Anyone built Troels Jensen 1071?

[Steve Manning]

Well after living with, and loving my Modula MTM's for a few years, the urge to build something new has become too strong. Sometimes I hate this hobby, too addictive . Anyway, looking to improve on the sound of the last project, https://www.htguide.com/forum/showthread.php?t=28398, and try a three-way design as well, I have been looking at Troels Gravesen's Jensen 1071 design with interest, http://www.troelsgravesen.dk/Jensen.htm. I was wondering if anyone has built this design and had any comments on it? Looking at his site those that have built it seem very pleased, and like those folks I would add my touch to the cabinet. Thinking of a Wilson Sophia 3 inspired look, keeping as much of the important dimensions and volumes of course. Any comments and thoughts appreciated.

Thanks,

Steve

[Rolex]

I've had my eye on this design for a while, but most likely won't pull the trigger. The other designs I have built from Troels have been very nice. Especially the ekta grande. If you have the room, I imagine it would be a really fun project. Let us know what you decide.

[5th element]

There's nothing wrong with the design as per say, my only main gripe is with the off axis response. Troel's likes to cross his tweeters in higher then the typical crowd and this does cause off axis 'issues'. I find it interesting that he says 'it doesn't get much better then this' when describing the horizontal off axis performance.

Even though the off axis performance looks decent, there is still a reasonable amount of deviation around the mid to tweeter xover point. Troel's has also only measured out to 30 degrees, which is very good way of helping to disguise what happens if you go out to 45 then to 60 degrees.

This may seem like nit picking, but even Troel's extols the benefits of good off axis behaviour in his TJL3W design, even if he crosses the W12 over slightly too high to get it's peak in third harmonic completely out of the way.

It seems strange that he would then use a normal 6.5" mid/bass in a three way design. From my point of view one of the benefits of going three way is that you get to improve the off axis response by using a smaller driver. Of course there are ways to use a 6.5" inch up high and that's with a proper wave-guide (or you cross low), but Troel's doesn't here. Nor does he keep the centre to centre spacing on the mid to tweeter particularly tight. Now the 1071s vertical off axis looks alright, but it could be better. Troel's specifically went out of his way to get the C-C spacing down in the Prelude and with a ~3khz xover in the 1071 it would certainly offer some improvement.

As I said there's nothing really 'wrong' with the design as per say, it's just that if I am going to put in the effort and shell out the rather sizeable amount of cash that a speaker like this costs, then I was all the details to be as perfect as possible.

[JonMarsh]

There's nothing wrong with the design as per say, my only main gripe is with the off axis response. Troel's likes to cross his tweeters in higher then the typical crowd and this does cause off axis 'issues'. I find it interesting that he says 'it doesn't get much better then this' when describing the horizontal off axis performance.

Even though the off axis performance looks decent, there is still a reasonable amount of deviation around the mid to tweeter xover point. Troel's has also only measured out to 30 degrees, which is very good way of helping to disguise what happens if you go out to 45 then to 60 degrees.

This may seem like nit picking, but even Troel's extols the benefits of good off axis behaviour in his TJL3W design, even if he crosses the W12 over slightly too high to get it's peak in third harmonic completely out of the way.

It seems strange that he would then use a normal 6.5" mid/bass in a three way design. From my point of view one of the benefits of going three way is that you get to improve the off axis response by using a smaller driver. Of course there are ways to use a 6.5" inch up high and that's with a proper wave-guide (or you cross low), but Troel's doesn't here. Nor does he keep the centre to centre spacing on the mid to tweeter particularly tight. Now the 1071s vertical off axis looks alright, but it could be better. Troel's specifically went out of his way to get the C-C spacing down in the Prelude and with a ~3khz xover in the 1071 it would certainly offer some improvement.

As I said there's nothing really 'wrong' with the design as per say, it's just that if I am going to put in the effort and shell out the rather sizeable amount of cash that a speaker like this costs, then I was all the details to be as perfect as possible.

+1

Still, they are good drivers with fairly good distortion characteristics. I do agree that the 8531 is not an ideal midrange driver. The best part of this design, IMO, is the woofer choice. I do really love the character of the 10" SS revelators.

[craigk]

i have had the chance to listen to this speaker, and it is very good. i would not think twice about building it if i were looking for an upgrade from your previous speaker. there may be benefits to using a small mid in a three way, but there is also benefits to using a 6.5 inch driver also. this speaker has the ability to produce some of the most natural vocals/mids i have listened to. off axis may suffer a bit, but in this case the benefits out weigh the loss of a little off axis performance. do not let "nit picking" scare you away from these speakers. they are very good.

[numberoneoppa]

Those speakers make me excited.

[5th element]

but there is also benefits to using a 6.5 inch driver also.

