That was easy to try except I only have a few resistors to play with. I was already using .47 ohm extra padding on the tweeter. I tried 0.69 ohm which does not make a lot of difference from .47 and the next step up I had from there was 2.7 ohms and that pretty much ruins the sound making the highs really dull. It also again makes me lean towards the tweeter not being the problem judging from how it sounded with the extra padding.
I hear what you are saying about hyperacusis. I don't know if that is what is going on here. I recently went listening to a few speakers at some electronics stores and did find some OK and some quite pleasant sounding. Really nothing that I listened to was bothersome although even the best ones lacked a bit of clarity compared to the ZRTs.
I also have to take my room into account. I will soon have a few large acoustic panels that I will make and move around to try dealing with first reflections and the back wall behind the speakers and see if I can get some improvement.
Really I appreciate all the ideas and there are a couple things here I can try, but I will probably need some measurement equipment to pinpoint the problem.
Last edited:
The 18W8531G00 is a great driver and well liked by builders and manufacturers.
I've used it in several designs from 2006 and have never had any issues in crossover or box design and build. It's been used in sealed (midwoofer & mid), OB (mid), vented (midwoofer), aperiodic (mid). Partnered with various tweeters such as XT25TG, 810921, D2608/913000, S2905/970000. I did look at using the D3004/660000 and it modelled well (see below) but too expensive for me so stuck with the 810921 or D2608/913000 as they were a good match.
The 18W8531G00 does have it's issues such as it can do great deep bass and mids but not at the same time. Removing low bass from this driver pays off in the mids which is why I ended up using it as a mid from upper bass onward to 2.4kHz xo with the tweeter. I found that it can have a nasal quality but a light coating of LTS-50 sorted that out. The usual vented modelled box size of 37 litres is just too large and found 20-22 litres with the F3 well over 40Hz the best when used as a midwoofer.
This doesn't help merlinx76 as he's still not happy. I assume all the drivers are rebated and maybe if you post some pics, someone might spot something. I suggest you run some more YPAO signals through the amp but only use 1 speaker... as is and as a 2-way by disconnecting the lower woofer. Do it at 0.5m, 1m, 2m and 3m with the YPAO mic in line with the upper woofer. This may pick up some artifacts or something odd. You could always just try the 2-way crossover and temporarily blank off the bottom woofer hole and install the port there.
There is always the possibility that you may be sensitive to certain frequencies which makes them feel like a knife in the ear. Originally the issue was around 1kHz but moved onto 3kHz or so. I only mention this as other builders of this speaker find them fine. Certain frequencies can drive me nuts (hum & buzz for example).
These Yamaha HT amps (I have one for movies) usually sound best on music when using Pure Direct and I find their implementation of the ESS DAC to be quite aggressive when compare to say a Cambridge CXN. You do have another amp and even if you run a iPod through it, this removes the Yamaha from the equation.
Hoping you get this sorted as it must be a major disappointment.
I've used it in several designs from 2006 and have never had any issues in crossover or box design and build. It's been used in sealed (midwoofer & mid), OB (mid), vented (midwoofer), aperiodic (mid). Partnered with various tweeters such as XT25TG, 810921, D2608/913000, S2905/970000. I did look at using the D3004/660000 and it modelled well (see below) but too expensive for me so stuck with the 810921 or D2608/913000 as they were a good match.
The 18W8531G00 does have it's issues such as it can do great deep bass and mids but not at the same time. Removing low bass from this driver pays off in the mids which is why I ended up using it as a mid from upper bass onward to 2.4kHz xo with the tweeter. I found that it can have a nasal quality but a light coating of LTS-50 sorted that out. The usual vented modelled box size of 37 litres is just too large and found 20-22 litres with the F3 well over 40Hz the best when used as a midwoofer.
