An RPi would be the way to go if you'd want to make a setup for adoption by many users. Getting an RPi as a controller for the entire system is cheap, and you could focus on making software run smoothly on just the RPi.
I wouldn't be surprised if the RPi would be totally fine to also run the data processing.
I wouldn't be surprised if the RPi would be totally fine to also run the data processing.
Easily even.I wouldn't be surprised if the RPi would be totally fine to also run the data processing
Probably even the RPi Zero will be totally fine.
I have no direct experience, but found this discussion of the R Pi Pico and its application to audio. Here's a snippet:
The Raspberry Pi Pico is a microcontroller board released in 2021 that has proven very popular for its small size, powerful features and affordability. Measuring just 6x20mm, the Pico packs a capable dual-core microprocessor and ample I/O pins into a tiny package. At its heart lies the RP2040 chip designed in-house by Raspberry Pi which includes specialized hardware for digital signal processing. With its DSP capabilities, versatile programming options and thriving ecosystem of add-ons, the Pico has emerged as an accessible development platform for designing embedded DSP applications. The dedicated DSP block contained within the RP2040 makes the Raspberry Pi Pico well-suited for Raspberry Pi Pico DSP tasks involving real-time audio, image and signal processing.
The Raspberry Pi Pico is a microcontroller board released in 2021 that has proven very popular for its small size, powerful features and affordability. Measuring just 6x20mm, the Pico packs a capable dual-core microprocessor and ample I/O pins into a tiny package. At its heart lies the RP2040 chip designed in-house by Raspberry Pi which includes specialized hardware for digital signal processing. With its DSP capabilities, versatile programming options and thriving ecosystem of add-ons, the Pico has emerged as an accessible development platform for designing embedded DSP applications. The dedicated DSP block contained within the RP2040 makes the Raspberry Pi Pico well-suited for Raspberry Pi Pico DSP tasks involving real-time audio, image and signal processing.
I was making some progress. I could communicate with the tic, the arduino and the audio interface. However, at the moment only one of the usb ports of the RPI is functioning. Searching the internet for non-functioning usb ports results in a lot of hits.
Does anyone here have any idea what to do? What to check for?
Does anyone here have any idea what to do? What to check for?
After seeing Tom's prptotype it made me think of a different approach to how the movement could be accomplished.
I now like the idea of using a Core XY belt system (like is used in drawing robots), just more sturdy and reinforced in the up down axis.
The in and out movement could use a very similar setup with linear rods as the weight of the mic is low. The mic would be attached in place of the pen.
This keeps the motors on the fixed parts of the chassis and wiring would be mostly fixed in place.
I now like the idea of using a Core XY belt system (like is used in drawing robots), just more sturdy and reinforced in the up down axis.
The in and out movement could use a very similar setup with linear rods as the weight of the mic is low. The mic would be attached in place of the pen.
This keeps the motors on the fixed parts of the chassis and wiring would be mostly fixed in place.
So a motor on the bottom near the floor, a motor on the top and the mic boom gets actuated back and forth. You can get high lubricity plastic bearings that slide well along those rods if you want to 2D print the rest.
https://www.aliexpress.com/w/wholesale-plastic-linear-bearing.html?spm=a2g0o.home.search.0
https://www.aliexpress.com/w/wholesale-plastic-linear-bearing.html?spm=a2g0o.home.search.0
Yes, depending on which direction each motor is moving changes the motion from up down to left right or some combination of the two. Core XY is supported by GRBL as well so it should still fit within the same setup as Tom has been using.
Here is a version on a pretty low budget, I would prefer to make something more rigid, there are off the shelf linear rods and aluminium housing ball bearing slides from the usual CNC vendors.
https://www.instructables.com/Dual-Task-CoreXY-DrawBot/
Here is a version on a pretty low budget, I would prefer to make something more rigid, there are off the shelf linear rods and aluminium housing ball bearing slides from the usual CNC vendors.
https://www.instructables.com/Dual-Task-CoreXY-DrawBot/
There are already many designs using off the shelf parts that could be followed or altered. The scond link is about as simple as it could be.
https://makerhardware.net/wiki/doku.php?id=manuals:drawbot_v1.2
https://cs448m.github.io/assignments/tbot/
https://makerhardware.net/wiki/doku.php?id=manuals:drawbot_v1.2
https://cs448m.github.io/assignments/tbot/
Hi guys,
just a helpfull pointer,
if you buy this as pdm to usb convertor;
https://www.audiohobby.eu/en/usb-au...multi-channel-asynchronous-usb-interface.html
with 4x this;
https://nl.mouser.com/ProductDetail...130FLEXKITTOBO1?qs=t7xnP681wgXdd%2BkWvZ2roA==
you get 16 channels of very high qualitymeasurement mics for +- 250 euro
you can even use 2 mchstreamers at once to get 32 simultanious channels.
since all mics are very light and small you can move them around very fast.
only thing to watch out for, use short, twisted pair, shielded cables for the pdm signals.
best regards,
Kees
just a helpfull pointer,
if you buy this as pdm to usb convertor;
https://www.audiohobby.eu/en/usb-au...multi-channel-asynchronous-usb-interface.html
with 4x this;
https://nl.mouser.com/ProductDetail...130FLEXKITTOBO1?qs=t7xnP681wgXdd%2BkWvZ2roA==
you get 16 channels of very high qualitymeasurement mics for +- 250 euro
you can even use 2 mchstreamers at once to get 32 simultanious channels.
since all mics are very light and small you can move them around very fast.
only thing to watch out for, use short, twisted pair, shielded cables for the pdm signals.
best regards,
Kees
Have you managed to get this working with any of the ideas you posted in the ASR thread?you get 16 channels of very high qualitymeasurement mics for +- 250 euro
You could also have the top motor replaced by an axle and have a second belt on the other side and place that motor also at the bottom. Or even fancier: belts all the way to the central turntabel. keeps all the mass centred.