What exactly would they be though?

this speaker has the ability to produce some of the most natural vocals/mids i have listened to. off axis may suffer a bit, but in this case the benefits out weigh the loss of a little off axis performance. do not let "nit picking" scare you away from these speakers. they are very good.

This doesn't make any sense though, any 5.25" will also be able to give you 'the most natural vocal/mids' as that type of thing is handled mainly in the voicing of the loudspeaker, this has very little to do with the drivers diameter.

No doubt the loudspeakers are very good, but the important point here is that things could be better and considering the cost and effort, I'd want them to be as good as they could be. Ie, mount the mid and tweeter closer together and push down the xover frequency. Linkwitz has shown that the SS D2905/97, which is the 99 but with a smaller faceplate, can handle down to around 1500hz with a steep xover, so there's a lot of room here for technical improvement.

Troel's isn't adverse to using wave-guides either. This would be a terrific design if Troel's were to use the WG300 + SS7100

http://www.troelsgravesen.dk/7100-waveguide.htm

I think a lot of the time with Troel's though, is that he has built so many loudspeakers that he builds things to satisfy his curiosity, rather then builds things for the best technical specification. I know that if in his position, I would most likely build a few things that don't follow typical conventions too, just to see what it sounds like.

The three way designs of Troel's that I like the best are the classic three way, the poor mans strad (paper version) and the Ekta grande, although I would have let the midrange in the grande play down some-what lower.

Given the option though, when considering systems of similar concept and end goal (ultimate high fidelity). I would rather build Jons Ardents over the Ekta Grande and I would also rather build his Module Xtreme (if the bass units were available) over the 1071. The big question would be, can you afford the accutons (not that the scan drivers are cheap either!)

[craigk]

your warm prickly attacks of anyone that doesnt agree with you is very amusing. i don't know what the advantage would be either, but since wilson uses 7 inch mids for all his top end speakers , along with eggleston and mythos, etc. your right, i must be wrong. there can not be anything to using a 6.5/7 inch driver as a mid.

[Steve Manning]

Thanks for the responses guy's, I appreciate the inputs. I suppose part of me is looking for a warm and fuzzy before spending what I consider a large bit of cash on parts and what not. I figured you guys would be a good sounding board before I made a final speaker choice, some of Jon's new designs are pulling me as well. The thing about this design is it seemed like it would fit well into a cabinet design I have been working on , here's a few pictures .......







The lines are Wilson Sophia 3 inspired, it would still have separate bass and mid cabinets internally, driver spacing and cabinet width, volumes, driver placement, all the same as the 1071. I did angle the front a little towards the tweeter, which I understand will have some affect on sound, hopefully not too much. Trust me the main goal is the sound, but I have to give a few percent to aesthetics, can't have something sound great looking like crap in the middle of the house can I?

As for some of the other suggestions, I have been following the Jon's Ardent project with great interest, I guess I have been waiting for where the final version will end up? As for the Modula Xtreme's, from what I can tell that's a fair bit pricier than the 1071 which is pushing it for me already .....
Steve

[kendomusic]

I like Troels work. I built his Ekta's and love them. As you can see feedback on the 1071 builds are good. I wouldnt worry about the technicalities.
Its like a critically acclaimed TV show that no one watches. In this case, I would go for the feedback on the listening pleasure that builders have experienced.

[5th element]

Steve, I wouldn't use that cabinet design for the 1071. The front baffle shape and size is critical to the design and will alter both the box diffraction and the baffle step transition. These are corrected for in the crossover and altering the baffle will undo some of Troel's design work.

My warm prickly attacks are nothing but hard scientific fact, I am not saying these things based on personal opinion, I am simply offering a more objective point of view. It is very easy to create loudspeakers that people will rave about subjectively yet fair pretty poorly from a design perspective.

As far as I am aware there have been very few, if any, loudspeakers designed, that measure like champs, but sound like crap. On the other hand there are plenty of designs out there that measure poorly and also get poor reviews.

I would also say that there is only a small difference between a great design and a very good design. Troel's design falls into the very good category, there are things that he could do to make small technical improvements. Are these going to make the difference between a loudspeaker that makes you go wow and a loudspeaker that also makes you go wow? Of course not, the 1071 will sound very good as would a slight redesign that gives you a reasonable improvement in the off axis behaviour.

The point I am trying to make here is that there are designs out there that do satisfy all design criteria.

This thread was started asking for peoples comments on the given design and I have done so based solely on the technical details about that design. It's good, but you can do better. For this amount of money I'd rather build something where, for the given drive units, you can't really do better.

As to other manufactures also using 6.5" mid range drivers? I am well aware of these, most of the designs also have a compromised off axis response so they are technically flawed also. One of the reasons for going DIY is so we can do away with these small technical flaws.

