I'm probably going to upset somebody here, but this MarkK Dayton metal woofer RS225 is IMO absolutely dreadful in implementation:
Dayton Reference RS225
That crossover is way too complex and expensive. Dissi knocked up a much simpler way of doing it: http://www.diyaudio.com/forums/multi-way/220465-new-drivers-my-2-way-boxes.html#post3181822
Another piece of audiophoolery IMO is Zaph's ZD5: Zaph|Audio - ZD5 - Scan Speak 15W8530K00 and Vifa XT25. This speaker will work just as well with a simpler 4th order tweeter filter.
I suppose it proves that if you baffle people with enough complex nonsense, they will buy it! 😀
FWIW, I still think I could do something with that Epos M12i crossover to get away from the daft first order tweeter filter. 😎
Dayton Reference RS225
That crossover is way too complex and expensive. Dissi knocked up a much simpler way of doing it: http://www.diyaudio.com/forums/multi-way/220465-new-drivers-my-2-way-boxes.html#post3181822
Another piece of audiophoolery IMO is Zaph's ZD5: Zaph|Audio - ZD5 - Scan Speak 15W8530K00 and Vifa XT25. This speaker will work just as well with a simpler 4th order tweeter filter.
I suppose it proves that if you baffle people with enough complex nonsense, they will buy it! 😀
FWIW, I still think I could do something with that Epos M12i crossover to get away from the daft first order tweeter filter. 😎
Linkwitz would claim otherwise, as would the many hundreds of people who praise the Orion so highly.
The RS225s distortion issues hit around 1.8kHz, one needing to cross a little before this, and steeply, to make them a non issue. MarkK handles this with the unusual quasi 8th order xover @ 1.2kHz in one of his xovers and ~1.5kHz is in his revised version, that uses notches to push the xover into 8th order slopes around the xover frequency and then transitions to 4th order once significantly within the stop band.
These both provide more than 10dB attenuation before the 3rd order uglies from the RS225 hit, keeping them well out of the picture.
Mark is no idiot and knows that the design pushes the tweeter extremely hard. There's a reason why he posts punishing 3 tone non linear distortion tests for the completed loudspeaker at the end of the design page. These were done on purpose to make sure that the tweeter could cope. The measurements show that it can.
While you wont get any argument from me about the use of a wave guide, having the xover frequency at around 1.5kHz does limit the off axis issues of such a design. Mark shows some lateral off axis measurements and they are as flat as a pancake, the low xover making these far less of an issue than they would otherwise be.
Yes except that Dissi's version has the xover set too high and doesn't completely address the RS225s main issues.
If you call getting the job done properly audiophoolery then so be it.
The idea behind the ZD5 was to purposefully use 2nd order acoustic slopes. When one wants to do this you usually have to use a stepped or sloping baffle, like the designs by Jeff B and Joachim using the Satori. If you don't want to use a stepped/sloped baffle, then you have no choice but to use a delay network of some sort on the tweeter. Zaph chose to go the second route and introduced the ladder delay network.
Yes, Zaph could have used a 4th order network, but then the design would have simply been just another '4th order LR' Zaph speaker and the ZD5 was an example of how to do something different in the correct way.
You have to remember, it gets boring when everything is just another 4th order LR design. It isn't the only way to skin the proverbial cat, but it does happen to work well in almost all situations. Mark Ks RS225/RS28A design is one such design where this isn't quite true and going with the quasi 8th order scheme provides certain technical benefits that improve the end design. Zaphs ZD5 is just another example of how you can do things differently should the drivers either require it, or allow you to do something simpler.
Btw you might be interested to know about this. 😀
Attachments
I lack the information to say with any authority in this particular instance but expensive complex crossovers and metal drivers crossed close to their breakup do tend to go hand in hand. An 8" paper or poly cone would be a more suitable candidate for a cheaper simpler crossover and, perhaps, a better 2 way speaker. However these decisions are not made in isolation. What about the bass? What about reusing the woofer and tweeter in a 3 way? What about...
Selection of driver material is very important to keep a crossover complexity down. I often think either a driver has a built-in mechanical filter, or you engineer it in the crossover. So paper is easier than metal in that respect. Polycones even simpler.
One of the things you do in engineering a completed design towards production is strip out components. I can comment that there is a great deal that you can strip out with the two designs I mentioned!
A two way is a variously compromised design, but if you can't implement it cheaply, forget about it! As it goes, there are some elegant solutions that sound very reasonable. But Zaph and MarkK are having a laugh here. At YOUR expense! 😀
Spending more cash best goes towards a three way, which we are all agreed is a better way of doing things. Troels has published a new design which looks close to ideal: ScanSpeak-3W-Discovery
I'm not sure I see huge advantages with a stepped baffle, and perhaps Mr. Troels too is indulging in a bit of fashionable audiophoolery there, but the ring radiator certainly appeals to me. 😎
One of the things you do in engineering a completed design towards production is strip out components. I can comment that there is a great deal that you can strip out with the two designs I mentioned!
A two way is a variously compromised design, but if you can't implement it cheaply, forget about it! As it goes, there are some elegant solutions that sound very reasonable. But Zaph and MarkK are having a laugh here. At YOUR expense! 😀
Spending more cash best goes towards a three way, which we are all agreed is a better way of doing things. Troels has published a new design which looks close to ideal: ScanSpeak-3W-Discovery
I'm not sure I see huge advantages with a stepped baffle, and perhaps Mr. Troels too is indulging in a bit of fashionable audiophoolery there, but the ring radiator certainly appeals to me. 😎
Last edited:
No, one of the things you do in engineering is work towards an end goal that satisfies the design requirements. Stripping components out of Zaph's ZD5 would not make it a 2nd order acoustic 2 way design any more. You'd have failed at your initial task. Zaph's design is very simple apart from the ladder delay network.
From an engineering point of view though, you've got two options. Go with a stepped baffle, or go with a ladder delay network. The stepped baffle/delay is necessary such that you don't have to use asymmetric slopes and can go with a perfect 2nd order filter.
No doubt Zaph weighs this up in his design. 1) Make a cabinet that is significantly harder for people to build, or 2) Go with a very easy cabinet and make the crossover more complex/expensive.
I'd wager most people would prefer 2. I know I would.
Mark Ks design is a no compromise, brute force approach towards working two drivers at the limits of their usable frequency ranges. You'll get no argument from me that it would be better to insert a mid range into the design, a la the Finalists, but that again means you've failed at your initial task. And the Finalists are significantly more expensive than the two way.
Even if you were to go with a poly cone 8" driver you're still going to want to cross it over at ~1.5kHz to minimise off axis issues, so you've hardly changed the goal posts. You're still going to require 4th order acoustic slopes at the least here and as far as I know there aren't any 8" paper or poly cones out there that can do what the RS225 does, up to around 1.5kHz, as well as the RS225 does, for the price that you can have the RS225 for.
Yes I linked to that in my previous post.😉
Because without it you would not be able to use perfect 2nd order slopes and keep the drivers perfectly in phase with one another.
in my experience, sound quality and measurments do not go hand in hand.
The tannoy red 15 concentric driver is a perfect example. It measures relatively poorly, yet sound exquisite, more delicious then any ''well engineered'' speakers.
I'm blown away by statement around here like you can know how a speaker will sound by looking at the measurments as this has not been my experience, ever.
The tannoy red measures badly, yet the presence, emotion, dynamics, musicality of the tannoy will never show up in a measurments. The sound is so powerful, so visceral, yet will not measure good. What does that say?
All my favorite speaker i ever heard measured less then perfect, while design that measures flat and correct is not the best sound ive heard in my life.
Coherancy may be somehow a a little bit measured, but not musicality. and well I have listen to many many speakers. What makes a good speaker is not all the details, soundstage, ect. its much more about the tone rightness and how you get drawn in the music listening to the speakers, and that cannot be measured.
I do not trust reviewers, I only trust peer review.
I consider a speaker that has been compared to 50 speakers and won every competition to be trust worthy. (continuum)
I would consider multitude of commercial standmount speaker that would be trustworthy. Audio note k, ref 3a, coincident, proac, merlin, ect.
The tannoy red 15 concentric driver is a perfect example. It measures relatively poorly, yet sound exquisite, more delicious then any ''well engineered'' speakers.
I'm blown away by statement around here like you can know how a speaker will sound by looking at the measurments as this has not been my experience, ever.
The tannoy red measures badly, yet the presence, emotion, dynamics, musicality of the tannoy will never show up in a measurments. The sound is so powerful, so visceral, yet will not measure good. What does that say?
All my favorite speaker i ever heard measured less then perfect, while design that measures flat and correct is not the best sound ive heard in my life.
Coherancy may be somehow a a little bit measured, but not musicality. and well I have listen to many many speakers. What makes a good speaker is not all the details, soundstage, ect. its much more about the tone rightness and how you get drawn in the music listening to the speakers, and that cannot be measured.
