Well, I've been up all night. I opened up the keyboard about 10pm last night and put the last screw back just after 6am this morning.

Verdict, PERFECT!!!

It repeats like a Renner action now! I'm not kidding either! Also not a single sign of velocity problems.

The size of the time lapse video is over 21GB!!! Probably because it's recorded in HD 720dp and the file is in .avi format!!!

I'm so pleased with the job I'm not even tired anymore. It went without a single hitch. The expression has improved no end and there is complete evenness throughout the whole range. I will have to make adjustments for the later strike point etc., but that's going to be a walk in the park now!

I'm going to convert the video file now, chop it up if need be and put it on youtube. Not sure when it'll be ready, but it'll be soon!!!

See y'al shortly

Chris

Could you measure the distance between note-on and note-off @ keyfront on a few keys and let me know the average measurement.

Well, I compared about 12 keys in the middle, and few on the ends:

- Total travel = 13-14 mm
- Note On = 10-11 mm
- Note Off = 4-5 mm

I used the built-in midi monitor of Pianoteq as well as Midi OX just in case...same results.

So, suffice it so say, if I were dependent on the note-off it might be touchy as there is only a 5 mm window (or so). That is something to look out for in any future keybed. I imagine that the note-on happens before the contact to allow for the propagation delay I mentioned...the effect is that the midi information hits the target at about the same time you dig into the keybed. I am not using the USB interface, but it's probably about the same.

As I stated, the only problem I have is the aftertouch...the range (or window) really should be adjustable in software.

John, you are right. I am a little bit of a perfectionist in some areas. I have enjoyed measuring and analyzing, and the problem solving process that has been whirring in my head! It won't be long before I realize the benefits of all this.

It is very true what you said about the differences between the design and purpose of the action of a real piano compared to that of a controller/midi keyboard. When I read this research paper, it really brought that home to me - the fact that even the "best" hammer action controller... keyboard available at the moment is nothing like the real thing, even if it is hooked up to PianoTeq Pro with the best sound system, acoustic environment, etc. etc.. If you want to "experience a Grand Piano", you must go "EXPERIENCE A GRAND PIANO!!!". Nothing is like playing a Grand Piano other than playing a Grand Piano!

If however you want to make fine recordings of a Grand Piano, then there are a large number of keyboards out there that, when hooked up to PianoTeq, will enable you to do so. Granted, there may be some setup needed - sometimes intensive - but on the whole, you can do a fairly nice job out of the box most of the time!

I realise, now more than ever before, that I'm not going to ever find a substitute for sitting at a concert grand. But, more importantly for me, I realise now, more than ever, that I can get exceptional sound/recordings/performance/dynamics/expression etc. from a small slab of plastic and metal that cost less than a "grand" rather than a "Grand" that costs 100 X as much!

The main purpose of my posts, other than venting my fierce passion for music/musical science/musical physics/obsession?, is to help anyone who would wish to do with their keyboard/controller what I am doing with/to mine, but without the hassle of experimentation, research, measurements, analysis, mental torture! No seriously, I wish to write this stuff to make it easier for others who have either had similar problems with their boards, or wish to understand and/or modify them in the same way, or similar, by providing detailed measurements, instructions, photos and finally, a time lapse video of the whole process I will go through when I open up the my Nero for the second time to put into practice what I have been preaching! Kind of like writing a guide.

I have enjoyed the experience so far immensely and would desire that others would also enjoy any aspect(s) that would interest them.

Just one question, John, concerning repetitions. Could you measure the distance between note-on and note-off @ keyfront on a few keys and let me know the average measurement. Also, would any others of you who have the Nero, who are happy to, do the same, in order that we might see if there is a consistency or lack thereof in this area also, as with the varying conditions of cleanness (grimefreeness) - and resulting corresponding variances in velocity performances - of the sensor strips in different Nero boards. I would be exceptionally grateful.

