Perhaps I've created some confusion, or perhaps I'm the one who is now confused.
My description of a 3.5-way speaker as a 3-way with a built-in Sub, was somewhat metaphorical. I was the easiest way to describe the configuration in the fewest possible words. It's conceptual, not literal.
The second 0.5-way woofer re-enforces the primary woofer at very low frequencies. So, Tinitus description as identical woofers crossing over at different points is correct within conceptual limits.
But I was under the impression that at very low frequencies the 0.5-way and the main woofer were working together with the 0.5-way fading as frequency increases leaving the upper bass to the main woofer. In short, the 0.5-way re-enforces the main woofer at low frequencies; they work together.
A true Sub, to my limited understanding, handles all the low bass with that same low bass blocked from the main speaker. So, in that sense, a 3-way with a true Sub could be considered a 4-way system with 4 discrete frequency bands.
Again, I was simply striving for a short conceptual description.
Or, if this has nothing to do with me ... never mind.
Steve/bluewizard
My description of a 3.5-way speaker as a 3-way with a built-in Sub, was somewhat metaphorical. I was the easiest way to describe the configuration in the fewest possible words. It's conceptual, not literal.
The second 0.5-way woofer re-enforces the primary woofer at very low frequencies. So, Tinitus description as identical woofers crossing over at different points is correct within conceptual limits.
But I was under the impression that at very low frequencies the 0.5-way and the main woofer were working together with the 0.5-way fading as frequency increases leaving the upper bass to the main woofer. In short, the 0.5-way re-enforces the main woofer at low frequencies; they work together.
A true Sub, to my limited understanding, handles all the low bass with that same low bass blocked from the main speaker. So, in that sense, a 3-way with a true Sub could be considered a 4-way system with 4 discrete frequency bands.
Again, I was simply striving for a short conceptual description.
Or, if this has nothing to do with me ... never mind.
Steve/bluewizard
BlueWizard said:
So, with a bit of difficulties we seem to agree...but I suppose he will still need a sub though 😀
So we are looking at Sub/WWMTM...estimated xo points, 70(sub)-350(0.5)/1k/5k
AE makes some nice 10" woofers, TD10X, 92db...man, why cant I just leave it alone...dont get your hopes too highj, it wont work
Tinitus-
So, with a bit of difficulties we seem to agree...but I suppose he will still need a sub though
So we are looking at Sub/WWMTM...estimated xo points, 70(sub)-350(0.5)/1k/5k
Yes, that's roughly what I was trying to describe. But, and again, it was only a design concept somewhat consistent with the original desire, I was thinking more of Passive 8-ohm subs with two 16ohm woofers in one cabinet combined with a multi-driver tower in the other cabinet.
So, in my conceptual view, I envisioned -
(Sub+WWMTM) + (WWMTM + Sub)
Or two Subs + two Towers.
He has an amp with plenty of voltage and current reserve, so I though we could bypass the separate Sub amp. Again, I was just spit-balling possibilities for consideration, and to show that the original 4-way dual parallel speakers was not the only way to accomplish the goals.
I'm sure there is someone in this group that has a system similar to what I describe, though certainly they have a dedicated amp for the Subs. I think the speaker is visible on the right every time this person posts, but I don't know who it is.
Steve/bluewizard
So, with a bit of difficulties we seem to agree...but I suppose he will still need a sub though
So we are looking at Sub/WWMTM...estimated xo points, 70(sub)-350(0.5)/1k/5k
Yes, that's roughly what I was trying to describe. But, and again, it was only a design concept somewhat consistent with the original desire, I was thinking more of Passive 8-ohm subs with two 16ohm woofers in one cabinet combined with a multi-driver tower in the other cabinet.
So, in my conceptual view, I envisioned -
(Sub+WWMTM) + (WWMTM + Sub)
Or two Subs + two Towers.
He has an amp with plenty of voltage and current reserve, so I though we could bypass the separate Sub amp. Again, I was just spit-balling possibilities for consideration, and to show that the original 4-way dual parallel speakers was not the only way to accomplish the goals.
