[sdiy] Very cheap SPI pressure sensors.. anybody ??

Wed Mar 24 18:43:43 CET 2021

>I take it back, actually. You could do exactly what I propose with very light foil suspended so it can move depending on the air. It would be extremely, extremely cheap, and given you >only really need on/off >you should be perfectly fine there.

Yeah.. I was trying to figure out some vacuum/mechanical movement sensors
that would do exactly what you bring as a solution..

But how to achieve that ??
I thought about a small moving stick that is inside a vertical hose
resting down when no air flows but sucked up on any vacuum.
Then optics could check for any ON/OFF movement..

De : cheater cheater [mailto:cheater00social at gmail.com] 
Envoyé : 24 mars 2021 13:10
À : Jean-Pierre Desrochers
Cc : Didrik Madheden; SDIY List
Objet : Re: [sdiy] Very cheap SPI pressure sensors.. anybody ??

I take it back, actually. You could do exactly what I propose with very light foil suspended so it can move depending on the air. It would be extremely, extremely cheap, and given you only really need on/off you should be perfectly fine there.

On Sat, Mar 20, 2021 at 5:30 PM cheater cheater <cheater00social at gmail.com> wrote:

oh, you meant air pressure sensor... sorry... i thought you meant polyphonic aftertouch.

On Fri, Mar 19, 2021 at 10:34 PM Jean-Pierre Desrochers <jpdesroc at oricom.ca> wrote:

>As for the option of light sensing, the assumption of freedom to design, would allow you to simply have an array of LEDs directly on the top and a array of sensors directly under, >mounted on boards. To me at >least, using optical fibers for this purpose and placing the sensing elsewhere seems like an overcomplication. Unless there's some design requirement I'm not seeing

An optic based reader would look about like this:

Notice the big horizontal LEDs bar over the row of holes in reading..

Not very nice..

On the other hand a vacuum based reader would look like the original thing:

The rolling mechanism will be using a DC brushless speed controlled motor
connected to the right gears assembly.

>I would consider designing boards where you have the sensors in a row on the board. You could for example have 15 sensors per board and 6 groups for the target of of 90. You could >then either >have a multiplexing arrangement like the one I suggested, or a microcontroller on each board for local processing. 
This is about the way I was thinking to do it..

>You could then have a physical piece (3D printed or >otherwise fabricated) that fits over the board to create a chamber for each sensor.

Very good idea ! My first ideas were to glue very small silicon hoses over each sensors
but a separate chamber for each sensor is much better..
These chambers could have a small pointing upward ‘chimney’ to insert
the vacuumed hoses. About like this housing:

And by the way, this reader will have its own audio player
playing ‘honkytonk’ sounding notes with a MIDI output !

De : Didrik Madheden [mailto:nitro2k01 at gmail.com] 
Envoyé : 19 mars 2021 16:13
À : Jean-Pierre Desrochers
Cc : SDIY List
Objet : Re: [sdiy] Very cheap SPI pressure sensors.. anybody ??

So that gives me one important piece of information: the sensors could be placed in a linear pattern, as opposed to, for example, being spread out and needing to be wired up individually. 

I would consider designing boards where you have the sensors in a row on the board. You could for example have 15 sensors per board and 6 groups for the target of of 90. You could then either have a multiplexing arrangement like the one I suggested, or a microcontroller on each board for local processing. You could then have a physical piece (3D printed or otherwise fabricated) that fits over the board to create a chamber for each sensor.

I'm assuming here that this is a machine that you are either designing from scratch, or have relative freedom to modify. In particular that you control fully what's above and below the line of notes that's currently being read. But from the rest of your description I realize that this may not be the case. Are you for example retrofitting an old machine, or you feel that you want to be true to the original principles? Or do you just want a solution that works for as cheap as possible?

As for the option of light sensing, the assumption of freedom to design, would allow you to simply have an array of LEDs directly on the top and a array of sensors directly under, mounted on boards. To me at least, using optical fibers for this purpose and placing the sensing elsewhere seems like an overcomplication. Unless there's some design requirement I'm not seeing...

/Didrik

On Fri, Mar 19, 2021, 20:27 Jean-Pierre Desrochers <jpdesroc at oricom.ca> wrote:

Good idea here Didrick !
But a little too complicated for my design..
This futur project is a piano roll scanner/player.
This has been a project done many times by DIYers all over the place
but they mostly used optical approach to read paper holes
(fiber optics + photo-transistors).
This needs a LED array bar put over the rolling paper holes rows
for the fiber optics holes detections,  which I don't like at all..
I'd like better to read the punched holes using vacuumed air
like antique piano did. A small vacuum pump will do that.
This is why I'd use pressure sensors to detect any holes
coming up while the paper roll is playing.
But I'm still looking any design possibilities.
I'll need around 90 sensors (SPI communications) to get the job done.
This will obviously need some multiplexing to access each of them.
Doable..

-----Message d'origine-----
De : Didrik Madheden [mailto:nitro2k01 at gmail.com] 
Envoyé : 19 mars 2021 14:16
À : Jean-Pierre Desrochers
Cc : SDIY List
Objet : Re: [sdiy] Very cheap SPI pressure sensors.. anybody ??

Here's an idea. It's a bit janky, and requires some extra circuitry, but it might work for you. The basic idea is to use a chain of 74hc595 serial to parallel shift registers to generate either a chip select or clock signal. You feed in a single 0 into the shift register, which you move forward one step at a time.

