Agreed about the excursion but less so the power handling. The SEAS 5" Prestige midrange for example has a power handling RMS/peak of 100/300W whereas the SEAS 5" Excel midwoofer has 70/200W. In addition the higher sensitivity of a midrange means it needs less power to reach the same SPL. However a high Fs short stroke midrange is unlikely to be easily crossed 2nd order like in 5th elements detailed example above. So it can be a case of working to the strengths of what you have got. A common solution these days to the problem of using midwoofers for the midrange seems to be to use two of them.
If the NeoX plays well with the midrange, we can consider that as well.
Though i am not sure a round mounting front plate is available for that model.
Honestly somehow i am a bit skeptical of another ribbon. May be because the distortion and mid breakup of current one sound so awful
Navin, please see some of the pictures i have posted earlier in the thread. I can post some more if it helps understand the problem more.
Yes experts, Please also offer some views and comments on the exact effects i seem to notice. Can all that be attributed to cone breakup and distortion? Then why do slower music seem to sound very good?
I know the subs are hard to blend well into the rest of the speakers for all types of music. I do notice the subs throw some slow laggy thumps in some tracks, maybe the dayton is not upto driving them connected in series. I dont know.
But the subs are nicely sealed away in an enclosure at the bottom of the cabinet, so they are not affecting the 8"woofers directly. Though as you can see from one of the pictures, the mounting hole of the lower 8" woofers isnt even a proper curcular hole. It hits the upper edge of the subs cabinet below it. I am sure early reflections are muddying up some of the bass there, not sure if that would be a clearly audible issue.
Maybe, maybe not. If Matt (5th Element) can tell us the values of the components in his simulation you can compare them to the values you currently have.
Do you have access to a good impedance bridge to measure the components you have?
Agreed, I did mean "mechanically handle more power"
Jojip,
I got the NeoX with a round mounting plate. Maybe you can too.
However if you are sceptical feel free to use a dome. Although I suspect the breakup might well be because the existing tweeter is being pushed too low (if indeed the HP XO is 2kHz). I am happy with the NeoX (more than the NeoCD 3.0 which in my case is being used by a less critical rear speaker).
I did see the pictures. The cabinets are lovely, the leather wrapped baffle, the woodwork is just exemplary and I can see why you don't want to discard them. I was just wondering about the edge of the baffle and the way it meets the side walls. I could not see this detail clearly.
I think the snap you miss is because of the upper bass or mid bass response and might extend into the midrange as well. Slower music does not have that "snap" that rock has hence it sounds good.
Maybe, just maybe the movement of the subwoofers also vibrates the main cabinet? Maybe, the phase response of the side mounted subwoofers is part of the problem? I don't have enough domain knowledge to categorically state if the subwoofers are causing an issue or not (not without hearing the speakers anyway). Running the speakers with the subs off might shed some light on this.
Thanks to Andy, Matt, and the others we have gotten this far. If we can determine the cause of the problems, it would be easier to find the requisite solution(s).
Is there any expert around the south bay area willing to come over for a listening session to judge these better?
thanks
Joji
thanks
Joji
ARTA is shareware in that you can use it (almost) fully without paying. It will still measure everything you want it to and it will still export graphs, the only thing you cannot do is save impulse responses.
Here is a quick measurement guide I wrote using ARTA, describing how one needs to go about making measurements that are suitable for xover work.
measurement guide
I went with the 2kHz xover as this is what the W15CY really needs to work properly and to show how bandpass gain would work in such an situation. The NeoCD3.0 is completely unsuitable, imo, for this design. I don't care what other people have done with this tweeter, it's distortion profile at high drive levels (typical of large three ways) shows that it really needs a 6kHz 4th order acoustic target to remain clean. Clearly this is not anywhere close to what the W15CY needs to shine and it really should be made to shine as it is a formidable performer.
Yes the ribbon can probably be made to work well in small two way systems crossed much lower as it will never be turned up that high, but from these measurements...
??-????????
It's clear that its struggling at anything below around 6kHz.
The cabinet facets will help to improve anything related to diffraction troubles so I strongly doubt that it's that. It simply sounds like a pair of speakers with an inappropriately designed crossover. Phase issues cause sound staging to collapse and for the sound to appear confused, straining is a classic example of drivers working outside of their useful bandwidth and/or simple linear distortion problems (ie non flat frequency response).
Well if the Subs and W22s share the same air space then that's a major problem. If they do not then it will probably not be the issue.
Not when crossed this low, the subs are only used from around 60Hz and below right?
