They sound great, extremely flat and well balanced. Great detail, not harsh, encourages singing along. Vocals are nicely forward.
I am not sure if they will replace, or be used with, my current speakers. Still playing with them. Existing speakers are also DIY. Nearly 8' tall towers with 4 Dayton woofers, 6 SB mids, and 75" of Newform Research tweeters. Active using Hypex 3 way plate amps.
Hey Joe! Again really hope everything is okay with you.
So given I had the Purifi drivers laying around, I re-read some of the previous posts that you had posted. In one, you had indicated that the original crossover changes would be minor (before all the discussion of extending the low end by tuning the box low and reducing the sensitivity via voltage divider), and in another post you mentioned that the mid range inductor would likely be either 0.33mH or 0.22mH. Incidentally, the Mark V (which is the version I have set up) has 0.33mH as the midrange inductor, so, in preparation for the final version (since the new drivers had to be mounted in anyways!), I decided to pop in the Purifi drivers (surface mounted, not rebated) using the stock Mark V xover - listening to them right now, and although I don't have any of the equipment to verify frequency response, etc they do actually sound pretty good to me right now
!
So given I had the Purifi drivers laying around, I re-read some of the previous posts that you had posted. In one, you had indicated that the original crossover changes would be minor (before all the discussion of extending the low end by tuning the box low and reducing the sensitivity via voltage divider), and in another post you mentioned that the mid range inductor would likely be either 0.33mH or 0.22mH. Incidentally, the Mark V (which is the version I have set up) has 0.33mH as the midrange inductor, so, in preparation for the final version (since the new drivers had to be mounted in anyways!), I decided to pop in the Purifi drivers (surface mounted, not rebated) using the stock Mark V xover - listening to them right now, and although I don't have any of the equipment to verify frequency response, etc they do actually sound pretty good to me right now
!
Just want everybody to know that things are happening...
You guys may have heard about our weather down here. It's been rain, rain, and more rain. This is our forecast eahead:
Ten days of rain ahead.
This time of the year we are mostly sunny and not much rain.
Thankfully, the worst of it is up north of here. The neverending rain is like nothing I can remember. We are seeing the damage caused, it is all over the place and we have been fortunate. It has made a mess of our roads, hard rain makes the surfaces crumble and you have to look out for holes.
By the time the rains are gone and we get the usual sunny skies, I will be ready. Will be able to set up the equipment outside and start doing the acoustic measurements that will alow me to model the crossover. I may at that point need to buy parts for the crossover if I don't have what I need here. We shall see.
In the meantime, I can do work inside, the damping pieces have been cut and over the next week all the drivers will be mounted and I can work on the bass/box alignment. When the weather comes good again, I will be ready.
On the personal front, with mental illness issues in the family, things are looking better than they have for a long time. Paranoia is slowly being overcome and more accepting of treatment, this is proving the way forward.
Cheers, Joe
You guys may have heard about our weather down here. It's been rain, rain, and more rain. This is our forecast eahead:
Ten days of rain ahead.
This time of the year we are mostly sunny and not much rain.
Thankfully, the worst of it is up north of here. The neverending rain is like nothing I can remember. We are seeing the damage caused, it is all over the place and we have been fortunate. It has made a mess of our roads, hard rain makes the surfaces crumble and you have to look out for holes.
By the time the rains are gone and we get the usual sunny skies, I will be ready. Will be able to set up the equipment outside and start doing the acoustic measurements that will alow me to model the crossover. I may at that point need to buy parts for the crossover if I don't have what I need here. We shall see.
In the meantime, I can do work inside, the damping pieces have been cut and over the next week all the drivers will be mounted and I can work on the bass/box alignment. When the weather comes good again, I will be ready.
On the personal front, with mental illness issues in the family, things are looking better than they have for a long time. Paranoia is slowly being overcome and more accepting of treatment, this is proving the way forward.
Cheers, Joe
Mental illness is tough. I had a neighbor who ran into some issues with coming off of her treatment. She was destroying her house and all sorts of things... She got help and now you would never know. She is a very nice person. People in the neighborhood talk to her often. Thank you for sharing!
Really glad to hear that the family side of things is getting better! Very tough situation for certain and we all wish you guys the best.
FYI since mounting the Pruifi's I made the two modifications you had discussed in post #4616 - I extended the port to 16cm to reduce the box tuning and then used an L pad to reduce the midrange by 1.5 db. I'm using a Mark V crossover currently - however the tweeter coupling cap is a 1.8uf Audyn true copper, which I think from your previous posts should reduce the output level relative to the originally specced 2.0-2.2uf. Currently the speakers sound truly fantastic - the mid range especially gained clarity vs the Peerless HDS drivers used in the Mark V. I really can't wait to have your final revision set up - thank you very much for all your work!!
FYI since mounting the Pruifi's I made the two modifications you had discussed in post #4616 - I extended the port to 16cm to reduce the box tuning and then used an L pad to reduce the midrange by 1.5 db. I'm using a Mark V crossover currently - however the tweeter coupling cap is a 1.8uf Audyn true copper, which I think from your previous posts should reduce the output level relative to the originally specced 2.0-2.2uf. Currently the speakers sound truly fantastic - the mid range especially gained clarity vs the Peerless HDS drivers used in the Mark V. I really can't wait to have your final revision set up - thank you very much for all your work!!
I'm going to post up my unfinished Elsinore project. As much as it pains me I've come to the conclusion that I'm just going to build these.
I can post pics or send pics to anyone who might be interested.
8 of the MFC drivers BNIB
2 of the 810921 tweeters with Joe's waveguides
Xovers sourced and built by a forum expert.
Will only sell this gear as a complete set. Would love to find a local buyer and could meet within a reasonable distance.
I also have 12 x 11.5" x 48" Bamboo boards that was going to use. They are nice and HEAVY! Local pickup on these only.
PM me if you interested. Ad Posted today but I thought someone in this thread might be interested.
I've sold under Pilk on DIYMA and GolfWRX and under Huskerpilk on Ebay for references. I ship UPS ground from my office (Springfield Eyecare in Springfield MO) so you know who you're dealing with.
I can post pics or send pics to anyone who might be interested.
8 of the MFC drivers BNIB
2 of the 810921 tweeters with Joe's waveguides
Xovers sourced and built by a forum expert.
Will only sell this gear as a complete set. Would love to find a local buyer and could meet within a reasonable distance.
I also have 12 x 11.5" x 48" Bamboo boards that was going to use. They are nice and HEAVY! Local pickup on these only.
PM me if you interested. Ad Posted today but I thought someone in this thread might be interested.
I've sold under Pilk on DIYMA and GolfWRX and under Huskerpilk on Ebay for references. I ship UPS ground from my office (Springfield Eyecare in Springfield MO) so you know who you're dealing with.
Last edited:
Sorry to hear, but I know the feeling with regards to unfinished projects and finding time/energy. If I was in your hood, I'd offer up to help you build them. I have a very similar crossover to you in terms of parts and I know for a fact that once put together they sound spectacular. Someone is going to be very happy!
OK, for all you guys who have been patient, but speaker design is about getting to the right place in the end and not how fast to get there. So progress is being made.
The rains that you may have heard about, have continued and last week were horrendous. Up to now we have been lucky, but finally, something had to give and yes we have now had property damage and some things have been lost or damaged. I won't go into any detail, but it's the water ingestion into structures that leave quite a pungent smell. But things could have been worse, but I estimate that damage was around $1000.
The "ULD" speaker boxes have been assembled. The "ULD" moniker is what I have settled on and that is how Purifi describes the drivers, as "Ultra-Low Distortion" and below shows where we are up to.
Elsinore Mk6 "ULD"
Not bad for a quick picture using my Note 10 phone. It looks even better in real life.
Yes, the boxes are done, the internal damping all fitted, part filled and part lined. The instructions for that are not likely to change. The port is 150mm deep and it will now be interesting to see where the box tuning will end up. The aim is a bit under 30 Hertz and maybe as low as 28 Hertz. The port length may yet be tweaked.
To short-circuit any recurring topic, will the tweeter be good enough to match the quality of the Purifi drivers, keeping in mind the utmost importance is very low distortion. The resounding answer is yes.
Even stock it is a top-class soft dome tweeter, it has design features that were influenced by Russell Storey of Stones Sound Studios, right down here in Australia. But an additional level of performance is derived by using a waveguide and good crossover design. Besides, I know this tweeter like the back of my hand!.
The waveguide is not used to increase the bandwidth (this entices using very low crossover frequencies), but here it is done for the most basic reason of all, to reduce distortion. Yes, the waveguide has other advantages and thank you very much we will take them too, but the rationale here is distortion reduction. The crossover additionally reduces distortion by using a carefully tuned "Q" LC null at Fs that minimises and virtually eliminates excursions of the tweeter. It goes from 1st order into 5-6th order in two octaves.
Brian Gurr in Auckland rebuilds electrostatic speakers and is ultra-sensitive to HF distortion and his pair of "NRXC" Elsinores, he says the top end is right up there with his custom-built electrostatics. Yes, it is that good. It will be right home in this design.
There are six Teflon/Silver/Copper wires coming out of the back. The crossover will be developed externally first and then later into the base of the box.
What yet to be done?
We have been told the weather will continue until June, but there will be days when watching the rain radar we use will give us a break. I can do all the electrical measurements as well as nearfield acoustic measurements indoors in the meantime.
The farfield acoustic measurements have to be done outdoors. When that is finished, then all the data will be edited in Clio, merging and summing nearfield and farfield measurements. These and the electrical measurements (impedances) will be imported into SoundEasy and here the very strictly (and calibrated) collected data will be put to good use and we can construct the crossover and model/observe the behaviour of the crossover. BTW, the acoustic response will be referenced to 90dB-SPL @ 1 Metre. But the actual acoustic farfield measurement will be made at 2 Metres. This is because I want to capture the driver integration and the relative phase differences at a distance where beyond I am confident that it will be similar to most people's listening distance. So if the stimulus is 1 Watt @ 1M, I use 4 Watt @ 2 Metres or +6dB. But the main drivers will not see the full 4 Watt but 2 Watt, but the tweeter will, so some care is needed when doing that measurement. Very low frequencies need to be kept out or damage may happen.
(Above: These were also checked and modelled at 2 Metres)
On Axis Versus Off Axis:
The acoustic measurements will be taken (1) on axis, (2) 15 degrees off axis and (3) 30 degrees off axis. So when we model the crossover in SoundEasy, we are actually doing it three times side-by-side. Yes, I am running the programs at the same time on a wide-screen computer monitor. This way we can monitor what the crossover is doing across a sixty-degree arc in front of the loudspeaker. The 15 degrees off axis response will have primary importance over the other two.
So the above gives some idea of what is ahead.
I need to give some mention to Troels Gravesen and what he did in his Purifi-6661 speaker design, even if he does not know it. I looked carefully at what he has done. Particularly with attention to the box alignment and what I have in mind for the Elsinore "ULD" alignment looks to be similar to what he has done. By tuning lower than before and because the Purifi driver has significant lower Q and also somewhat lower Vas, the idea is that we can arrive at a desirable response by reducing overall system sensitivity by 2dB or maybe slightly less. I will use the current Elsinore Mk-6 "MFC" as a reference using pink noise tests and adjust until the right balance is arrived at. The Elsinores have always been praised by you guys for having that near-perfect balance and that has to be achieved here or else we will have a failure, and that is not an option. But the overall sensitivity will not be as high as "MFC" and "NRXC" versions, but it will still be high. I am not aiming for lower LF extension, but that may well be the result, but again the final bass alignment tuning will determine that.
Talking about Troels, I also have been following his gradual journey towards 1st order crossovers. Not the Butterworth kind (only -3dB summing) but summing at full -6dB, which BTW also reduces reduction and improves power response into the room. Butterworth tends to have very small windows, study Duntech and John Dunleavy's work, where speakers became more like giant headphones because that is how small the sweet spot was, focussed array about 20cm wide at the listening position. But some will argue that you lose the phase advantage. I am on the side that says it is not worth it because you will sacrifice something more important. I believe that the sound that Troels is getting has to do with distortion that happens on the electrical side and much less on the purely acoustic side. That may be considered for some time yet as controversial. But I also hope that this will eventually become better understood. This is about getting current through the voice coil as clean as possible and giving the transducer a decent chance to convert current into its force-factor equivalent.
Cheers for now, Joe
The rains that you may have heard about, have continued and last week were horrendous. Up to now we have been lucky, but finally, something had to give and yes we have now had property damage and some things have been lost or damaged. I won't go into any detail, but it's the water ingestion into structures that leave quite a pungent smell. But things could have been worse, but I estimate that damage was around $1000.
The "ULD" speaker boxes have been assembled. The "ULD" moniker is what I have settled on and that is how Purifi describes the drivers, as "Ultra-Low Distortion" and below shows where we are up to.
Elsinore Mk6 "ULD"
Not bad for a quick picture using my Note 10 phone. It looks even better in real life.
Yes, the boxes are done, the internal damping all fitted, part filled and part lined. The instructions for that are not likely to change. The port is 150mm deep and it will now be interesting to see where the box tuning will end up. The aim is a bit under 30 Hertz and maybe as low as 28 Hertz. The port length may yet be tweaked.
To short-circuit any recurring topic, will the tweeter be good enough to match the quality of the Purifi drivers, keeping in mind the utmost importance is very low distortion. The resounding answer is yes.
Even stock it is a top-class soft dome tweeter, it has design features that were influenced by Russell Storey of Stones Sound Studios, right down here in Australia. But an additional level of performance is derived by using a waveguide and good crossover design. Besides, I know this tweeter like the back of my hand!.
The waveguide is not used to increase the bandwidth (this entices using very low crossover frequencies), but here it is done for the most basic reason of all, to reduce distortion. Yes, the waveguide has other advantages and thank you very much we will take them too, but the rationale here is distortion reduction. The crossover additionally reduces distortion by using a carefully tuned "Q" LC null at Fs that minimises and virtually eliminates excursions of the tweeter. It goes from 1st order into 5-6th order in two octaves.
Brian Gurr in Auckland rebuilds electrostatic speakers and is ultra-sensitive to HF distortion and his pair of "NRXC" Elsinores, he says the top end is right up there with his custom-built electrostatics. Yes, it is that good. It will be right home in this design.
There are six Teflon/Silver/Copper wires coming out of the back. The crossover will be developed externally first and then later into the base of the box.
What yet to be done?
We have been told the weather will continue until June, but there will be days when watching the rain radar we use will give us a break. I can do all the electrical measurements as well as nearfield acoustic measurements indoors in the meantime.
The farfield acoustic measurements have to be done outdoors. When that is finished, then all the data will be edited in Clio, merging and summing nearfield and farfield measurements. These and the electrical measurements (impedances) will be imported into SoundEasy and here the very strictly (and calibrated) collected data will be put to good use and we can construct the crossover and model/observe the behaviour of the crossover. BTW, the acoustic response will be referenced to 90dB-SPL @ 1 Metre. But the actual acoustic farfield measurement will be made at 2 Metres. This is because I want to capture the driver integration and the relative phase differences at a distance where beyond I am confident that it will be similar to most people's listening distance. So if the stimulus is 1 Watt @ 1M, I use 4 Watt @ 2 Metres or +6dB. But the main drivers will not see the full 4 Watt but 2 Watt, but the tweeter will, so some care is needed when doing that measurement. Very low frequencies need to be kept out or damage may happen.
(Above: These were also checked and modelled at 2 Metres)
On Axis Versus Off Axis:
The acoustic measurements will be taken (1) on axis, (2) 15 degrees off axis and (3) 30 degrees off axis. So when we model the crossover in SoundEasy, we are actually doing it three times side-by-side. Yes, I am running the programs at the same time on a wide-screen computer monitor. This way we can monitor what the crossover is doing across a sixty-degree arc in front of the loudspeaker. The 15 degrees off axis response will have primary importance over the other two.
So the above gives some idea of what is ahead.
I need to give some mention to Troels Gravesen and what he did in his Purifi-6661 speaker design, even if he does not know it. I looked carefully at what he has done. Particularly with attention to the box alignment and what I have in mind for the Elsinore "ULD" alignment looks to be similar to what he has done. By tuning lower than before and because the Purifi driver has significant lower Q and also somewhat lower Vas, the idea is that we can arrive at a desirable response by reducing overall system sensitivity by 2dB or maybe slightly less. I will use the current Elsinore Mk-6 "MFC" as a reference using pink noise tests and adjust until the right balance is arrived at. The Elsinores have always been praised by you guys for having that near-perfect balance and that has to be achieved here or else we will have a failure, and that is not an option. But the overall sensitivity will not be as high as "MFC" and "NRXC" versions, but it will still be high. I am not aiming for lower LF extension, but that may well be the result, but again the final bass alignment tuning will determine that.
Talking about Troels, I also have been following his gradual journey towards 1st order crossovers. Not the Butterworth kind (only -3dB summing) but summing at full -6dB, which BTW also reduces reduction and improves power response into the room. Butterworth tends to have very small windows, study Duntech and John Dunleavy's work, where speakers became more like giant headphones because that is how small the sweet spot was, focussed array about 20cm wide at the listening position. But some will argue that you lose the phase advantage. I am on the side that says it is not worth it because you will sacrifice something more important. I believe that the sound that Troels is getting has to do with distortion that happens on the electrical side and much less on the purely acoustic side. That may be considered for some time yet as controversial. But I also hope that this will eventually become better understood. This is about getting current through the voice coil as clean as possible and giving the transducer a decent chance to convert current into its force-factor equivalent.
Cheers for now, Joe
I would like to add that, in the photos, the colour is brighter than it looks. The typical phone cameras sex up photos, and hence they look more technicolour than they are. And while I am at it, I actually wanted a red to be a bit darker, but I nevertheless like it. Then it occurred to me how close the colour I got looks close to the colour that Purifi have adopted in their corporate scheme - I promise, I did not end up with this colour so as to look like I am sucking up to Purifi.
For the US builders I'm in contact with Madisound about a 'group buy' on the Purifi drivers. If we get 20 pieces ordered they are offering $293.50 ea which is almost 30% off their MSRP of $411. I'm waiting until Joe gets some things settled on the ULD build before I commit but if anyone is interested in putting their name in the hat for drivers you can respond here or PM me. I think they would have to order the drivers once enough units are purchased so there may be a bit of a wait.
That's the one.
Next week or so all the measurements that can be done indoors will be done. We have reported rain during that whole week. Then we can find a day to set up the equipment outdoors and do the acoustic farfield measurements. Immediately it will all go into the computer for modeling (SoundEasy) and that's the exciting bit - and if that all comes good, then a few days later we will have the crossovers working externally. Then when all the final adjustments (and listening), the crossovers will be installed in the bottom of the box.
Then all the details will be published here and on the website.
Next week or so all the measurements that can be done indoors will be done. We have reported rain during that whole week. Then we can find a day to set up the equipment outdoors and do the acoustic farfield measurements. Immediately it will all go into the computer for modeling (SoundEasy) and that's the exciting bit - and if that all comes good, then a few days later we will have the crossovers working externally. Then when all the final adjustments (and listening), the crossovers will be installed in the bottom of the box.
Then all the details will be published here and on the website.