The "Elsinore Project" Thread

Hi, Joe,
We haven't corresponded with you in a while, hope all is well.

Hi Ken

Hey, good to hear from you, been wondering how you have been doing. No way I can sum up what I have been dealing with in the last few years, mental health of my son has been at times harrowing (schizophrenia) and a system that is broken. But don't want to go there, on other fronts I have been my usual busy self.

Yes, I get across the LR2 comparison from time to time and it is seeing the tweeter's reverse electrical phase is that catches the eye. But it is not LR2 and the step response looks like this and is the key:

1205lipfig.7.jpg

Lipinski L707

Now let us look at the tweeter in red:

1205lipfig.8.jpg

Lipinski L707

I am not the only one who has worked out, the above is from a speaker designed by Andrew Lipinski who worked with the late John Dunleavy who was wedded (welded?) to 1st order Butterworth. But Andrew and John just did not agree that was the best way.

Neither did I (I have a relationship with John's earlier company Duntech - an interesting story there). We, Andrew and I, independently ended up with a different way of doing 1st order as electrical crossovers and the step response looks like above, which is John Atkinson's measurement of the Lipinski L707. (But I have used a few tricks that AL hasn't, particularly reducing tweeter distortion - an obsession with me.)

Below is the Elsinores, compare with above:

elsino1_part-1.gif


Now we are starting to understand why the tweeter's phase needs reversing:

elsino12.gif


The perfect tweeter will have an infinite rise time, hence all its output is reversed. The ideal tweeter inverts its phase. Of course there is no such thing, but it supplies the clue that we can get the reversing out of the way (going negative) as quick as possible and line it up in time with the mid/bass driver, so that in time they both go positive together. Yes, it mans you have introduced a pole, but it is well above 10KHz most of the time and in one case got it up over 20KHz with 1 Watt input. Obviously, the pole swings in frequency as power changes, but in practice does not prove to be a problem.

Also, the off-axis response improves greatly, because now you get 100% vector summing there when you only get 50% with Butterworth. Off axis better, power response better.

That's just the tip of the iceberg, a phone or skype discussion is needed to get into the questions and answers part, as I can imagine you would have a number of those.

We try (with difficulty) to keep things here as theory-free as possible. This thread is mainly for constructors. It also helps to keep the 'usual suspects' from trying to crash in and use their textbook questions as weapons. Textbook jockeys I now call them.

But I am delighted to hear from you, comrade in arms - I still wear battle scars, maybe you do too?

Cheers, Joe

PS: Look below, a recent Stereophile review of a Swedish speaker.

521MPTriofig8.jpg


That step response is not by chance, it's done by somebody who knows something and when you see it, that if the speaker is done properly, it will have something about the sound. The review gave it top marks! Marten Parker Trio.
 
Yeah, and that speaker needs the protection with that tweeter. But the thing that brought a smile to my face was "Marten's 'Multi-Slope' crossover technology" and guess what, that's a good explanation for what I use in the Elsinores, that's the L3-C2 that I use after C1. That step response makes it pretty certain they are doing the same thing. :D

Note the price USD $37.000.
 
diyAudio Moderator
Joined 2008
Paid Member
Yes, it's a versatile topology. I often demonstrate it when members ask for a high order filter with a low component count. You get a faster rolloff with a slower phase change.

In this case a sixth order rolloff for reduced excursion, and a third (or forth) order fitting phase response.
 

Attachments

  • 6th.png
    6th.png
    121.2 KB · Views: 96
  • 6th phase.png
    6th phase.png
    102.7 KB · Views: 100
Now we are starting to understand why the tweeter's phase needs reversing:

The perfect tweeter will have an infinite rise time, hence all its output is reversed. The ideal tweeter inverts its phase. Of course there is no such thing, but it supplies the clue that we can get the reversing out of the way (going negative) as quick as possible and line it up in time with the mid/bass driver, so that in time they both go positive together. Yes, it mans you have introduced a pole, but it is well above 10KHz most of the time and in one case got it up over 20KHz with 1 Watt input. Obviously, the pole swings in frequency as power changes, but in practice does not prove to be a problem.

Is this explanation and example of when a speaker creates high and low pressure the same as what you are describing between electrical and acoustic phase?

Improving the loudspeaker through simulations with Rene Christensen of Acculution! - YouTube
 
Just got an email that FedX has landed my Purifi drivers.

