Let’s say that I’ve designed a low frequency alignment for a bass unit using theory, simulation and finally tweeking by ear, arriving at something I’m happy with (eg, an IB enclosure of 25 litres). I recently read that when I add a low pass crossover, the loading on the bass unit will modify the units T/S parameters, which I’m concerned will alter the loudspeakers bass alignment. For example, increasing Qtc by 10% could require an enclosure volume of 33 litres for the same bass alignment. Is this true and if it is, how is this normally dealt with?
The T/S parameters are not affected by the crossover.
The bass alignment can change if the crossover changes the response, which is what a crossover is there to do. However the part which sometimes is overlooked is the series resistance of an inductor.
Do you use a simulator? This makes it easy to see.
The bass alignment can change if the crossover changes the response, which is what a crossover is there to do. However the part which sometimes is overlooked is the series resistance of an inductor.
Do you use a simulator? This makes it easy to see.
The R in the first inductor we typically see will bump up the effective Q of the bass driver
Qt'=Qt+ any added series resistance [Rs]: http://www.mh-audio.nl/Calculators/newqts.html
dave
Qt'=Qt+ any added series resistance [Rs]: http://www.mh-audio.nl/Calculators/newqts.html
dave
That’s a great calculator. Series resistance can make a big difference in the Qts and change bass alignment. You can use it to your advantage if you want. Also many people don’t realize the resistance of the inductor in the woofers low pass can really mess up the alignment which is why this calculation should be taken into account. Many times people wonder why they arent getting the results from the cabinet they expected, this is why!
What everyone else said. As you add components to the driver and/or mount it in a box to create a system, the equivalent TS parameters of that collection of components or system will change. To fully beat the topic to death => The Qe, ( Q electrical ) the damping provided by the motor, assumes the driver is connected to a zero impedance voltage source. So if you are connected directly to a low output impedance amplifier, the motor damping that produced the datasheet value of Qe is what you get. In the opposite extreme, if your amplifier behaves like a current source, a transconductance amplifier with very very high output impedance, the electrical damping goes to zero, so the Qe goes to infinity as the back emf of the moving motor can't change the current in the coil, and your Qt tends to the value of Qm, the mechanical Q of the driver. Of course Qm will be influenced by the box volume or what ever baffle a driver is mounted in. My last speaker build I used transconductance amplifiers for three of the four drivers and use a Linkwitz transform to tame the resonance. It worked beautifully. You can model all the interactions of electrical components and boxes with software like boxsim provided free by Visaton. https://www.visaton.de/en/literature-software/software
Last edited:
Thank you, everyone!
I used mica.de to give me a ball park figure for Vb, then built a slightly larger box and gradually reduced the volume with 0.5 & 0.25 dm3 blocks. Then I used XSim to design the crossover. From your answers, it sounds like I need to use something like the Visaton simulator to understand how the crossover affects the bass alignment?
PS. Does the XSim “Lookback” graph show the effect of adding the crossover, as seen by the amplifier?
I used mica.de to give me a ball park figure for Vb, then built a slightly larger box and gradually reduced the volume with 0.5 & 0.25 dm3 blocks. Then I used XSim to design the crossover. From your answers, it sounds like I need to use something like the Visaton simulator to understand how the crossover affects the bass alignment?
PS. Does the XSim “Lookback” graph show the effect of adding the crossover, as seen by the amplifier?
If you double click on an inductor, you can enter the coil resistance and the response will change accordingly.
Theoretically it is correct that the Q changes, but a calculated solution for an enclosure size easily contains 10 percent of wriggle room.
Thank you. I’ve noticed previously that if an inductors resistance is below about 150mV the effect on the drivers frequency response is minimal and above about 200mV the effect gets increasingly pronounced. So, is the answer to my question “Maybe; it depends”, based on the particular drivers response to the inductors resistance level?
In my experience XSim will reflect the changes due to the electrical circuit on the bass alignment. As you load measurements made with the drivers in the box and baffle, it will give accurate results. The super fast response to part value changes with the mouse scroll wheel make Xsim really great. I have had some schematics in Xsim become corrupted so that I could no longer use them, so save different versions often to be safe.
Visaton will do what Xsim does, operating from your measured results, but it also allows entering TS parameters and includes modeling the physics of the box so you can see the effect of changing box volume or ports, diffraction effects due to driver positions on the baffle and it will calculate horizontal and vertical dispersion. Boxsim also has a parametric optimizer built in, so you can select a few components in the crossover and it will adjust the values to optimize the crossover for you. You can not add to the boxsim library, but you can overwrite the driver name and all the parameters and load response files for any driver in a project. So your project becomes the secondary library for out of catalog drivers.
I'm not certain but I think the normal impedance graphs are the amplifier sees as the load. The lookback I think is what the driver sees at the source impedance. The manual should spell it out.
Last edited: