That's a good idea; they seem to be mostly MKP. It looks like I can get 30-40uF for under $10 each, which is quite acceptable. I'll try with and without protection caps on the first pair and see how they go.
It's only going to sum flat if the phase difference between the LP and HP sections is the usual LR4 360 degrees which it isn't going to be because the 1st order filter that the series cap introduces adds another (eventually) 90 degrees (45 at it's nominal crossover frequency and about 70 if you shift it down 2 octaves).
If you select a cap to provide a 1st order filter at the same crossover frequency as the main crossover (now 3rd order BW) you end up with the required 360 degree phase shift and and pretty much a LR4 response. Or near enough that you can't fix an deviations easily.
You also get to use a much smaller value cap so you can use a better quality one. Using a big cheap cap to shift the freq down to get around the response and phase issues (which it doesn't anyway) sounds okay in theory but the signal still has to go through it.
But then you have the susceptibility to impedance changes in the driver, and reduction of damping, two reasons for going active in the first place.
Hello, in my 3 way active loudspeakers, i'm protecting my tweeters by putting a capacitor in the negative feedback loop of my HF chip amp based power amplifier.
By doing this I also prevent the amplifier to ring.
By doing this I also prevent the amplifier to ring.
Especially important if there's a resonance well below the crossover frequency, as with some compression drivers. My TD-2002s have a resonance at 400Hz, yet are crossed at 800Hz. A series cap will introduce reactance right where good damping is needed.
The frequency response differences don't seem like they're correct- do you have an impedance measurement of the driver? I'd expect a lot more than 6dB loss down near 250.
No impedance measurement for the nominally 16 ohm Paraline horn/driver using an 18 MFD capacitor.
The capacitor does measure at it's rating, as does the DCR of the driver.
The frequency response is definitely correct, it matches (nearly) exactly at the HF.
As previously mentioned, the response was not what I'd have expected either.
So far, putting a capacitor in series has never resulted in an exact 6 dB per octave acoustical response in any driver I have measured.
Some drivers get close, but this combination not at all.
Art
Francis,
Years ago, Dave Rat was experiencing shattered diaphragms in his Tad drivers from the SPL of adjacent mid/bass drivers in his PA system.
After removing the "protective" series capacitors which eliminated amplifier damping, he had no more problems with the Beryllium diaphragms.
The low resonance in Be is problematic in that regard, other driver materials don't have that particular problem.
Objectively, removing the series capacitors in my PA made alignment between cone mid and Titanium HF drivers easier (less aggressive PEQ filters and other "tricks") but subjectively I did notice better transient response.
Art
Art,
Mine are used in a home system at very reasonable SPL, but removing the caps definitely improved the clarity.
Mine are used in a home system at very reasonable SPL, but removing the caps definitely improved the clarity.
A 2 cents question (sorry for that) : does a 1:1 transformer able to stop the dangerous DC before a tweeter to protect it ?
Transformers won't pass DC, but a high quality, high bandwidth, high power transformer costs far more than most HF drivers.
The biggest DC danger comes from amps with no output protection, when output transistors fail the entire power supply DC voltage can be dumped on the output.
That much voltage could easily be enough to fry the transformer primary, and could possibly arc over to the secondary and still take out the driver coil.
If you are using "Flame Linear" type amplifiers, a capacitor is still the way to go 😉.
Capacitor provides no damping. Damping requires power dissipation. This requires resistance across the load.
Not sure where you're going with that, since that's what I just implied. Read it again.
I'm always paranoid that if the driver sounds like it might be "more clear" without a protection cap, it's actually issues caused by increased 60/120Hz noise in the signal stressing the driver, creating harmonics, etc. I'm not saying that's necessarily common, just saying I worry about it. The only active speakers I have set up right now have an L-pad-like resistor network, and no cap, with DE250. They used to have a 40uF, but I stole it for something else.
Low frequencies getting into the HF driver is a valid concern. My HF is coming out of a balanced DCX2496 output, through a balanced pad/HF filter, into a balanced input of the J2, so I doubt I'm getting much LF junk. Generally protection caps are sized big, to not unduly attenuate the passband, with the result that 120Hz rejection isn't real large.
Last edited:
A series cap will not dissipate energy unless it has significant series resistance, in which case it is a crappy capacitor. Reactance is not resistance.
Reactance will reflect energy back to source.
Low impedance amplifier provides active resistance via feedback.
- Status
- Not open for further replies.