What is the best way to stretch Mylar for ESLs? What tools or machines would be beneficial to handle this with consistent tolerances? Lastly instead of a sub in the base - would having a cant of a few microns allow enough extension for low end articulation? Perhaps not deep sub bass but maybe something below 50hz?
For flat panel ESL's, a pneumatic (bike tube) jig will stretch Mylar to a very uniform tension, and it's precisely repeatable. For example; if I want to stretch a 6-micron diaphragm to exactly 1.5% elongation, I first use a fine tip felt pen to place reference marks on the Mylar exactly 12" apart. Then tension the filme (pump air into the tube) until the reference marks measure exactly 12 3/16" apart. It's a very easy jig to construct and to use.
Some prefer heat shrinking; which is easy but less precise.
For a curved panel, you must either stretch the film manually or with a mechanical jig (can't use heat or a pneumatic jig) because practically all of the tension must applied normal to the curvature (any tension applied across the curve tends to pull the diaphragm into the rear stator).
Some builders manually pull the film to the desired tension using strips of 2" wide adhesive tape connected to a spring tension gauge.
Whatever the method, the tricks are knowing how much tension to apply and being careful enough to not tear the fragile film.
Last edited:
Hi Jared
I built a wooden frame and used the tape method Charlie mentioned. Worked fine for me. Of course the tensioning would not have been perfect throughout the panel, but so far I've had nothing but sweet music out of it.
Attachments
You might think a bit about what it is that you want to keep consistent. One target might be the resonance frequency of the diaphragms. I'd argue that reaching a target tension is largely a means toward reaching a target resonance frequency. Bear in mind, though, that the diaphragm tension will also strongly affect how much bias voltage you can apply without causing the diaphragm to collapse into a stator. High tension stabilizes the diaphragm but increases the resonance frequency. It can be a tricky balance.
If you have measurement tools you can measure the resonance frequency of the tensioned diaphragm even before you've installed it (perhaps irreversibly) between the pair of stators. Tapping on the frame holding the diaphragm excites the fundamental resonance of the diaphragm, which typically has a high enough Q so that its frequency isn't hard to identify. You could probably even tap two diaphragm frames and judge the consistency of their diaphragm resonances by ear. That gives you the opportunity to tune one diaphragm to match the other.
In theory, the stretching and measuring method Charlie described should yield consistent resonance results as well. You might adopt his method and use the drumhead resonance test as a way to verify that all is well.
That said, since it sounds like you're aiming for low frequency operation (not an easy task), you really want to know what the resonance frequency of your pair of panels is. Their response will plummet below the resonance so you'll never reach 50 Hz if your panels resonate at 150 Hz.
Few
If you have measurement tools you can measure the resonance frequency of the tensioned diaphragm even before you've installed it (perhaps irreversibly) between the pair of stators. Tapping on the frame holding the diaphragm excites the fundamental resonance of the diaphragm, which typically has a high enough Q so that its frequency isn't hard to identify. You could probably even tap two diaphragm frames and judge the consistency of their diaphragm resonances by ear. That gives you the opportunity to tune one diaphragm to match the other.
In theory, the stretching and measuring method Charlie described should yield consistent resonance results as well. You might adopt his method and use the drumhead resonance test as a way to verify that all is well.
That said, since it sounds like you're aiming for low frequency operation (not an easy task), you really want to know what the resonance frequency of your pair of panels is. Their response will plummet below the resonance so you'll never reach 50 Hz if your panels resonate at 150 Hz.
Few
Thanks a ton sorry for the delayed response. Been working and learning about other things and having had the time to dig into this yet. Can the same jig handle both curved and flat panels? Who sells the tooling or has diagrams for the equipment needed where consistent tolerances are achievable? I have access to a full machine shop with manual and CNC tools - would the same practices and equipment be used for AMTs and ribbons?
I have always wanted to build a stretcher for use on larger panels, but for small panels like I make using just a heat gun gives me very consistant results in the 70Hz to 90Hz range.
jer 🙂
jer 🙂
It takes a lot of displacement to get a good response into the bass range and there are many tradeoffs.
The width of the panel depicts what kind of High Frequency Horizontal dispersion characteristics the panel will have.
Narrow is preferable but lacks the displacement for good bass.
Wider panels have an increased displacement and can give good bass as well as a lower resonate frequency that is less annoying.
But, The wider panels also make the Higher Frequency's form more into a sharp beam and gets more focused as the frerquency is raised.
This is for a panel that is driven by the same signal over its entire surface area.
Mechnically segmenting them helps this a bit and can be just fine for most, in most cases.
This is the common method used by many.
When listening to my (little) panels in fullrange they sound great but I don't particularly care for the low end due to the high Q bump in the response due to the diaphragms resonate frequency.
Being that they are only 3.25" wide the dispersion is very good, much much better than the 7.5" wide ones that I had made originally.
I had to sit with my head in a vise with the wider panels to enjoy a perfect stereo image.
Not so with the more narrower ones.
Search the threads in Electrically Segmented designs and you will find a ton of great info there on such subjects.
This trade off will help you get the best of both worlds if you are planning a very wide panel in order to get some good low end as well.
In a nearfield measurement most any panel size will give you a flat response, it is the dipole cancellation and width dimensions (front to back distance) that determine what knid of actual low end you will get along with that bump caused by the Diaphragms resonance (Q).
The general consences is to crossover above that point (at an octave or two) and use a good wooofer for the low end.
jer 🙂
The width of the panel depicts what kind of High Frequency Horizontal dispersion characteristics the panel will have.
Narrow is preferable but lacks the displacement for good bass.
Wider panels have an increased displacement and can give good bass as well as a lower resonate frequency that is less annoying.
But, The wider panels also make the Higher Frequency's form more into a sharp beam and gets more focused as the frerquency is raised.
This is for a panel that is driven by the same signal over its entire surface area.
Mechnically segmenting them helps this a bit and can be just fine for most, in most cases.
This is the common method used by many.
When listening to my (little) panels in fullrange they sound great but I don't particularly care for the low end due to the high Q bump in the response due to the diaphragms resonate frequency.
Being that they are only 3.25" wide the dispersion is very good, much much better than the 7.5" wide ones that I had made originally.
I had to sit with my head in a vise with the wider panels to enjoy a perfect stereo image.
Not so with the more narrower ones.
Search the threads in Electrically Segmented designs and you will find a ton of great info there on such subjects.
This trade off will help you get the best of both worlds if you are planning a very wide panel in order to get some good low end as well.
In a nearfield measurement most any panel size will give you a flat response, it is the dipole cancellation and width dimensions (front to back distance) that determine what knid of actual low end you will get along with that bump caused by the Diaphragms resonance (Q).
The general consences is to crossover above that point (at an octave or two) and use a good wooofer for the low end.
jer 🙂
Last edited:
- Status
- Not open for further replies.