| looneybomber |
First off I am new to the forum, but I am not a complete n00b when it comes to HT stuff, however, I still have a lot to learn. If you do not feel like answering a question directly but instead want to give a URL, I will gladly do more reading! I just can't find the info I am looking for. :( I am probably searching for the wrong stuff but when searching for subwoofer preamp,
My first question.
When using winISD pro, if I input a speakers T/S parameters it keeps giving me some error message telling me none of my parameters are correct and gives me a list of everything I need to correct. How can I get that to stop when I am putting them in exactly right. I will even use other speakers that are currently in the data base as a template to make sure I am putting them in correctly and it still doesn't work.
Question #2
What are some good preamps to use for a subwoofer. What I am wanting to do is use a Crown K2 as my amp, but want at least a 7band EQ and a crossover with a variable frequency to use as my preamp. I do not want to waste money by buying a 31band EQ when I am only going to use the 7 lowest bands.
Question #3
Where can I find box and frequency response info on an Alliance 18, Sound splinter RL-p15 D4 and the Dayton Audio Titan 15. (If I had my winISD working, I'd be fine)
My goal is to produce 115db @ 20hz anechoic. At first I wanted 120db's but after further investigation, realized it would cost too much $$$ and the enclosures would take up too much space in my room.
If you have recommendations on any other sub, or even reasons to stay away from a Crown K2, I am all ears.
Thanks for your help |
|
|
| nunayafb |
| quote: | | When using winISD pro, if I input a speakers T/S parameters it keeps giving me some error message telling me none of my parameters are correct and gives me a list of everything I need to correct. How can I get that to stop when I am putting them in exactly right. |
Unfortunately, you are not inputting everything right. Most of those values are calculated from other values-Qts from Qes and Qms etc...
and since winisd calculates some fields even though you entered data, it compares the two and reports an error if there is so much as roundoff difference.
So what you need to do is read the help file on "entering new driver data" and it should guide you as to what to leave blank and what needs to be filled in. |
|
|
| richie00boy |
Post up the parameters you are using and maybe somebody can see where you are going wrong.
Just to add that 115dB @ 20Hz anechoic is unlikely to happen unless you use probably about 4 drivers. If you aim for maybe 102dB you might not be dissapointed. |
|
|
| looneybomber |
| quote: | Originally posted by nunayafb
Unfortunately, you are not inputting everything right. Most of those values are calculated from other values-Qts from Qes and Qms etc...
and since winisd calculates some fields even though you entered data, it compares the two and reports an error if there is so much as roundoff difference.
So what you need to do is read the help file on "entering new driver data" and it should guide you as to what to leave blank and what needs to be filled in. |
I will probably have to read that whole thing and experiment with what I can omit.
| quote: | Originally posted by richie00boy
Post up the parameters you are using and maybe somebody can see where you are going wrong.
Just to add that 115dB @ 20Hz anechoic is unlikely to happen unless you use probably about 4 drivers. If you aim for maybe 102dB you might not be dissapointed. |
At 102db's I would cry myself to sleep. If you meant 112, I could live with it, but would always crave more.
I was hoping to use 2 drivers to help with the inroom FR and to keep from having so many localized spots of standing waves. Even with 2 Alliance18's or 2 RL-p15 D4's 115db's @ 20hz (anechoic) isn't possible? Another good question, would those be quality drivers for music?
As for posting up the T/S parameters, I was wanting to to a lot of experimenting with different tuning frequencies, Vb ect while trying to get a good EBS. If you still want them, I can go to the websites and get them for you.
So many more questions I could ask. Thanks for the help thus far. |
|
|
| richie00boy |
Well all I can say is that WinISD seems optimistic on predicted SPL levels by upto 10dB in the bass.
2 18 inch drivers in a vented box might see you hit 108dB anechoic.
Your target might be met if you have some usable room gain, but for anechoic I think you have little chance with just two drivers. |
|
|
| looneybomber |
| quote: | Originally posted by richie00boy
Well all I can say is that WinISD seems optimistic on predicted SPL levels by upto 10dB in the bass.
2 18 inch drivers in a vented box might see you hit 108dB anechoic.
Your target might be met if you have some usable room gain, but for anechoic I think you have little chance with just two drivers. |
Looks like I may have to set my sights a little lower.
