You mention that the mic position is the same relative to the speaker, but I would take it a step further and measure both speakers in exactly the same place, keeping everything as similar as possible, including the position of you in the room.
Edit: looks like it's a bit late for this suggestion...
Edit: looks like it's a bit late for this suggestion...
Actually it is not too late. I think I did exactly as you suggest. I used the same right channel of the amp for both speakers, without touching the volume control or anything else on the preamp. They are floor standing so I put masking tape on the floor and marked the exact position, and then just interchanged the speakers. The stand for the mic or the mic was not touched between the test of the left and right. For the last test I even put some throw rugs on the floor as it is hardwood, but it did not seem to make a difference. To read the iPod Touch I used a stool about 4 feet behind the mic and stood on that to read the dB level without getting close to the mic. That in fact seemed to be the biggest influence. Totally different readings if you stood right behind the mic. Even from 4 feet back leaning forward and back would change the dB level with some of the frequencies. Others it was rock solid no matter what you did. Those rock solid responses were the ones I depended on the most, and in most cases those were 85 dB. I suspect 85 dB is very close to the true sensitivity at whatever voltage I was using (did not measure it).
In any case it is not hard to see why an anechoic chamber is used with a remote mic to measure frequency response accurately. It is not an easy thing to do accurately. Even turning everything off I could in the house I did not get background noise below about 33 dB. I think if I could have found pink noise bands instead of pure notes it would have been less sensitive. I do have those on vinyl but that is too much of a pain to use...
Guys, Im new here and I have some strange idea but need someones opinion. Its about tweeter in my two way system. I have xo for 4 ohm tweet but now Im putting in a 8 ohm driver. Beside that, this tweeter is about 5db louder than midbas. The idea is to put 8 ohm 20 w resistor in paralel to driver after xo and fix my problems. Or not?
True. Pure 12db xo at 2.5kHz for 4 ohm tweeter. I think that splitting drive current in half would atenuate sound pretty much, maybe even for 5 db. But i question sound quality in this case. I've never done this before but I don't have much time for testing, that's why I ask for other opinions. I know my own test would be the best option.
Thanks
Thanks
4 ohm tweeters are often very loud. That it is not attenuated sounds strange. Perhaps your speakers were too bright in the past? I think you should try my recommendation. 2.5 ohm in series with the capacitor and 8 ohm across the inductor. Most crossovers are so screwed up from the factory anyway. There is nothing there to "guard". If the tweeter becomes too attenuated from this, move the 2.5 ohm to the side of the inductor, with the other resistor. Make the 2.5 ohm 20 watt. For example two 5 ohm resistors in parallel that are 10 watt each.
Last edited:
I understand L-pad.
The point of my issue is -I don't want to keep nominal impendance of driver, but cut it in half + attenuate for about 5 db.
For the crossover to work as designed, it needs to see the same impedance with the new tweeter as with the old, so the L pad needs to do two things - maintain the total impedance, and incorporate the attenuation you require. (It may be simpler to use a variable resistor & replace with two fixed resistors once you've found the level required)
The beauty of an L-pad is that it will partially flatten out the impedance, bringing it closer to the L-pad design impedance. The more you attenuate, the more it will do this. For example, if you pretend the tweeter is a four ohm tweeter and calculate the L-pad for 4 ohms and attenuate by a fraction of a dB, it will still look like an 8 ohm tweeter with all its peaks and curves....but if you attenuate by several dB then it all begins to look just like a 4 ohm resistor without the peaks and curves.
Hi, About 2R in series and about 3R in parallel will do it, rgds, sreten.
(1.8R series and 3R parallel gives 4 ohms and -5dB.)
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.