Which is yet another reason why an upfiring or backfiring tweeter makes so much sense.
But, I am curious; at what distance were the on-axis/off-axis tests conducted? If they were at the 1" distance you recommended in another thread, it seems there might be more physics at play to skew the results than plain old "beaming".
The reason I say this is because my dual EL70 microTowers do not sound anything like your graphs look, if that makes any sense. Perhaps this is due to the fact that my speakers are in the "Castle' configuration (one driver firing up, one forward).
Either way, I do not find that there is much difference at all in the high end whether I am perfectly on axis with them (via appropriate toe-in of each) or standing in the adjacent dining room. Laid-back, rolled-off highs is just the way they are.
For close to 18 months, this has been fine for me. Lately I've noticed, however, that cymbals and such just don't have the "presence" that they should (again, regardless of the orientation of my ears to the drivers).
With the upfiring tweeters, the presence is there.
Last edited:
I said measure at 1" to get a nearfield measurement. The measurements I've posted were at 1m distance. Maybe 0.8m because I was indoors for these and a small single driver can be measured at 0.8m without any issues.
It's likely related to your castle configuration that you're not hearing what I'm hearing, because I've found them to behave like this since the first time I have the EL70 at home (Dave lent me a couple pairs quite some time ago). I am accustomed to high dispersion designs though. Lately I've been doing controlled directivity designs, and the beamyness of the EL70 is less noticeable.
I however disagree this is a reason to use upfiring/rearfiring tweeters. Those just spray sound everywhere we don't want them to. I could write a page about why that's a bad idea. Admittedly though, I've never done it.
It's likely related to your castle configuration that you're not hearing what I'm hearing, because I've found them to behave like this since the first time I have the EL70 at home (Dave lent me a couple pairs quite some time ago). I am accustomed to high dispersion designs though. Lately I've been doing controlled directivity designs, and the beamyness of the EL70 is less noticeable.
I however disagree this is a reason to use upfiring/rearfiring tweeters. Those just spray sound everywhere we don't want them to. I could write a page about why that's a bad idea. Admittedly though, I've never done it.
There seem to be as many people who feel front-mounting a tweeter with a fullrange is a much worse idea than back-mounting it. Pit Hinder's explanation makes a lot of sense to me. The point is not necessarily for the tweeter to handle the entire upper portion of the spectrum (as in a true multi-way design) but to simply supplement. Back-firing (and top-firing, it seems) allows one to do this without the corresponding time domain issues of overlapping a full range and a tweeter in a traditional manner. The "air" is there, but without the negative effects. The full ranger is still providing pretty-much everything, so imaging, sound-stage, etc. remain.
At least the way I have deployed it, the change is very subtle and inconspicuous, but immediately obvious when disabled.
BTW, I did try rear-firing the tweeters last night for a brief time and I found it unpleasant (relative to up-firing, which is completely non-fatiguing). Perhaps I should have reversed polarity on them in that configuration. I don't know...
Last edited:
More anecdotal mumbo-jumbo, of course, but I should tell you something that happened last night.
I've played Patricia Barber's Cafe Blue (88.2, 24-bit) dozens of times. Last night I am sitting there beside my wife appreciating the liveliness of the cymbals and the tapping of the sticks on the edges of the drums, etc. and I look over at my wife and she is sitting there playing "air drums", tapping the imaginary drum sticks on the cymbals, etc.
This is extremely out of character for her! So I sort of laughed under my breath and she glanced up at me.
"Who is this? Is this new?" she said.
"No, you've heard it dozens of times... You're just hearing it differently."
"Oh, those twitters things?" (She's Japanese)
"Yes, dear. The twitters."
I've played Patricia Barber's Cafe Blue (88.2, 24-bit) dozens of times. Last night I am sitting there beside my wife appreciating the liveliness of the cymbals and the tapping of the sticks on the edges of the drums, etc. and I look over at my wife and she is sitting there playing "air drums", tapping the imaginary drum sticks on the cymbals, etc.
This is extremely out of character for her! So I sort of laughed under my breath and she glanced up at me.
"Who is this? Is this new?" she said.
"No, you've heard it dozens of times... You're just hearing it differently."
"Oh, those twitters things?" (She's Japanese)
"Yes, dear. The twitters."
I personally don't agree with the super tweeter concept in general. IMO, the EL70 is a one man show. No help up high or down low. That's what it's good at. That doesn't mean it's perfect. But if you need one driver that does it all, this is it.
As for up firing, Pit Hinder's explanation of time domain would conclude that up firing is a bad idea. There will be some direct sound, albeit delayed from the full range drivers sound. Because you have experienced the up firing to be better, then it may be reasonable to conclude time isn't the issue. When you up fire, you fill in the off axis, but at a time delay. That's my problem with it. There is likely major nulls in the response that aren't audible, but it's missing information. So it should be audible.
As for the castle configuration, the reason it works, is because it's not omni above about 2khz. If it were, then it would also be bad. Even so, I believe there is likely a null around 1khz due to the offset. Probably not so bad though, and the advantage of baffle step compensation and higher dynamics outweighs what ever little driver interferance there is.
When you rear fire, you are not getting any sound other than reflected sound off the rear wall, which is likely around 10msec delayed. Ok, fine, but why do we need reflected sound from the rear wall? Will that add spaciousness? I doubt it.
EDIT - Have you tried front firing? What tweeter are you using? If it's the ERT26, I doubt it's putting out much of anything above 8khz at a 90degree angle like that. Most treble we hear is in the <8khz range anyways.
The few rear tweeters I've heard were on commercial speakers and I didn't like them at all. But I wasn't privy to the design details, so maybe they were just crappy speakers. If I had to choose between up or rear firing, I'd take rear firing. But again, I'm just talking theory here, I've never tried to do it.
As for up firing, Pit Hinder's explanation of time domain would conclude that up firing is a bad idea. There will be some direct sound, albeit delayed from the full range drivers sound. Because you have experienced the up firing to be better, then it may be reasonable to conclude time isn't the issue. When you up fire, you fill in the off axis, but at a time delay. That's my problem with it. There is likely major nulls in the response that aren't audible, but it's missing information. So it should be audible.
As for the castle configuration, the reason it works, is because it's not omni above about 2khz. If it were, then it would also be bad. Even so, I believe there is likely a null around 1khz due to the offset. Probably not so bad though, and the advantage of baffle step compensation and higher dynamics outweighs what ever little driver interferance there is.
When you rear fire, you are not getting any sound other than reflected sound off the rear wall, which is likely around 10msec delayed. Ok, fine, but why do we need reflected sound from the rear wall? Will that add spaciousness? I doubt it.
EDIT - Have you tried front firing? What tweeter are you using? If it's the ERT26, I doubt it's putting out much of anything above 8khz at a 90degree angle like that. Most treble we hear is in the <8khz range anyways.
The few rear tweeters I've heard were on commercial speakers and I didn't like them at all. But I wasn't privy to the design details, so maybe they were just crappy speakers. If I had to choose between up or rear firing, I'd take rear firing. But again, I'm just talking theory here, I've never tried to do it.
Last edited:
We are at odds because I have zero theory but I have actually tried it and I like what I hear.
I think there are simply too many variables to consider in the real-world applications of loud speakers, which pushes the final implementation firmly into the realm of art as much as science.
In an anechoic chamber, all of your theories would "ring true" I think. With all the other variables involved, including luck, I find it less than surprising that a pair of 1978 Sansui tweeters stuck on the top of my microTowers adds something very wonderful to my experience of the music I listen to. But that's the way I usually roll. I just try sh!t. Sometimes it works, sometimes it doesn't, but I usually break all the rules one way or another.
I think there are simply too many variables to consider in the real-world applications of loud speakers, which pushes the final implementation firmly into the realm of art as much as science.
In an anechoic chamber, all of your theories would "ring true" I think. With all the other variables involved, including luck, I find it less than surprising that a pair of 1978 Sansui tweeters stuck on the top of my microTowers adds something very wonderful to my experience of the music I listen to. But that's the way I usually roll. I just try sh!t. Sometimes it works, sometimes it doesn't, but I usually break all the rules one way or another.
I didn't create this thread to start an argument, but I take on board everyone's point of view. I've ordered the tweeters and caps, 2 x 1uF and 2 x 0.33uF. This will give me the 20KHz cross over point for the tweeters. I will experiment, by placing them facing in different directions and see what I prefer. I'll post my finding when my parts turn up.
Thanks again for the help with the sizing of the components.
Thanks again for the help with the sizing of the components.
Guys,
Do what gives you the greatest emotional connection.
I just finished wiring drivers in a box that would fir this discussion. MicroTower II inspired bipole microTowers with CHP70.2eN and ERT26eN on the previously unused faces. CHPs wired for 4 ohm, ERTs wired for 16 ohm with 1.8 uF series cap.
dave
Do what gives you the greatest emotional connection.
I just finished wiring drivers in a box that would fir this discussion. MicroTower II inspired bipole microTowers with CHP70.2eN and ERT26eN on the previously unused faces. CHPs wired for 4 ohm, ERTs wired for 16 ohm with 1.8 uF series cap.
dave
You're right, we're at odds. But I don't disagree you probably get a lot of enjoyment out of your added tweeter. I just don't think it's because of additional top octave output
We'll have to agree to disagree (apparently we appear to be arguing, but I haven't lost any respect for you throughout the discussion).
Anyway, I have the tweeter hooked up the the "B" terminals on my receiver.
When I enable and disable the "B" speakers, there is a subtle but significant difference in the highs.
I don't know what else I can say, but that's what I hear. My wife hears it and she didn't even know she was supposed to...
So, if it isn't the additional top-octave output, why are we enjoying the tweeters?
A quick google; "Typical hi-hats are usually between 300-3000 Hz dominant frequencies, and can extend up to 10-17k Hz for crispness, “air” and sparkle."
Ride cymbals and crash cymbals go there as well.
Not to mention the harmonics of fingers sliding on guitar strings, etc. which I have been able to hear much better with the tweeters.
I am convinced it is this +10k area that I am hearing. I know I can hear all the way up to 15k, as I have tested it several times.
Ride cymbals and crash cymbals go there as well.
Not to mention the harmonics of fingers sliding on guitar strings, etc. which I have been able to hear much better with the tweeters.
I am convinced it is this +10k area that I am hearing. I know I can hear all the way up to 15k, as I have tested it several times.
Last edited:
While I'm waiting for my tweeters to arrive I decided to get a little bit nerdy. I downloaded a demo of a piece of software called FuzzMeasure and installed it onto my MacBook Pro. I connected the output of my MacBook to my LM4780 chip amp which runs my Derwent full range speakers with the CSS EL70 drivers.
I placed the MacBook 1m from the speaker and ran the test. Here is the result.
I was only using the built in microphone of the MacBook as I haven't got an expensive mic to test speakers with, but before I tested my speakers, I ran a quick test using the MacBooks internal speakers and it's microphone and the frequency response was pretty flat. From from the graph, you can see a drop off starting at about 7Khz, which is significant after 10K. I think the addition of tweeters will definitely help my speakers.
I didn't have the amp or output of the MacBook turned up full, so the magnitude isn't as high as it could have gone, but it still gives me an idea to how the speakers perform.
I placed the MacBook 1m from the speaker and ran the test. Here is the result.
I was only using the built in microphone of the MacBook as I haven't got an expensive mic to test speakers with, but before I tested my speakers, I ran a quick test using the MacBooks internal speakers and it's microphone and the frequency response was pretty flat. From from the graph, you can see a drop off starting at about 7Khz, which is significant after 10K. I think the addition of tweeters will definitely help my speakers.
I didn't have the amp or output of the MacBook turned up full, so the magnitude isn't as high as it could have gone, but it still gives me an idea to how the speakers perform.
Last edited:
In the last month I have added rear and upward firing Piezo horn tweeters crossed over at 6K with a 1.22uf stacked film feeding a 22 ohm resistor and the Piezo is connected across the 22 ohm, with a 20 ohm connected to the + lug of the Piezo. This is powered by the main amp. The Piezo's are mounted higher that the main speakers. These very inexpensive horns have made a good improvement in "air" and decay, makeing the room sound larger that it is. The whole project cost less that $20 US. Can't beat a Piezo for cost and high freq SPL per dollar.
I got a little nerdy today too. Found REW online and was happy to find it runs in Linux perfectly.
Measured from about 1m away, on axis.
Blue = just EL70
Red = just tweeter
Green = EL70 + tweeter
My tweeters are coming in way earlier than thought. I suspect the old caps are no good. Going to recap with 2.2 uF and maybe even 1uF, and do more measuring. Although it appears my EL70s can use a bit of help starting at around 4k.
Measured from about 1m away, on axis.
Blue = just EL70
Red = just tweeter
Green = EL70 + tweeter
An externally hosted image should be here but it was not working when we last tested it.
My tweeters are coming in way earlier than thought. I suspect the old caps are no good. Going to recap with 2.2 uF and maybe even 1uF, and do more measuring. Although it appears my EL70s can use a bit of help starting at around 4k.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.