These do not determine how it sounds. Hovewer, all what you mentioned will determine how much it will cost and I bet most DIY people here will not like it
haha
Only here because you can’t start a thread in multiple forums. The subjext goes beyond multi-way.
dave
CharlieLaub, Thanks for your comments. You are correct, a push-pull-slot-loaded woofer is not a good choice for a 500Hz crossover to a fast horn. A musical 18" woofer like the Faital Pro 18FH500 would be a good value choice to blend with the coaxial BMS 4594HE in the SEOS24 waveguide. The Big Easy.
In case a couple pictures would help the discussion:
Big Classic 3
Big Easy
In case a couple pictures would help the discussion:
Big Classic 3
Big Easy
Attachments
You can even get away with a 2-way (monitor) if you use such a large format horn like the SEOS-24. I have project slated that will use an 18" woofer and the SEOS 18 horn with the DIYsoundgroup DNA-360 CD plus an active crossover. The particular 18 I am using is pretty good up to 1kHz - almost unique in that way. I am hopeful that I can cross over at 1kHz and get smooth on and off axis response. At the same time, the woofer won't do well below about 100Hz, so it might need to be paired with subs to dig low. But the main goal was pattern control and not LF extension.
Also, if you are planning to use pro drivers or make a high efficiency vented-box 3-way with a 15 or 18 woofer be prepared to use a rather large cabinet. I have always been a bit skeptical of the "Maximus 18" design from DIYsoundgroup. Seems a bit too small of a cab... Something more like the JBL 4545 looks to be more on target. That could really be a sperm killer!
Also, if you are planning to use pro drivers or make a high efficiency vented-box 3-way with a 15 or 18 woofer be prepared to use a rather large cabinet. I have always been a bit skeptical of the "Maximus 18" design from DIYsoundgroup. Seems a bit too small of a cab... Something more like the JBL 4545 looks to be more on target. That could really be a sperm killer!
Last edited:
Is that because your wife kicks you in the nuts when she see's the size of the thing!
Last edited:
If a speaker is designed for a digital crossover then it is likely there isn't a reasonable passive crossover. The constraints are much less concerning driver sensitivity, phase matching, slopes, handling response bumps and dips, etc... This tends to mean a well designed speaker is either designed to use a passive crossover or to use a digital one but not both.+1 design should propose both options : analog and digital crossover
One could take a passive design and get it to work probably a touch better with a digital crossover but why would you want to? A better speaker could be designed by changing the driver sensitivities, removing stepped and sloped baffles and the like,... and all the other fun stuff that a passive design requires to be juggled to get things to fit reasonably well.
I agree that an analog or digital xo perform not the same, but both can perform well.
Indeed, the digital one can be designed more accurate on target w.r.t. SPL and phase.
But that doesn't mean an analog version will not be good anymore and should perform worse per definition.
I think today it is interesting to offer both, because not everybody likes to choose for a digital concept.
Some people want to keep their stereo solid state or tube amplifier, they don't want digital now or maybe never, they will do this move to digital later on,...
For systems with a low frequency crossover point, it is more recommended do make it digital or analog active.
For a classic 3-way it still can be made passive in a very good way.
Indeed, the digital one can be designed more accurate on target w.r.t. SPL and phase.
But that doesn't mean an analog version will not be good anymore and should perform worse per definition.
I think today it is interesting to offer both, because not everybody likes to choose for a digital concept.
Some people want to keep their stereo solid state or tube amplifier, they don't want digital now or maybe never, they will do this move to digital later on,...
For systems with a low frequency crossover point, it is more recommended do make it digital or analog active.
For a classic 3-way it still can be made passive in a very good way.
Indeed but my point was that the designs that perform well are normally different because a digital crossover has significantly less constraints. So a design that differs only in having a passive or digital crossover is limited by the constraints of a passive crossover. Why should people that wish to use a digital crossover include these unnecessary constraints?
Obviously people can go for whatever they like but I posted after two posts expressed an interest in speaker designs with passive or digital crossovers as if it did not affect the choice of drivers and their configuration. It does.
I would argue the opposite. Those constraints you are talking about lead to having drivers that reasonably integrate without any heroic efforts. Surely if you go to the trouble of carefully selecting your drivers to NOT need any major manipulation to get something good sounding, then you should be able to get a better result than taking some drivers that are not particularly well suited, and equing the crap out of them to get a decent result?
