Puzzled by tweeter response

[joeling39]

Hi,

I purchased a speaker kit recently but found that I did not like it even after more then 1 week of break-in. On a whim, decided to measure the frequency response using a TACT RCS 2.0 that I have on hand. To my dismay, the response fo the higher frequencies is very alarming indeed.

Basically there is a drop of more then 10dB from 10000hz to 20000hz. To a novice such as myself, this would seem to unacceptable. What do you think ?

The pic below shows the frequency response from approx 10khz to 20Khz. The amplitude between the 2 extreme is about 11dB (sorry about the poor picture quality)

Regards,
Joe Ling
Malaysia

[joeling39]

This picture shows the crossover network. HF is for the tweeter section.

[joeling39]

This is a schematic that I have drawn up based on the crossover board. Anyone can point out whether there is any obvious errors? Unfortunately, I do not have any spec. on the tweeter nor do I know how to figure out whether this network is ok or not. For some reason, the red wire is soldered to the negative speaker binding post while the white is on the positive. Can't tell whether the red goes to the positive of the tweeter or not as there is no marking on the tweeter.

[joeling39]

This is the crude setup used to make the measurement on the TACT

[5th element]

Is that an audionote loudspeaker? Looks a bit like one to me from the drive units to the cabinet.

A response like that could be due to several things,

1) the tweeter could actually have a response like that.
2) the microphone has a response like that.
3) the equipment used has a fault somewhere.
4) the xover network causes that response.

[5th element]

Looking at teh xover it shows nothing that would cause the reponse to droop like that. Simulating the xover on a scan speak d2905/97 doesnt show any signs of high frequency attenuation either.

Altho the scan is a completely different tweeter the overall impedance and frequency reponse should be close enough to see what the xover actually does.

Needless to say the xover doesnt do a good job of anything with the scan, but that wasnt to be expected.

[wigginjs]

What software are you using to measure the response?

[kelticwizard]

For all you know, the frequency response of your sound card might be 10 dB down at 20,000 Hz.

If you are going to use any software to measure frequency response, you must be sure your soundcard's frequency response is correct.

To do this, go to the site below and download RightMark Audio Analyzer:
http://audio.rightmark.org/index_new.shtml

All that is necessary is a simple, inexpensive stereo 1/8" to 1/8" plug. This is also known as a Walkman-to-Walkman plug. Very cheap.

Even a good soundcard can be installed incorrectly, and give some sound but not good sound.

When I first used RightMark, I was astounded to see that my soundcard's high frequency response was just terrible. I simply removed the soundcard driver from my Control Panel, then re-installed the driver. Then I retested with the RightMark Audio Analyzer.

The frequency response was very good, about 3 dB down at 20 kHz and 6 dB down at 20 Hz. Not perfect, but now I knew what to factor into my frequency response measurements.

So, it is very important to test your sound card, preferably with the Right Mrk Audio Analyzer.

[joeling39]

Thanks for all of your response.

I am using a TACT RCS2.0 room correction device for the measurement. Initially, I am was also thinking that there is something wrong with the measurement setup (crude it might be). I was thinking maybe my amp frequency response was poor. Anyway, I measured a SEAS tweeter used together with my Diatone & it wa fine (+_ a few dB right up to 20kHz). So, the TACT is fine as is the rest of the amplification chain.

If like what 5th element says - the crossover being ok, then I have more cause for concern as it would imply that the tweeter is the culprit. Serves me right for trying to save a few bucks & order a kit instead of going for audition & buying retail.

Yes, it is a AN speaker Kit 03.

Regards,
Joe Ling

PS : FYI, I was comparing this speaker against my DIY Diatone PM610MB with SEAS helper tweeter. Needless to say while I was expecting the AN to trounce the Diatone but the opposite occurred & I was a bit depressed by that. That sparked the investigation.

[5th element]

If the crossover schematic you have drawn is correct then that does not do anything to curtail the tweeters high end extension. Could it be that you have connected something up incorrectly?

I looked at the audionote website and thats a lot of money to pay for a kit!

Having studied your xover picture more closely it seems you have indeed made no mistakes. I am appaled to see really tiny non polarised electrolytic caps in that xover.

The one thing that puzzles me is that odd looking device you have labelled 1.2ohms. If it is a resistor then it looks like a wirewound made out of very thin wire, which would serve as a sort of fuse. If too much power passes thru it the wire would get hot, melt and protect the tweeter. However I think it would be a good idea to bypass that component and see if it makes any difference to the high end response.

If it is indeed acting like an inductor it could be reason why, altho that would not give you a 11db roll off over one octave.

Have you tried measuring from further away? say at 60-80cm?

Lastly contact audionote and tell them of your situation, say you have measured other tweeters with this system and that it is flat to 20khz, but when you measure the kit it is not at all pretty.