[jmimac351]

Tyson, thanks for the advice and I bet you're right.  For sure, I can play with it On and 10 degree off axis and the peaks will swap... I bet it's great. If I play with the felt, I think I figured it out.  Measuring the thickness on the other speaker here at the house, another pair of the tweeter pads will stack up nicely to the foam... easy.  It'll look factory... Perfect.

Also, here is the revised Duette crossover overlayed with the revised Sony SS-CS5.  That may seem like an odd comparison, but that revised Sony is one of the best measuring speakers... and the current Duette crossover revision compares well.  So, you're probably right... it's done.

It'd still like to get "measurement" figured out, though.  I've got other stuff around here I want to play with.

[jmimac351]

Umbilical Cords...

I asked Danny to please check on differences between the 2 different umbilical cords used for connecting the crossover to the speaker.  There is a gray cable and a brown cable.  The gray cable is the "Free Space" cable for use with the 5.3Ohm resistors.  That is for when the speaker is pulled >24" from the front wall.  The brown cable is the "Near Wall" cable for use with the 4.2Ohm resistors.  That is used when the speaker is placed <24" from the front wall. 








In the Stereophile review, it's noted that the Inductance is different... "warranty impact", "part of the crossover design" etc. Who told them that?...

https://www.stereophile.com/content/wilson-audio-specialties-duette-loudspeaker



Here is what the Wilson manual says about the topic:



Here is the section of the manual referred to above:







So, you read that in a Stereophile review, you see the manufacture of 2 different cables, labeled for specific purposes, called out in the manual in a very direct / explicit manner... it makes a guy wonder, at the very least be curious about what's going on with that issue.  If it's having an impact, we need to know that when working on a crossover, right?

So, Danny measured the 2 umbilicals and plotted their performance on top of each other... their performance was exactly the same: 



Well, at least now we know... about more than just the cables...

[Hobbsmeerkat]

What's interesting is that the only parts affected in both of Wilson's graphs are the top and bottom of the response..
Neither of which are due to the cable/umbilical.

The "rise" shown in the bass of the first graph is simply due to the boundary reinforcement provided by the wall behind it. But then they also opt to have you roll off the top end by using a higher value to bring down the tweeter level, which seems really silly to me.

Then for the "free space" graph, it simply shows the lack of any boundary reinforcement, and they keep the tweeter a little higher.

I'd be curious to know how they came up with this explanation, cuz it doesn't make sense to me, especially when the measurements show no change between the cables. 

[Tyson]

So, how long before the speakers make it back to you and are up and playing?  Inquiring minds want to know!

[jmimac351]

So, how long before the speakers make it back to you and are up and playing?  Inquiring minds want to know!

I expect the parts will go out early next week to Frederick so they are pretty, oh so pretty...  Crossovers for 2 pairs of speakers will be built, as a good friend of mine also has a pair of Duette.  I'm going to make custom enclosures for them to sit in the same spot on the Duette stand... but the enclosure will need to be taller than the original Wilson unit, given the crossover parts involved.  I won't be waiting on the enclosure to be done before listening, that's for sure!  I have a milling machine and haven't decided yet what to make the enclosures out of.  Billet aluminum, aluminum painted to match the speaker, wood... the enclosures will have tube connectors in them and umbilicals to mimic the original setup.  The Wilson binding post plate will remain, to start.  I may end up pulling that and making a plate with tube connectors in it.  Anything done will be a bolt-on thing such that the original Wilson setup can be plugged back in.  No permanent modification.  It would be a quick swap to A / B the setups.

I'm really excited about it.  The speaker has great drivers and the extension from the 8" woofer make me chuckle when you remember it's a 2-way on a stand.  I think it's a special speaker that now will be "Right".

This is the space we have to play with for the crossover, but we're going to make it fit in the machined spot for the original Novel crossover, and just go higher vertically.

[Tyson]

Who knows, you might even be able to toe the speakers in now!!

[jmimac351]

Who knows, you might even be able to toe the speakers in now!!

I'll listen somewhere between ON and 10 Degree Off Axis. For those 2 different response measurements, the thinking is the height of the foam that is in a circle at the perimeter of the felt diffraction pad immediately around the tweeter may be causing the ripple.  You can see the on and 10 degree off axis peaks / valleys are inverse to one another.  So, that felt is supposed to be helping, but the raised foam may be hurting.

I'm going figure out REW with my UMIK mic and take some measurements, then play with some felt to build up the height around the tweeter and make the height transition to the foam more gradual, and see what happens.  Regardless, I don't know what the +/- dB swing is across the response, but it's obviously a transformed speaker.  Eyeballing it, it looks like the +/- variance at 10 Degree off axis is about 1.5dB.  Maybe Danny / Hobbs know the exact number.