I do not trust reviewers, I only trust peer review.
I consider a speaker that has been compared to 50 speakers and won every competition to be trust worthy. (continuum)
I would consider multitude of commercial standmount speaker that would be trustworthy. Audio note k, ref 3a, coincident, proac, merlin, ect.
Last edited:
Audio note, are you having a laugh? Their speakers tend to be a prime example of how not to design a good loudspeaker.
I did post a response to your comment on the Tannoy's before though.
I did post a response to your comment on the Tannoy's before though.
Karl, what is it about the designs that have already been recommended and that meet your selection criteria - and let's include Troel's latest because that one hits all the right notes as well and hits them very well indeed - that is making you hesitate?
Just be aware that you will end up locked into probably the Finalists with these driver choices (the Dayton 2-way) unless you want to go a different way after that which in the end is going to cost you more money. My thoughts are that Mark K's speaker is going to be better than what you currently own but that it won't be as good as the Finalists which has the much better, dedicated midrange.
In that case, you might as well just build the Finalists.
Just be aware that you will end up locked into probably the Finalists with these driver choices (the Dayton 2-way) unless you want to go a different way after that which in the end is going to cost you more money. My thoughts are that Mark K's speaker is going to be better than what you currently own but that it won't be as good as the Finalists which has the much better, dedicated midrange.
In that case, you might as well just build the Finalists.
have you heard a audio note speaker? So now any audio note speaker are bad? come on now
do you deliberately omit to answer the most important point of my posts?
Can you measure musicality, coherancy, tone? those three aspects are the most important part of any loudspeaker, and besides coherancy that you can slightly measure, you cannot measure those things.
basically i feel you recommend speakers you never heard and bash speaker you never heard.
Last edited:
In my opinion too many of the designs out there looks terminally DIY, and I really puts me off wanting to build them.
I know they primarily are for listening too and not for looking at but as I have the ability I'd like to exploit it.
So yeah my thoughts with the 8+1 is I can eventually add a mid and build the finalists if I end up going that direction but in the meantime gets me a couple of quality drivers that I can cut my teeth on and that could go into a future design too.
What I think I would really like would be some simulation software too.
I do not need to listen to a speaker with a right wonky frequency response to know that I will not like it.
Coherency and tone are direct results of a flat frequency response with correctly integrated and phase aligned drivers. It's as simple as that.
Musicality is not an attribute that one attaches to equipment though it is down to what's played through it. But even the must 'musical' piece of music is going to be unenjoyable in the long term if you've got driver integration issues and large deviations from flat, in the frequency response, in key locations that make music annoying to listen to. Such as over emphasised bass or sibilance in the treble.
I have designed enough loudspeakers of my own to know when a design has been done competently and what that means in terms of how it will sound. No, I do not know the specifics about its overall tonal balance in the room in which it will be used, but like I said before, this is DIY, you can tune the tonal balance to suit your preferences.
What I can say is that the flat frequency response, flat off axis response free from the usual holes, low distortion and well integrated drivers used within their optimum ranges, will result in a very clear and detailed sound that is easy to listen to. There will be nothing that sticks out apart from maybe a bit too much bass, not enough sparkle etc that will be down to positioning and level matching between the drivers. IE adjustable by the end user to suit their needs.
The only loudspeakers that I've tended to bash are the ones that measure badly. I have never seen an Audio Note loudspeaker and gone, wow that's well designed.
You on the other hand have thrown criticism towards designs that measure superbly without any other reason except that you do not know who the designer is and whether or not you can trust them. The measurements show that you can trust the design or not.
I have not thrown any criticism toward any design. if I did, that was not my intention at all.
However, you seem really comfortable bashing a speaker that you have never heard based on measurments and you think that you can determine the sound of a speaker beased on meaasurments, which has not been my experience at all. not even close seriously.
I suggested to the op to be really sure of what the op is getting into.
Spending 1k for a design that wasnt heard by many seems really risky, especially if the op is limited in his ability to change the speaker if after everything he is not happy with the sound. Dont you agree?
However, you seem really comfortable bashing a speaker that you have never heard based on measurments and you think that you can determine the sound of a speaker beased on meaasurments, which has not been my experience at all. not even close seriously.
I suggested to the op to be really sure of what the op is getting into.
Spending 1k for a design that wasnt heard by many seems really risky, especially if the op is limited in his ability to change the speaker if after everything he is not happy with the sound. Dont you agree?