Sincerely to all on this forum (and those unsuspecting trespassers!!!),

Regards,

Chris

well you sound enthusiastic about this to say the least :-)
This may not be a topic I'm really getting into although the note-on point of any keyboard should be interesting...
I use a CME UF80 keyboard and I do like the action and everything myself but sometimes when I play a note REAL soft, so not even in 'normal' play, you can 'feel' where the note on point is reached which when you really follow through that slow, seems to appear a bit sooner than I would expect...
but when I play normally it feels like an average piano, not like if the notes are there 'too soon' or something...

Anyway - good luck in your quest for perfection!!

cheers
Hans

http://www.ofai.at/~werner.goebl/papers...Action.pdf

This reminds me of music theory...I would rather play than analyze! However, all of that theory is influencing what I do (at least on a subconscious level) so maybe it was worth it in some respects.

I suppose it is impractical to develop a keybed that meets these requirements and still make it affordable. Maybe your best option is a "real" piano with the retrofit.

I don't expect this type of behavior from the Numa...or any keybed. They are, after all, digital controllers. We are using them to trigger digital software too, so the mindset is different for me. For example, how does latency fit into this model? How about propagation delay?

The piano keybed could be considered an interface to what is basically a harp. That was the technology of the day when it was designed. I am not so sure that a piano, were it designed today, would be anything like it is. I am comfortable with the fact that the technology (and the underlying mechanics) is changing.

The Pianoteq software itself does not sit right with some people because they are caught up in some intricacy that is in direct comparison to a physical piano. What a shame. It is a very expressive piece of software, and that is all I care about. In the end I don't think the listener is going to be concerned with these details...they just want to hear the music!

You are obviously a perfectionist, and that may be what drives you to be a better musician. Your 'quest' is interesting to say the least!

Excerpt from research paper on the internet

Blow - 45mm
Let Off - 3-1.5mm
Drop - LO+1.5mm
Key Dip - 10mm
Back check - 15mm

These are purportedly factory recommended measurements for the regulation of an Estonia grand. Interestingly, and particularly happy news for me, the recommended 'Drop - LO' is +1.5mm. Even though this dosen't look much, you can certainly feel a drop of 1.5mm when depressing the key. The 3mm drop I have at present on the Nero feels more like 4-5 (and I'm no princess!!!).

As I mentioned in my last post, I was reluctant to go to 1.0mm and prefered if possible to set a drop (note-on point) of 1.5mm. Finding the above info has certainly helped me to see that I'm on the right track!

I apologize for the confusing terminology here, I know I haven't explained this aspect very well. I will do my best to do this now.

When I use the term escapement or letoff I mean that point on a piano action when the jack or knuckle leaves the hammer butt. Because this point is so close to hammer strike point, I mentally see the point at which the note-on midi signal is sent i.e. the point at which the note sounds and the point at which escapement/letoff occurs as the same. This might have been a cause for misunderstanding in my posts here. As the Nero has no mechanical escapement / letoff, when I depress a key slowly to find it's note-on position, as soon as I hear the note sound, I stop depressing the key. Then when I depress the key the rest of the way, I experience what I perceive as the key drop (that which happens after escapment - after the jack has tripped out of the butt and the hammer has been thrown in a real piano action.

I'll try to put simpler still. I have chosen to put note-on in the same place in the keydrop of the Nero as one would find escapement/letoff on a real piano's keydrop.  THAT'S BETTER!!!

SO, I'll aim to set the sensors so that note-on occurs at 1.5mm from bottom.

I am actually very surprised that drop is set so low. I thought it was something like 3mm, but I realise now that it is half that. If you look at 10mm on a ruler it seems so small, but you can do so much with that small distance when that 10mm is magnified by the levers in the mechanism that is the action. Also fingertips are extremely sensitive - take braille for example!!! and have any of you had a finger jab? My that used to hurt!!!