I'm sure there is someone in this group that has a system similar to what I describe, though certainly they have a dedicated amp for the Subs. I think the speaker is visible on the right every time this person posts, but I don't know who it is.
Steve/bluewizard
BlueWizard said:
Isn't that slightly redundant. If the Sub driver is mounted in the same box, than it is infact a 4 way.
All of the drivers need their own seperate Chambers in one box.
Weather the Sub driver is crossed over actively with a seperate amp or passively crossed over. It would still make it a 4 way.
Again the .5 comes from Surround sound LFE (which is its own channel Low Frequency effects.
There is a compromise called Bass Management, when the full range speakers can't go low enough to meet X curve. Then the signals from the full range are crossed over (normally around 80Hz), and mixed down into the LFE channel.
If you have a single external sub that your stereo channels are crossed over and summed into I still don't think you can consider it a 0.5 way as there is not discrete 3rd audio channel for Extended bass.
a 2.5 way refers to a speaker in which the .5 driver is used to compensate for the baffle step loss. I would consider a 3.5 way the same thing, a 3 way in which the .5 driver is compensating for the baffle step loss. For instance, if the midrange only goes down to 600hz, but baffle step starts in around 450hz, then you would need to compensate for the loss, and could not do so with the crossover points chosen. Zaph has a design like this on the parts express sight. The reason its called a .5 way design is that the speakers have 2 or more drivers covering the same range for a period of time, but a portion of those drivers cuts off before the others at the baffle step point. In an actual 2 way, 3 way, 4 way you would not have two drivers covering the same range for a portion of their response.
While I'm sure this amplifier can theoretically produce the power you are saying, in real life, you have a lot of issues to contend with. The Current consumption will be huge. If you were to design a speaker with say a 1 ohm load, even a 2 ohm nominal load with the amp you are talking about, it likely would be drawing between 15 and 20 amps from the wall quite frequently. Because its a Class A amplifier its going to be producing huge amounts of heat which you will need to dissipate. I really think you will find that things are more reliable and work better if you don't push things so close to the limit. Designing a speaker to be intentionally 3 or 4 ohms nominal is fine and quite common, but below that is very demanding of an amplifier.
I mean, that amplifier will have to have its own dedicated 20 amp line, at least. There may be someone else who can speak a little more accurately on exactly how much current this amplifier would draw under this kind of load. I haven't read every single post, so if I misinterpreted something I apologize, but did I read correctly that you are running 40 volt rails for this? That seems pretty low for an amplifier of this power capability. It would seem to me that it would have to produce considerably more current then, which again, is going to be much harder on your wire, fuses, capacitors, etc. My current amplifier project is running 75 volt rails to produce roughly 350 watts into 8 ohms and 600 into 4 ohms. It's a Class AB design and I still have massive heat sinks to dissipate the heat effectively (Roughly 20"x8"x1.5"). I also have a power supply capable of roughly 1600 watts rms, so i don't even want to think about how big yours is going to need to be (40+ amps per rail).
Have any pictures of this amplifier, I'm curious what it looks like. Mostly of the chassis and internal layout. I like looking at others work in that area, mine is usually not the greatest, so it helps give me more ideas.
While I'm sure this amplifier can theoretically produce the power you are saying, in real life, you have a lot of issues to contend with. The Current consumption will be huge. If you were to design a speaker with say a 1 ohm load, even a 2 ohm nominal load with the amp you are talking about, it likely would be drawing between 15 and 20 amps from the wall quite frequently. Because its a Class A amplifier its going to be producing huge amounts of heat which you will need to dissipate. I really think you will find that things are more reliable and work better if you don't push things so close to the limit. Designing a speaker to be intentionally 3 or 4 ohms nominal is fine and quite common, but below that is very demanding of an amplifier.
