Aa007,
That's a very nice looking pair of Elsinore speakers, congratulations on your successful build.
What amp are you using with your Elsinores, and what genres of music have you tried out with them?
That's a very nice looking pair of Elsinore speakers, congratulations on your successful build.
What amp are you using with your Elsinores, and what genres of music have you tried out with them?
Thanks! I have them currently connected to two very different amps via a speaker switch box I made: Pioneer SC-65 (140W class-D amp) and a DIY KT88 tube amp (~20W output). There are no issues with driving these speakers with either amp, while SC-65 has many more room EQ features of course. I had them in use only for about 3 days now, and have noticed that the imaging has gotten even better after more break-in hours. I have taken them through all the decades, starting with Miles, Nina and Chet, got hang up on TDSOM for a while (those clocks sounds just unreal!), then on to the Thriller, and more recent things like Daft Punk - they did everything very well with a presentation that just makes you seat down and listen closely again to the things you've heard a million times before. 🙂
While checking out the new Hamlet updates, I noticed that for my EL-Mk6 crossover layout above I followed an earlier schematic posted in https://www.diyaudio.com/community/threads/the-elsinore-project-thread.97043/page-179#post-6475993 as well as at the canonical https://www.customanalogue.com/elsinore/elsinore_index.htm URL. It looks like there was a new EL-Mk6 crossover schematic posted recently in https://www.diyaudio.com/community/threads/the-elsinore-project-thread.97043/page-203#post-6753166 though.
Joe, it would be great if you could briefly clarify the need and effects of the recent EL-Mk6 crossover changes. Thank you!
Joe, it would be great if you could briefly clarify the need and effects of the recent EL-Mk6 crossover changes. Thank you!
Where to start?
Yes, it is acoustical. The phase inversion only happens well above 10KHz and everything below that pole is in phase and that includes the crossover frequencies. They are way out of range.
It is not LR2 and the crossover is electrical 1st order. Unlike Butterworth, this sums -6dB as LR crossovers and not -3dB. Butterworth goes sum, cancel, sum, cancel and that means it only sums 50% of the time and that is the reason Butterworth has to sum at -3dB and that is not good. This was also realised by LR where the idea is 'let us get that phase right.' They sum at -6dB.
But this is not what we aim and what we are doing (because clearly, I am not alone) is quite different. We can reverse the tweeter and yet acoustically it inverts the phase. This only applies to the tweeter and not other lower frequency drivers. Indeed the perfect/ideal tweeter should have infinite upwards bandwidth by having infinite rise time:
You might as well turn it around, physically locate the tweeter to match up with the midrange - and the tweeter is a 100% inverter.
But of course, in the real world the tweeter will be fast, but it will not be infinite. With little power input, the reversing will be at a high frequency, often above 20KHz with music. With a lot of power to the tweeter, it will still be over 10KHz. The more power, it takes time to get the peak, four times the power will lower the pole by an octave, assuming linear acceleration.
One of the huge advantages is that we can, at the crossover, get both drivers 100% in phase. This is still first order except that it sums flat even if the drivers are -6dB at the crossover frequency. Conventional first order Butterworth has to sum at -3dB because the drivers only sum 50%, every second quarter of a cycle. Summing -6dB also gets you lower distortion.
Note Butterworth on the left and perpendicular to the drivers only gets to 87dB whereas on the right, get the phase right and you have 90dB
Below is a phase response measurement I did to an earlier version of the Elsinores, but it still applies:
I think this was a hefty 4 Watt input, you can see the pole a bit above 10KHz.
You can see the phase tailor off above 10KHz.
Does that help?
Last edited:
Joe, thanks, for that detailed response,
I think I understand where I was confused. Based on the system step-response graphs, I was erroneously thinking the crossover was an LR2, with the tweeter and woofer wired in typical anti-phase. Instead, the crossover is actually 1st order, except crossed at -6dB (0 degree acoustic inter-phase between the drivers), instead of at -3bB (90 degree acoustic inter-phase between the drivers) of a Butterworth alignment. The drivers must also be wired in relative anti-phase to achieve the 0 degree acoustic phase difference at the crossover. Does my understanding sound accurate?
I think I understand where I was confused. Based on the system step-response graphs, I was erroneously thinking the crossover was an LR2, with the tweeter and woofer wired in typical anti-phase. Instead, the crossover is actually 1st order, except crossed at -6dB (0 degree acoustic inter-phase between the drivers), instead of at -3bB (90 degree acoustic inter-phase between the drivers) of a Butterworth alignment. The drivers must also be wired in relative anti-phase to achieve the 0 degree acoustic phase difference at the crossover. Does my understanding sound accurate?
Last edited:
Yes, that is basically it. But there is something else and it's a difficult subject as things are fluid at the moment. Some have rightly pointed out that speakers with this step response have a certain sound, and even those with Butterworth 1st order have it to a degree, based on how well they are designed. But our solution, especially the tweeter being down -6dB means lower distortion and better summing etc. So what am I getting at? That maybe the real advantage is not so much on the acoustic side, but on the electrical side, the way that the speaker reacts to the amplifier on the current side. The force factor in the speaker (called BL) is directly related to the current of the amplifier - this is the force that creates sound. Back in 2018 I sat down with a certain person (at ETF18) and said, looking straight into his eyes, "I am certain we are listening to the current of the amplifier and not the voltage" and he said "yes, and the good news is that I think we should be able to prove that." I can't identify the person by name on social media, I don't think that is prudent. But he is arguably the foremost authority on audio transformers and says things like "thinking voltage is easy, thinking current is hard." I would say that thinking voltage is English and current is more like Chinese. It tricks you up so much that people just go back to thinking voltage because one is intuitive and the other counter-intuitive. For example, the way that current sources work with loudspeakers is not the way that most people think. And there really is no such thing as voltage-drive, and that was figured out in 1947 and then forgotten. That last statement is controversial and I won't discuss it with trolls (a lot wiser), but when the physics are pointed out to you, it becomes obvious and you can't unthink it.
It's Monday here and my Purifi drivers should hopefully be delivered today. They have blogs and also YouTube videos discussing "Force Factor Modulations" and in one video there are references to "impedance modulations" (unstable inductance with changes in excursion) I got confirmation that they agreed with me that "impedance" is the current of the amplifier that is being modulated, hence "impedance modulations." An impedance is related to current and not voltage, it is the current of the amplifier that is being impeded by Force Factor Modulations because they are inextricably linked. So here is another controversial statement, that amplifiers produce distortion on the current side that does not show up on the voltage side. The amplifier can have a voltage distortion profile with negligible distortion and another current profile with a lot less distortion. This is not speculation, we have the measurement to prove it; totally and conclusively. Guess which profile we end up listening to? The Purifi drivers reduce that kind of distortion. The fundamentals of the Elsinores reduce it to because FFM is related to excursions and four drivers now only need to go -/+1mm (2mm peak) excursion, where a single driver would need -/+4mm (8mm peak) and that is a big deal already. Since FFM is "impedance modulations" are related to the inductance of the driver, a single series inductor larger than the inductance of the driver, also reduce FFM. Reduce excursions, increase external series inductance, we suppress FFM distortions, yes, we reduce that amplifier current distortion.
Whew! This is a big subject, in first-order crossovers, the single series element being an inductor and a cap, this suppresses these current effects unknowingly, but if we adopt techniques that target current effects, and we can do even better. You can hear this stuff. John Atkinson of Stereophile talks about these speakers sound more "accessible" and I know what he means. I believe Elsinore builders here will also understand, that it means the speaker is more connected to what we hear. Easier on the ear and the sound just seems to come out of the box.
Now if the trolls turn up because of the above, then I will just shut up. I have long worked out that there are some who read everything I post. To some this may sound like paranoia, but you know some of the things that have happened in the past.
I will PM you that name.
Cheers, Joe
PS: My son is in a psychiatric hospital, they need to get his medications right, if they can get that right, he can live a fairly normal life - but to convince a person with schizophrenia that he needs to make the right decisions when they really are not fully able to do that... the system here is broken, but still needed. There is now real hope because he is beginning to understand the benefits so far. So we are now full of hope. But to me, the extent just how common this is, that is truly shocking - so many have come out of the woodworks and spoken to me, is this some kind of pandemic of its own? That is a frightening thought.
To Allen.
I have read the forum rules and I wish to abide by them in good spirit. I have just added the 'Signature' and hope it is taken as well-intended. Despite the efforts of others to paint the Elsinore Project as something commercial, the truth is that the Waveguides I had made, as well as Crossover PCBs and the main 18mH inductor etc, has set me back a huge number of dollars with no guarantees that I will ever get it back. I did this fully in the spirit of DIY and keeping the Elsinore Project alive. I don't know how many times I have had to say this, but to me this was never about the money. I make my income from other sources, like making equipment so people like the guy below can make great sound music. This was done with an all-tube 6-channel front-end that I custom made for Garth Porter, an old friend, and former keyboard player for Sherbert, who produced this:
WOW, isn't that amazing how the player turns up in the post. For those that have TIDAL, you can listen to this in full resolution.
I have read the forum rules and I wish to abide by them in good spirit. I have just added the 'Signature' and hope it is taken as well-intended. Despite the efforts of others to paint the Elsinore Project as something commercial, the truth is that the Waveguides I had made, as well as Crossover PCBs and the main 18mH inductor etc, has set me back a huge number of dollars with no guarantees that I will ever get it back. I did this fully in the spirit of DIY and keeping the Elsinore Project alive. I don't know how many times I have had to say this, but to me this was never about the money. I make my income from other sources, like making equipment so people like the guy below can make great sound music. This was done with an all-tube 6-channel front-end that I custom made for Garth Porter, an old friend, and former keyboard player for Sherbert, who produced this:
WOW, isn't that amazing how the player turns up in the post. For those that have TIDAL, you can listen to this in full resolution.
I can assure anyone concerned that if the cost of money is considered, and not even his time, Joe is not going to break even, let alone make money on the Elsinore-related bits.
That throat protrusion on the back side of the waveguide is a bitch to get right, even on CNC.
That throat protrusion on the back side of the waveguide is a bitch to get right, even on CNC.
Yes, it’s always been a mystery to me why some have singled you out for commercial accusations, while some of the biggest names on the site similarly have commercial interests. I do not understand the source of their hypocrisy. I see nothing improper by any of you. By the way, I really appreciate that your approach to audio is a bit unique. Lateral thinkers like you are interesting, whether my own thinking always agrees with yours or not. You may have a few detractors, but you have admirers as well.
This situation is extremely challenging to the patients family. It’s a big plus that he understands the benefits of taking his medication. The mental health system here is also broken, but seems to be improving. Largely, I think, because of increased public awareness, and the lessening stigma long associated with it. Best wishes.
Last edited:
As an aside, I notice that you’re looking at 6.5 inch mid-woofers. You might considered the Anarchy XBL motor 7-inch drivers. Price is $66USD each. They’ve been around a number of years and vendors, under somewhat varying model numbers for apparently the same driver, but don’t seem to be that well known. Current model numbers are Anarchy-704 (4-ohm), and Anarchy-708 (8-ohm). I haven’t personally laid hands on them myself, but test data (especially, deep basss distortion) and user reports seem very encouraging. Might be worth considering as an much less costly alternative to the Purifi.
https://www.diysoundgroup.com/speaker-parts.html
https://audioxpress.com/article/test-bench-exodus-audio-w06-017r-anarchy
https://www.diysoundgroup.com/speaker-parts.html
https://audioxpress.com/article/test-bench-exodus-audio-w06-017r-anarchy
Attachments
There are a number of reasons this driver is a no-go, but the stand-out is the inductance, which either has to be very low (which applies to our current driver) or is constant with excursions (very rare, but Purifi fits into this camp).
The inductance of Anarchy driver relative to 1KHz and changes in excursion:
I can do this measurement in a rough kind of way with my inductance meter set to 1000Hz (you will hear a tone). Check the inductance at the rest position, then push the cone gently in 4mm and hold still (imagine you have glued the position of the voice coil) and check the inductance. It will go up. Now do it the opposite way, push fingers behind the cone and push out 4mm. The inductance goes down. You don't want this, the force factor goes up when the cone is in the (-) position and goes down in the (+) position. This is not what you want to see, or as little change as possible (Purifi does this).
We are talking about reducing FFM, Force-Factor Modulations caused by excursions, and AM distortion.
Articles to read from Purifi's website:
https://purifi-audio.com/2021/10/14/some-speaker-problems-that-needed-solving/
https://purifi-audio.com/2019/12/12/imd/
Check the sound samples in the second article, AM vs FM (Doppler) IMD distortions - note how this is related to excursion issues.
EDIT: Internal inductance is bad unless it does not change with excursions. But inductance externally is really good and can used to swamp and flatten the internal inductance. This is why the MFC driver we use now is so good. At 2KHz the inductance is 0.1mH and we use 0.5mH inductor, FFM will be low because of the high external inductance, which changes neither with frequency or excursion, it is a purer inductance (yet not perfect, of course). We shall also use some external inductance with Purifi drivers, I am thinking 0.33mH would be good and 0.22mH also good. Higher the better.
Attachments
Last edited:
FedX is driving me crazy. They should have been here on Monday and the other two shipments to Australia arrived then, but not mine. Now it is Friday morning, will I get them today. They are only twenty minutes by car away. I offered to pick them up, but they said they did not know where they were and were likely still in the container. Lovely!
I am also working on getting the boxes done before Xmas holidays, but they cannot paint them, so I will grab them as they are and find somebody who can. I am thinking of a darker red satin finish, a two-pack paint. Maybe early January?
I just got off the phone again and they have only found one of the two boxes. They arrived at the depot 5AM last Friday and now 8.45AM today, Friday week.
All kinds of excuses about being busy is not cutting it. It just seems that this Erskine Park depot is a mess of a place (ie mismanaged), the others are working OK and they got their deliveries on time. You can't actually call FedX, you just get a lady with a heavy accent and they say that all that they can do is put through a "request" and that's it. The lady with the heavy accent is basically a tool to stop you from getting further up the chain, so they are the meat in a one-sided sandwich. Corporate structures work like that these days. Like computer games where it is hard to get to the next level, which is really about insulating you from getting to the people who can actually do something. It's them versus us, and they know how to win because they make the rules.
I will report later in the day how this saga is unfolding... or rather not unfolding...
All kinds of excuses about being busy is not cutting it. It just seems that this Erskine Park depot is a mess of a place (ie mismanaged), the others are working OK and they got their deliveries on time. You can't actually call FedX, you just get a lady with a heavy accent and they say that all that they can do is put through a "request" and that's it. The lady with the heavy accent is basically a tool to stop you from getting further up the chain, so they are the meat in a one-sided sandwich. Corporate structures work like that these days. Like computer games where it is hard to get to the next level, which is really about insulating you from getting to the people who can actually do something. It's them versus us, and they know how to win because they make the rules.
I will report later in the day how this saga is unfolding... or rather not unfolding...
Fingers crossed Joe, hope you get all the drivers by tomorrow, and all of them reach you in pristine condition.
They did arrive safely.
I have taken a sample out and done some tests and basically compared it to a run-in SB17MFC35-08, the current Elsinore driver. Even though the Purifi driver hasn't been run in, the Thiele-Small Parameters are where I would expect even if it had not run in. Maybe they have a;lready done some burning in already before packing.
I will connect it up to my run-in setup for 10 hours and retest, then another 10 hours. That way we shall see if the driver parameters settles in or whether they change much at all. I will post the results here.
But basically, it seems to work out what I have been expecting. The Purifi looks like not quite have the same LF extention, but maybe that is not going to be obvious when you listen to it? It may not have the same extention but it will also be a bit better damped. So it may come down to swings and roundabouts.
As expected, this points to a slightly higher tuning frequency of the box, I expect the 90mm OD port (around 87mm ID) needs to be shortened to 60mm. This will be confirmed later.
_________
Now I am getting the boxes done and should be available in the first week of January. Then Monday 10th I am hoping to drop it off for a two-pack spray painting being organised by my brother-in-law and should have the completed box during that week. The drivers (and damping) will then be fitted without crossovers. This is where a series of highly disciplined measurements will take place with the drivers in the box. This will be done using my ClioFW-01 system and Earthworks M32 Microphone. The results of those measurements (transfer function files) will be imported into SoundEasy and that is where the fun starts. We will load the existing MFC crossover and now we can see what changes need to be changed and accomodating the new Purifi drivers. When that is done, we hook up the new crossover with new values. Now we can listen and make measurements too, making sure that everything hangs together. The ear will always be the final arbiter, make no doubt about that, but I know the behaviour of the Elsinores so well, I know exactly what I want to see. There is definitely science involved that translates into results.
I have taken a sample out and done some tests and basically compared it to a run-in SB17MFC35-08, the current Elsinore driver. Even though the Purifi driver hasn't been run in, the Thiele-Small Parameters are where I would expect even if it had not run in. Maybe they have a;lready done some burning in already before packing.
I will connect it up to my run-in setup for 10 hours and retest, then another 10 hours. That way we shall see if the driver parameters settles in or whether they change much at all. I will post the results here.
But basically, it seems to work out what I have been expecting. The Purifi looks like not quite have the same LF extention, but maybe that is not going to be obvious when you listen to it? It may not have the same extention but it will also be a bit better damped. So it may come down to swings and roundabouts.
As expected, this points to a slightly higher tuning frequency of the box, I expect the 90mm OD port (around 87mm ID) needs to be shortened to 60mm. This will be confirmed later.
_________
Now I am getting the boxes done and should be available in the first week of January. Then Monday 10th I am hoping to drop it off for a two-pack spray painting being organised by my brother-in-law and should have the completed box during that week. The drivers (and damping) will then be fitted without crossovers. This is where a series of highly disciplined measurements will take place with the drivers in the box. This will be done using my ClioFW-01 system and Earthworks M32 Microphone. The results of those measurements (transfer function files) will be imported into SoundEasy and that is where the fun starts. We will load the existing MFC crossover and now we can see what changes need to be changed and accomodating the new Purifi drivers. When that is done, we hook up the new crossover with new values. Now we can listen and make measurements too, making sure that everything hangs together. The ear will always be the final arbiter, make no doubt about that, but I know the behaviour of the Elsinores so well, I know exactly what I want to see. There is definitely science involved that translates into results.