Printable View
I'm very confused as to why this printer is using sound waves in order to transfer the data? Why can't it just work like a regular printer, and send the data directly to the printer like a PC send data to a regular printer? What are the benefits of using sound waves? Seems kind of unecessary to me. Am I missing something?
Derek
It's simpler. You need to convert the digital data into analog for the mirror actuators and soundcards, even integrated ones, contain very good (high resolution, low noise) digital-to-analog converters, so why not use that, rather than having to add more components and more cost?
yup as Compro01 has pointed out the system aims to use analog control over digital control, granted in most systems some DAC/ADC (digital to analog conversion/ Analog to digital conversion) needs to take place this is not always the case as some motors like stepper motors (more expensive then a simple motor) are driven from a digital signal. If any one is not sure of the area or is interested I would suggest spending a small amount of time looking into the differences between analog and digital control systems and sensors. If you do explore this area and take into account the design of the peachy in the process you will see how cleaver the idea of using an analog signal in this context actually is. One final thing I would like to point out (or make more confusing :D ) for anyone who is not used to this subject is that the question here is slightly misleading, the question asked about using a sound wave over data, a sound wave (an analog signal) is carrying data, I assume that the OP meant the data in this case to be a digital signal, any encoded signal is carrying data, even a hand gesture or facial expression :D.
(It looks like the question was answered while I was writing up my reply; I'm leaving it because it goes into a little more detail.)
It's not using sound waves exactly, it's just using the sound card to send an (analog) electric signal.
The printer has to somehow move the mirrors to control its laser. It does that using electromagnets, and it moves the electromagnets using an electric signal on a wire. It needs some way to convert commands that describe what to print into signals for the electromagnets.
Now it could do what you say: receive commands over USB (or some other digital interface with the computer) and have electronics on board that interpret that digital signal and convert it into an analog signal that controls the magnets. But just about every computer already has something that converts digital signals to analog: their sound cards. Sound cards generate an analog signal that moves an electromagnet in the speaker, which generates sound -- almost the same job the Peachy Printer needs done, just moving a speaker diaphragm instead of a mirror. By using a host computer's sound card to do the job, the printer doesn't need any electronics on board: it can just hook the electromagnets directly to the speaker wire. It ends up being a lot simpler and cheaper.
So in short, the sound card does a job that the printer would have to do anyway. By letting that job be done on the host computer, the Peachy Printer can be less expensive and less complicated.
And just as FYI, a speaker dosent get a "sound wave" too. he's moving to an electric signal to produce a sound wave! :)
The most popular option would be to use a micro controller, going with the sound card may have saved us a small amount of money but the true reason I stuck with the sound card is that in the 3rd world, people are far more likely to have a smart phone than a laptop.
So it was largely a moral decision to include everyone I could in the use of this printer!
It seems that 3d printers become even more use full if your in a country that does not have the infrastructure to provide you with lots of products.
Rylan Grayston
Rylan, you out of the box thinking is why I backed you. I look forward to the next year. Congratulations to you and your team
Quote:
Originally Posted by rylangrayston
I actualy thought about this just right now,
Will we be able to use speakers to listen to music as its printing? (Even tho we have 2 3.5mm audio jacks.)
I am guessing not? ^^
Not very likely. But if you have a smart phone, you can print from that and listen to music on your computer, or vice versa.
If I need to, I'll buy (or use an old one) a SmartPhone dedicated to the 3D Printer.
But there's more chance I'll fit the Pro version Microcontroller.
As you can imagine, the sound card won't be able to play music while it's busy printing (it won't even be plugged into the speakers at the point), but if you want, you can still play your music by using multiple sound cards. You can get a USB sound card for as little as $5. You can also use other devices to play music, such as your smartphone or CD player.
JamesCooper: Will the software "take over" the sound card to prevent any sound? Like if you receive an email ?
That's a great question. I don't think that's actually something a program has a choice over -- the OS typically deals with mixing multiple audio streams into a single output. I'll keep in mind to see if there's a way to force exclusive output, but I'm confident that there isn't a portable way to do it -- certainly not with the PortAudio portable audio library I'm currently using. I know Linux and Windows both have relatively simple interfaces for the user to mute other programs and system sounds. You can also programs to play to specific audio devices (not necessarily the default), so a second sound card would again be a way to ensure that isn't a problem.