This doesn't help merlinx76 as he's still not happy. I assume all the drivers are rebated and maybe if you post some pics, someone might spot something. I suggest you run some more YPAO signals through the amp but only use 1 speaker... as is and as a 2-way by disconnecting the lower woofer. Do it at 0.5m, 1m, 2m and 3m with the YPAO mic in line with the upper woofer. This may pick up some artifacts or something odd. You could always just try the 2-way crossover and temporarily blank off the bottom woofer hole and install the port there.
There is always the possibility that you may be sensitive to certain frequencies which makes them feel like a knife in the ear. Originally the issue was around 1kHz but moved onto 3kHz or so. I only mention this as other builders of this speaker find them fine. Certain frequencies can drive me nuts (hum & buzz for example).
These Yamaha HT amps (I have one for movies) usually sound best on music when using Pure Direct and I find their implementation of the ESS DAC to be quite aggressive when compare to say a Cambridge CXN. You do have another amp and even if you run a iPod through it, this removes the Yamaha from the equation.
Hoping you get this sorted as it must be a major disappointment.
Attachments
I had a look at the Solen Experience kit which seems to use the 18W/8531G00.
EXPERIENCE - Solen Electronique Inc.
MTM or TMM style. Seems to work and addresses the real issues. I used the 18W/8531G00 FRD and ZMA files. But it's not quite the final optimisation, because I just used another tweeter.
You might give Solen a ring. They seem to be posting the wrong 15W/8530K01 PDF on their website at the bottom of the page. Well, it confused me anyway! But nothing I did is wildly different.
EXPERIENCE - Solen Electronique Inc.
MTM or TMM style. Seems to work and addresses the real issues. I used the 18W/8531G00 FRD and ZMA files. But it's not quite the final optimisation, because I just used another tweeter.
You might give Solen a ring. They seem to be posting the wrong 15W/8530K01 PDF on their website at the bottom of the page. Well, it confused me anyway! But nothing I did is wildly different.
Attachments
Last edited:
This thread is exactly why I went the full range driver route. The mid/woofer I was using was an expensive Eton that had a peak similar to the SS you are using. For over a year (maybe more!) all I did was noodle around with expensive crossover parts, swapping amps, cables, etc. The results were audible but what sounded good on some recordings sounded bad on others. I'm not suggesting you go the full range driver route but I will say trying another mid/woofer (completely different from the one you currently have) and making another project might provide more musical enjoyment. Your ears simply might not like the SS sound. When your budget permits, maybe try another woofer like Seas or Peerless? Ask Madisound. Good Luck and I hope you find musical enjoyment. This is part of the DIY process and will benefit you long term... I think.
Attachments
That looks like it needs an elliptical or Cauer filter, Godzilla. An LCR on the shunt of the bass filter around 4kHz.
Even so, it might work better with 4th order, because notches worsen ultimate rolloff. Gives you more variables to adjust, and pure rolloff is the best solution IMO.
Back to this wretched Zaph Audio 2.5 way:
And the woofer from hell:
Ha Ha, just kidding. It can be fixed, but not as a 2.5 way on a simple filter. I think the Solen Experience is a good point of departure.
It's an easier problem than you think to adapt for a different tweeter. The D3004-660000 is a 92dB 3 ohm tweeter. Which is very low, as it goes. And I would try and get it higher to keep the amp comfortable.
What Steen Duelund brilliantly realised is that a two way and a three way are essentially the same animal in LR2 or LR4. Just a spectrum of change, albeit you get a power hole every time you cross over. Just how the Universe works. His slopes work well on phase too.
You can keep Solen's bass filter and its slope, and just fiddle with the tweeter slope and level to match for whatever tweeter you use.
The thing is, the only fix you can apply here is in the crossover. So I think you need to get on with it. 🙂
Even so, it might work better with 4th order, because notches worsen ultimate rolloff. Gives you more variables to adjust, and pure rolloff is the best solution IMO.
Back to this wretched Zaph Audio 2.5 way:
And the woofer from hell:
Ha Ha, just kidding. It can be fixed, but not as a 2.5 way on a simple filter. I think the Solen Experience is a good point of departure.