I sometimes feel like I've hijacked the PianoTeq forum. I write on it nearly every day, often several times a day, and my posts aren't particularly short are they? (I hear a yawn!!!). Please don't let me put anyone off here. If my posts are something of a sleeping tablet to you, then please let me know. I am just so passionate about the whole PianoTeq/playing/posting/uploading/reading/anticipating(PT4)/discussing THING!!

Just so you know!!!

Regards,

Chris

Above, the durable cotton buds I will be using for the cleaning procedure

Below, measurement of the nylon washers shows 0.78mm (not 0.3mm!)
[handy mind you as I will not have to do as much piling of washers]]

I have measured the depth a piece of the adhesive foam I used when I adjusted the height of the sensors last Wednesday. I used the micrometer to compress the foam to simulate the compression when I tightened up the strip. I got a measurement of approximately 0.5mm. I will still need to check this before deciding how many washers I will require (1 X 0.75mm or 2 X 0.75mm). If I lower the sensor strip by 0.75mm, this will cause the note-on point measured at the keyfront to show 1.5mm lower. Accordingly, lowering the sensor strip by 1.5mm will result in a keyfront note-on drop of 3.0mm. Bear in mind that the adhesive foam strip will have been removed and any adhesive residue cleaned so I will be packing from original factory height. More than likely I'll be using 2 x 0.75mm washers just to make sure.

Here's the math

Original note-on measured @ keyfront

(if compression of adhesive foam results in 0.5mm) = 4mm above key travel end

(if compression of adhesive foam results in 0.75mm in practice) = 4.5mm above key travel end

Lets take a 4mm note-on position for example. Note-off is 3.5mm above this so the key would have to come back up by at the absolute minimum, >7.5mm. That's >2.25mm higher than half way [actually almost 2/3's of the way back!]. Yamaha keytouch is set at 10mm. That would mean that if this were a Yamaha, I'd have to come back by at least 3/4's of the key depth before I could repeat! (Which by the way wasn't far from what I experienced with my P120S and CP33 also).

Now with this example, using 2 X 0.75mm washers, note-on is lowered by 3mm @ keyfront. Therefore, note-on position would be approximately 1.0mm. This is the lowest position I would allow. I would prefer 1.5mm note-on position. 1.0mm note-on will result in 4.5mm note-off (I would have to let key come up to only 4.5mm!) 1.5mm note-on = 5.0mm note-off. Both of these would allow for very rapid repetitions.

My only concern would be that note-on @ 1.0mm were too low. Obviously, if I set the sensors too low, I could get to the point where no notes trigger at all. This will not be the case, but the point is that I'm not sure how low I can comfortably go before I get missing notes in my playing simply because I'm not catching note-on? Therefore, I will not go lower than 1.0mm note-on.

Regards,

Chris

sigasa wrote:

May I ask anyone with a NUMA / NUMA Nero to measure the various deployments of the full keystroke for yourself. See at what point the note sounds when slowly depressing a key and how far it is between that point and 3mm from key travel end (escapement). Bearing in mind what I've said here, I think you'll be suprised at how early Fatar have set the strike point (note-on)!

Chris,

I do read all of your posts...I just don't respond if I have nothing relevant (or helpful) to add. I can tell you that I have been recording midi fervently for over two years now with my Numa Nero and I am getting the expressiveness I want...from triple pianissimo to triple forte, and everything in-between. For example, I play a version of Maple Leaf Rag (by Scott Joplin) that is very fast on purpose and I have not experienced anything out of the ordinary.

Do you have any particular fast passage that I could try on mine to get it to produce these 127 bursts? I have tried hand-over-hand single note rapid succession with no negative results, but maybe there is something in particular I can try?

The keys do seem a bit high to me compared to the travel range, and I measure my travel range (measured at the very front edge of the key) to be 13-14 mm to strike, and then another 2 mm  for aftertouch. Even with the key pressed all the way down (and into the AFT) I still have about 11-12 mm (or so) of the key showing. I don't recall what a "real" piano is like, but I thought there was more travel range.