Just curious - is there any way to have the Peachy utilise some of the unused channels on most soundcards? Pretty much every soundcard (or mainboard with integrated sound) has at least six channels these days (eight is more common) and yet a lot of people only use two (headphones or standard 2-channel desktop speakers). Could the Peachy just grab some of the channels that are meant to be for surround-sound speakers and leave the main two channels for everything else? That way you could keep using the soundcard while the Peachy was working.
Slatye: It looks like it might be possible in some cases, but not in a general, portable way. Windows seems to give you the option of specifying which channels to use, which would allow you to play on only the rear channels, while programs would normally request only the stereo (front) channels. At the same time, if Windows was configured for 2 speakers, it would copy the front to the rear, defeating the purpose, so people would have to configure their computer appropriately. SDL has support for surround sound, but it doesn't allow using only a subset of channels. I suppose Peachy could just avoid playing on the front channels, though there's still no way to ensure another program isn't writing to those rear channels).
Honestly though, this is much ado about nothing. Expect that the Peachy will require the dedicated attention of the sound card while it is printing. If that's a problem for you, I recommend buying a cheap USB sound card.
USB soundcard seems to be the best solution. Or even better with a Microcontroler! :)
Another solution for those running Windows Vista, 7, or 8 is to right click the Volume icon in the taskbar while potential sources of sound are in use (example: watching youtube with your broswer, listening to music with windows media player, etc). From there you can open the Volume Mixer, and adjust an individual program's volume. Just set your browser, video games, music, and flash player to mute, and you are all set to do whatever you want while your system prints.
Really excited to be a part of the Peachy as a beta backer. I'm especially interested in the electronics now it has progressed from a DC coupled soundcard into some circuitry with an AM demodulator and I'm wondering if you've somewhat overestimated the cost and benefits of a complete micro system compared to what you now have....
I've had some previous experience with STM32 micros and I've currently got my eye on the STM32L100R8T6 which is under £1 (+ resistor, linear reg, crystal, a handful of caps and resistors) and would provide you with USB data connection and two 12Bit DACs onboard (I suspect you would want cheap buffer amps to drive the mirrors). It should be relatively easy to get a few 10's of kpoints/s on each DAC channel from USB you would also then get all the device controls via this mechanism; control for laser, drip feedback, calibration storage on device.
This might also offer an easier upgrade path for the basic unit, it would be easy to incorporate on the PCB but not populate a better DAC, SD storage, LCD status, user controls/feedback etc.
This of course could work with Smartphones and Tablets that have USB host capability which are becoming far more common.
I thought the entire system was dependent on microphone feedback from a capacitive sensor that discharges every time the resin drips through? How will it work well via smart phone if it does not have the separate I/O channels to control the laser and the resin level? I assume it's not needed if the drip reservoir is a specific shape and volume but I thought that was intended to be a separate part and not included with the kit?
I think it works by reading a certain part of the 3.5mm phone connector.
http://i.imgur.com/DsZ9cZ7.jpg
Ever notice how some of the earphones included with smartphones has a microphone in it? Most smartphone "headphone" jacks are actually TRRS connectors. They function as both headphone jacks and microphone jack.Quote:
Originally Posted by BrockMcKean
Hmm, I wonder whether you could use the phone's camera instead. Just mark the container at, say, 10mm increments (for calibration), and then point the phone at it. Easy to use a threshold to detect the resin (assuming it's one of the brighter colours), and that gives you both the water level (bottom of the resin layer) and resin volume remaining (total height of the resin layer). The advantage would be that it provides an absolute height - whereas the microphone method is an integral and may or may not suffer from drift.
@Compro01. Ah! I see. I wasn't aware of this. I don't know very much about peripherals of these kinds of handheld devices, but I guess it makes perfect sense with products like Square's card reader and PaypalHere. I wonder if the noise is approximately the same on phones, and from phone to phone? I know in many laptops the output and input can be very noisey depending on whether it is plugged into a charger or not. If this is a real concern, to print identical objects with identical resolution from multiple devices, more hardware and a filter would be necessary.
