sooo I made 4-way speaker design.
to avoid confusion i will tell the full story.
I dont really have much knowledge of designing speakers, the drivers im using are from commercial speakers and i use the original crossover plans for the drivers I`m using,
the only thing i did (and needed to do) is changing the cross point of the low mid and adding resistors before the crossover, the rest of the crossover is untouched.
as i was doing measures inside the room i noticed there is a very big dip at 3khz-4khz.
outside measures shows there is something at that area, but nowhere near the same dip size as I got inside the room...
(note that the crossover resistors slightly change from measure to measure, and the mic position and such,
which is why the measurements differ from each other. I use the measurements just to to see the general response,
note that this dip has nothing to do with what i changed in the crossover, its the exact
cross as the original design Acoustic Research did, and I tried phase and what not, that dip was there to stay...
and this is where the rear tweeter enter the story, and this is the result:
I listened to the speaker for hours and it sounds bloody amazing, but its just one speaker,
making the other crossover for stereo gonna cost $$$ and time,
so I guess want to make sure i wont get some surprising acoustical problems in stereo.
also, Is if its okay to do such thing?, using a rear tweeter to make up for the lost frequencies?
more info to make the picture clearer:
to avoid confusion i will tell the full story.
I dont really have much knowledge of designing speakers, the drivers im using are from commercial speakers and i use the original crossover plans for the drivers I`m using,
the only thing i did (and needed to do) is changing the cross point of the low mid and adding resistors before the crossover, the rest of the crossover is untouched.
as i was doing measures inside the room i noticed there is a very big dip at 3khz-4khz.
outside measures shows there is something at that area, but nowhere near the same dip size as I got inside the room...
(note that the crossover resistors slightly change from measure to measure, and the mic position and such,
which is why the measurements differ from each other. I use the measurements just to to see the general response,
An externally hosted image should be here but it was not working when we last tested it.
note that this dip has nothing to do with what i changed in the crossover, its the exact
cross as the original design Acoustic Research did, and I tried phase and what not, that dip was there to stay...
and this is where the rear tweeter enter the story, and this is the result:
An externally hosted image should be here but it was not working when we last tested it.
I listened to the speaker for hours and it sounds bloody amazing, but its just one speaker,
making the other crossover for stereo gonna cost $$$ and time,
so I guess want to make sure i wont get some surprising acoustical problems in stereo.
also, Is if its okay to do such thing?, using a rear tweeter to make up for the lost frequencies?
more info to make the picture clearer:
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Adding a rear tweeter changes the energy bilance within the room, the reflected (indirect) sound also adds to the tonality. You've measured with the room influence. The dip in 3-4kHz is most likely the beaming of the drivers in that range, likely a too big midrange driver. The problem is, your rear driver adds quite a lot sound pressure level, which adds more presence and room depth but also impacts the exact location of the reproduced stage. To avoid the back tweeter become too dominant, there is often a voltage divider (1 series and one parallel resistor) in place. It is ofcourse dependent on your personal taste what's 'too much' and what's not. Maybe put an L-Pad in front of the back tweeter and try to find the right dose for your taste, you'll likely take a while to listen to it and to adjust the L-pad. You should also try if the best location of the diffuse sound is really on the back, in some cases (especally if the speaker is close to the wall or got furniture close to it) it's better to put the extra tweeter on the top of the speaker. Maybe that doesn't work great but it's still worth a try.
The back tweeter is an effect. Like any effect added to the sound, it might run boring or annoying after a while. Maybe put an L-Pad in place for the tweeter aswell as an off-switch to make it easier to find the right dosage for you and to compare with and without the back tweeter. In the most cases that helps a lot to decide what's best for you.
The back tweeter is an effect. Like any effect added to the sound, it might run boring or annoying after a while. Maybe put an L-Pad in place for the tweeter aswell as an off-switch to make it easier to find the right dosage for you and to compare with and without the back tweeter. In the most cases that helps a lot to decide what's best for you.
thanks for the answer!
I doubt its beaming, the highmid\tweeter were basically made to each other (they share the same magnet structure
)
The first thing I did is place LPAD (the good old type that u can adjust) on the rear tweeter but surprisingly i liked it without the LPAD,
but will probably take off 2\3DB from the tweeter on the final design.
