New Tweeters: In or Out of Phase?


I fried a tweeter in a KEF 104.2 speaker and replaced it with a VIFA unit that has .5db higher sensitivity than the stock unit. The VIFA replacement does not indicate + - leads. Does anyone know how do I tell them apart and what are the sonic effects of incorrect wiring? The new tweeters are a great improvement over the orignal 1990 tweeters but are a little bright. I hope this reduces after burn in or is it that they're out of phase? Should I have gone with Morels whose sensitivty is rated at 2.5db less than the Vifa, but might have better matched the 2 midrange elements (from 1990) that might have lost sensitivity since their manufacture in 1990.
Thanks for any responses.
somec59
Thanks for all the replies, here's the latest: The Vifa's do have a very small + - stamped on the inside of the plastic face frame, but I had to remove it to install the tweeters (I tested then with a 9V battery to determine correct polarity). I let them burn in for a week so they don't sound as bright, or is it that I've grown used to it. My girlfriend said she likes the mellow sound of the Shahinian/Theta living room system but she says that the 7:1 KEF hometheater with an 8 foot screen is overload. Please, no suggestions on getting a new girlfriend.
Thanks
My suspicion is that the Vifa tweeter has lower impedence than the KEF T27.Even if they are rated at the same nominal impedence their actual impedence could be quite different.

Impedence affects crossover points so there are other factors to consider other than just sensitivity.

The KEF T27s do sound a bit dull.I have been told this is due to fatigue-so a new tweeter probably should sound a bit brighter.
I've replaced tweeters on several occasions and find that in some cases using a series resistor will increase the impedance and shift the crossover point slightly -more noticeably with higher order networks (such as the 3rd & 4th order filters most often used by KEF, or low impedance tweeters(4ohms).

I would recommend using an "L-Pad", which is two resistors (one in series, one in parallel) to maintain the correct tweeter impedance. here is a link to a page where there is a calculator for the correct values.

http://www.lalena.com/audio/calculator/lpad/

Try it both ways and see which you prefer.

FWIW, another consideration is that KEF may have used a Zobel/conjugate network to tame the fundamental impedance peak of the KEF tweeter. Unless the Vifa has the same Fs resonant frequency/impedance peak,there may be a mismatch(Just my 2 cents)

Good luck and good listening.

-chas
Thanks for the great replies. It sounds like I should start with a .5 ohm resistor and put it in the line just before the tweeter (on the positive feed??). The crossover is buried somewhere inside between the upwards and downwards firing woofers and I can't get to it.
Don't be afraid to put a resistor in series with the tweeter. The crossover probably already has such a resistor (called the "tweeter padding resistor). Just increase its value until you like the sound.
Hi

All Vifa tweeters have a positive indicatuion of +/- its "moulded" into the face plate on the back can be hard to see but it is there - another way to chack is apply a battery to the terminals of the tweeter, positive to the positive terminal and negative to the negative terminal will result in a outwards movement of the dome.

Thanks

Peter
.5 dB is most likely the cause of the brightness you're hearing. If you know what you're doing, you could add a resistor to bring the level on the tweeter back to normal. A less optimal option would be to adjust the treble on your preamp.

Phase is crucial. Checking it is easy. Take a C or D cell battery and connect the + to one lead, the - to the other. If the dome moves forward, you have the + to wired to the + on the tweeter, if backward, then you know that the - is connected to the + on the tweeter. Usually there's some indication of which lead is the positive.

You also need to know whether the original tweeters were wired in phase or out. Depending on the crossover, out of phase might be correct.

If you have the phase is wrong, there will be a very audible drop in the frequency response in the crossover region. Use an SPL meter and a test CD with sine waves to double check.