@Slayte. Using the phone's camera would largely depend on several things that would be basically impossible to configure identically. Consider the following:
1) Position of the lens is not identical across most phones. You would have to have a way of ensuring the camera was positioned identically every time you began a print. This would mean a physical device to ensure it is physically some defined distance from the printer and oriented identically as well, probably orthogonally. You could attempt to account for the differences in position and orientation through software, but it may not be possible depending on the distance and orientation of the camera. Some extreme examples would be if it was pressed directly to the print area so that the entire reservoir was outside the scope of the image, or if it was turned away. Even neglecting these rather macroscopic concerns, if the resolution of the printer are to approach several hundred micrometers, this is tenths of a millimeter. 200 um = .2mm, so your 10mm markings would have to turn into 0.1mm markings.
2) Quality of the image and/or video is not the same. Most newer smart phones are ~20MP, but not all are this quality and some regular cellular phones are significantly lower. I'm not going to go into a physics and geometry explanation of this particular problem, but basically MP is a measurement of how many millions of pixels a camera produces in a single image. For a given object at a given distance with a given number of MP, you have a resulting actual size pixel dimension. This means, one phone may see 0.1mm where another phone may see 0.2mm depending on all of these things, assuming it is even high enough resolution to see 0.1mm markings and when liquid rises beyond it. The same problem of reading a beaker or graduated cylinder (the miniscus) would be present here as well. The surface tension of the resin would cause the mixture to cling to the sides and this would likely make it very hard to read in 0.1mm increments.
3) Lighting changes reflective properties of objects, especially liquids. As you noted, you may have to use a colored resin for this kind of approach. What if your resin is red and the background is red? What if the lighting is such that the surface of the liquid is very very hard to see? For example, in ultra low light conditions when you turn your lights off and the print continued?
There are too may design problems to correct with this method, in my opinion. Not to mention the same phone, and nearly all phones use this same TRRS connection Compro mentioned. The prints would be wildly unreliable and differ greatly for each camera device used.
You know it's going to sound like your commodore64 tapes loading!Quote:
Originally Posted by DrLuigi
Just hook up a Y connector.Quote:
Originally Posted by DrLuigi
Y wont work.. you can't play music while printing, except if you have a 2nd sound card, like a USB SoundCard.
Nevermind. I was misreading what Dr. Luigi was saying as a result of what Geoff said. I was thinking Luigi was saying something about listening to the sounds the printer is operating off or something.Quote:
Originally Posted by nka
Oh!
But with a Y, you could print 2 time the same things at the same time! :D
I've been uploading a bunch of videos of my Makibox printing, and YouTube keeps taking them down because apparently they sound like copyrighted music. Maybe the Peachy will be the same, but it's even easier to record the music it produces!
It'd be an interesting bit of steganography - who can design an object that results in the Peachy software generating music? Can the object be banned for copyright infringement?
Very well worded, infact our instructions will say exactly this!Quote:
Originally Posted by Anuvin
So many interesting, unintended consequences.Quote:
Originally Posted by Slatye
Governments may let people enforce all kinds of things, The question is dose it make sense....
All I know for sure is i love that you brought this up :)
Seems unlikely, given that the Peachy runs on amplitude modulation. You'd only get the same note over and over, except for the frequency shifts used to shut off the laser.Quote:
Originally Posted by Slatye
Ah, yes. Good point. Still, I'll bet that YouTube's automated "sounds like copyrighted music" system will find something similar and happily remove the videos.
I am more interested in the reverse relationship. Can you make a song that results in the peachy printer printing an object? Can you secretly embed an object in a song such that you wouldn't know it was there when you listened to the song?
I know nothing about this, but the guy who makes music as Aphex Twin embedded an image of his face in a song that could be seen in an spectrgraph. Well written article here http://www.bastwood.com/?page_id=10
That's a pretty cool idea. Presumably if the Peachy's amplifier has a pretty tight band-pass filter on it then you might be able to do something like that.
Obviously scale would be completely messed up. The X-Y scale depends on how high the Peachy is above the resin (which depends on the installation) and the Z scale depends on the drip rate. However, many objects (eg. a 3D model of a face) would be recognisable even with very substantial scale errors.
I suspect the bigger indicator that there was extra data embedded would be the length of the track - for a nice Peachy print you'd probably have tracks at least half an hour long. That'd be suspicious when most tracks are more like 3 - 5 minutes.