Loudspeaker Measurement Turntable
I've had several people ask about my loudspeaker measurement turntable setup and the info I've posted is somewhat scattered in different threads. So the following is some documentation on what I've designed and built to measure the off axis data on the speakers I design and/or test. This certainly isn't the end all be all solution as there are some things I still want to improve upon but this is what I have build and I find it is easy to build, use and produce good results.
What is a speaker measurement turntable and what use does it provide?
Put simply it is a device that allows the speaker being tested to be rotated in order to capture the frequency response data at various angles giving you a more complete picture of the performance.
When testing loudspeakers the default method is to take frequency response measurements of the speakers directly on axis, that is pointed straight at the mic since this would often be the axis on which you listen and thus the same frequency response as the direct sound. However taking only the on axis frequency response provides a rather limited picture of the overall performance of a speaker as they do not radiate sound only straight ahead but at all angles. The sound emitted above/below, to the sides and behind the speaker all add up and influence the overall sound signature of the speaker in normal rooms. Therefor it is very important to capture frequency response data from not only on axis but off axis as well, knowing the off axis behavior gives you much more insight into the performance of the speaker allowing you to see dispersion and directivity behavior as well as calculate overall sound power.
Original Simple Design:
Originally my turntable was a simple Lazy Susan style bearing that I picked up from the hardware store sandwiched between two large circular subwoofer cutouts with a small stand mounted to it to keep the speaker elevated and baffle aligned with the rotational axis. Originally I just rotated it manually by hand and had placed some angle markings every 5 degrees on the disk to judge position. While this certainly worked well for several years it had two minor problems, it was tedious walking back and forth to move the turntable to the next position after each measurement and it was difficult to repeat accurate positioning while moving the turntable manually.
I knew this was something I had to improve upon soon as I was planning on taking many hundreds/thousands of measurements in the near future for the waveguide shootout tests.
My current smaller indoor measurement turntable.
Automating the Turntable:
With that large scale of waveguide testing planned I was looking for a simple solution to automate the movement of the turntable both for ease, accuracy and consistency. Having built and repairing a few 3d printers I had spare parts from those on hand and using some of those parts made logical sense and I knew it could work the way I wanted and would not be too complex to design or build.
The turntable would be moved by using a stepper motor and some GT2 belt wrapped around it, this would be driven by a stepper driver, I used an A4988 as I had several spare and everything controlled using an Arduino with fairly simple code utilizing a few buttons for inputs which would be move forward, move back, return to home, and select angle step (5 or 10 degrees). I also added a simple 4 digit 7 segment display (another spare part I had) to give a readout of the current turntable position.
The original breadboard prototype controller can be seen on the embedded video, below is the basic circuit layout I used. It is missing the 12v to 5v converter to power the Arduino which was built into the I/O shield board I ended up using when mounted everything in a box. I also added an additional switch connected to A5 in order to select between 5 and 10 degree movement.
Arduino code for original design:
Here is the code for that design if you want it, though I must warn you the number of steps will need to be modified depending on the gearing and size of the turntable. Since I was just trying to get things working I didn't spend much time cleaning up the code. The current steps are set as the following in the code and would need to be adjusted:
122 - this is the number steps for single degree of movement for shown on the display, adjust this value if you need more/less movement, the others should be based off this value.
610 - steps for 5 degrees of movement (single degree x 5)
21960 - steps to limit the turntable at 180 degrees of movement (single degree x 180)
I am working on a version 2 controller which is easier to use/assemble adds additional features and is easier to program/adjust for different sized turntables with more streamlined cleaned up code.
Here you can see the some of the first testing I was doing once I had the controller design done and had built the larger outdoor turntable.
Schematic for the controller:
Not shown is an additional button connected to A5 I added later for selection between 5 and 10 degrees of rotation. I also forgot to show a power source for the Arduino, I used a I/O shield style board for the nano that had built in regulated 5v output, in my case a single 12v supply powered both the stepper driver and the Arduino.
Automated Controller v2.0 - design in progress
Currently working on v2 of the turntable controller, this one built on an Arduino Uno and will be easier to assemble and program with cleaner code and will include more features like a 16x2 lcd display for easier readout of the current position and rotation step selected with more angle options for movement as well.
I also hope to give it an input for automatic detection of a sweep allowing it to move automatically without human input. Basically if REW is setup with the number of sweeps, delay and the measurement numbering correctly set a single mouse click should be all that is needed and the controller will listen for the sweep on the input then automatically move to the next position before that sweep starts and so on until the desired number of angles and sweeps have been captured. Basically automated capture of a full axis around the speaker.