Incidentally if I were to buy a pair of high end loudspeakers, I'd pick something from KEF or Revel, but probably KEF due to the constant directivity offered by the concentric mounting of the tweeter within the mid unit.

[craigk]

this will be my last post on this subject, but when you are getting over 150,000.00 dollars for a pair of speakers i am pretty sure that the little design flaws are taken care of.
when you make the first speaker with no technical flaws, please let me know because i have a bank full of money and i will make us both very wealthy.

[Steve Manning]

Wow ..... "This is exciting, I like this ship", oh wait that's a different movie .... Again the comments were asked for and appreciated.

Steve

[5th element]

this will be my last post on this subject, but when you are getting over 150,000.00 dollars for a pair of speakers i am pretty sure that the little design flaws are taken care of.
when you make the first speaker with no technical flaws, please let me know because i have a bank full of money and i will make us both very wealthy.

I believe the important part to this would be, and I quote myself...

I'd rather build something where, for the given drive units, you can't really do better.

I never said this would equal building a pair of loudspeakers without any technical flaws. What I said was that you attempt to minimise the technical flaws as much as is possible for the given drive units.

Sometimes you cannot escape certain technical issues so you compromise, like C2C spacing with wave-guides. The physical separation that you're generally forced to use, due to the size of the wave-guide, means that if you want to crossover so as to get a directivity match between the mid and the wave-guide that you're going to crossover a little too high for the driver separation and you sacrifice the vertical off axis response. This cannot be helped, ie for these drive units you're forced by physical constraints into sacrificing the vertical off axis. In Troel's design though, there is no reason for keeping the mid and tweeter that far apart, it's a bad design choice, not a killer, but it's something that could easily be improved, so why not improve it?

This has nothing whatsoever to do with me or Troel's, as I said before the science speaks for itself, there's no reason to get unhappy about what I've said.

this will be my last post on this subject, but when you are getting over 150,000.00 dollars for a pair of speakers i am pretty sure that the little design flaws are taken care of.

Apart from this comment looking mainly like a personal attack, I am having trouble trying to figure out exactly what it is you're trying to say?

Cost though has nothing to do with this. You can design a technically competent pair of loudspeakers for very little money. Just because something is expensive doesn't mean it will be well designed.

[djkest]

I agree, it seems that using a 5" revelator instead of a 6" would improve the off-axis response, and slightly lowering the crossover point would also improve this. There's a pretty big dip in frequency response in both the horizontal and vertical axis around the mid-to-tweet crossover point, in a frequency range where you could notice. And like someone else pointed out, if you are dropping $1500+ USD on components for a single set of speakers, it's not really a trivial matter.

On the flip side, the 6" has a little more SPL capabilities and the distortion measurements Zaph posted indicate it's a little cleaner in the upper midrange.

Of course, that would require completely changing the crossover, the baffle, the internal volume, etc.

[5th element]

Well the dip in the vertical axis is pretty much impossible to avoid when using drivers that aren't co-incident, or are crossed over at a frequency where the wave-length is still significant compared to the driver separation.

The dip occurs because the relative difference in path length between the mid and the tweeter increases/decreases as you go off axis and this acts to pull the drivers in and out of phase with one another. This creates a standard lobing pattern and the further apart the drivers are spaced, for a given frequency, the narrower the main frontal lobe (what you generally listen to) becomes. This means that the seating height becomes more and more critical as the driver separation increases. Looked at from a different point of view, the closer the drivers are together, for a given frequency, the wider the main lobe becomes and this decreases the change in tonal balance that will occur when one decides to listen whilst doing other things.

A good set of vertical off axis curves is one of those things that's nice to have, but is one of the less critical parts of a design imo. Of course it does depend on the end application. If you're designing a pair of loudspeakers for a kitchen, then keeping the main lobe as wide as possible, and keeping the nulls on either side of the main lobe as small as possible, is going to play a much larger role then it normally would. The loudspeakers will probably be positioned at a less then optimal height and the listener is likely to be in a variety of different positions etc, so using a small mid/bass with tight C2C spacing and a low steep crossover would definitely be a good idea.

As you have noted though, changing Troel's design in any way would require a complete change in crossover.

Personally I have no problem with him using the 6.5" scan speak driver as a midrange as it measures superbly and due to a well designed cone and motor, keeps its composure up high. The only issue is the off axis response. If I were to improve upon the design then I'd definitely go for the wave-guide approach.

http://www.troelsgravesen.dk/7100-waveguide.htm

Then mount the wave-guide similar to how Troel's does in the DTQWT, by slicing into the wave-guide to get a closer C2C spacing. This should bring the drivers to around 15-16cm apart allowing for a crossover around 2.2kHz that would match the directivity of the mid/bass to the wave-guide reasonably well and be satisfactory from a vertical off axis point of view too.