• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Greetings from a New Member

cshake

Member
Forum Donor
Joined
Jul 20, 2019
Messages
37
Likes
41
Location
NY State
Hi all!
I've been lurking for a few weeks and figured I'd make an account and start to occasionally contribute.

Some introduction and my audio background:
I'm a mechanical engineer by training (BS+MS) and trade (~6 years so far), focusing on robotics through school (alas, not exactly what I'm working on now). Exposure to the E&M side of things, playing percussion since 4th grade band, plus working in technical theater throughout college got me interested in good sounding audio. Our technical director in the college theater department built speakers as a hobby and had a very nice sound reinforcement setup in both the traditional auditorium-style and black box spaces, which I had the chance to run soundboard for a few times including live mixing of a show or two with <10 mics. The large mixer boards and full 19" rack of amps, players, EQ/DSP boxes, and a patch panel was great for learning how it all worked and how to do proper gain structure to minimize noise. Running department-produced shows with proper source material from CDs and then doing student choreography dance shows with student-provided burned CDs they cut themselves and sourced from ripping low bitrate youtube videos was also ... enlightening.

In my dorm room I got a small Tapco 6-channel mixer where I could plug in my electric bass, desktop, and laptop and listen to everything with my headphones at the same time without bothering others too much. Once I got out of school and into a salary, I could then afford my own space and got to unpack my dad's old silver Kenwood receiver with the giant knob that moved the radio carriage and plug in the 1980s Wharfdale speakers that he had with it. Big improvement in enjoyment over my headphones, though I did have to cut into one of the speaker cones when years of handling/use dislodged a tiny ball of solder that started bouncing around and buzzing and had to be removed, and the volume knob potentiometer wore out and the left channel became intermittent depending on knob position. A year or two and lots of research later (and moving into a house with my now wife), I upgraded to a 5.1 system with a basic Onkyo AVR and the best price-to-performance speakers I could find at my budget (the Pioneer Andrew Jones system including floorstanding fronts).

Outside of movies and TV, all my music is though my computer and headphones nowadays (AKG K612 at home, Senn HD280 at work). I found this forum when I started to wonder about what my next upgrade step was for my listening. I'd been happily listening to my desktop's onboard DAC (based on a Realtek ALC889) through the mixer, with the only downside I noticed being a slight hiss with nothing playing if the volume was turned up to higher-than-listenable levels. Since I'm not using multiple inputs on the mixer anymore, I decided that its 25ohm output impedence was a potential downside if my next headphones are lower impedence, so I decided to swap it for a JDS Atom (based on reviews here), which I just got yesterday and like already because I know it's future-proof for anything I plug in, and there's no hiss at any volume anymore. Of course there's no sound difference, as there shouldn't be when going from something that was already in the prosumer range.

Glad I found this place and didn't waste my money on a DAC or spend too much on an amp that would have zero audible impact. I can save my money for the subjective links in the chain - more CDs and better/different headphones.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,385
Location
Seattle Area
Welcome to the forum and thanks for the intro. Good to see other types of engineers on board!

Seeing your robotics interes and putting you to work already :), how easy would it be to build a small platform (2x2 feet) that could move in X and Y dimension and carry up to 100 pounds of weight? Would need to control it to 1 inch increments or so.
 
OP
cshake

cshake

Member
Forum Donor
Joined
Jul 20, 2019
Messages
37
Likes
41
Location
NY State
Hi Amir! I thought about that question a bit during dinner and had some ideas. The ease of building this would depend on how many obstacles there are in that X-Y space, if you can have an gantry, or if you need it to be completely self-contained with no external lines or guides. Based on the weight capacity, it being this forum, and the ongoing thread about testing speakers, I'm assuming this is for moving speakers around?

The simplest way of doing single axis motion for heavy platforms, which we used onstage often, is a platform with wheels, a guide rail, and using a pulley system like a clothesline with the platform fixed to a spot on the cable and a motor to move the cable. You also get opposing motion for two platforms for free by attaching the other to the other side of the cable (cheap "automatic" sliding doors using a stagehand, just like curtains). If you can ensure no slip on the pulley (adequate tension or using a chain drive or timing belt), you can stick a rotary encoder on the drive motor and have extremely good precision. Using just lumber, hard rubber wheels, twisted steel cable, and a three-phase motor, we were able to quickly and almost silently move large platforms with 10+ people on them for one production. That doesn't work for 2-axis out of the box though.

