I usually assign the buttons to the dry/wet of the effect - bipolar seems to work best, I'm not totally sure why. You'll need to tweak the amount if you don't want 100% effect 🙂

I am being stupid, sorry. Set the dry/wet of the insert effect to 0 then a unipolar option on the switch works fine. Then set the ratio to the amount of effect you want.

thanks a lot for your help. Thanks to you I understood how to program the A Switch. But unfortunately the insert effect I wanted to use does not have a dry / wet setting. I wish I could act on its button on / off from the ASw. I doubt that the on / off buttons are assignable.

When asking for support - often the details matter. All problems should probably list the firmware version used. Most support will begin with "upgrade to the latest" just to cut the fat off of possible items which have been fixed, to align your keyboard with the latest documentation/features, and to match the firmware that most support personnel are going to have themselves loaded (ignoring firmware transition times such as today).

Second - many users when having problems with one "thing" (a specific effect in your case) will leave off naming the specific "thing" (effect) used. This will have the forum guessing at how to guide you. Why not convey this detail so the feedback can be more directed?

Ignoring the Reverb Block (it's not insertion) - Insertion Effects which do not have wet/dry:

VCM Flanger (and Control Flanger) - Note: use mix=0. From the parameter manual ( https://europe.yamaha.com/files/download/other_assets/1/812531/montage_en_pm_a0.pdf )

Mix Determines the volume of the effect sound

I'm assuming 0=no volume for effect and 127=maximum volume for effect. Use your ears and adjust assumptions.

All other flangers have either "Dry/Wet" or "Mix"

... I haven't tested on my Montage (which is packed up) but for VCM Phaser Mono - I'm assuming there may be a number of settings which have to be adjusted to turn off the effect using only parameters.
Depth = 0 (turns off LFO for phase modulation)
Color = 0 (I imagine even without phase modulating - the color non-zero would change the sound)
Feedback = 0 (may not be necessary, not testing right now)

... and so on. Please provide more details so your case can receive more directed feedback.

==============

An aside on the discussion of uni vs bi polar ...

Regarding uni-polar vs bi-polar. When using the normal/default curve which is a ramp - then there's no effective difference between unipolar and bipolar (EXCEPT for the reset condition - covered in a bit). Yes, they work differently with respect to how the math works (all positive or positive and negative). But assuming you adjust the parameter being controlled by the modulation source - they end up being the same (again, ignoring the initial value). Given the same slope of the ramp - there's no difference in the range and affect of the parameter by position of the knob between bipolar and unipolar.

Uni-Polar:
Non-offset destination programmed value = 0
Curve range = +0 through +127
When modulator - say a knob is full counter-clockwise, then value will be 0. When at 12 o'clock, value will be 63. When full clockwise, value will be 127

Bi-Polar
Non-offset destination programmed value = 64 (note: I'm fuzzy off the cuff if this should be 63 or 64 - lets not sweat that detail right now)
Curve range = -64 through +63
When modulator - say a knob is full counter-clockwise, then value will be 0. When at 12 o'clock, value will be 63. When full clockwise, value will be 127 ** same as uni-polar. So assuming you account for uni vs bi polar in your programmed destination value (without offset) then the relationship of the position of your controller vs. the resulting value of the destination after offset is the same.

What's different between these two (and there is a real difference) is the default value. One will default at 0 and the other will default at 64 when you load (recall, select) the performance.

Yes, you can "compensate" for this by making your programmed destination value something other than "0" for the uni-polar case and something other than "64" for the bi-polar case. But for the ranges of offsets given in the above examples - you will lose "full range" of 0-127 if you change the offset. Making the slope more aggressive can compensate - but then you change the rate of change of the knob so there are side-effects (not bad - but it helps if you know how the system works so you have a better "feel" for how to coax results you want out of the programming choices available).

So generally, the rule of thumb is that if you want a value to START in the middle - use bi-polar. If you want the value to START at polar values (0 or 127) - use uni-polar. That's a rule of thumb (meaning it doesn't always apply). It's easiest to discuss the default ramp - then you can extrapolate for other curve types. Note: to start at 127 - you'd need to have a reverse ramp which can be accomplished by using user curves, linear type, with first point: output=127 for input=0, for last point (8th): output=0 for input=127, and points 1-6 with outputs as points along the line from 127 ramping down to 0. Yes, there's a formula you can use. I would just make point 2 input=1/output=126, point 3 input=2/output=125, point 4 input=3/output=124 ... and so on - keep point 8 as input = 127 output = 0.

