Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards Triple Sensor

I came across this video https://vimeo.com/50118333 when looking for info about triple sensor keybeds.

Can someone please explain whether, in this setup, it is possible to trigger a note by pressing it fast, but not completely bottoming out (simulating the continuing inertia of the hammer when thrown in a real piano action)? I would assume that you could take the velocity between Sensor 1 and 2 for this.

If not, then can you explain why Sensor 1 is required at all in this setup? (If the true note velocity is just that at the time it hits sensor 3 - this could be measured between 2 and 3 alone).

Thanks,

Whether you could get a note without the key reaching the bottom of the keybed depends entirely on whether the 2nd and 3rd sensors are coupled with the key or the hammer. If it's the hammer, yes, if it's the key, no. It varies between different makes and models.

Sensor 1 is for the damper simulation. It is still needed. otherwise you would get each note being separated (staccato) when doing rapid repeats. Techniques like tremolo involved keeping the hammer/key between sensors 2 and 3, but not lifting up far enough to trigger sensor 1, so the sound is smooth and continuous, rather than each note being damped and retriggered with each strike.

I hope that gives you a better understand of why triple sensors are used. Of course, in the real playing world, triple sensors are only as good as their implementation. For this reason you should always play such instruments before buying to make sure it really does feel right. Don't just buy from a spec sheet.

Thanks for the explanation. It always felt to me, though, when playing these popular actions that the hammer isn't loose and can't be thrown as such, further beyond the key position, but is tied to the keys height. So if the key doesn't reach key bottom then the hammer wouldn't reach sensor 3 either. This is one of the sensory dissonances I have with real vs simulated actions - especially noticeable when you reduce the pressure on the key immediately after "throwing" the hammer. Perhaps it does "throw" though and you just can't feel it as much on a digital. I will have to test it out next time I play one.

The best actions certainly do throw the hammer - but you are correct that you can't feel it as well as a real piano. What you are describing is called "escapement", this is when the hammer escapes to mechanism that pushes it and moves under its own momentum. Digital pianos tend to feel less real because the mechanisms are scaled down - everything is smaller, even though the apparent weight of the action and hammer is similar. I think the issue with having the action scaled down is that the weight of the hammer, being smaller, has the ability to change direction more easily. So it will seem to follow the movement of the key more closely. It still has escapement though - the key and the hammer can be moving in opposing directions if you hit it fast and shallow. Whether this is close enough to a real action feeling is up to the individual. But it's less of a sensor issue and more of an action design issue.

Most 3 sensor DP actions put all three sensors on the hammer. The mechanism itself is usually a simple dual lever arrangement, where the hammer is free to fly into the cushioning pad, but when it returns to the key it comes back to the launching position, not to a back check, so in that sense there is no escapement (and I think this arrangement may more easily lead to injury since any remaining hammer energy is geared down and sent back to the finger / hand / arm). Keys designed for portability likely use lighter weights that are more geared up, which makes the key return time more sluggish.

Does anyone know which three sensor actions allow a note to sound without bottoming the key on the key bed? I would really like to know this,as the lack of this ability is a major distinction between acoustic grands and most DPs. I just hate the necessity of bottoming out keys on a DP; it feels so unnatural and for me, and it really slows things down.

As per the above discussion the lack of comparable hammer inertia to an acoustic grand is also something I miss; on the acoustic grand I find it really helps control...

But it does have escapement in the sense that the hammer does escape the DP equivalent of the jack. It is this simple form of escapement that the OP is particularly interested in because it enables the key to impart enough momentum to the hammer to hit the 3rd sensor and sound a note without necessarily having to bottom out with the key. Whether it happens in a natural and convincing way is a whole different question, but most hammer action DPs do at least have this form of simple escapement. Perhaps we should call it single escapement as opposed to double escapement.

It does seem like Roland's PHA III works this way. If you give the hammer enough speed, you don't need to hit the bottom with the key itself to produce sound. How well it works in practice I don't know, as I'm not conscious of this when I'm playing.

Roland's PHA II works this way, too. It is unnecessary to push the key more than 50 percent of the downward journey to trigger the note. This seems a matter of sensor placement, though this is done with 2, not 3 sensors.

