I'm not going to pretend that I really understand this subject, but I'm intrigued by the idea of Zobel networks and would like to learn if it would improve my speakers, which are currently leaning toward a harsh top end.
They are small T-M-M towers, with the 25mm silk dome CSS LD25x XBL tweeter above a pair of CSS LDW7 7" midwoofers, crossed at 1800Hz. Crossovers have good quality components by Erse and Jantzen.
Actual construction of a Zobel network seems pretty simple, but I don't really understand how to design one that's right for my speakers. Can anyone help?
Thanks!
They are small T-M-M towers, with the 25mm silk dome CSS LD25x XBL tweeter above a pair of CSS LDW7 7" midwoofers, crossed at 1800Hz. Crossovers have good quality components by Erse and Jantzen.
Actual construction of a Zobel network seems pretty simple, but I don't really understand how to design one that's right for my speakers. Can anyone help?
Thanks!
People do put one in place and report a reduction in treble.. however it's not because of the Zobel, it's because of the change in response. This is something you could simulate, or use tone controls, try reducing tweeter level via resistance or use a Zobel like circuit to reduce only the upper frequencies.
The (original) purpose of a Zobel network is to stabilize the local feedback of CFP/ Sziklais outputs, including "quasi-complimentary" outputs (All NPN). An EF output stage should not require a Zobel network, but it doesn't hurt. More important for all amplifiers is an inductor+resistor series build-out so that no speaker cable and speaker impedance can short high frequencies and de-stabilize the amplifiers. An alarming number of amplifiers have stability problems and a Zobel network can help them, but the Zobel network is a high frequency load that needs to be connected before the series inductor.
I believe zobel networks are used in loudspeakers to stabilise the impedance seen by the crossover network in cases where the driver impedance varies significantly at the crossover point. The crossover would
have to be designed to take this into account. Just adding one to an already designed crossover would alter its performance. The effect could be simulated or measured and might or might not be beneficial.
have to be designed to take this into account. Just adding one to an already designed crossover would alter its performance. The effect could be simulated or measured and might or might not be beneficial.
Looking at the specs for that tweeter, the FR seems to rise at higher frequencies. The slight rise in impedance will result in a rise in voltage as well. A zobel can tame the rise in impedance, but not the natural rise in FR. I've used a parallel trap circuit to tame a similar bump in FR. There may be a better way?
That said, you shouldn't assume I know what I'm talking about 🙄
That said, you shouldn't assume I know what I'm talking about 🙄
Place a 2.5uF and 4 ohm across the tweeter. Vary resistance to taste.
True zobels are for impedance comp, but if voltage divider is present upstream then it will affect response. My suggested filter is not a true Zobel, but good for reducing upper treble if the series cap is present prior this network.
True zobels are for impedance comp, but if voltage divider is present upstream then it will affect response. My suggested filter is not a true Zobel, but good for reducing upper treble if the series cap is present prior this network.
I think of putting a zobel on a woofer, in the case where there is significant inductive impedance rise at high frequency in the woofer, which would reduce the expected attenuation of the woofers low pass. A zobel would counteract the woofer inductance rise, and in conjunction with the woofer LP provide some additional attenuation on undesired HF features of the woofer. I used a zobel to good effect once on a speaker with a 1st order electrical LP on a woofer with a breakup that was intruding on tweeter performance. That being said, it looks like the CSS LDW7 doesn't have a whole lot of impedance rise at HF (fancy motors in modern woofers have features to control inductance), so I wouldn't expect one in this application.
FYI/FWIW: reactance annulled zobels for flattening any/all driver's impedance, which among other things often allows textbook XOs to be close enough to go straight to in room 'voicing' either by ear or measurement.
As noted though, with this driver's very low inductance (Le) it will have no audible impact on its usable HF passband.
As noted though, with this driver's very low inductance (Le) it will have no audible impact on its usable HF passband.
All Zobel Network components have to be calculated according to the individual speaker's characteristics. Consequently, there is no such thing as typical values.
You could use the Speaker Zobel / Impedance Equalization Circuit Calculator.
The Dayton Audio Test System DATS-V3 is well worthwhile if you want accurate measurements. As a bonus, it can also measure components, so you can be certain they are within acceptable tolerance.
If you are really keen, have a search for an "Improved Zobel Network".
You could use the Speaker Zobel / Impedance Equalization Circuit Calculator.
The Dayton Audio Test System DATS-V3 is well worthwhile if you want accurate measurements. As a bonus, it can also measure components, so you can be certain they are within acceptable tolerance.
If you are really keen, have a search for an "Improved Zobel Network".