Great news.

And a new speaker is going to me as a replacement for the damaged one. And I hope this time will be fine.

I think the speaker packaging is very badly made up. It holds the frame of the loudspeaker and the magnet has no support. In my opinion, this is a big mistake. So, as I wrote earlier, be careful when collecting the parcel.
 
We are back "on air" and the editor side looks very different and now you can use keyboard strokes like a word processor.

Some will actually like that.

Question: I will have my drivers on Monday I have been told, there is also acg and Andrew, so that means 24 drivers to Australia.

JarBar in Poland has his eght drivers. That makes it 30.

I believe Helmut and paid and placed his order. Is that 48?

Who else has ordered and paid for their drivers, I would like a total so far. It must be 40 or more?
 
We are back "on air" and the editor side looks very different and now you can use keyboard strokes like a word processor.

Some will actually like that.

Question: I will have my drivers on Monday I have been told, there is also acg and Andrew, so that means 24 drivers to Australia.

JarBar in Poland has his eght drivers. That makes it 30.

I believe Helmut and paid and placed his order. Is that 48?

Who else has ordered and paid for their drivers, I would like a total so far. It must be 40 or more?
I've been in contact with Madisound. Its a month or so before they will have any stock. I'm waiting to hear what the build/test outcome is but I'm 80% sure I'm along for the ride.
 
We are back "on air" and the editor side looks very different and now you can use keyboard strokes like a word processor.

Some will actually like that.

Question: I will have my drivers on Monday I have been told, there is also acg and Andrew, so that means 24 drivers to Australia.

JarBar in Poland has his eght drivers. That makes it 30.

I believe Helmut and paid and placed his order. Is that 48?

Who else has ordered and paid for their drivers, I would like a total so far. It must be 40 or more?
I've committed as well - in fact the first 2 of the shipment of 8 showed up today!
8A1A49DF-06DA-4672-803C-4DE0CF245FEB.jpeg
 
Member
Joined 2009
Paid Member
Well, I placed my order and payed, but I didn't hear anything else since then. Now the christmas holidays are coming, time to build another pair of Elsinores as well as a set of Hamletts. Material will be beech multiplex 25 mm and for the front Panzer multiplex 20 mm, 2 m² for about 900 €.
My tweeters are also pending, difficult to buy speakers these days.
 
Is! New! My precious!

And this is what the shipping box looks like. No support for heavy magnet. See how hard the speaker basket works. Bent edges. I don't like this technique of packing items.
 

Attachments

  • IMG_20211210_133307.jpg
    IMG_20211210_133307.jpg
    175.4 KB · Views: 147
  • IMG_20211210_133226.jpg
    IMG_20211210_133226.jpg
    224.9 KB · Views: 158
  • IMG_20211128_115240.jpg
    IMG_20211128_115240.jpg
    130 KB · Views: 148
Joe,

Sorry to hear about your son’s situation. He is fortunate to have a dad who cares about him. Please accept my best wishes for your family during this time. Me, I’ve been making it okay. Life does bring challenges, as you know.

Regarding the phase response of the Elsinore, it is an actual measured acoustical phase? If so, it doesn’t show the expected 180 degree phase rotation across the all-pass band. Is this because of the expected 180 degree rotation of an LR2 with inverted driver is only an electrical phase response, while the speaker’s resulting relatively flat acoustical phase response stems from your ‘renegade tweeter’ view?

Also, is the frequency of the negative going spike in the step-response a function of the crossover frequency, or is it only due to the tweeter’s bandwidth?

By the way, I had forgotten about Duntech.
 
Last edited:
Reporting another successful build of the Elsinore Mark 6 speakers. I used 18/25mm Baltic (Russian) Birch plywood that you can find at many plywood distributors in the US. I am attaching a couple of pictures showing finished boxes and the crossover layout. The speakers perform above all expectations, and absolutely deserve the praise they get on this forum and beyond. Thank you, Joe and everyone else who share their knowledge and experience with the DIYers everywhere!
 

Attachments

  • elsinores_front.png
    elsinores_front.png
    916.9 KB · Views: 237
  • elsinore_back.png
    elsinore_back.png
    639.8 KB · Views: 248
  • elsinore_crossover.png
    elsinore_crossover.png
    818.8 KB · Views: 244