If you can do a gantry and the movement space is clear, my first thought would be to implement the next logical step up from a single guided pulley and do something like the CoreXY motion system used in 3D printers - https://corexy.com/theory.html. Two motors, two cables for control, the gantry gets fixed wheels to keep it in one line (or guide rails), and the stage/platform goes on rotating caster wheels. Position is very accurate, just needs rotary encoders on the drive pulleys, and the occasional absolute measurement by hand to the platform to verify. The control scheme is laid out on that website and should be easy to implement with a tiny microcontroller. With the desired 1" accuracy, you can get by with fairly inexpensive caster wheels and lumber from the hardware store for everything structural. If the movement space isn't clear, you can raise the motion system to the ceiling and have the platform hang down, but then the gantry and the rails it slides on have to be much stronger since they're carrying the whole weight and not just providing position. The biggest downside that I see (besides the obvious problem of this turning into a monster that takes over the room) is that you can't easily have two independent platforms with this system with overlapping movement - but if each platform only needs to cover half the space, put two next to each other and share the rail in the middle between gantries. I'd also expect that this is far too much permanent installation for whatever space you're putting it in.

If it has to be multiple platforms in the same area, or you can't have any external load-bearing cables or installation, I'd default to a 4-wheel RC platform using Mecanum wheels, which allow you to move the platform in any direction while independently controlling rotation (using a wire-remote not RF, since the speakers you're moving need power and/or signal anyway). Determining the absolute position is harder here, and may be less accurate. My first idea for cheapest positioning would be to use a handheld laser measure and watch it while moving the platform via remote, but then you can only really do one axis at a time. I can't tell without starting to spec it out if this would be more or less expensive - you've got four independent motors per platform vs. two, but no gantry system to build - likely this is more expensive for a smaller space but cost is fixed, while the CoreXY system price increases with served space size.
To get the positioning more automated, there are some very good item tracking vision systems using a bunch of cameras around the room and attaching either little motion capture white reflective balls or invisible IR beacons on the platform, but that gets expensive (not sure if there is any open-source version out there yet using commodity webcams - I used a PhaseSpace system in my graduate work). The inverse method, putting some small cameras on the platforms and tracking known point light sources around the perimeter of the space, could be accurate enough to work too, but it's much better for rotation than X-Y accuracy (my MS thesis work was essentially trying to do that with hardware as small and cheap as possible, in the context of small satellites aligning themselves for docking using known light patterns on another nearby satellite). Or, maybe you only need it to move between known target positions, then you can just put printed targets on the floor and a camera on the bottom of the platform (or targets on the ceiling and a camera facing up, more light that way).

... yeah, I'd be happy to help with this :) Let me know more of the design constraints and I'll look into it deeper. We can go to PMs or email if you want.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,385
Location
Seattle Area
Yeh, there is a huge "marker" movement fueling development and availability of such parts. I just don't have the time and patience to search and put them together. :)

And oh, this needs to be light and as low profile as possible.
 

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,459
Location
Australia
Yeh, there is a huge "marker" movement fueling development and availability of such parts. I just don't have the time and patience to search and put them together. :)

And oh, this needs to be light and as low profile as possible.

And, hopefully, inexpensive. :) Aluminium and your milling machine are a good combination.
 
Last edited:

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
4,685
Likes
10,700
Location
Hampshire
Seeing your robotics interes and putting you to work already :), how easy would it be to build a small platform (2x2 feet) that could move in X and Y dimension and carry up to 100 pounds of weight? Would need to control it to 1 inch increments or so.
How much load can a Roomba handle?
 
OP
cshake

cshake

Member
Forum Donor
Joined
Jul 20, 2019
Messages
37
Likes
41
Location
NY State
After some PM discussion, turns out a small wheeled frame will probably be best. I like the idea of a Roomba, but without major software changes for the positioning and directional control, especially with some sort of closed-loop program, it's probably not sufficient (also weight capacity). There are some warehouse robots doing this same job, but I doubt they're cost-effective for what Amir is looking to do.

Luckily, basically all the parts, and possibly even a complete frame design concept, are available due to the popularity of the nationwide VEX and FIRST robotics competitions. Coupled with more and more college curriculums and lab handouts being available online, some directly applicable, we might be able to do this without to much inventing.
 
Top Bottom