What about a preamp/eq for the Crown amp, any recommendations? |
|
|
| Collo |
| quote: | Originally posted by richie00boy
Well all I can say is that WinISD seems optimistic on predicted SPL levels by upto 10dB in the bass. |
The WinISD figure is for radiation into "half-space". Since most subs are not mounted flush with a wall, the "full-space" value is closer to what would actually be measured. Subtract 6dB to see what would actually be measured @ 1m. |
|
|
| richie00boy |
Half space is when the radiating plane is cut in half, with a sub this would be the achieved by the floor. Fitting close to a wall would achieve quarter space. If in a corner it would be eighth space.
WinISD is just optimistic. |
|
|
| looneybomber |
| quote: | Originally posted by richie00boy
Half space is when the radiating plane is cut in half, with a sub this would be the achieved by the floor. Fitting close to a wall would achieve quarter space. If in a corner it would be eighth space.
WinISD is just optimistic. |
How many db's of gain to get from each reduction of space, half vs quarter vs eighth?
That looks interesting. 5 variable bands would be awesome. |
|
|
| richie00boy |
| You are effective doubling sound intensity each time you intersect the radiating plane, so you get +6dB sound pressure level each time. |
|
|
| owdi |
| quote: | Originally posted by richie00boy
WinISD is just optimistic. |
I've found the exact opposite. Peak in-room SPL seems to be higher than what WinISD predicts.
114db at 20hz in a HUGE vented enclosure is very realistic with the Alliance 18. Even in a sealed enclosure, this sub could easily manage 120db once it's placed in a room. The quarter wavelength of 20hz is around 14 feet, so you don't even have to place it into a corner for it to behave like it is.
Dan |
|
|
| owdi |
| quote: | | Peak in-room SPL seems to be higher than what WinISD predicts. |
I forgot to add "even after estimating room gain with a Linkwitz transform."
Dan |
|
|
| looneybomber |
| quote: | Originally posted by richie00boy
You are effective doubling sound intensity each time you intersect the radiating plane, so you get +6dB sound pressure level each time. |
Thanks!
| quote: | Originally posted by owdi
I've found the exact opposite. Peak in-room SPL seems to be higher than what WinISD predicts.
114db at 20hz in a HUGE vented enclosure is very realistic with the Alliance 18. Even in a sealed enclosure, this sub could easily manage 120db once it's placed in a room. The quarter wavelength of 20hz is around 14 feet, so you don't even have to place it into a corner for it to behave like it is.
Dan |
But just how musical would that be? |
|
|
| badman |
| quote: | Originally posted by looneybomber
Thanks!
But just how musical would that be? |
How musical? If you're concerned with having 115dB @ 20Hz, music doesn't seem to be on your list of priorities for the sub. There's next to nothing musically at 20Hz.
And a big vented enclosure can sound musical, though I prefer sealed (for a variety of reasons: low Q sealed rolloff better matches room gain, IMO, and and doesn't 'unload' below Fb) |
|
|
| richie00boy |
| Not only that but one mans musical is another mans monstrosity. |
|
|
| looneybomber |
| quote: | Originally posted by badman
How musical? If you're concerned with having 115dB @ 20Hz, music doesn't seem to be on your list of priorities for the sub. There's next to nothing musically at 20Hz.
And a big vented enclosure can sound musical, though I prefer sealed (for a variety of reasons: low Q sealed rolloff better matches room gain, IMO, and and doesn't 'unload' below Fb) |
But it wouldn't solely play from 15-25hz, it would be my main source of bass for my classical, jazz, rock and progressive house music that I typically listen to.
| quote: | Originally posted by richie00boy
Not only that but one mans musical is another mans monstrosity. |
Valid point.
After reading the "help" file on WinISD, I can get the values inputed correctly and will be able to tell what the group delay is. How can I figure out what its' transient response will be though? |
|
|
| richie00boy |
| Transient response is not really relevant as it will never play any transients, being a sub-woofer ;) however, from a purely theoretical point of view, WinISD has a graph of transient/impulse response I think. |
|
|
| looneybomber |
| quote: | Originally posted by richie00boy
Transient response is not really relevant as it will never play any transients, being a sub-woofer ;) however, from a purely theoretical point of view, WinISD has a graph of transient/impulse response I think. |
I am confused. I thought transient response was the time it took the speaker to go from one position (sitting still) to another (moving). It's that transition that I thought was measured and called "transient response". If that is not transient response, what is that called?
