Cello and Violin.

Kind of like this:

With rack panel layout, it can be more obvious what voice is being pointed at by the long tail.

Except now that there keyboard markers have gone I need to swap the numbers for the actual notes, like so:

Split - The Difference

Much better! But, there is a centrality to this element now, so that anything on the same vertical axis, or group, is inclined to be grouped in common with it. The point of moving it in between Cello and Violin is to control the balance between them. Otherwise the pointers of the longer control become confused. Its location is either between, with the longer pointer, or it's location is almost arbitrary and the knob can be smaller.

That's great, but there could be a problem of space. The control takes up far more room now and the layout demands more separation.

Ideally there would be a control that is common that could also be in the centre. Pitch and vibrato are but that is a lot of controls to fit in. This will take some thought...