Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards Yamaha P105 teardown

Hey everyone,

I tried to look for a couple of pictures of the P105 guts because I was looking for ways to make it lighted/foldable/shorter so it is easy to take for travels. I didn't seem to find any photos of a teardown so decided to do it myself ;-).

As you might see, there is almost no way to make it foldable as the key assembly is another unit to the chassis. it will be too fidgety to muck around with.

Apart from taking the speakers out, there is nothing really for weight savings. I'll be only using headphones anyway so it's ok without speakers.

Anyway, here are some pics.

Wow, some great pictures!

dewster's going to love you.

James
x

Incredible shots. Hope ya can reassemble it . . . If I`m seeing things correctly, the keys don`t have a pivot as such; they bend as in a cantilever action, about the thin section near the end . . .Doesn`t auger well, but I`ve used this action for years without any problem . . . gulp

Thanks for pics. How much weight did you save by removing the speakers?

Recently I wondered if it is theoretically possible to cut off a few keys of p-35 and adjust the casing. What you think?

Great pictures.

As I was inspecting the innards it dawned on me, if someone did not like the onboard speakers, how easy would it be to replace the factory installed ones with something bigger (wattage-wise)?

Thanks for taking one for the team zerox61! You beat me to it.

Added to the Nekid Pichures thread.

Wow, the keys are really unexpected: switches on the key rather than the hammer, the hammer appears to be somewhat captive - and that funky, funky hinge! Yuck! I suppose we all knew the fulcrum to playing end distance was pretty short, it's quite obvious if you play it.

Tweeters look like they are aimed up into empty space? The fairly cheap looking speakers could be replaced, but they have semi-custom metal frames which could complicate things. The trick would be to find something more efficient with better frequency response in an open baffle setting (but after seeing the keys I'm thinking why bother). I'm moderately curious as to the woofer / tweeter crossover component(s) as this doesn't appear to be captured by the pix. Full range speakers like this don't tend to benefit much from the addition of tweeters, I'm kind of surprised they even bother but I suppose it looks good on the spec sheet.

It's like someone woke up with a severe hangover / radiation poisoning and designed the thing in like 5 minutes.

yeah, the mini speakers/tweeters are blasting into empty space, there is no box but it is pointing up to the metal grills on the DP.

It is now reassembled and plays exactly the same as before, nothing is broken

I didn't end up removing the speakers. I estimate 1kg for the four speakers. A 1kg saving didn't mean too much to me so I left the speakers inside :P.


As for that hinges.. I think I have seen this in my Casio WK3200 and it is the same style as the P105

Ha! Dewster, you're an incorrigible but cheerful cynic ....or simply a purveyor of the truth?

ICs:
- ESMT M12L64164A-7F : 4M x 16 SDRAM
- ESMT M12L128168A-7F : 2M x 16 x 4 SDRAM
- cFeon EN29LV640B-90TIP : 8MB FLASH
- Spansion S29GL128S90TF102 : 16MB NOR FLASH

I believe the cFeon part holds the processor instructions and the Spansion part holds the data (piano, etc.) samples.

For 16 bit stereo, 44.1k sample rate, this gives:

16MB / 2 bytes/sample / 2 stereo samples / 44100 samples/sec = 95 seconds of sample time. /88 = 1.08 seconds per note. To get any kind of realism out of it requires a stretching factor of at least 3x (and indeed there are 29 stretch groups on the P105 : 88/29=3.03).

The Spansion FLASH is at mouser for $2.49 a pop (quantity 1). A company like Yamaha could almost certainly buy these for a dollar or less.

Imagine if Yamaha had sprung $8 (quantity 1 at mouser) for the S29GL01GS10TFI010 1Gb part with 8x the room? ~8 seconds of sample time per note could remove most of the stretching, could relieve much of the horrid looping, and might allow another layer or two. But no, we can't have this because the lowly P105 would then be internally competing with other, higher end, Yamaha DPs (which are similarly hobbled with too-small sample memory).

And I wonder why I'm having trouble getting it up for the DPBSD project lately. It's this same old, same old grind of intentionally inferior products that never ends in this stupid market. Every NAMM is like a smack in the face. Beautiful CFX sample sets butchered down to <16MB (~1/1000 original size).

I think you may have invalidated your warranty :-o

"Beautiful CFX sample sets butchered down to <16MB (~1/1000 original size)."

Because some of us can`t tell the difference . . .

I guess you had to do a lot of work to get there, but you've at last arrived at a clear conclusion: It's all crap. The high-end goods may stink a bit less, but they all rather stink, eh?

Some day I'd like to have an actual digital piano designer/engineer answer these types of questions. As far as we can see they are working with decades old constraints using embarrassing technology, but we don't know all the details of what the reasons for this are, and presumably there are good reasons for all these types of choices.

I'm guessing that the reasons are marketing related. That is, if people will buy the stuff, they'll keep making it. (The designer/engineer can't speak to that, eh?)

Same here. Though I wonder how many "actual digital piano desisners/engineers" even exist anymore. The inmates (marketing types) are running the asylum.


Some of it might be that the engineers who have been doing this for decades are slow to change. Another might be that sudden technology upgrades tend to strand inventory.

I think two of the biggest feeders are: (1) old technology is cheap, and (2) there are diminishing returns in larger sample sets, particularly with an uninformed public doing the lion's share of the purchasing. This unfortunately leaves DPs stuck in something of a time warp (cue The Rocky Horror Show).

Great set of pictures. The gray rubber contacts look a lot like the ones in the VPC1, except two contacts instead of three. I wonder if they're from a common vendor, or if there's just one good way to make them.
Thinking about the tweeter in the middle of the box- maybe it gives some diffusion of the sound, making it less like a point source.

I've wondered this too. They always look just the same, regardless of the brand. I'm thinking there's only one easy way to make them and everyone follows the same basic pattern but I don't know.

Hello, couple of points- questions-
I assume the SDRAM is for the song recorder?
If indeed it is 16mb total (not that memory is everything of course) then my 1998 Korg SG Rack had the same amount pretty much with 15mb for the main piano sample.
(I wonder how much the SP280 uses, or any of the recent Casios)

Also, curious as to how much you think the speakers and amplifiers in this unit cost?

My P95 had a so-so piano sound but probably $6 OEM speakers.

Also, to the OP- before or after you put it together, are you experiencing any of the rattling that occurs at the treble end of the keyboard when playing both hands below middle C?

Looks like the Yamaha I am using for now is the 1.3gb Yamaha U3 from addictive keys!

And it is driven my marketing and how much they can get away with and how naive the consumer is.
I made 2 bad purchases- luckily returned them both

There were two big squishy foam parts on either sides of the "action" assembly when it is assembled to the frame. This prevents the rattling sounds on the ends.

you can see the big foam on the frame in picture #6