In the case of a chip select signal, this selects one chip at a time.
To select the next chip, clock the register chain one step to forward the 0 bit one step.

In the case of a clock signal, it's a bit more complicated. But this could be used to multiplex I2C. In this configuration, all the data pins are commoned. (Or commoned in multiple groups if needed for signal integrity.) You have a similar idea as in the example above in that a 0 bit travels one step at a time. You could do this the slow way and send out the full chain of bits just to flip one bit and clock the local I2C bus. Or, you could do it the slightly quicker way and use the output enable of the 74HC595's to clock the selected device.
It would work like this:

Most bits in the register are ones. For those bits, turning the output enable on or off would do nothing because you're switching between driving the clock signal high, or leaving the clock signal with a pullup resistor which also keeps at logic high.

One or multiple bits are zeros. When output enable is off, that output is high from the pullup that you add, or which exists on the module you use. When you turn on the output enable, voilà! The clock signal for that device goes low, and you have fed clocked one I2C bit. I think you can fill in the rest. All the other devices will also see the data line change, but they're in high impedance mode, and will not react to data without a corresponding clock.

If you are able to design a board and have it ordered from JLCPCB using their PCB assembly service, you could use SPL06-007. It costs, currently, $0.6622/pc in 100 quantity. There's some extra cost for the board and assembly service, but that should be relatively cheap in comparison.

To give even more precise advice, it would also help to know more about the details of the project. In particular the physical arrangement of the sensors. If they need to branch off from a central point using cables, if the cables need to be detachable etc. If you want to discuss these things closer in private, feel free to contact me off-list.

/Didrik

On Fri, 19 Mar 2021 at 16:14, Jean-Pierre Desrochers <jpdesroc at oricom.ca> wrote:
>
> I'm looking for very cheap SPI pressure sensors (around $1.00 a piece 
> or less)
>
> For a future project. The pressure value doesn't matter but the size and cost do.
>
> I will need around 90 of these sensors.
>
>
>
> For example I found this:
>
> BMP180 GY-68 GY68 3.3V 5V BMP-180 Temperature Pressure Sensor Module 
> Barometric IIC I2C Interface Sensor Module Replace BMP085  $0.90CAD
>
> https://www.aliexpress.com/item/32832330585.html?src=google <https://www.aliexpress.com/item/32832330585.html?src=google&albch=shop> &albch=shop
> ping&acnt=708-803-3821&isdl=y&slnk=&plac=&mtctp=&albbt=Google_7_shoppi
> ng&aff_platform=google&aff_short_key=UneMJZVf&&albagn=888888&isSmbAuto
> Call=false&needSmbHouyi=false&albcp=7386552844&albag=80241711349&trgt=
> 743612850714&crea=en32832330585&netw=u&device=c&albpg=743612850714&alb
> pd=en32832330585&gclid=CjwKCAjw9MuCBhBUEiwAbDZ-7mU2hTS0J2DLxvsPsj8F40L
> -zV8vtApw9ds_ks1c2Qjg7dLiKgsy3RoCq-EQAvD_BwE&gclsrc=aw.ds
>
> But because I need 90 devices and the BMP180 is I2C and has only one 
> address value I would need somekind of multiplexer to address 90 devices (one at the time obviously).
>
> So.. SPI devices would fit better here.
>
>
>
> One other possibility would be to use this type:
>
> MPS20N0040D MPS20N0040D-D Sphygmomanometer Dip Air Pressure Sensor 0-40kPa DIP-6 For Arduino Raspb   $0.74CAD
>
> https://www.aliexpress.com/item/4000354356373.html?src=google <https://www.aliexpress.com/item/4000354356373.html?src=google&albch=sh> &albch=sh
> opping&acnt=708-803-3821&isdl=y&slnk=&plac=&mtctp=&albbt=Google_7_shop
> ping&aff_platform=google&aff_short_key=UneMJZVf&&albagn=888888&isSmbAu
> toCall=false&needSmbHouyi=false&albcp=9604672960&albag=100679328364&tr
> gt=296904913880&crea=en4000354356373&netw=u&device=c&albpg=29690491388
> 0&albpd=en4000354356373&gclid=CjwKCAjw9MuCBhBUEiwAbDZ-7kff0q--0F6YUo73
> aVqE_HAtqkRyIslFbEMRTWz7QICR2szctlC7zxoCrMUQAvD_BwE&gclsrc=aw.ds
>
>
>
> But I’ll need some analog IC to manage the output..
>
> Again the part’s pressure specs value do not matter.. I will manage a way to use it.
>
> but the size and cost do.
>
>
>
> So far the best would be a SPI equivalent of the BMP180 IC..
>
> Let me know if someone googled something I did not spot !
>
>
>
> JP
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy
> Selling or trading? Use marketplace at synth-diy.org

--
/Didrik

_______________________________________________
Synth-diy mailing list
Synth-diy at synth-diy.org
http://synth-diy.org/mailman/listinfo/synth-diy
Selling or trading? Use marketplace at synth-diy.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20210324/82f27c2a/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.jpg
Type: image/jpeg
Size: 17467 bytes
Desc: not available
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20210324/82f27c2a/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.jpg
Type: image/jpeg
Size: 12019 bytes
Desc: not available
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20210324/82f27c2a/attachment-0001.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image003.png
Type: image/png
Size: 41748 bytes
Desc: not available
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20210324/82f27c2a/attachment.png>