For now I'm happy with what I have. I first need to find out why my usb ports don't work anymore. I already ordered some parts to test for back voltage or some grounding issue.
For now I'm happy with what I have. I first need to find out why my usb ports don't work anymore. I already ordered some parts to test for back voltage or some grounding issue.
Well I just checked the number and it's about $740 Canukistani shekels for 16 channels. The idea is interesting. A 16 mic array would definitely speed up the test procedure. But I am not sure of how to call out to each mic. I am guessing the wiring is rather simple. It looks like each Evaluation module has solder tabs and most likely a connection header connection point on the multiplexing card.Hi guys,
just a helpfull pointer,
if you buy this as pdm to usb convertor;
https://www.audiohobby.eu/en/usb-au...multi-channel-asynchronous-usb-interface.html
with 4x this;
https://nl.mouser.com/ProductDetail/Infineon-Technologies/EVALIM67D130FLEXKITTOBO1?qs=t7xnP681wgXdd%2BkWvZ2roA==
you get 16 channels of very high qualitymeasurement mics for +- 250 euro
you can even use 2 mchstreamers at once to get 32 simultanious channels.
since all mics are very light and small you can move them around very fast.
only thing to watch out for, use short, twisted pair, shielded cables for the pdm signals.
best regards,
Kees
I'll look up those mics and see what their response looks like. Potentially a calibration curve against a known good mic is possible. The only thing that I would wonder about is distortion. I am in the midst of doing some comparative measurements with this exact aim in mind. A few of my mics in a bit of a shoot out to see which ones are the lowest distortion of the lot. Piezo electric or metal diaphragm or electret.
The mics are exellent for this puspose, they completely match up to the spec ( see datasheet)
I have not used these in the context of a nfs system ( life happened, never managed to spend more time on it sadly) but I have used them on other mic array projects.
the minidsp asio driver is excellent, you just get a 16 channel audio interface in your pc which you can record.
the best thing about the mems mics ( apart from their tiny size) is their tolerance, they are all +-1dB out of the box, which is unheard of with electrecs etc.
This is all to thank to the automotive and home alexa/google stuff; they all use mic arrays and dont want to do calibrations for obvious reasons.
I have not used these in the context of a nfs system ( life happened, never managed to spend more time on it sadly) but I have used them on other mic array projects.
the minidsp asio driver is excellent, you just get a 16 channel audio interface in your pc which you can record.
the best thing about the mems mics ( apart from their tiny size) is their tolerance, they are all +-1dB out of the box, which is unheard of with electrecs etc.
This is all to thank to the automotive and home alexa/google stuff; they all use mic arrays and dont want to do calibrations for obvious reasons.
https://www.infineon.com/dgdl/Infin...N.pdf?fileId=5546d462602a9dc801607a0e46511a2e
A quick look and I see about $740 Canadian dollars in a 16 channel setup. Not cheap. But also not crazy. Each Evaluation module is $40 plus shipping. If 16pcs are ordered I don't think shipping is charged. But that's $540 plus the $156 MiniDSP multiplexer plus shipping.
Interesting idea. More or less twisted wiring bundles and you connect to a header on the MiniDSP board. As long as you are careful in your lay out. This certainly could speed up measurement.
But we need to put the horse in front of the cart first. Tom almost has a functional system. We need to first be patient. ANd be as useful as we can on this side. Not over burden the guy!
A quick look and I see about $740 Canadian dollars in a 16 channel setup. Not cheap. But also not crazy. Each Evaluation module is $40 plus shipping. If 16pcs are ordered I don't think shipping is charged. But that's $540 plus the $156 MiniDSP multiplexer plus shipping.
Interesting idea. More or less twisted wiring bundles and you connect to a header on the MiniDSP board. As long as you are careful in your lay out. This certainly could speed up measurement.
But we need to put the horse in front of the cart first. Tom almost has a functional system. We need to first be patient. ANd be as useful as we can on this side. Not over burden the guy!
https://www.ita-toolbox.org/
I just learned about this. Might be of use to some of the people on this thread.
I just learned about this. Might be of use to some of the people on this thread.
Missed that part. Thanks for the direction. So the mounting board is what we would need to make duplicates of to envision your idea.each evaluation module = 5 mics
That is what my lighted magnifier is for. I used to do this with ease. 30 years ago I worked as a repair tech for about a year in between contracts. Now my eyes need help. But, Can still be done! So truly this is something to consider once we have a good programmable movement system. I like the idea.No, you can just solder the wires directly to the flexpcb header pins ( that is if you're comfortable soldering 1mm pitch)
- Home
- Design & Build
- Software Tools
- Klippel Near Field Scanner on a Shoestring