Also, do ported boxes increase that time from not moving to moving, or is that more controlled by the amp and its damping factor? |
|
|
| v-bro |
http://www.acoustisoft.com/basics.html
This may be of interest to you?
Transient response in this case is measured by a microphone, on a certain distance in a specific room.
You give the speaker a pulse with virtually no wavelength(only amplitude) and "hear" with the microphone how long it takes to fade away(go back to zero) and how many times it crosses zero and in what way..
Don't know exactly what a standardised setup for such a measurement
should look like(supposedly at 1 meter and a certain voltage?)
The box and the room will give reflections that will influence the transient response, the amplifier dampening factor of an amp is of little influence to this... correct me if I'm wrong (think any modern amp will not produce dramatical differences in transient response graphs of a loudspeaker).
With simulation software you can maybe make it possible for you to compare things a little bit, but It'll never predict the real life situation..
But I mostly find speakers placed nearer to walls "BOOM" and would
prefer a more "decoupled" aproach, when it is preferred to sound a little more "exact". |
|
|
| nunayafb |
| quote: | | The box and the room will give reflections that will influence the transient response |
actually the room will provide "transient reflections" and dont actually affect the transient response of the driver. These show up as the secondary and tertiary(and so on) peaks on the impulse response graph.
Slam and boom and fast and slow are all laymen terms used to describe the relative levels of upper bass and lower bass, in addition to any uncontrolled cone motion(or lack of) due to the enclosure tuning. |
|
|
| v-bro |
quote:
""The box and the room will give reflections that will influence the transient response"
actually the room will provide "transient reflections" and dont actually affect the transient response of the driver".
True, filtering can also affect the bass(or indeed actually mid-bass) to be experienced "slow" or "fast"(or however you want to call this experience) when things were not done right.
In my previous post I was solely writing about the experienced sound in a particular room. Maybe I should have added that in a transient test graph the line would have shown several (reflected)peaks...
Waves reflected in the room I think will only cloud the experience in relation to the source signal. As it will reach your ears in another timing, and interfere with directly radiated waves and other reflected waves, combining them to amplify or dip and may create the "boom experience" on some places. Most of all when standing waves occur amplified low frequencies can occur too.
Though wanting to produce 120 db's it might help indeed....;-)
Bass is a tough sucker though, goes through flesh and bone.
Where mid and high frequencies benefit from a good dosage of reflections in order to create a "pseudo" realistic reproduction of truth, bass is better off outside or in a dead room. Everything in a room will remain a compromise to reality. Getting it as right as possible in a room might involve eq-ing or having everything placed and designed specifically for a certain room when it comes to experiencing a "flat" frequency response. Experiment experiment experiment.....
Afterall we're not dealing with poor enclosure tuning cone slowing matter here I think, we're dealing with a person listening in a room in the future who'se speakers are yet to be built...
And you can unleash a lot of theory on the design, accurate sound does not give everybody more listening pleasure, listening and experimenting right mostly does the trick better. I've had greater breaktroughs by "hunches" than with my calculator..
I like richie00boy's comment:"one mans musical is another mans monstrosity".
No offense to anybody, just trying to help determine what important influences are in acoustics(the whole picture)and what aren't... |
|
|
| looneybomber |
Lot of good information here guys. If you have any more links to some good reads, I will gladly pay my dues just like everyone else.
| quote: | Originally posted by v-bro
quote:
Afterall we're not dealing with poor enclosure tuning cone slowing matter here I think, we're dealing with a person listening in a room in the future who'se speakers are yet to be built...
And you can unleash a lot of theory on the design, accurate sound does not give everybody more listening pleasure, listening and experimenting right mostly does the trick better. I've had greater breaktroughs by "hunches" than with my calculator.. |
Very true. This is my next project that I am planning/thinking about. Currently I am getting ready to build my first sub, just for practice, then will give it to my sister for christmas.
I have an old JBL GT1200 sitting around NIB, so I figured, hey, let's do something with it. I ordered up everything I needed including the 300w BASH amp from Partsexpress. It was on sale!