My hope is that improved keybeds will be coming to the market soon.

First of all, a big thank you for reading my posts!

I am very pleased you are getting the expression you want from the Nero. It IS very dynamic and expressive - one of it's key features. I never had anything like this kind of range on my Yamahas (P120S and CP33). That was one of the first things that struck me when I got the Nero, the ability to bring so much more out of PianoTeq.

Maybe you could look at the repetition area in Rhapsody in Blue? But I don't think it will bring any problem to light on your Nero that hasn't already shown up. I do think however, that I need to get in there anclean out the sensors and switches/contacts with the proper stuff (isopropyl alcohol) rather than the baby wipes I used before!!!
This will sort out the errant velocity business.

As to the travel range, and the note-on and note-off trigger points within that range, it dawned on me this morning how to accurately calculate the distance by which the sensor strip needs to be adjusted. The distance between note-on and note-off points at the keyfront on my Nero is 3.5mm. Now all I have to do is measure the difference in length of the two carbon contact lugs located in the silicone switches and this measurement (which I suspect is around 1.75mm) can be used to calculate how much packing (0.3mm thick nylon washers) I need to place between the sensor strip and it's supports.

At the moment, I couldn't tell you how much the packing has raised the sensor strip because I used foam !!! (Daft eh?). What I shall do when I open the board to clean the strip and adjust the strip height is first of all measure the distance of the board from the metal base and then again when the foam packing tape is removed and cleaned up. The difference between these particular measurements will show me how thick the packing was and indicate to me how many washers I need to pack it to that distance. Then I suspect all I'll need to add on top of that is 2X0.3mm washers. This will lower the note-on / note-off points by approximately 1.2mm which will be perfect for the rapid repetitions. This will mean that note-on will occur at approximately 1.8mm above bottom of key travel and note-off 3.5mm above note-on at approximately 5.3mm. This means that the key only needs to rise to a minimum of >5.3mm rather than it's present minimum of >6.5mm. This makes a bigger difference than one might think.

Further testing last night revealed that I am actually getting better repetition since lowering the sensor strip a week ago? I think it was Wednesday last week?!!! Also some notes on rapid repetion now absolutely refuse to spurt out random errant 127 velocities no matter how hard I try to make them!!! As I've said a few posts back, I am now convinced this was a dirt/grime issue and that even after wiping with 'baby wipes!', there has been a significant improvement in performance. Lowering the sensors has nothing to do with this, it is just to aid ease of rapid repetition on a given note.

Regarding your measurements, you say you're getting 13-14mm before strike (note-on?) and 2mm aftertouch (escapement?). That's a travel of 15-16mm! Yamahas have 10mm! On my Nero I'm calculating 11.5mm full touch range broken down as 8.5mm before strike (note-on) and 3mm escapement so I'm not quite sure why your touch figures are so high? What I can concur is that the key top level is indeed very high, higher than a normal piano I am convinced!

Some of my washers have just now arrived. Just waiting for isopropyl alcohol and more washers and then I'll be ready to open up the board again. However, it may be next Wednesday before I do.

Regards,

Chris

May I ask anyone with a NUMA / NUMA Nero to measure the various deployments of the full keystroke for yourself. See at what point the note sounds when slowly depressing a key and how far it is between that point and 3mm from key travel end (escapement). Bearing in mind what I've said here, I think you'll be suprised at how early Fatar have set the strike point (note-on)!

Chris,

I do read all of your posts...I just don't respond if I have nothing relevant (or helpful) to add. I can tell you that I have been recording midi fervently for over two years now with my Numa Nero and I am getting the expressiveness I want...from triple pianissimo to triple forte, and everything in-between. For example, I play a version of Maple Leaf Rag (by Scott Joplin) that is very fast on purpose and I have not experienced anything out of the ordinary.

Do you have any particular fast passage that I could try on mine to get it to produce these 127 bursts? I have tried hand-over-hand single note rapid succession with no negative results, but maybe there is something in particular I can try?

