Active cross speakers and the tweeter protection cap

[prickears]

TRUE STORY NO BS

.........so anyhoo I disconnected my sub today, refiddled the inputs on the back of my Behringer CX3400 to operate in two way mode, instead of three way mode and left it at that. When I got home, the kids were watching a movie, sound system shakin' the walls, as they do. It was then it suddenly struck me!!! I forgot to re-range the XO dial by depressing the buttons on the back of the XO, those poor tweets had been crossed @ 270 Hz (instead of 2700hz, depress for 10x button on rear) for an hour and a quarter at ridiculous SPLs, somehow they are still alive, but it just goes to show how easy it can be to overlook something!!!!!

 

[badman]

I had this issue to think through when I was working on my OSWG crossover. The conclusion I came to was that since it needed CD EQ anyway, the best choice was to combine the CD EQ and protection cap, and accept that it required additional impedance compensation thereafter.

 

[KlipschKid]

Hi,

I was in a similar situation. Speaker Workshop results and my own listening indicated I needed to drop the tweeter's response at, and just below, the xo frequency to get a flat response, as well as dropping the tweeter's output a few dB. The tweeter has an impedance of 6.8 ohms at 2260 so I used a bridged-t attenuator (to dampen any tweeter peaking) and a 6.8uF and a 8.2uF in parallel. It wasn't cheap but it is just about as good as I can get it.

So my protection cap is necessary because it also peforms another essential function.

 

[automaticmojo]

I know this is a old thread, but any capacitor won't pass dc, as it is essentially a open circuit, i.e. two plates with insulation between then.
Essentially they don't really "pass" Ac current either-just signal information.
If your speakers have caps in the tweeter, mids, woofer circuits-your fine from a amp going dc.
Just go back to electronic basics-no magic here.

 

[weltersys]

I know this is a old thread, but any capacitor won't pass dc, as it is essentially a open circuit, i.e. two plates with insulation between then.
Essentially they don't really "pass" Ac current either-just signal information.
If your speakers have caps in the tweeter, mids, woofer circuits-your fine from a amp going dc.
Just go back to electronic basics-no magic here.

The capacitor in a woofer circuit typically is in parallel,it offers no protection from DC to the woofer.
Audio "signal information" is AC (alternating current). Depending on the speaker impedance and the capacitor value, a capacitor will reduce the AC (or DC) voltage the speaker absorbs.

Having experienced catastrophic failure from "Flame Linear" amplifiers that would burn up woofers when the output transistors failed, dumping their full DC power supply to the speaker output, I continued using protection capacitors fot the HF drivers even though the amps i presently use have a relay circuit protecting the speakers from DC.

A few years ago, after some extensive testing, I eliminated the protection capacitors in my 3 way sound reinforcement system. Elimination of the capacitors simplified the active crossover requirements for flat response below the crossover point and improved response above.

Added ten 18 mFd capacitors to my passive crossover construction inventory.

 

[automaticmojo]

I agree if you don't have capacitors in the stock crossover-then I probably wouldn't add them, because of possible phase shift, etc issues.
Where I think your mistaken-even a cap in parallel-the dc current needs to have a complete circuit to do it's damage-one "side";+ or - won't do anything, as the capacitor is essentially a open circuit to DC, it's not really absorbing anything DC.
Even a small cap is a open circuit to DC.
If you don't believe it, try a experiment. It's one of the basic experiments in electronics class(s).
I also has a amp go DC years ago- the woofers (2 12 " in each speaker fried because they only has a inductor in the circuit-the mids (8" mid woofers) and tweeters where fine because of the capacitors in the passive crossover, which where parallel in the case of the mid woofer.
Also having old amp periodically checked up and refreshes will help reduce this issue. Healthy, well made amp with good parts-even direct coupled one rarely go DC-Phase Linears are another story unfortunately

 

[automaticmojo]

Just to add, -capacitors won't fully block DC in theory (a very small amount) until the cap fully charges-(the cap having a voltage that exactly cancels out the applied DC-the sum output voltage equaling zero) but in audio use this isn't a problem.
This explains why only the woofers fried-a simple high pass has a cap in parallel, and still protects the driver from applied DC doing damage.

 

[audiobomber]

If your speakers have caps in the tweeter, mids, woofer circuits-your fine from a amp going dc.
Just go back to electronic basics-no magic here.

I think you've misunderstood the topic of the thread. The OP was asking whether to add a cap to protect the tweeter in an active system. There is no cap in the tweeter circuit unless he adds one, because the power amp is an active system is connected directly to the tweeter:

preamp => active crossover => power amp => tweeter

There are no line level crossover elements. The purpose of the cap would solely be for DC protection, not to roll off the tweeter response. IOW, a large cap.

 

[KlipschKid]

The purpose of the cap would solely be for DC protection

This is the theory, and in this case the very large cap is used to keep the high pass frequency far below the crossover freq so that tweeter peaking is avoided.

However, I have found that using a bridged-T attenuator after the cap dampens tweeter peaking so the protection cap can also serve as a part of the crossover if needed - but I'd recommend anyone considering this to test it first. I did a large amount of testing using Speaker Workshop. I'm using 15uF as my "protection" cap.

It worked well for me - I used 1% Mundorf silver/oil caps and Dale resistors in a passive line level xo. The tweeter has a 24dB octave roll off @2259Hz and the woofer has 18dB @2262Hz. I spent time measuring/selecting caps and resistors to get it this accurate. It isn't perfectly phase coherent and I haven't done any time offsets either, but I've made it possible to add this later.

I know others prefer to connect amp-tweeter directly and use line level gain to adjust tweeter output but I have had amps go dc on me so I'm not willing to risk it. I've destroyed 2 pairs of headphones and an amp because of mistakes (not all mine I'm sorry to say - some very reputable vendors - so I'm aware everyone can make mistakes).

Anyway, at the end of the day, the aim is to get a phase coherent seamless crossover, with minimal noise and distortion. Is there a perfect way ?