New On Axis vs 10 Degree Off Axis Response

[jmimac351]

YouTube Link:
https://youtu.be/o_PJzPBCaOY?si=ifmtRTWPEpDx84nx

[jmimac351]

UPDATE:

Crossover parts should land tomorrow with Frederick... so he can make them pretty.  Then I will make an enclosure for them and a set of umbilicals to connect the crossover enclosure to the Duette.

In the meantime, I got the Duette back and decided to play with something, and it's been interesting.  Given the lack of baffle step compensation, and the PEQ adjustments available with the dual SVS SB16-Ultra subs, I tried scaling up the sub crossover to 200hz on the high end, and changing slope to 6db, which I expect will throw it out further than 200hz for roll off...400hz?  I added 4db of boost, and a second slope at the very low end, while trailing them both off a bit as they met each other.  I tuned by ear for level. 

What did I end up with... Much Better Balance.  Much more enjoyable to listen to.  I think I just dialed in a taste of the balance I expect to end up with that offsets the lack of baffle step compensation... and I'm very pleased.

Obviously I could have already done this, but it is very helpful to see measurements, see the problem, and have an idea of what to try to make it better.

[jmimac351]

Frederick has parts in-hand, and after having a nice talk with him, I had a thought about how to make the outboard, Novel-like crossover a bit different than how Wilson implemented it.  As these were (technically) able to go on a (huge) bookshelf (more like maybe built-in cabinets / shelving), the outboard crossover was referred to as a "Novel", with separate connections for input, and outputs for each driver, to accommodate the, well... gimmick... of having separate colored umbilical cords that were to be employed based on "Near Wall" (<24" inches to front wall) or "Free Space" (>24" inches from front wall) setups.  As Danny found no measurable difference between those umbilicals, it begs the question:

Why do I need to put additional connectors on the enclosure for the crossover when the wires aren't going to change? 

Why not just make the leads to each woofer running directly off the crossover long enough to connect to the Duette woofer / tweeter connectors? So, I think that's what I'm inclined to do.  Have a single set of inputs on the enclosure for amp hookup, and have a pair of leads for each driver, of some appropriate length, running out of the crossover enclosure to hookup directly to the Duette woofer / tweeter connectors. In fact, as the binding post panel of the Duette is a removable piece, I may make a plate to replace it, and install tube connectors.  Overall, this arrangement is simpler and eliminates additional connection points.  Just because Wilson put additional connectors in the audio path doesn't mean we need to. 

Frederick also made a funny... he advised that I should set the crossovers up to start burning in while I made the enclosure.  I told him that was funny, because about 5 minutes after they show up from UPS, they are going to be laying on the floor / shoe box / zip tied... whatever... and hooked up to the speaker and playing music.    We're not waiting on no stinking box to hear what this sounds like. 

[jmimac351]

Crossover layout coming together.  I've really enjoyed talking to Frederick about what I'm thinking for the crossover board.  In fact, I've enjoyed talking to him, in general.  What a nice guy...

[jmimac351]

Coming along...

[jmimac351]

Final Wilson Duette crossover board. I will make an acrylic enclosure for this and it will lay flat at the rear of the Duette stand, where the current Novel crossover sits. The board is sized to fit in that space.  I plan to make it out of acrylic / lexan, because I think it looks cool, and I want it to be visible. The parts will not be hidden / potted in the enclosure, like the original Novel crossover. Output wires from the crossover will exit thru the rear and out of the top of the enclosure thru a grommet. They will run up the column for the stand, under the stand shelf, and up to the back of the Duette... instead of hanging out of the rear of the crossover in free space, like the original installation setup. The output wires from the crossover will connect directly to the Duette - no connectors in between the amp inputs and outputs to the Duette.  There will be one set of connectors on the enclosure for inputs from the amp.

I'm very happy with how this is coming along.

[jmimac351]

[KTS]

Frederick,
You do some very nice crossover work, super nice and clean! Well done!


Kelly

[Danny Richie]

Nice work Frederick!

[fre11111]

Thank you very much. I am very excited for this project. I just need to test these crossovers and sen them home.

Frederick

[jmimac351]

Thank you very much. I am very excited for this project. I just need to test these crossovers and sen them home.

Frederick

Funny you mentioned that... I was thinking about what I was going to hook them up to first.  What do you normally do?

[jmimac351]

I have shared this info on another spot where some Wilson owners may frequent more.  It's gotten a fair amount of interest. One person (non-owner), expressed disappointment that I obscured the values of the crossover components.  "Why did you blackout the values of the capacitors? Not seen that done before on all the crossover pics I have viewed over the decades."

Yeah, well you've probably also never sent a speaker to a guy who measured it and fixed it, for FREE. You want the parts / knowledge?... Pay The Man. 

Thanks again, Danny.

[Rocket]

Hi,

Yes there are always someone who wants something for free. Anyway, please let us know how the new crossover sounds?

Cheers Rod