No. It should be called "half" escapement, as opposed to "single" escapement.

"Single" escapement (just referred to as "escapement", normally) which is present on all acoustic pianos (both upright and grands) is mechanism that separates the hammer from the key during the stroke, which allows the hammer to strike the strings freely, allowing it to bounce off the strings, rather than being held against the strings. No standard DPs that I'm aware of have this - if you simply press down all the way on the key such that it bottoms out, the hammer will be hard up against the key. (obviously I'm excluding high end DPs that have real actions)

"Double" escapement has already been described earlier and that matches my understanding, and on acoustic pianos, it's only present on grand pianos. (and that's what the third sensor is there to emulate - double escapement)

I think it would be extremely confusing to refer to the ability to "throw" the hammer "single" escapement.

Btw, are we absolutely certain that when we "throw" the hammer on these DPs that the keys don't also keep moving in the same direction, at least to some extent?

Greg.

I've looked at the VPC1. I tried putting in a piece of wood under the lip of a key, that prevented it from descending far enough to actuate a tone for a normal keypress. Then pressing that key hard would produce a tone by making the hammer jump ahead of the key motion. However, the key will keep moving unless something stops it and catch up to the hammer.
I tried playing a little to produce sound without the key hitting bottom. It's possible, but not easy, or very useful, I think.

Also, note that without the weight of the hammer pushing back, it seems that piano keys naturally want to fall down, so IF a standard hammer-action DP behaves the same, it seems to me that this is evidence that there won't be much separation of the key and the hammer, if one is playing without the keys bottoming out.

You can just make this out at time 2:10 here: http://www.youtube.com/watch?v=YmreTCoRIBw#t=2m10s

and the same applies for an old electro-acoustic piano (the Rhodes) - see time 8:00 here: http://www.youtube.com/watch?v=sLraP5ha_Kg#t=8m0s

Does this sound reasonable?

EDIT: Ok, I just tried it on my Kawai MP9000 (which has an action that is similar to the current Kawai wooden actions) - yes - when I lift a hammer, the key goes down. If I hold the hammer up, the key can be freely lifted up, proving that the action of lifting the hammer does not force the key down - it is falling down due it's own weight.

Greg.

Maybe it's wrong, but I think of "escapement" as the hammer breaking away from the linkages that geared it up and launched it, with the key losing control after that point until it is re-cocked by raising the key. AFAIK no DPs outside of the Yamaha AGs have this. "Double escapement" is just a method of re-cocking it at a lower point of key-up.

Yes, that's a reasonable description, and as crude as it is, I think most DPs can do this these days. The only question is how well implemented it is. Most DPs don't give you the sensation of really throwing a hammer towards a string - even though I think all would be capable of triggering a note without the key hitting the keybed.

I also think all the 3 sensor actions are somewhat capable of "re-cocking", though not necessarily with enough velocity on subsequent strikes to feel anything like a real grand piano. The AGs are interesting here because even though it has a proper double escapement action in it, they have done an awful job on the sensors (I don't even know if they have 3-sensors). Have a look at the Yamaha AG promo video of Cyprien Katsaris trying to get a rapid tremolo out of the AG - it's seriously embarrassing. Here is a guy with monster technique who could tremolo all day long on a real piano, but he can't execute a tremolo to save his life on an AG. It goes to show that the placement and number of sensors is at least as important as the mechanical action it is coupled to. Btw, I tried rapid repeats on the AG myself too - it doesn't work well at all.

I notice DP companies love to present demos of Chopin and avoid pieces that require fast repetition like the plague. But to be fair to Yamaha and the AG, in the second half of this video Katsaris really cranks a version of Gottschalk's The Banjo (with ad libs) and hardly misses a repetition. I have this piece firmly in my brain and do however hear a certain odd but minor latency somewhere in the sound that obviously is not the result of the player's technique.

I don't think companies avoid rapid repetition in demos because of piano limitations. For one things, the demos are almost certainly done with MIDI files, where action limitations would not be a concern. Moreover, as discussed elsewhere, I don't even think lack of third sensor generally affects same-note repetition.

http://www.pianoworld.com/forum/ubbthreads.php/topics/2171115

It might be more of an issue for rapid pianissimo trills, though.

