Driver Polarity on the KHorn

[StickMan]

I recently had my room analzyed for bass response and in the process we measured the polarity of the drivers. According to my Tech, on my 1987 Khorns the Tweeter polarity is inverted to the midrange and woofer. (Same in both speakers). The tweeters are 'polarity' and the midranges and woofers are 'inverted polarity'. Anyone know if this is normal spec for this model?

The 87's use the Type AK2 network

K77-M Tweeter

K55-M Midrange

K33-M Woofer

Also, thought that I might switch both sets of speaker leads (I'm biwiring) to put the Tweeter in inverted polarity and the midrange & woofer in polarity. Wonder if there would be a difference in sound?

I'm little concerned because I had the networks re-wired with clean wire several years ago. At the time the networks were modified to enable biwire also. Just want to make sure a mistake was not made to the normal spec.

Thanks

[WMcD]

It is tough to tell what you tech was meauring.

Let me start by saying that all the Klipsch crossovers I've seen show the drivers being wired in phase. If yours are not wired that way, then perhaps it is an anomaly.

It should be pretty easy to trace out that the minus or black outputs of the crossovers are connected to the proper terminal of the driver. On the other hand, you'll have to look for markings. Perhaps others can advise. I only have home built versions.

Some years ago I looked at the effect of the different path lenghts of the horns and the cross over frequencies.

The biggest issue may be in the woofer to mid. The 400 Hz crossover corresponds to a wavelength of about 34 inches. Then you look at the path length of the bass horn starting from the driver, through the passages, unit you reach the back of the cabinet, where the mid driver is located. That is pretty close to 34 inches.

Therefore, at the crossover point, where both the bass and mid are putting out sound, there is a delay of 1 wavelenght. That is okay. Things are "in phase". If there was a half wavelength difference they arguably might benefit by a reversal of one from the other. But you don't have that.

The tweeter to mid difference is a bit harder to take stock of. The crossover is at 6000 Hz and therefore the wavelenght is just over two inches. So we're playing with a 1 inch path length difference making some difference between good and bad, in theory.

Maybe the earlier tech made some estimations and decided to invert the mid relative to the tweeter and also realized he'd have to invert the woofer to maintain polarity between them.

However, please let me point out that some common rules of thumb do not apply to the K-Horn. It is true that the phase, and polarity, of the hook up must be taken into consideration in theoretical designs. Those classical designs assume the two drivers under consideration are putting out sound for, say, an octave or so on either side of the filter point. So having the drivers out of phase can lead to a deep notch in that octave near the crossover if the polarity is out of phase there.

For good or bad, the three horns in the K-Horn generally have good output to the crossover points, and then the acoustic output drops like a stone. The crossover filter is mostly there to make sure electrical power is not fed to a driver which is not sounding to begin with. So there is/are not going to be much acoustic interference(s) no matter how they are wired.

I made a two way horn system where the pathlength / polarity did come into play. This was with 12 dB per octave crossovers. Playing with polarity of the woofer versus treble did show an anomaly just at the cross over but it was about 10 dB deep and about 1/3 octave wide as shown with an LMS system.

On the other hand, I didn't hear it in some casual listening.

There is much more to crossover design and results than what I've spoken about above. The path lenght calculations are easy. However, the drivers themselves have phase issues at the edges of their performance. Big math predicts just so much.

The bottom line is that I'd suggest you check wiring, put it back the way the factory made it, and don't worry.

Gil

[JohnA]

Here is a schematic of the AK-2. You can use it to chack the wiring of your modified crossovers. It is very close to the well liked AL/AK-3. The simple answer to your "problem" is to reverse the leads to the tweeters. The whole system will be "in Phase". If desired, you could then reverse the leads to the each imput terminal of eack K-horn and everything will be in "polarity". I can't imagine the latter is audible.

I am in agreement with Gil. Everything is probably correct and the measurement detected a "problem" in one spot. The tweeter may not test inverted at another place in the room. This is due to the path length differences in the K-horn.

Type AK-2.pdf

[StickMan]

Thanks for the feedback. You guys lost me on the 'path length' stuff. I have a friend that's an EE perhaps he can explain further. My tech used a small handheld meter that had a green and red light. He played a repeated test signal and held the meter next to each driver. I don't remember the exact sequence of flashes for the test lights but for example the tweeter signal produced a red red green flash while the mid and woofer had a different sequence something like green red green (you get the idea). This is how he tested the polarity.

I wonder if anyone at Klipsch can tell me the factory spec for the polarity of the AK-2 and the drivers for the 87 horns? Moderators, can you help me with this? I just want to know that it's correct. I want to make sure that when the mod was done to bi-wire enable the network something didn't get screwed up.

Also, thanks for the diagram. Does it show how the AK-2 should be set for bi-wire? One set of my speaker wire goes directly to woofer network. The other set goes directly to terminals on the mid/tweeter network. The is no "jumper" between the two networks.

Thanks!

[JohnA]

Path length is the horizontal distance between the voice coils of 2 drivers, or the distance the sound must travel before it reaches the "front" of the speaker.

Both of your crossover networks should be identical in all respects.

[lovemyhorns]

Oddly, on my AK-2 crossover, off of the tweeter (+) connection where that 2uF cap and 500uH go to tweeter (-) my cap and coil are opposite of the scematic (the cap goes to tweeter (+) and the coil goes to tweeter (-)

Does this present a problem??

[Q-Man]

I believe some networks reverse the polarity of the midrange driver and or the tweeter. This will eather increase or decrease the output at the crossover point to give a flatter responce.

[JohnA]

Since the schematic I posted was published by Klipsch, I'd make my crossovers match it. If you want to confirm the schematic is correct, call Klipsch. If you find it in error, please post the information on the Forum.