After tinkering with WinISD for a while, I've come up with this. It's not the best and buying a different speaker would work a ton better, but I am just using what I have sitting around.
The db number in the upper right corner is with a 250w signal. It's roughly 9db down from it's peak. Also I added a 2nd order LP x-over @ 110hz. I could get a better response with a bigger box, but my group delay would go crazy and I would run out of excursion at 35hz with just 70w of power. So 3.2ft^3 is the largest I can go. |
|
|
| richie00boy |
| quote: | Originally posted by looneybomber
I am confused. I thought transient response was the time it took the speaker to go from one position (sitting still) to another (moving). It's that transition that I thought was measured and called "transient response". If that is not transient response, what is that called?
Also, do ported boxes increase that time from not moving to moving, or is that more controlled by the amp and its damping factor? |
Yes that's what transient response is, but think about it - a sub is fed from a low-pass filter so it will never see a fast changing signal.
A vented box generally has more overshoot of transients/settling time.
Just off to try and find kelticwizard's excellent graphs...
edit: here it is for sealed
http://www.diyaudio.com/forums/atta...tamp=1069945507
and now vented
http://www.diyaudio.com/forums/atta...tamp=1081342894 |
|
|
| v-bro |
Exellent graphs indeed,
but to keep the birds eye view, some closed speaker designs have poorer experienced accuracy than some vented ones.
Always keep in mind the whole picture, for some vented speaker designs use specifically adapted drivers and some closed designs as well. Though most drivers are designed for several type of enclosures and give fine results in both, this is mostly done for commercial reasons. It would be always preferable to design a driver for the housing it's intended to be used in.
Read a good speaker design book on "thiele and small" parameters and you will probably find a table showing the relation between mechanical, electrical and acoustical physical quantities.
You may find a translation in your language of H.H. Klinger's "Lautsprecher-baubuch fur Hifi-Amateure und Musikfreunde"1989).
I bought it(dutch translation) in 1990 and advise anyone who wants to start understanding the deeper knowledge of loudspeakers to read it. In the book you find a very clear table like I just mentioned(with symols). Many designs use combinations of either mechanical, electrical or acoustical quantities that can only be calculated or tweaked right knowing the schematic in which all factors are incorporated.
Example: a bandpass subwoofer acoustically filters (like an inductance) mid and high frequencies (by air mass) and may save on electrical filtering.
Merely calculating this would have to incorporate so many factors that it would very likely not give accurate results in the end.
The value in mh resulting from this acoustic filtering can be better predicted than calculated I think. By knowing the slope will become steeper. Building an electric filter out of the enclosure and unwind/wind up the coil can help you get it right. And knowing larger values lower x-over frequencies and smaller values make them higher both for high and lowpass filters(for beginners).
Just using calculation software many times won't lead to good results unless you know what you're doing.
Once you know what you're doing solving problems is easy!
Or problems won't even occur...
My brother bought a new house, placed his speakers on the floor, straight in a rectangular room. He called me asking to help him out because it sounded awfull. I rotated his whole "relax corner" 45 degrees, put the speakers on a heavy tile on 3 "speaker toes" and made dramatical improvements(less standing waves for instance, still have to hang some carpets and stuff).
3 toes cant waver btw(hint).
That's why experimenting is so important, but always investigate your findings unless you're sure about them...
And see things in prespective, my brother would have noticed less if I had shrunk the content of his speakers 5 liters (from 46) or if I had put Van den Lul cables in there(dutch joke, I know it should be "Van den Hul")....
Your ears shall seperate the ******** from the truth.
Also never go for first impressions (the best materials used in loudspeakers often take longer periods to "break in". (therefore they live long)).
Some will "grow" better, some deteriorate fast.
Foam woofer cone edges for instance "sound" great almost from the start, rubber ones though live much longer and-if properly manufactured sound good too.
I find paper woofer cones mostly "sound" best, and often handle heat better than poly-whatever cones, dampen better than metal cones(domes!), but when made light they tend to fold quicker (fail mechanically).
Have to admit though that I am still learning every day and will keep making mistakes(I can learn from). Nobody will probably ever invent the perfect loudspeaker.