I don't think my Kawai MP9000 can "do it" to any meaningful extent whatsoever. An action with an escapement has a dead zone during the key's return, such that when the key is pressed down again, in any area of that dead zone of the key's return, the hammer does not restrike the string again. In a grand piano, in any area of that dead zone, the hammer actually moves AWAY from the strings as the key is pressed down again. (and moves towards the strings as the key is released, by virtue of the double-escapement mechanism) In my MP9000, when the key is down, the other end of the key is pressed up against the hammer, and any upward movement of the key results in a downward (i.e "away" from the striking point, or away from what would be the strings) movement of the hammer, and regardless of how small or large the key's return is, if the key is then pressed down again, the hammer will move towards the striking position/strings. The only "re-cocking" that is present in this DP is the obvious one of having to let the key return sufficiently for a key release to be sensed, allowing another note to be played.

Note also that it's my understanding that in real pianos, they try to DELAY the point of escapement as much as possible, because this results in hammers that can be controlled up to a point that is very close to the strings, allowing very soft and controllable playing. Grand pianos can generally go closer to the strings than uprights, too - that's one of the advantages of a grand.

I see that in that other thread that Anotherscott referred to there was also discussion about the fact that the keys will keep going downward if the hammers are not pushing back on them.

Greg.

Agreed. This is an idea propagated frequently by users of particular products (PianoTeq forums, I'm looking at you) but I haven't seen any validity in it. Demos are chosen to sound impressive and appealing. Frequently that means frenetic, romantic pieces.




I didn't see the relevant post in that thread that demonstrates this. I think you had better clarify your position on this. Rapid repeats of the same note are the classic example of a case where triple sensors help. While playing, not having to lift the key as high before restriking is necessary to repeating a single note very quickly. In the MIDI you can send notes out as fast as you want, as long as you don't want that note to be undamped while the other keys are damped, but for the most part the third sensor is related to being able to input stuff from the fingers, not just recreate a MIDI stream that may not be physically playable on that keybed.

In that thread it seems that your position is that the number of sensors may not be the limiting factor in some actions. That's something I can agree with. But if we are searching for where it helps (where implemented well) I think repeating the same note quickly is the first place we should look.

gvfarns: FWIW, my piano technician's reference book does say that the double escapement is mainly of benefit for rapid soft repeats.
(which of course includes trills).

Greg.

I think my 3 posts on that thread (
http://www.pianoworld.com/forum/ubbthreads.php/topics/2171115 ) are pretty self-explanatory. But yes, I've given this more thought, so let's continue...


And, in my view, wrong. Or at least incomplete and misleading. I think if you can play the intro to Billy Joel's "Angry Young Man" on a three sensor board, you can play it on a two sensor board. Each release would be expected to lift the key above the highest sensor (unless you have a really sluggish action, in which case you have a bigger issue than the lack of a sensor). So that "classic example" as you put it is, in my view, inaccurate.

I'd say the benefit has more to do with repeating notes softly. In that case, I think there may be a subtle benefit even when playing slowly, in the feel (of not having to raise your finger as high, which may feel more expressively natural when playing very delicate passages), and possibly in the sound (if you're not using pedal, the third sensor provides the only way to re-strike the key without silencing the note first). So in terms of same-note strikes, the benefit has much more to do with playing soft than it does with playing fast. That said, yes, the faster you need to repeat that soft note, the more you may appreciate the feature, but it is still tied to the need to repeat the note quietly, rather than the need to repeat the note quickly (as the "Angry Young Man" example illustrates).

The place it would come most into play would be in same note repetitions that are both quiet and fast. The most common occurrence of that would be trills.

To the bigger point, you and I do agree that "the number of sensors may not be the limiting factor in some actions," and I would not rule a piano purchase in or out based on this feature, as it alone doesn't really tell you which of two actions you may find more playable overall. As has been mentioned, upright acoustic pianos can be seen as suffering from the same "shortcoming" as two-sensor DPs, and there are wonderfully playable uprights, and nobody is saying "never buy an upright" because of this.