============

Also, you may want to familiarize yourself with the "Send" effects. These are pooled effects - global for every PART to participate in or not. You do not have as many of these since they are global. However, each PART can dial up the send effect or cut it off completely. This applies to Reverb (more limited) or Variation (which includes all the effect types that insertion effects have to choose from). If you're not using sends - or can stand to reassign say the Variation Send effect (for all PARTs) - AND you only have one effect you want to turn on/off with the switch -- OR just have one "problematic" effect to turn on/off with the switch - perhaps consider using the variation send.

Many presets may utilize variation send as the default programming for the performance. When you combine performances - if PART 2 was using a different variation send than PART 1 - then PART 1 "wins" and claims the effect type for the variation effect. PART 2 would inherit the variation effect that PART 1 uses. This may slightly alter sounds when combining "stand alone" performances into a single performance.

All of this detail is to understand what's going on and to give more confidence in "blowing away" variation send if that would help you achieve an effect modulation (switch on and off) goal. YMMV depending on how much your PARTs rely on the variation send effect and how dramatic change is to the sound by using the default/programmed variation send vs. using something different for the variation (send) effect for your own purposes.

Adrien wrote:

... unfortunately the insert effect I wanted to use does not have a dry / wet setting. I wish I could act on its button on / off from the ASw. I doubt that the on / off buttons are assignable.

That is correct, they aren't. I was at the NAMM show when I first noticed your post. You mention in the original post you want to "trigger", as in start/stop, an Insert Effect. It's not a correct thing to say... because you can "trigger" an event, like triggering a note-on, trigger an Arpeggio phrase, trigger a Motion Sequence.

So it is strange to say 'I want to trigger an Insertion Effect'. Technically speaking, the Insertion Effect is "triggered" as soon as signal arrives... and as you know, an Insertion Effect is, literally, inserted in the signal flow path of the entire signal. So I was hoping you would provide a bit more info about what you want to do.

_ Some Insertion Effect Types have a Dry/Wet parameter, but only those where there is a subjective decision about "how much" to be made
_ Some Insertion Effect Types do not offer a subjective balance decision, these are your more utility Types like EQs and Amplifiers like Compressor/Limiters, etc. these devices typically work on 100% of the signal, so you wouldn't expect to find a Dry/Wet parameter on these Insertion Effect Types.
_ Some are physical models of classic devices... where either a DEPTH parameter can be real time controlled, and on some the SPEED csn be reduced to 0.00Hz which affectively kills the application of the "effect".

But we can say one general thing about Insertion Effects as a whole in the MONTAGE: Each Part (plus the A/D Input Part) has its own two Insertion blocks... these can be assigned real time Controllers and are interwoven with the sound at the Element level, and are considered as much apart of the instrument as the filters, and Envelope generators. When you select a Part, the Insertion Effects, and the assignment to control the Insertion Effect parameters, are automatically brought along.

A B3 Organ sound might have a Rotary Speaker as one Insert Effect, and an Overdrive as the other, anytime you 'wheel' that B3 into a new Performance it will come in with that Rotary Speaker, and that Overdrive. They are considered a part of the Part.

So on a TYPE by TYPE basis you would need to decide *how* you want to realtime control it. Remember, you can route the signal on a per Element basis to one or the other of the Insert blocks. Each AWM2 Part has eight Elements. You could send four of them through Insert A, you could send the other four through Insert B.

Set Elements 1-4 to sound only when both Assign Switch 1 and 2 are in the Off status. (XA CONTROL)
Set Elements 5-8 to sound only when Assign Switch 1 is On. (XA CONTROL)

Depending on what is comfortable for your idea, set the [AsSW1] to momentary or latch...

This type of assignment would allow you to play the sound without whatever Insert Effect being applied, and then by pressing the Assign Switch you would bring in the Insertion Effect you wish to apply.

The above concept is just one of many ways to approach bringing in an effect. It can be expanded where you use XA CONTROL to toggle OFF an entire eight Element PART and simultaneously toggle ON a second eight Element PART. Each Part could be either with or without the Effect condition you are designing.

You could morph between conditions... we can only talk in theory... there are many ways to

Why not a simple ON/OFF as an assignable parameter?
There is a very good reason... it was specifically decided not to allow this. It is deemed one of those changes that would/could cause a significant change in level, it would not be musically friendly (you want the change to be sonically invisible not a loud pop or click!)