One area where I can see passive ruling out a particular driver would be where you have a very sensitive woofer that needs to be "padded down" This is not really doable with a passive setup but easily achieved with an active setup (not necessarily digital).
Tony.
Andy,
Yes I agree on that point, with a passive filter some drivers are more difficult to filter than a digital can. For instance equalizing sharp resonances.
Compensating for acoustical offset is also a big advantage of dsp of course. But despite that the passive version remains interesting as an option to be chosen.
In the same way there can be discussed about options of different filter topologies.
Especially for a digital xo, Linkwitz Riley versus Butterworth, elliptical, different orders. Very interesting to compare them and choose your preferred one.
IMO for such forum speaker we have to split up the design in a driver/enclosure and a crossover part.
In a way to reach more people that are interested to built such speaker.
Having a wider choice of appropriate drivers is a significant benefit, being able to place the drivers on the baffle anywhere and get exactly the acoustic offset required is another, as is using the crossover slopes to get precisely the change of beamwidth with frequency that you require, as is the greater ability to deal with bumps and dips from the drivers, guides and/or baffle, plus no doubt several other factors that I have missed which could move the driver choice and cabinet design away from what a passive crossover could handle in order to improve the performance of the speaker.
That is what I am arguing against. Design a speaker for a passive crossover (and it will be straightforward to use with a digital crossover) or design a speaker for a digital crossover and use the flexibility to gain some performance (in addition to that from the lack of nonlinear passive components, easier amplifier load, etc...).
That is just something you can do with digital delays, I don't see the relevance to driver choice there...
Are you saying you can't get whatever slope you want with a passive crossover? The only thing I can see possibly here is if a driver is rolling off too quickly you may try and boost it to get a gentler slope, but I can't see that as being a particularly useful thing to do, better off choosing a more appropriate driver
Yes getting a perfectly flat response will always be easier with digital, but again if the driver is really nasty and you are having to do all that correction, is it really the best driver for the job in the first place?
If a particular set of drivers can't be made to sound good with passive, then IMO you shouldn't be trying to make them sound good by digital means.
At the end of the day you want the drivers to integrate well and provide a seamless sound (however you decide to do that, passive, analogue active, digital active, whatever). Sure you can try to do that with brute force, or you can use a bit of finesse. I know which approach I'd rather take
Tony.
It is not clear for me, are you against this statement or is it the way you propose to follow for the design of a forum speaker.
I think we are all convinced of the benefits of a digital concept, but to get a wide interested public for this speaker you cannot omit the passive filter.
So the driver set has to be chosen to make for both analog and digital a good speaker. Therefore design both
.I mean that a digital filter isn't better than a passive filter, because dsp is more powerful for designing.
A passive system is more sensitive on components, amplifier load, etc...indeed, but well designed with the good drivers it will perform also excellent.
Last edited:
Obviously. Passive filters work in 6dB/octave chunks whereas the filters for beamwidth control are usually of variable slope. Beamwidth control is one of the growing uses of digital crossovers in speakers and soundbars. It is also one of the significant factors when it comes to how a speaker sounds in the home.
Drivers are starting to get designed to have a smooth frequency response and not a flat frequency response. Waveguides introduce smooth changes to the frequency response. Correcting such things is trivial with a digital crossover but non-trivial with passive crossovers and usually with less than perfect results. The more you try to control the directivity of a speaker in order to raise sound quality the less attractive a passive crossover becomes.
It will rather depend on why a driver with a nasty response is being used? If it is unavoidable then a digital crossover could address it fairly well whereas a passive crossover could not.
I have no problem with speaker designs using passive crossovers. They are wanted by substantial numbers of DIYers.
I don't see the point of a digital crossover version of a passive crossover speaker but if people want one then fine.
I am aware that a high quality design with tightly controlled beamwidth using waveguides and multiple drivers will in practice require a digital crossover. To disqualify the best performing speakers because they cannot be made to work with a passive crossover seems daft to me.
But to slight the analog readership and require the remainder to use multiple or multi-channel amps would not be?
IMHO, this design needs to work both ways to appeal to the widest audience.
Just like the many amplifiers and other speakers here, whatever design is ultimately decided on won't be in place 5 minutes before new iterations show up. Leave room for folks to upgrade or simplify as they see fit- that's why this forum is successful.
There can't be just one single "diyAudio.com speaker", we all know that. If Aatto manages to stay the course and coax a successful design from this exercise, surely there will be more to follow and each will be an opportunity to explore other avenues. There is plenty of room for everyone here.
- Status
- Not open for further replies.