I mean, that amplifier will have to have its own dedicated 20 amp line, at least. There may be someone else who can speak a little more accurately on exactly how much current this amplifier would draw under this kind of load. I haven't read every single post, so if I misinterpreted something I apologize, but did I read correctly that you are running 40 volt rails for this? That seems pretty low for an amplifier of this power capability. It would seem to me that it would have to produce considerably more current then, which again, is going to be much harder on your wire, fuses, capacitors, etc. My current amplifier project is running 75 volt rails to produce roughly 350 watts into 8 ohms and 600 into 4 ohms. It's a Class AB design and I still have massive heat sinks to dissipate the heat effectively (Roughly 20"x8"x1.5"). I also have a power supply capable of roughly 1600 watts rms, so i don't even want to think about how big yours is going to need to be (40+ amps per rail).
Have any pictures of this amplifier, I'm curious what it looks like. Mostly of the chassis and internal layout. I like looking at others work in that area, mine is usually not the greatest, so it helps give me more ideas.
sumsound said:
Ah...not quite.
It hinges on how you define 'WAY'.
In one sense each 'way' is a separate band of frequencies primarily covered by one speaker, or one purpose specific set of speakers as in multiple woofers or multiple mids; ie: WWMTM is a 3-way even though there are two 'W's and two 'M's.
An 0.5-way is not the same as a 0.1 , which is what you are referring to. In Surround Speaker/Amp Systems, the 0.1 is not a 'way'. It simply designates that the system has a true separate self-powered Subwoofer as in 5.1 AV systems or 7.1 AV systems. So, in this sense, a 5.0 AV system is a 5-channel AV Surround System that doesn't have a dedicated Subwoofer.
A stereo system with a true Subwoofer would be referred to as a 2.1 AV system, but neither the '2' nor the '.1' is a 'way' in the sense of speaker cabinet design as we are using it hear.
The 0.5-way is a not a full 'way' because it doesn't have it's own dedicated frequency band. It shares a band with the main woofer. So, it doesn't get to be a 'way' on it's own, it just gets to be a half-way or 0.5-way.
A 3-way speaker with a true Sub, is a 4-way system because each set of dedicated speakers has its own dedicated band of frequencies to cover.
The 3-way system with an additional 'assist' woofer is just a 3.5-way system.
See?
steve/bluewizard - who commonly creates far more confusion than he ever clears up.
Eva said:
The amplifier was specifically designed as an extravagance to drive difficult loads. It will drive an Apogee no sweat. It will drive Apogee speakers in parallel. It will also drive, without current limiting, any of the many high-end speakers out there (Watt/Puppy, etc) with nominal impedances around 4-8 ohms, which dip to 1 ohm or less.
This is what it was designed to do, and this is what it will do – with other peoples speakers. It is not intended to be a PA amplifier and “real loud” was never on the agenda.
However, do not think that an amp rated at 200W per channel into 4 ohms will have a problem with making substantial noise in a HiFi setting with most speakers regardless of their efficiency.
I’d really appreciate it if people here could stop second-guessing the design criterion.
I am interested in designing a pair of speaker for my own demonstrator use (I cannot afford at pair of speakers in the Watt/Puppy league). If it isn’t practical to build a high power handling HiFi speaker with an impedance substantially less than 4 ohms, then I will just build one with 4 ohms and be done with it. It is no big deal and I’m not particularly worried.
As for my general ignorance WRT to speaker design, I made this perfectly clear in the opening post. I have absolutely no intention on embarking on any speaker project before studying up on the topic a great deal more.
I am just throwing up some concepts ATM to get an idea of what is feasible and what isn’t.
Cheers,
Glen
BlueWizard said:
pjpoes said:a 2.5 way refers to a speaker in which the .5 driver is used to compensate for the baffle step loss. I would consider a 3.5 way the same thing, a 3 way in which the .5 driver is compensating for the baffle step loss. For instance, if the midrange only goes down to 600hz, but baffle step starts in around 450hz, then you would need to compensate for the loss, and could not do so with the crossover points chosen. Zaph has a design like this on the parts express sight. The reason its called a .5 way design is that the speakers have 2 or more drivers covering the same range for a period of time, but a portion of those drivers cuts off before the others at the baffle step point. In an actual 2 way, 3 way, 4 way you would not have two drivers covering the same range for a portion of their response.