It's an easier problem than you think to adapt for a different tweeter. The D3004-660000 is a 92dB 3 ohm tweeter. Which is very low, as it goes. And I would try and get it higher to keep the amp comfortable.
What Steen Duelund brilliantly realised is that a two way and a three way are essentially the same animal in LR2 or LR4. Just a spectrum of change, albeit you get a power hole every time you cross over. Just how the Universe works. His slopes work well on phase too.
You can keep Solen's bass filter and its slope, and just fiddle with the tweeter slope and level to match for whatever tweeter you use.
The thing is, the only fix you can apply here is in the crossover. So I think you need to get on with it. 🙂
Attachments
Nope. 🙂
But looks pretty unpromising material to me. I would much prefer this SEAS U18-RNX-P driver.
Attachments
Ugh. I hate the U18. It's the only mid woofer I cannot get to sound natural. It's got a suckout centered around 750 Hz. It makes for a boxy presentation that I can't listen to. I'm sure the SS driver in question is vastly better.
The devil is always in the detail, my friend. I am quite surprised you hear a suckout at 750Hz, usually it's too much presence in that region. Perhaps you could post a schematic?
This is one of my 6" efforts, which I have fiddled about with quite a lot. I'd actually like to put it in a bigger reflex box to get some bottom end.
The breakup on the woofer is around 5kHz, and I notched it just a tidge for -4dB. It's a very detailed speaker on vocals and you can hear fretting on guitar. So it's doing something right. But this approach just wouldn't work with the Scan woofer, IMO.
Attachments
<< Scratching my head >>
I seem to have lucked out compared to a lot of the criticism lobbed at the 18W/8531. The two 4531s I have display none of the complaints lobbed at them here. Easy to crossover at 2 kHz, easy to fix breakup mode, fabulous deep bass and midrange.
Best,
E
I seem to have lucked out compared to a lot of the criticism lobbed at the 18W/8531. The two 4531s I have display none of the complaints lobbed at them here. Easy to crossover at 2 kHz, easy to fix breakup mode, fabulous deep bass and midrange.
Best,
E
It's because you, seemingly like everyone else confusing the issues here, haven't grasped that a two way is much easier than the 2.5 way. 😀
Zaph 2.5 way:
Solen MTM:
As it goes, 4 ohm 6" woofers are easier than 8 ohm ones too.
Anyway, merlinx76, you are doubtless befuddled by all this technical talk. I have some good news and some bad news in that I will be ducking off 'teh Interwebs for a while. 😀
I spent this afternoon reminding myself what I like about 6" speakers and what is difficult about them.
We've mentioned the 5kHz cone breakup (and every darn one has this...) forcing a lowish crossover and some phase issues.
The lack of natural rolloff making for more complex filters.
Time-alignment problems forcing difficult compromises.
It's a hard speaker at the best of times, IMO.
Your one is not my preferred sort of project. This scanspeak 6" project (and I have heard stuff like this...) appeals to me at many more levels:
I think that you can fix this speaker. IMO, the Solen idea is not far off the money even as a TMM. It is essentially a theoretically correct 2.2kHz BW3 filter with impedance correction. Impedance is uncomfortably low, but most amps should cope. Or you series wire the basses and make some adjustments to crossover values, which would be my more controversial preference. Impedance matters, IMO.
2-Way Crossover Calculator / Designer
A very interesting idea, IMO, would be to try the H1499-06 27TBCD/GB-DXT H1499-06 27TBCD/GB-DXT 104mm driver. A very different sound. But might be what you are looking for.
I spent this afternoon reminding myself what I like about 6" speakers and what is difficult about them.
We've mentioned the 5kHz cone breakup (and every darn one has this...) forcing a lowish crossover and some phase issues.
The lack of natural rolloff making for more complex filters.
Time-alignment problems forcing difficult compromises.
It's a hard speaker at the best of times, IMO.