The keys do seem a bit high to me compared to the travel range, and I measure my travel range (measured at the very front edge of the key) to be 13-14 mm to strike, and then another 2 mm  for aftertouch. Even with the key pressed all the way down (and into the AFT) I still have about 11-12 mm (or so) of the key showing. I don't recall what a "real" piano is like, but I thought there was more travel range.

My hope is that improved keybeds will be coming to the market soon.

"My Studiologic VMK188plus arrived last night, and I spent some time with it after rehearsal tonight. My patience has paid off, as this is the best keybed action I have ever felt on a digital keyboard. Very solid construction, no side to side wiggling, no annoying click sound or stickiness.

Just as the Keyboard review said, escapement is slightly slower than on some other boards like the CME's, but if you simply re-strike a bit further back on the key, there is no problem at all doing fast multi-strike patterns."

Pretty much explains why I have been in the process of lowering the strike point!!!

               "But if you want to do it better, you can check with a scale (= "ruler") to see that the rise of the hammer from rest position to its adjusted let-off (on bench again) is about 1 3/4 inches, which is the Steinway recommended figure. That will give a nice firm strike to each hammer stroke, and with the LET OFF (escapement) SET CLOSE TO THE STRIKE, the sound of the piano will be fine while the action is less heavy than with the two inch stroke which my piano had at one point."

After reading this again this morning, (bear in mind I heard the first bird tweet earlier as I went sleep!!!) I realised that having to set note-on position somewhere around 2, 2.5mm above key travel end/key bottoming out position is not going to be a problem like I thought it would be.

So in a correctly regulated piano (according to the above quote), escapement should be "set close to the strike." This minimises the delay between when the hammer butt leaves the jack and the hammerhead strikes the string making it possible to play fast repetitions. But note this - the strike point (note-on) in a real piano CANNOT POSSIBLY OCCUR   B-E-F-O-R-E    LETOFF / ESCAPEMENT!!! (you can never score a goal before you've kicked the ball - unless someone else kicks it for you!!! )

Then why is/was 'note-on point' factory set so high above escapement point in my (and may I presume, everyone elses!) NUMA Nero??? No wonder I had difficulty with rapid repeats!!!

So lets say for example that letoff occurs 3mm before key travel end. And letoff always occurs before hammer strike. Seeing as 2.75mm, 2.5mm, 2.25mm, 2mm, are all after the 3mm point, then I can set note-on at any of these points (or anywhere between these points) and acheive a genuinely convincing and accurate strike point. Of course I wouldn't want to go much less than 2mm, and preferably NOT less than 2mm. We don't want to have to dig into 'aftertouch' just to get a note-on now do we!!!

Frivolity aside, what it appears to me is that I'm actually setting up my keyboard as it should have been set up originally. 'Regulating it' if you will.

May I ask anyone with a NUMA / NUMA Nero to measure the various deployments of the full keystroke for yourself. See at what point the note sounds when slowly depressing a key and how far it is between that point and 3mm from key travel end (escapement). Bearing in mind what I've said here, I think you'll be suprised at how early Fatar have set the strike point (note-on)!

Regards,

Chris

Touch Depth = 11.5mm

Note-On point = 8.5mm down from keytop / 3mm up from end of key travel

Note-Off point = 6.5mm up from end of key travel / 5mm down from keytop

From these measurements it is clear that I can only lower the sensor strip a little.
The note-on point is already only 3mm from end of key travel so I think I need to aim for between 2 and 2.5mm, preferrably nearer 2.5 than 2, but I may need to set it nearer or at 2mm in order for the repetitions to work best. It would be a shame to lose any kind of sense of letoff/escapement/hammer delay etc. due to having to set the strip lower than I would have liked. The final decisions will be based on the present height of the sensor strip from the frame (as set last time). I shall measure the current height before removing any screws.

Regards,

Chris