While I'm sure this amplifier can theoretically produce the power you are saying, in real life, you have a lot of issues to contend with. The Current consumption will be huge. If you were to design a speaker with say a 1 ohm load, even a 2 ohm nominal load with the amp you are talking about, it likely would be drawing between 15 and 20 amps from the wall quite frequently. Because its a Class A amplifier its going to be producing huge amounts of heat which you will need to dissipate. I really think you will find that things are more reliable and work better if you don't push things so close to the limit. Designing a speaker to be intentionally 3 or 4 ohms nominal is fine and quite common, but below that is very demanding of an amplifier.
I mean, that amplifier will have to have its own dedicated 20 amp line, at least. There may be someone else who can speak a little more accurately on exactly how much current this amplifier would draw under this kind of load. I haven't read every single post, so if I misinterpreted something I apologize, but did I read correctly that you are running 40 volt rails for this? That seems pretty low for an amplifier of this power capability. It would seem to me that it would have to produce considerably more current then, which again, is going to be much harder on your wire, fuses, capacitors, etc. My current amplifier project is running 75 volt rails to produce roughly 350 watts into 8 ohms and 600 into 4 ohms. It's a Class AB design and I still have massive heat sinks to dissipate the heat effectively (Roughly 20"x8"x1.5"). I also have a power supply capable of roughly 1600 watts rms, so i don't even want to think about how big yours is going to need to be (40+ amps per rail).
Have any pictures of this amplifier, I'm curious what it looks like. Mostly of the chassis and internal layout. I like looking at others work in that area, mine is usually not the greatest, so it helps give me more ideas.
The heatsinks for each channel of the amplifier measure 1500mm X 220mm X 75mm. It has rails of +/-47V. These rails are electronically regulated. It has a f^%$&^% huge power supply. Aluminium bus bars and used in lieu of output stage and series pass regulator wiring. I am currently finishing off the mechanical assembly (a huge job). All details of the finished product will go on my webpage under the K800AB title when I’m finished.
The amp cost me in the vicinity of 4k. I can spend a comparable amount on a pair of demonstrator speakers.
Cheers,
Glen
I’d really appreciate it if people here could stop second-guessing the design criterion.
We understand the underlying purpose of having an amp with capabilities like this...we really do.
It has to be massively voltage and current stable, which means it is equally massively load stable. All of great benefit to anyone listening to music. I wish I could afford an amp like that.
Our problem is what seems to be your idea that you should use this capability directly. That you should commonly connect 1 ohms loads to it. That is not why this capability is generally there.
Most people don't realize that when it come to power supplies, you can have voltage or current, but you can't have both. When the current load jumps massively, internal losses means the voltage drops proportionally. Your amp by your estimation is immensely voltage and current stable. It really can supply masses of current and within reason maintain the voltage - THAT IS A GOOD THING, A VERY GOOD THING.
But, it represent reserve capability for when more common 4 ohm and 8 ohm loads become demanding. It doesn't mean the capability exists to be used in the long term to drive 2 ohm loads.
So, what you have is a massively transient stable amp. When huge music transients occur, you amp doesn't falter in voltage or current. It continues to give the best possible and most stable dynamic range to the music. Again, this is all very good, and the kind of stuff we dream about.
But just because your amp can drive 1 ohm loads, doesn't mean it should. That's all we are saying; keep the impedance up in normal impedance range, and take comfort in knowing you have this massive stability available to you.
Now there are exception, as you rightly point out, an 8 ohm or 4 ohm speaker drops much lower than that. In fact, I've see 8 ohm rated speakers that drop as low as 3.2 ohms. Your amp doesn't have to worry about that because it has plenty of reserve. If you had two of these speakers on each channel, it could drop as a low as 1.6 ohms, which spells trouble for most amps, but not yours. again, that's a good thing.