With the W15 needing a low xover point I would strongly recommend going for a replacement that can handle that with aplomb, not a tweeter that is likely to need delicate handling to work correctly.
Probably because slower moving and simpler music excites less issues. Brash sounding rock music tends to require better designed speakers to sound decent.
That's good to know.
This is unlikely to be any issue as the wavelengths of the sound produced by the W22s are large compared to the size of the obstruction.
This wont work as my simulation hasn't been done in anyway as to be comparable to what's going on here. It is in effect a completely different design.
I can help with xover design if measurements can be made in the correct way.
This is a very good tutorial. Thanks a lot for providing this. I am reading through it.
Yes i am all for replacing the ribbon with seas soft dome
https://www.madisoundspeakerstore.com/seas-soft-dome-tweeters/seas-excel-t25cf-001-e0006-tweeter/ (one you recommeded)
Or this if the added price is justified in this application
https://www.madisoundspeakerstore.c...eas-excel-t25cf-002-e0011-millennium-tweeter/
The only thing i will probably miss is that the ribbons seem cable of going way higher. Much of which i am not likely to be capable of hearing anyway.
I have noticed some high end stuff using a ribbon and a dome combo, likely an overkill for most people.
Yes the designer said that the subs need to be set only below 60Hz.
May be the measurements will show how low the 8" are really going.
Thats good to know. Then i may only need to improve the mid-range mounting holes with some fine chiseling and filing.
That would be really great. I will do my best to get useful measurements for this process.
Last edited:
I dont have one at the moment.
Here are couple of additional pictures
Yes many thanks to them. Over the coming weekend i will first make some measurements of the speakers
Last edited:
Will impedance measurements be needed? I dont currently have the jig for that so will need to build one or buy one.
Also reverse engineering of the existing XO may be limited to getting the circuit topology if values are not legible. I dont have an LCR meter or miliohm meter and this stuff is quite pricey.
If you can measure impedance, you can most certainly
determine LC values using formulae for inductive and
capacitive reactance. First measure the impedance of
the unknown part, then simply choose your value in the
range where phase is close to 90 degrees. I have included
textual files of a 1 uF cap and a 5,6 mH coil. These values
are marked on parts.
Xl=2*pi*fr*L; Xl(ohm); L(H); fr(Hz)
Xc=1/(2*pi*fr*C); Xc(ohm); C(F)
determine LC values using formulae for inductive and
capacitive reactance. First measure the impedance of
the unknown part, then simply choose your value in the
range where phase is close to 90 degrees. I have included
textual files of a 1 uF cap and a 5,6 mH coil. These values
are marked on parts.
Xl=2*pi*fr*L; Xl(ohm); L(H); fr(Hz)
Xc=1/(2*pi*fr*C); Xc(ohm); C(F)
There are lots of good tweeters out there, the reason why I picked that one was because the Neo3CD appeared to have a 110mm cutout. Most tweeters are 104mm, so wouldn't be a drop in replacement. The SEAS tweeter I recommended has a 110mm cut out.
You are right, similar to the one you recommended, the Excel Millennium is 110 mm as well
You are indeed correct for some reason I thought the Millen was 104! Used above 2kHz though I prefer the way the other SEAS performs.
With those magnesium drivers I would rather go with a metal dome tweeter like the Seas T29MF001.
Here's a thread of someone who used that tweeter with Seas magnesium drivers.
Here's a thread of someone who used that tweeter with Seas magnesium drivers.
The metal domed SEAS Excel tweeters border on the ridiculous when it comes to price performance. I too am partial to metal dome tweeters, but in that case I'd much rather go with the DXT from SEAS. It is not 110mm though.
Why not 104mm and fill the remaining cutout with some felt ?
Plenty of choices for good 104mm metal domes, then you can use the excellent Transducer Lab metal or oxide dome tweeters.
Plenty of choices for good 104mm metal domes, then you can use the excellent Transducer Lab metal or oxide dome tweeters.
Lovely find Matt.
Thanks, I was wondering is the sides that are not flush can interfere in any way. Matt? Andy? anyone?
My thoughts were that if you had a bridge, you could even salvage some of the parts for the new crossover Matt has offered to help with.
Yes I believe the flange is 120mm which is not a lot more. BTW Morel's ST11xx tweeter and ET338 also use the 110mm baffle as does the MDT33 I have (all are soft domes though).
Yep, you're right, I missed that. They use rather large magnets.
The SB Acoustics SB26ADC looks amazingly good and that for only 50$ 🙂
Last edited:
- Status
- Not open for further replies.