Your one is not my preferred sort of project. This scanspeak 6" project (and I have heard stuff like this...) appeals to me at many more levels:
I think that you can fix this speaker. IMO, the Solen idea is not far off the money even as a TMM. It is essentially a theoretically correct 2.2kHz BW3 filter with impedance correction. Impedance is uncomfortably low, but most amps should cope. Or you series wire the basses and make some adjustments to crossover values, which would be my more controversial preference. Impedance matters, IMO.
2-Way Crossover Calculator / Designer
A very interesting idea, IMO, would be to try the H1499-06 27TBCD/GB-DXT H1499-06 27TBCD/GB-DXT 104mm driver. A very different sound. But might be what you are looking for.
Attachments
Last edited:
I really don't want to take another design and try to modify it to work on a different tweeter (at least not without measurement equipment) and am reluctant to use these woofers for another design in case it's just the woofers sound in general that I am not partial to. Especially considering at this point, I have found a way to make them much more reasonable and I can live with them like this.
If I were to use these woofers in another design, I think I would like to move to Troels' 3 way which would be a big departure from the ZRT.
I am also considering whether I want to learn more on how to design my own and if I would enjoy the process. I have been reading some on the subject for years but still have a lot to learn. My upgrade-itis could be cured by building someone else's design again (as long as I am a bit happier than with this design). That would be a relatively quick process although it will ultimately be a far cheaper solution if I am only interested in a decent result.
If I were to use these woofers in another design, I think I would like to move to Troels' 3 way which would be a big departure from the ZRT.
I am also considering whether I want to learn more on how to design my own and if I would enjoy the process. I have been reading some on the subject for years but still have a lot to learn. My upgrade-itis could be cured by building someone else's design again (as long as I am a bit happier than with this design). That would be a relatively quick process although it will ultimately be a far cheaper solution if I am only interested in a decent result.
I think there is an awful lot of snake-oil salesmanship applied to loudspeakers. Various vested-interests making it out that it is more difficult than it really is.
Your idea of a genuine three-way will doubtless solve a lot of problems.
Ekta-Grande.
Tweeters are really the easiest problem in audio. Most of them are reasonably flat, although I have mathematical notions of what really works best, like cones and ring radiators.
People will never admit this, but ALL loudspeakers are imperfect. It's built into the Universe. Me, I err in favour of low distortion. 🙂
Your idea of a genuine three-way will doubtless solve a lot of problems.
Ekta-Grande.
Tweeters are really the easiest problem in audio. Most of them are reasonably flat, although I have mathematical notions of what really works best, like cones and ring radiators.
People will never admit this, but ALL loudspeakers are imperfect. It's built into the Universe. Me, I err in favour of low distortion. 🙂
Maybe I do misunderstand the complexity of a 2.5 way, never having designed one, but the break up mode o the 4531, easy peasy.
erik, I really don't like vague suppositions about troublesome cone-breakup, leading to distortion. What did you do with it? And DO get on with it. My internet expires at midnight.
After that, you are on your own. 😀
I just love the level of love I get here in telling people they are wrong and ignorant. 😀
Recall that I am a man who has seen Trolls in the wild. They are not quite what you expect. I went hunting them down in Norway. It's my nature. 😎
Attachments
You can say what you want, but it is at bad habit ditching a driver only by its specs - Which by the way is very good.
Steve,
It wasn't magical, but my measurements of the speaker's FR was a lot better than spec. I never saw the shelf present in the spec sheets. More of a broad bump around 1.5kHz.
I used a single pole low pass filter with a 5kHz notch, and Zobel. I'm not ready to share the specifics of the design, but I'm confused because I'm not that good, and the results were pretty easy to achieve in real life.
Best,
E
It wasn't magical, but my measurements of the speaker's FR was a lot better than spec. I never saw the shelf present in the spec sheets. More of a broad bump around 1.5kHz.
I used a single pole low pass filter with a 5kHz notch, and Zobel. I'm not ready to share the specifics of the design, but I'm confused because I'm not that good, and the results were pretty easy to achieve in real life.
Best,
E
- Status
- Not open for further replies.