So, if you were not really intending to make speaker with insanely low impedances on purpose, then ...oops... our mistake. But if you were, then ...oops... your mistake on several fronts.
Again, intensionally making excessively low impedances is like driving you car around all day in low gear. Certainly you can do it, certainly the car can handle it, but really, what is the point.
So, again, it is nice to have that huge stable reserve in your amp; an amp to be envied. But use it to give you reserve and stability at normal loads, not to simply consume power at pointlessly low loads.
Do, you see my point?
Steve/bluewizard
We understand the underlying purpose of having an amp with capabilities like this...we really do.
It has to be massively voltage and current stable, which means it is equally massively load stable. All of great benefit to anyone listening to music. I wish I could afford an amp like that.
Our problem is what seems to be your idea that you should use this capability directly. That you should commonly connect 1 ohms loads to it. That is not why this capability is generally there.
Most people don't realize that when it come to power supplies, you can have voltage or current, but you can't have both. When the current load jumps massively, internal losses means the voltage drops proportionally. Your amp by your estimation is immensely voltage and current stable. It really can supply masses of current and within reason maintain the voltage - THAT IS A GOOD THING, A VERY GOOD THING.
But, it represent reserve capability for when more common 4 ohm and 8 ohm loads become demanding. It doesn't mean the capability exists to be used in the long term to drive 2 ohm loads.
So, what you have is a massively transient stable amp. When huge music transients occur, you amp doesn't falter in voltage or current. It continues to give the best possible and most stable dynamic range to the music. Again, this is all very good, and the kind of stuff we dream about.
But just because your amp can drive 1 ohm loads, doesn't mean it should. That's all we are saying; keep the impedance up in normal impedance range, and take comfort in knowing you have this massive stability available to you.
Now there are exception, as you rightly point out, an 8 ohm or 4 ohm speaker drops much lower than that. In fact, I've see 8 ohm rated speakers that drop as low as 3.2 ohms. Your amp doesn't have to worry about that because it has plenty of reserve. If you had two of these speakers on each channel, it could drop as a low as 1.6 ohms, which spells trouble for most amps, but not yours. again, that's a good thing.
So, if you were not really intending to make speaker with insanely low impedances on purpose, then ...oops... our mistake. But if you were, then ...oops... your mistake on several fronts.
Again, intensionally making excessively low impedances is like driving you car around all day in low gear. Certainly you can do it, certainly the car can handle it, but really, what is the point.
So, again, it is nice to have that huge stable reserve in your amp; an amp to be envied. But use it to give you reserve and stability at normal loads, not to simply consume power at pointlessly low loads.
Do, you see my point?
Steve/bluewizard
Making speakers with low impedence is easy! Trying to make them with no less than 3 ohms is much tougher, unless you want a speaker box about the size of a refrigerator.
BIG questions for you: Do you want baffle step compensation, and if so, how much? (I always do a full 6 dB) Next question--do you have an idea of how many dB @ 2.83V/m you would like your speakers to crank out? And third question--what would be your target F3 for the bass response?
JJ
BlueWizard said:
Ok got it.
3.5 sounds like a terrible way to go overlapping the sub with the woofers yuck.
The amp cost me in the vicinity of 4k. I can spend a comparable amount on a pair of demonstrator speakers.
Now you are finally giving me what I've been asking for all along. If you had made this crystal clear then we wouldn't be having this problem.
So, you build this outragiously high current and voltage stable amp, and now you want to prove to people that it really can do what you say it can do, so you need a real dynamic speaker load to demonstrate it. If that is right, and you had stated that clearly in the beginning, this would have all been much easier.
So, here's my plan. Build three pair of WWMTM (or similar) speakers and build a switch box so you can select them.
Now, set up your demo with the first set of 4 ohms towers attached.
Everything is cool, every thing sounds good.
Now, flip the switch and the second additional pair of 4 ohm towers come on line -- 2 ohm load per channel.
Still every thing is cool, everything sounds good.
Now flip the next switch, and 2 more pair of identical towers come on line.