Key is to be satisfied at a certain acceptable degree of compromise and enjoy and listen to music..... |
|
|
| looneybomber |
| quote: | Originally posted by richie00boy
Yes that's what transient response is, but think about it - a sub is fed from a low-pass filter so it will never see a fast changing signal. |
How fast is "fast" when refering to "see a fast changing signal"?
| quote: | Originally posted by TwisterZ
Hello All
Here is another EQ solution that you might like even better.
http://www.behringer.com/DSP1124P/index.cfm?lang=ENG
12 band parametric per channel and $129 retail
Also check out the EP1500 & EP2500 power amps at the above link. |
Just minutes ago I purchased this.
Behringer DEQ2496 mostly because I can plug a mic into and annalyze my frequency response. From what I've been reading, that FBD you posted would work good too. |
|
|
| v-bro |
after being satisfied with your eq-settings you may consider building a better filter(in which signal will make shorter routes) that gives the same result.
Eq's are built with ap-amps and aren't exactly audiophile devices, this is mostly not very audible with bass-frequencies though...
Just do some a/b testing.... |
|
|
| TwisterZ |
V-Bro
I went to school with a guy that hooked up a 2" speaker from a transistor radio to a fender bass guitar amp. When he hit the low string on his bass we found the Voicecoil stuck in the wall 30' away. LOL |
|
|
| TwisterZ |
V-Bro is right about the low end amps in EQ's
It is best to use an EQ to cut undesirable frequences and leave the boosting to your amplifier. |
|
|
| v-bro |
twisterz:
30 ft away eh? maybe I should change my footnote..;) |
|
|
| v-bro |
Eq-ing=filtering for those who didn't know, look at this:
http://www.diyaudio.com/forums/showthread/t-56701.html
This will work between any line level connection and becomes very interesting when using multi-amping.
And a lot less complex than active filtering! (as used in equalizers)
With good quality components I find them sound great.(and produce far less noise, when well shielded(surrounded by metal connected to ground) offcourse).
Google on "pllxo"....
Years ago I would "create" a bassguitar amp from a "350 watt" polypropene woofer and a "40 watt" amplifier to be able to practise with my schoolband (me on drums). Also at the first string hit the cone landed on the floor meters away smoking...... |
|
|
| v-bro |
To maybe make things a little easier to understand from the start:
Resistor in series capacitor parallel cuts the highs, capacitor in series resistor parallel cuts the lows(same values, same x-over frequency).
Increasing values lowers x-over freq.
Decreasing values rises x-over freq.
With only two components per channel you can create a 6db's decreasing per octave filter!
With only four components per channel you can create a 12db's decreasing per octave filter!
Choise of component quality in order of preference:
Capacitor:"peanutbutter oil soaked paper-foil caps" (joke) "MKP","MKC","MKT".
Resistor:at least "metalfilm" |
|
|
| looneybomber |
| That is good information and I will use that later when I build my 2way bookshelfs, but how does that relate to an eq with an RTA and a bunch of other functions? Right now I can still use it to analyze my FR of my current setup. Later I can also use it as a preamp for my subwoofer(s) I will build. |
|
|
| v-bro |
a spectrum analizer can be useful to analize what you can later create in a less complex way (if desired, as for the lower freqs the quality of the behringer will do just fine, and makes the setup more easy).
Unless you want to keep shifting knobs (I don't, once I'm satisfied I never touch them again)..... |
|
|
| looneybomber |
You can make a passive crossover amplify or cut a signal by 2-3db at a narrow frequency? I thought they could only be high or low pass and not have any other effect on a speaker level signal.
Question four. When bracing a box, will 1" dowel rods work when glued & screwed to parallel walls or do I need to stick with sheets of MDF with window panes cut out? I was hoping to use a 1" dowel rod every 6'ish inches.
Question five. Would it be better to use 1 large port, or 2 smaller ports (with equal area as the single large port)? If I use 2 ports, do they both have to be the same length tuned for 24hz, or can I have one longer tuned at 22hz and the other shorter tuned to 26hz? |
|
|
| v-bro |
Virtually you could, making it function as a bandpass or notch type of filter, only you might have to tolerate very high insertion losses (especially when using steep slopes)..
I think 1" dowel rods won't work as well as an MDF cutout board, MDF is better (dampening) material for cabinets, and glueing them stretching from one side to another helps more than parallelling walls with a 1" rod. Though I'd say it's better than nothing...