Update (October 2023):
Finally got the prototype PCB designed and made, seems to be functioning as expected. Automatic sweep detection is still a work in progress but it's working perfectly using standard control.
Indoor Turntable:
I currently have two turntables that I use this is the smaller one that I use to perform indoor measurements. It allows me to quickly test designs but due to reflections from the floor/ceiling I can only manage about 5ms of window time at 1m distance which limits resolution of low frequencies basically making anything below 600 Hz almost useless as it falls below 1/3 octave resolution at that point which will smooth out and hide any frequency response issues. I'll usually start my designs using this turntable as it allows me to get a general sense of the dispersion behavior of the drivers/design and in general testing of drivers/horns like for the waveguide shootout where I do not need much low frequency resolution.
Generally I find this works well for up to two-way designs using 8" woofers, larger then that and the 1m distance is too close and 5ms gate too short to see correct far field behavior on any larger cabinets and driver diameters, so I stick with outdoor testing using the larger turntable for those.
Indoor Turntable with my VBS-10.2 on the test stand.
Stand is adjustable in height for testing difference sized designs to keep the mic near the center point between the floor and ceiling in order to maximize gate time.
Stepper motor on adjustable mount allowing for tension of the belt.
Electronics portion of the turntable.
Outdoor Turntable:
I built this turntable when I realized I was severely limited in the accuracy of my measurements that I could take indoors especially on larger designs. This turntable did provide some challenges in the construction process, I wanted a sturdy platform with a stand that could elevate the speakers quite high to get a good gate time at a further 2m mic distance to improve far field accuracy with the larger designs. I also didn't want to shell out a huge amount to build the thing or make it overly complex. Finally I wanted to be able to quickly assemble/disassemble the whole rig to allow it to be moved and stored easily.
While trying to balance what i thought would provide stability with potential weight and easy of construction/material usage I settled on a 36" diameter turntable. I chose MDF as I didn't want to chance plywood being slightly warped and not laying flat. Not to mention this was during the height of the Pandemic and wood prices were crazy but the cost of sheets of MDF had not yet jumped too much.
Initially I ordered a fairly large (24") lazy susan style bearing to use for the turntable portion, unfortunately it got lost in transit, not wanting to wait on shipping for another I moved to my backup idea, a smaller 12" bearing I could get at my local hardware store at the center and several small caster wheels providing support around the outside. Initially I tried roller ball casters around the outside but those surprisingly caused quite a bit of friction and noise when trying to rotate the platform in a circle, however they worked very smooth when moving it in straight lines. I swapped them for normal fixed wheel style casters and it moved much more smoothly with less friction.
Knowing that the ground outside was not going to be perfectly level I built a large triangle shaped base out of 2x4's. At the corners I have 3/8" T-nuts installed on the bottom with threaded rod inserted, feet are attached on one side while a 3d printed knob is on the other end allowing for leveling of the platform.
For the tower structure I had initially looked at using aluminum lighting truss sections that could be bolted together but that was looking to be a rather pricey route. I eventually decided on just using a simple open frame of 2x2's for the tower structure, a larger 12" x 12" section at the bottom and a 9" x 9" section that would fit inside and allow me to choose or adjust the height as needed. Both sections are 6' tall though I limit it to 9' maximum height. I have holes drilled with T-nuts inserted on the inner section at various lengths which allow it to be positioned at different heights but I normally just leave it fully extended. Mounted on the turntable and base overall height is just over 10 feet to the platform at the top. I have found it's sturdy enough for anything I can lift up there (Titan-815 being the largest yet) though I did need to add a brace that runs to the front of the turntable which minimizes some front/back swaying.
Originally I used a spare Pancake Nema 17 motor for movement but the tiny motor barely had enough torque and would often stall out missing steps and throw off the positioning. I ended up upgrading to a larger Nema23 which has zero issues, you can even stand on the platform with a heavy speaker up top and it will rotate no trouble. The smaller motor got used when I rebuilt the indoor turntable.
Base Platform for Leveling the turntable.
Adjustable Feet.
Turntable construction, small lazy susan bearing in center with 6 caster wheels for support around the edge of the circle.
Turntable is easily mounted to the base with three thumb screw bolts.
Tower Structure mounts using four bolts, 3d printed knobs make installation on this section tool free as well.
Tower extended to full height.
Tower extended to full height.
Nema 23 Motor for movement, platform is adjustable for tension.
A GT2 Belt wrapped around the disc allows for 180 degrees of rotation and give the motor necessary torque/leverage.
If I can figure out an easy solution for 360 degree rotation I will implement that. So far can't find belt loops long enough.
Some additional photos taken while testing various speakers:
Video showing the turntable in use testing a JBL PRX415M: