That used to be the case, but now the DE250 is so hard to get that I have had to squarel away my cache.
I hope that's temporary.
The other day during testing, my amp went into oscillation producing what sounded like a midrange square wave. It was probably +/-23V. Could this be enough to damage these drivers?
I did not say that they were not available elsewhere, only from me, because I cannot keep them in stock for my own usage. I use to sell them much cheaper than what Parts Express gets for them, but yes, they are always available at those prices.
And, there is no guarantee that Parts Express actually has them in stock. If not then they won't have any more until May either.
And, there is no guarantee that Parts Express actually has them in stock. If not then they won't have any more until May either.
Only just found these responses.
Earl, you said that you've found harshness comes from the waveguide, but I suspect you found that because you started with the better drivers. IMO the Selenium is an obnoxious driver. We wanted to find out where waveguides or drivers were at fault. We found relative differences between the waveguides, some were better than others. We also found the Selenium could be blamed. Put the B&C in the same waveguide and there really wasn't much harshness most of the time. Put in the foam (which I've tried but the others in the event haven't had a chance to hear yet) ... suddenly you get back to that easy on the ears sound of domes, yet with all the advantages of the compression driver! And like you said somewhere, it did sound slightly dull at first and that was a surprise because I found before the foam, it certainly wasn't dull. Yet there was a sound to them all, probably HOMs that once detected, the ears become sensitised (kind of like seeing the man in the moon, you then can't un-see him). Then I find you hear it even at lower levels and in more laid back tracks. Where at first you don't hear it, then once your brain locks in to the sound, you hear it all the time. Of course, with a bad example it just shouts at you, but with say B&C + an oblate waveguide which is already pretty good, you get the really obnoxious harshness only at certain points.
So far I have to say that where I've tested the things Earl talks about, they work! No great surprise there as the theory behind it all had me all but convinced, as much as you can be convinced without hearing it. Kudos to Earl for putting the info out there. I'm now more confident than ever that Earl is on the right track with his speakers. Those who want to just jump to a result and not tinker away like a nerd on a mission would do well to just buy one of the kits.
BTW, if anyone in Melbourne Australia does have a Gedlee kit, please let me know. I'd be interested to hear them.
When it comes to addressing the harshness, I find people often have all kinds of ideas, many of them dancing around the edge and some band aid style. My acoustics lecturer used to say "a problem is always best addressed at it's source. The closer to the source, the better the solution." That's why I favour the foam, it is as close to the source as you can get. You are addressing the cause. As you move away from it, even as you do anything that is outside the waveguide itself, you have allowed the problem to "ripple out" and now you are trying to cover the effects of the problem. The cat is out of the bag now and is about to get it's claws out. Would have been much easier to keep it in the bag!
Earl, you said that you've found harshness comes from the waveguide, but I suspect you found that because you started with the better drivers. IMO the Selenium is an obnoxious driver. We wanted to find out where waveguides or drivers were at fault. We found relative differences between the waveguides, some were better than others. We also found the Selenium could be blamed. Put the B&C in the same waveguide and there really wasn't much harshness most of the time. Put in the foam (which I've tried but the others in the event haven't had a chance to hear yet) ... suddenly you get back to that easy on the ears sound of domes, yet with all the advantages of the compression driver! And like you said somewhere, it did sound slightly dull at first and that was a surprise because I found before the foam, it certainly wasn't dull. Yet there was a sound to them all, probably HOMs that once detected, the ears become sensitised (kind of like seeing the man in the moon, you then can't un-see him). Then I find you hear it even at lower levels and in more laid back tracks. Where at first you don't hear it, then once your brain locks in to the sound, you hear it all the time. Of course, with a bad example it just shouts at you, but with say B&C + an oblate waveguide which is already pretty good, you get the really obnoxious harshness only at certain points.
So far I have to say that where I've tested the things Earl talks about, they work! No great surprise there as the theory behind it all had me all but convinced, as much as you can be convinced without hearing it. Kudos to Earl for putting the info out there. I'm now more confident than ever that Earl is on the right track with his speakers. Those who want to just jump to a result and not tinker away like a nerd on a mission would do well to just buy one of the kits.
BTW, if anyone in Melbourne Australia does have a Gedlee kit, please let me know. I'd be interested to hear them.
When it comes to addressing the harshness, I find people often have all kinds of ideas, many of them dancing around the edge and some band aid style. My acoustics lecturer used to say "a problem is always best addressed at it's source. The closer to the source, the better the solution." That's why I favour the foam, it is as close to the source as you can get. You are addressing the cause. As you move away from it, even as you do anything that is outside the waveguide itself, you have allowed the problem to "ripple out" and now you are trying to cover the effects of the problem. The cat is out of the bag now and is about to get it's claws out. Would have been much easier to keep it in the bag!
I like to view data both normalised and non-normalised. They have different purposes.
Here is the mentioned JBL+De250 without normalisation and 5 minute xo job (active).
No axial holes with these JBLs so it looks like a real winning combination for wg fans.
It looks superb! 
I'd love to use a pair of these in some speakers, but the thing that stops me is driver integration regards centre-to-centre spacing.
Earl, you obviously consider power response one of the most important factors in speaker design, how do you deal with / justify the serious suck out that would occur in the vertical axis with such a system as woofer - compression driver?
I'd love to use a pair of these in some speakers, but the thing that stops me is driver integration regards centre-to-centre spacing.
Earl, you obviously consider power response one of the most important factors in speaker design, how do you deal with / justify the serious suck out that would occur in the vertical axis with such a system as woofer - compression driver?
Hi Paul - thanks for the writup.
It can be very difficult to determine things like "harshness" because the drivers and the waveguides "couple", meaning they can act differently on different waveguides. I see the identical driver act differently on three different waveguides. What I have tried to do is elliminate all of the sources of harshness, I never really rank ordered them or tried to find which ones came from which source (I'd like to do that someday). So what I meant with my comment is that the "limiting" factor is the waveguide. I can see where some drivers are so bad that they can make the waveguide superfluous.
There is no hole because it is not round.
Sorry but all these plots look too good to be real. As I said, everything done with ARTA looks this way. When I measured this JBL waveguide it was no where near as good as you are showing. Having a sharp edge at both the throat and the mouth there should be some serious diffraction and resonance issues. There are in my data, but some how not in yours.
Are you referring to this?
It looks excellent to me. Did you use D220Ti or DE250 ?
As paul mentioned we mounted DE250 to the JBL and the harshness was gone. I eq the D220Ti to similar but not ideal effect. It would be interesting to add foam to these square WG. I may need to start getting them foams....
Apparently some people in poland dislike the sharp edges and came up with this one..!
An externally hosted image should be here but it was not working when we last tested it.
I'd love to hear your speakers one day. I heard the Orions last few weeks and the Summa is practically the only one I'd want to listen to next. Pity none in Melbourne (???)
Nope, none of my speakers in Australia at all.
We seem to see different things in the results. I see a lot of resonances that do not show in your results. Why is that?
We are still a long ways from measurements that reflect reality. Mine are all custom because I seek to see what it is that I hear. Other measurements seem to seek to show something that "looks good". One of us is right and the other wrong. I am comfortable with my position and why I do not use "canned" measurement software.
The "Poland knock-off" still has a huge discontinuity at the throat.
We seem to see different things in the results. I see a lot of resonances that do not show in your results. Why is that?
We are still a long ways from measurements that reflect reality. Mine are all custom because I seek to see what it is that I hear. Other measurements seem to seek to show something that "looks good". One of us is right and the other wrong. I am comfortable with my position and why I do not use "canned" measurement software.
The "Poland knock-off" still has a huge discontinuity at the throat.
I don't know. I simply look at the data and what's there.
These measurements do look insanely good. But not all my measurement looks good! As paul posted, this was the best of the day.
In post #36 which was the result of your measurement of the JBL I see that as excellent too.
To see resonances it's best to utilise FR plot:
DE250 + Econowave JBL
red=1.5khz LR2
These measurements do look insanely good. But not all my measurement looks good! As paul posted, this was the best of the day.
In post #36 which was the result of your measurement of the JBL I see that as excellent too.
To see resonances it's best to utilise FR plot:
DE250 + Econowave JBL
red=1.5khz LR2
Last edited:
Making some assumptions (not having heard it) about the plot in post 36 there are a few frequencies of interest. Such, for example, as 4kHz. Here there appears to be a problem with the waveguide as shown by the response at the further off axis points. This wouldn't be a resonance like one that could be fixed with EQ. If it were, the off axis plots would be consistent with the on axis ones. This one appears to be a mechanical issue that would need to be fixed accordingly.
It also doesn't appear to have constant directivity as indicated by the relatively level (not tilted) plots. By comparison, your sonogram in post 31 doesn't seem consistent.
It also doesn't appear to have constant directivity as indicated by the relatively level (not tilted) plots. By comparison, your sonogram in post 31 doesn't seem consistent.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.