Last edited:
Can you be more specific? Or can you link to the type of aesthetics that appeal to you? Because we've been trying to tell you that you can alter the cosmetics up to a certain degree to suite your tastes.
There is a whole slew of easily accessible free simulation software out there that uses Excel and a number of others that don't. If this is the way you want to go, I concur with everything Mr element has been saying his last few posts and suggest you first try to master an understanding of harmonic distortion and cone resonance (CSD) measurements that I posted before and that you can find at Zaph|Audio. These are of key importance when selecting drivers and understanding where to select the xo points and slopes.
Why is it so hard for you to accept that the measurements dictate how a design performs. If it has a wonky frequency response then it is NOT an accurate loudspeaker. If it has ill chosen xover frequencies then it will have issues off axis and possibly with non linear distortion. If the design doesn't include baffle step compensation, or enough of it, then the design will sound thin etc.
Now the degree to which deviations from 'perfection' affect the sound in a negative way depend heavily on what the deviations are. I have never said that a loudspeaker that measures badly will inherently sound subjectively unpleasing, just that it will not be the best choice of loudspeaker and that if given the choice it should be left aside.
Yes, but it isn't possible for people to listen to most DIY designs before building, it's a risk we all take. Having said that the risk is significantly reduced when you build a design that measures superbly and is built with excellent drivers. Like I've said a few times already, one can alter most DIY designs very easily to suit ones own tastes. The designer and co-devloper of the Finalists is also an active poster on the HTguide forum and is also quite friendly towards others. I am sure if Karl has a specific issue with the sound that they will chime in with some advice.
What more could you want with regards to picking a DIY design? Excellent drivers, excellent design, excellent measurements and the promise of support if things don't go entirely to plan.
This comment surprises me a little too as jReave pointed out, and as I did at the start, aesthetics are quite open to your own requirements. So show us designs you like, or let us know how you would want to alter some DIY design and we'll see if it's possible.
youknowyou is completely right about Tannoys. They are a classic paper speaker with musicality in spades. I used to build them, so I know. 🙂
Why are they so good? Troels knows and I know: TQWT-
Runs away giggling. 😀
Now really everybody, we are better moving forwards, than keep criticising the previous post, or banging on that this or that driver is the best thing since sliced bread. That's very boring indeed. Speakers are really quite easy. It's the compromises that hurt. 😱
Why are they so good? Troels knows and I know: TQWT-
Runs away giggling. 😀
Now really everybody, we are better moving forwards, than keep criticising the previous post, or banging on that this or that driver is the best thing since sliced bread. That's very boring indeed. Speakers are really quite easy. It's the compromises that hurt. 😱
Well I pretty much explained why earlier on when the Tannoy's were first mentioned 😀
I have no issue with people making recommendations towards designs that perform well from an objective position. But there's no sense in recommending something that measures poorly when there are designs that both sound and measure well.
I too have a thing for higher than average sensitivity designs 🙂
I have no issue with people making recommendations towards designs that perform well from an objective position. But there's no sense in recommending something that measures poorly when there are designs that both sound and measure well.
I too have a thing for higher than average sensitivity designs 🙂
Here's some interesting high efficiency news. I think I've figured out how WLM do the La Scala!
What is this crossover doing?
I guessed the WRONG driver! It's the W 200 S - 4 Ohm...😱
Confirmed it here:
HighEnd 2009
THAT, my friends, is a rather brilliant 5 element series crossover, which conceals a surprise if you think the coil is bafflestep. The rest, as they say, is just technique. 😎
An externally hosted image should be here but it was not working when we last tested it.
What is this crossover doing?
An externally hosted image should be here but it was not working when we last tested it.
I guessed the WRONG driver! It's the W 200 S - 4 Ohm...😱
Confirmed it here:
HighEnd 2009
THAT, my friends, is a rather brilliant 5 element series crossover, which conceals a surprise if you think the coil is bafflestep. The rest, as they say, is just technique. 😎
here is a sense when you recommend a speaker that measures ordinary but sounds much better then a well measured speaker that sound less good.
My recommendation is a used ref 3 a mm decapo, because the sound is beautiful, has great resell value, is efficient, and sounds really special imo. One of the best speaker I have ever heard. I doubt you can do much better for the price.
This might be the right build for you:
https://sites.google.com/site/undefinition/tarkus
The great thing about it, it's not expensive, drivers are decent and you
always have the chance to improve it one day if you get to that level.
I am convinced forum members would help you along.
- Status
- Not open for further replies.