One port would be better I recon. Two ports with different tuned freqs in one cabinet I think will more or less work as one port cut off at an angle. Never seen manufacturers do that. I'm afraid the res.freq. wont be properly supressed. |
|
|
| v-bro |
| Make it a port from two MDF panels ending up in the baffle will help you both ways. |
|
|
| looneybomber |
| Well I have a problem with that. Using a 5" port (one flanged edge), it will have to be 25" long and the box will only be 27" long. If I make the port 1.25" x 14" (the width of the box) it will only have to be 22" long, but that will completely separate the top portion of the box from the bottom portion, which seems like a bad idea. I'd draw a picture, but all I can use is Paintbrush... |
|
|
| v-bro |
| maybe you can glue it in there under an angle... |
|
|
| looneybomber |
Ok new question. How do you regulate a speakers impedance through its frequency range? For example, this sub I threw on WinISD, peaks at 215ohms at 36.3hz.
What does that do for sound?
What does it do to the amp? |
|
|
| v-bro |
It peaks on the resonance frequency, this is the frequency on which the cone takes the least power to move, still the frequency spectrum should stay even, because it only gets very little power on that freq.
Immagine it would stay an even load of for instance 8 ohms, the cone would simply pop out the basket when you play the resonance freq.
As the amp produces less power at higher impedance the combination of the dynamic loudspeaker and amplifier works! Providing the loudspeaker had been designed well.
The only thing it will do to the amp is giving it an easyer ride..:angel:
The effect of filter components between the amp and the drivers are much more dramatic. If you really want to screw up the soundquality use impedance correction networks (RLC network) in such a filter, or amplifier decoupling stages. The amp can "feel" a much greater load (and become hot!) from some type of filters.
But enough on that, explained before that I would advise to filter on the inputs of the power amp....
Input the driver impedance -to calculate the filter- the average impedance the driver shows round the crossover freq. and you'll be a lot more accurate on the calculations...:cool: ( just estimate this when reading the driver's impedance curve, most dynamic loudspeakers(drivers) have one or two peaks in the curve round the res.freq. and the impedance will mostly rise up to the higher freqs).
Dunno if you can read through this, hope I've been helpfull.... |
|
|
| richie00boy |
| A passive crossover for a sub is just not worth considering, for the reasons you are discovering. But wait until you see the size (and price) of the inductors required. |
|
|
| looneybomber |
| quote: | Originally posted by v-bro
Dunno if you can read through this, hope I've been helpfull.... |
It helped a lot, thanks. That spike to 215ohms made me nervous because I knew I would lose a lot of power, but since it wont effect my FR and just gives the amp a break, I am all for it!
| quote: | Originally posted by richie00boy
A passive crossover for a sub is just not worth considering, for the reasons you are discovering. But wait until you see the size (and price) of the inductors required. |
Yeah, I'm not considering anything passive, because like you said, the components would cost way too much. I bought a Behringer DEQ2496 that I will use as my sub preamp/eq/xover.
Boosting with an EQ increases your group delay, correct? Cutting, however, should be just fine? |
|
|
| v-bro |
Passive components are only expensive when they have to filter the hurricane-size power amp-out signal(to the speakers).
I am currently listening to a 4.60 US dollar (subwoofer) filter that works just fine!:wave2:
Many people misunderstand that the values needed to filter line-signal (very low current/voltage signal this is!) are so small that even when you filter very low freqs the size and price of even the most exotic is still very small...
Even the very best of filters like the marchand xm46 (24db/octave) cost only 65 dollars per channel for a kit. (no matter the XO freq.)
http://www.marchandelec.com/ftp/xm46man.pdf
http://www.marchandelec.com/xm46.html (look at the specs!!)
They also custom build them!(but this one can be built for under 20 dollars for sure)....
Funny thing is, I asked people on this forum whether they would know the schematic for this one, nobody came up with this link. :confused:
I just found it myself (strangely enough it must've always been there....).
I can tell you my neighbours have heard me cheer for sure! :cheers:
The filter appears to be more basic than I even thought, you can make any type of filter this way for very little buck.
And as this one is LC instead of RC (I'm currently using with 10db insertion loss) the losses will probably be very minimal....
Gonna build this one real soon!
Keep ya'll posted.. |
|
|
|