4ohms/3 speakers = 1.333 ohms final load.
If you are right, and your design works, everything should be cool and everything should sound good.
I really don't think you want to use a single low impedance load, there's no contrast. I think it is much more impressive for a prospective client to hear each new set of speakers come on line, and hear the amp taking it in stride.
For the most part, each speaker cabinet could be in the 100watts to 200watt power handling range. Since, while you will certainly crank it to high volumes, you will never truly be using full power.
If you do want to demo it at literally full power, get a bunch of heatsinked 50watt 8 ohm power resistors and gang them together into a huge 2 ohm high powered load.
Though admittedly three pair of towers is going to stretch your budget. But in the mean time, you've got some great towers to listen to.
How does that sound?
Steve/bluewizard
Now you are finally giving me what I've been asking for all along. If you had made this crystal clear then we wouldn't be having this problem.
So, you build this outragiously high current and voltage stable amp, and now you want to prove to people that it really can do what you say it can do, so you need a real dynamic speaker load to demonstrate it. If that is right, and you had stated that clearly in the beginning, this would have all been much easier.
So, here's my plan. Build three pair of WWMTM (or similar) speakers and build a switch box so you can select them.
Now, set up your demo with the first set of 4 ohms towers attached.
Everything is cool, every thing sounds good.
Now, flip the switch and the second additional pair of 4 ohm towers come on line -- 2 ohm load per channel.
Still every thing is cool, everything sounds good.
Now flip the next switch, and 2 more pair of identical towers come on line.
4ohms/3 speakers = 1.333 ohms final load.
If you are right, and your design works, everything should be cool and everything should sound good.
I really don't think you want to use a single low impedance load, there's no contrast. I think it is much more impressive for a prospective client to hear each new set of speakers come on line, and hear the amp taking it in stride.
For the most part, each speaker cabinet could be in the 100watts to 200watt power handling range. Since, while you will certainly crank it to high volumes, you will never truly be using full power.
If you do want to demo it at literally full power, get a bunch of heatsinked 50watt 8 ohm power resistors and gang them together into a huge 2 ohm high powered load.
Though admittedly three pair of towers is going to stretch your budget. But in the mean time, you've got some great towers to listen to.
How does that sound?
Steve/bluewizard
BlueWizard said:
Hi Steve.
I agree with and appreciate your comments, however my amp can drive a continuous sinewave into a 1 ohm load (800W) and deliver 1600W into 0.5 ohms at the onset of clipping.
However, that is only 80A!. The amplifier will not shut itself down until the load current exceeds 100A for 10mS and it current limits at 120A.
A nominal 2 ohm load (20A peak) really isn't straining the amp - the 20 parallel pairs of 16A ring emitter bipolar output devices are only passing 1A each!
Cheers,
Glen
Eva-are you suggesting that it is impossible to make a speaker play loud on only 40 volts? Even with multiple drivers and low impedence?
Doing things different is part of the joy of DIY.
JJ
can your wall outlet and heatsinking handle it? Again, I'm not trying to doubt your engineering ability, thats just a lot to ask of any amplifier. Are you using forced air cooling? How about a dedicated 30 amp line?
I tried to have a dedicated 30 amp line installed and was not allowed, apparently by code, to have an electrician install it where and how I wanted.
I tried to have a dedicated 30 amp line installed and was not allowed, apparently by code, to have an electrician install it where and how I wanted.
BlueWizard said:
Steve, this sounds good. Multiple towers would be expensive, but on the plus side a single pair of 200W units would be a lot cheaper than a single monster pair, and better to tackle for my first DIY speaker.
Cheers,
Glen
G.Kleinschmidt said:
Get yourself a pair of second hand VAF i93's or Equinox Jupiters and be done with it. They do appear on the market from time to time, not too far off that mark. Although not 4 ways, they are full range, low-ish impedance (~3 ohm nominal) and will work very well with 40 volt p-p high current amp.
You will never make anything even remotely close for that kind of money.
- Status
- Not open for further replies.