About position, they are 1 meter away from the front wall and i got skyline diffusers on the wall as well,
its the minimum distance from what i rear around to get the proper effect of both the diffusers and rear tweeter
Last edited:
1m away from the wall is likely a good distance for that setup.
If the speaker itself is linear and got a good, mostly even dispersion, it has to be something else. You've made the measurements, if that's not the result of beaming, then it's either an error in the crossover or a trait of your room/enviroment. That response is usually a very good indicator for a bad crossover or very uneven dispersion.
If the speaker itself is linear and got a good, mostly even dispersion, it has to be something else. You've made the measurements, if that's not the result of beaming, then it's either an error in the crossover or a trait of your room/enviroment. That response is usually a very good indicator for a bad crossover or very uneven dispersion.
You need to adjust your gates, they're too long
While that's true, it only affects mostly the lower frequencies, in the upper midrange (where the back tweeter works) it's not that greatly affected. Aside from that, it's the response in the whole room and that's what matters here. It's the reflections and the response in the room you need to measure here. If you gate the reflections out, you can't get the influence of the back tweeter at all. The problem is, you can't apply everything you've learned blindly to every situation, you need to differentiate what measurement you're actually doing and what's the goal of the measurement.
i really doubt its the room since other speakers I measure dont have this problem.
anyway either I made a mistake while reconnecting the crossover or that I`m doing something wrong and i lack the knowledge to realize what it is
either way i let my ears decide in the end, and they sure as hell like what they hear atm, and as long as that rear tweeter isn't something i shouldn't do thats all i needed to hear.
Well, with that as a given fact, which is certainly possible but also definitely means, either the speaker got that non-linear response inherited or it is, indeed, the bad dispersion. Or it is a serious deviation in the measuring setup.
anyway either I made a mistake while reconnecting the crossover or that I`m doing something wrong and i lack the knowledge to realize what it is
either way i let my ears decide in the end, and they sure as hell like what they hear atm, and as long as that rear tweeter isn't something i shouldn't do thats all i needed to hear.
That depends. If your reference is your ear and it's fine according to it, then the measurement doesn't matter at all. Or you go after the measurements which show you've still got work to do. It might also be the measurement is fine but you simply don't favour the linear response. The measurement itself might also be at fault. It might be either of these but they mutually exclude each other. If it fits your taste, everything's fine (but don't bring measurements for arguing then please).
It`s not that I`m ignoring the problem completely, I still plan to work on finding why that dip is there, but the speaker sound so good with the rear tweeter (while i can see that it makes up in the dip area as well)
so basically worst case the dip stays, i can still enjoy the hell out of this speaker, the rest is a bonus .
anyway I thought i never saved it but apparently i made some squiggles of the original AR speaker when had it
An externally hosted image should be here but it was not working when we last tested it.
Red is the tweeter
Purple is without the tweeter
Green is when everything working (ignore the blue)
Going to make a very calculated guess that i ****** up with the crossover somehow
I hope this thing gonna be solved by the end of the week.
Just updating, it seems that the problem is in the tweeter, it rolls of at 6Khz while it should roll of at 4Khz?
I have been playing around with the crossover and position, but the tweeter refuses to roll off after 6Khz, even played with the high mid cross which seems like it can go up to 6Khz flat still does that same dip.
gonna assume that this "problem" is something the designer meant to do at this point.
It somewhat makes sense as well, isn't 3Khz area is generally agreed as out most sensitive frequency for our ears?
measurement is without the rear tweeter and woofer inside the room ( mic is 1meter distance)
I have been playing around with the crossover and position, but the tweeter refuses to roll off after 6Khz, even played with the high mid cross which seems like it can go up to 6Khz flat still does that same dip.
gonna assume that this "problem" is something the designer meant to do at this point.
It somewhat makes sense as well, isn't 3Khz area is generally agreed as out most sensitive frequency for our ears?
measurement is without the rear tweeter and woofer inside the room ( mic is 1meter distance)
Attachments
Yes, it's pretty much preferable to have a dip there instead of a peak. The tweeter and the mid are so close to each other, a xo frequency of 6kHz should not be a problem. To get comparable measurements you should do a gated measurement in standard (1m) or close distance (30cm), including angle measurements, not room measurements.
Well, it's ungated, that means there are a lot of reflections in it. To get to the source of the problem, you need clean measurements without reflections, ie. outdoors or indoors but gated.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.