I was curious on any information anyone has or any studies that have been done on drivers where excursion is varied to see how distortion changes with excursion and if there is any relationship that can be generalized to drivers? For example for woofers as you reach xmax and beyond, distortion goes up dramatically, but is there a function for this? Is it approximately parabolic with excursion, exponential, or some other function? Or is it just completely random from driver to driver. I feel like there has to be some sort of relationship.
Also how does distortion decrease down to low excursions? As you get below 1/10th of xmax does distortion continue to decrease? Or is there a baseline distortion for a driver that decreasing the excursion below a certain point stops giving you any distortion benefits? I ask because I'm interested in bass arrays and getting very low excursion for low frequency signals to get extremely low distortion.
Also does anyone know how xmax is measured for drivers? I've heard 10%THD, but the excursion to reach that also varies with frequency...
Also how does distortion decrease down to low excursions? As you get below 1/10th of xmax does distortion continue to decrease? Or is there a baseline distortion for a driver that decreasing the excursion below a certain point stops giving you any distortion benefits? I ask because I'm interested in bass arrays and getting very low excursion for low frequency signals to get extremely low distortion.
Also does anyone know how xmax is measured for drivers? I've heard 10%THD, but the excursion to reach that also varies with frequency...
You may find some useful information here https://www.klippel.de/fileadmin/_m...linearities–Causes_Parameters_Symptoms_01.pdf
xmax is often given in the datasheets by (voice coil length - gap height) / 2 for overhung designs and (gap height - voice coil length) / 2 for underhung designs. Some use 70% of BL and I've seen others using (their own unexplained) constants to the first equation that may be related to BL.
xmax is often given in the datasheets by (voice coil length - gap height) / 2 for overhung designs and (gap height - voice coil length) / 2 for underhung designs. Some use 70% of BL and I've seen others using (their own unexplained) constants to the first equation that may be related to BL.
I believe the (voice coil length - gap height )/ 2 gives you xmech where the driver literally can't move further, but the 70% of BL makes more sense. And thanks for the link, I'll read it now 🙂
Nope. It gives you the linear x-max which is dictated by the constant number of voice coil turns inside the gap, there the BL curve remains somewhat linear and within a specified percentage of its nominal stated value.
Your question is far more complex - distortion at higher levels depends on many factors. At high power levels power compression would increase the voice coil resistance, the back emf is much higher and motors with no shorting rings would have demodulation of the field inside the gap. The backstroke of the driver also could be compromised if the motor is simple, this is visible on Klippel tests. Magnet strength and saturation of the top and back plates is different from driver to driver. There are other factors too. So, electrically caused distortion is dependent on many factors and I doubt can be predicted by some simple model.
Mechanical distortion would be dominated by spider/surround design and cone flexing. Again - probably too diverse to estimate by a simple model.
Maggiesnmacs,
1) I have studied distortion in a variety of woofers and cabinet designs, excursion is posted in post #1, distortion in post #12 here:
http://www.diyaudio.com/forums/subwoofers/185588-keystone-sub-using-18-15-12-inch-speakers.html
Other than distortion increases dramatically above Xmax, I would not say there are hard and fast rules that can be generalized to all drivers.
2) As the voice coil leaves the magnetic gap, control of the "spring" is progressively less.
3) Lighter cones are more prone to distortion than heavier. Attention to detail in the magnetic circuit is important.
4) In general distortion decreases at lesser excursion.
5) At 1/10th of Xmax distortion on a good driver is nearing the noise floor of measurement, and generally would not be an audible concern unless the speaker had a dragging voice coil, or ripped soft parts.
6) Same answer as above.
7) If you can find Klippel plots of the BL curves of the driver(s) you are considering, that can help. Well designed speakers can have very low distortion up to Xmax.
8) Xmax can either be at 10% THD, or different mechanical gap vs coil length formulas.
B&C includes Xmax (by formula) and Xvar, the excursion when the magnetic strength is reduced by 20% (IIRC). Depending on the magnetic structure design Xvar may be more or less than Xmax.
Art
Hi,
No. Those equations give ~ linear xmax, where the same amount of
coil remains in the stated gap. The driver literally can't move any
further when the coil completely leaves the gap, so xmech is gap
height bigger than xmax for overhung and coil height for underhung.
In reality all those numbers are very approximate, especially xmech.
rgds, sreten.
A very good speaker designer has a spreadsheet for calculating xmax.
Visit this link and post #6...
How to calculate excursion in terms of input voltage?
Visit this link and post #6...
How to calculate excursion in terms of input voltage?
This is what the spreadsheet's author wrote about it. FWIW.
"The input values, Fs and Qt are the values for the driver in the box. Sensitivity is the manufactures specified sensitivity for 2.83V input. The spreadsheet will give toy the SPL vs frequency for the specified input voltage and at what frequency Xmax would be met and exceeded below. Not exactly what you want, but should give you what you need."
- Status
- Not open for further replies.