| AudioGeek |
OK,
I want to design something with those Jordan drivers. Simple design and crossover a must.
Either:
a fullrange system with just the JX92s
or the JX53 with the JX125.
Haven't decided on dipole, IB etc.... looking for input on best way for these.
I am also looking to increase sensitivity by using multiple drivers.
Also may couple them in a push-push configuration to cancel distortion..
HELP!!! |
|
|
| ultrachrome |
I assume this is your first speaker project? I'd start with the Jim Griffin design, full range in a ported cabinet with baffle step correction. These sound really good...after much break-in. Simple design, easy project.
Then if you really want to design something yourself, you at least have a reference. Doing a google search, you can find several enclosure designs being used with this driver.
Do you have a table saw and a router? |
|
|
| AudioGeek |
Probably go with 2 JX92s per channel.
One run full range.
Other run enough to account for baffle step.
Will look into ported design... tuning is an issue yet. |
|
|
| planet10 |
A TL design i did for the JX125 (untried). It would have to be tweaked a bit for push-push.
dave |
|
|
| AudioGeek |
ported design.
1 cu foot enclosure tuned at 45Hz.
12in wide, 20in high and 7in deep or so.
2 drivers per channel. one run full, other run to baffle step corner frequency.
port on front. slot loaded. 12in wide, 1/2 inch high and 2 3/4 in deep.
mach number is 0.09
Will it matter if I put the port in between the drivers, or just at the bottom?
Thanks!!!! |
|
|
| planet10 |
| quote: | Originally posted by AudioGeek
2 drivers per channel. one run full, other run to baffle step corner frequency. |
If you are going to do that, then the best place for the 0.5 driver, by far, is on the back of the box, mounted push-push.
dave |
|
|
| AudioGeek |
can you explain to me the advantages of mounting on the back for a 0.5 driver???
Thanks! |
|
|
| planet10 |
| quote: | Originally posted by AudioGeek
can you explain to me the advantages of mounting on the back for a 0.5 driver???
|
As you roll off the 0.5 driver its phase rotates thru 90 degrees. As it rotates it gets more & more out of phase with the driver that is not rolled off.
When you mount a 2nd driver on the back of the box it will perfectly compensate for baffle step loas. As long as the 0.5 rolloff is above the bafflestep frequency, the phase roll is hidden in the shadow of the box (ie this part of the sound only radiates backwards -- by the time it gets back off the back walls, it is so delayed and so mixed with the room, that the phase roll has little if any effect).
Further, by mounting on the back, it is convienient to mount the drivers back-to-back (push-push) and, if rigidly coupled, actuvely cancels driver vibration and improves downward dynamic range.
of course you can also choose to not roll off the back driver at all and have a bi-pole (works well with FRs)
dave |
|
|
| navin |
| heard the JX92 has very limited HF response. in that case a 53/125 might be a better bet. Jordan recomends 4 x 53s line source. how does he handle comb? anyone know? |
|
|
| planet10 |
| quote: | Originally posted by navin
heard the JX92 has very limited HF response. in that case a 53/125 might be a better bet. |
Certainly go lower & higher... XO 1st order anywhere between 500 & 4k -- i've heard series at 3-4k works pretty well.
| quote: |
Jordan recomends 4 x 53s line source. how does he handle comb? anyone know? |
Would be very similar to the 7th viel.
dave |
|
|
| navin |
dave,
this might be a culmination of a lot of ideas.
think jx53+125.
jx 53 above 300Hz, jx 125 below 300hz. 6db PLLXO.
jx 53 driven by EL83 PPP (parallel push pull).
jx 125 driven by 30W SS amp (made in India group project)
use amplifier gain to compensate for BSC
next level.
2-4 x jx 53 in line array open baffle.
2 x jx125 in push push box
U see any problems? Any better ideas? A still simpler design? Fostex a better bet?
Sorry for the personal post but I have great regard for your opinion. |
|
|
| Colin |
I use a JX53/125 cobination in a sealed enclosure and they sound very good. The drivers are very natural and dynamic. Crossover is the recommended 500Hz. After first hearing the Jordans, I gave up bothering with other drivers.
My JX92s are still awaiting enclosures to see how the HF response matches (but the Pass Labs J-Low article claims they're fine into the HF). I have briefly heard the JX92 in a sealed enclosure and I thought it sounded fine (although I wasn't comparing it to anything at the time).
I've also heard Ted Jordan's 4xJX53 linear array system. This has an amazing ability to provide a rock steady image whilst you wander about the room. He uses the array without a crossover, so it operates to around 100Hz and the bass sytem picks up from there. I did notice a slight loss of HF sparkle when I stood up out of the vertical 'beam'. Fine for the favourite armchair though...
Hope this helps.
Colin |
|
|
| navin |
| i am told that tilting the array a bit reduces the loss of HF sparkle. My fear however is that the comb filtering might introduce other distortions not only a loss of hf sparkle. |
|
|
| AudioGeek |
so do I have room boundary issues putting that 0.5 driver on the back???
My enclosure is about a foot wide. What value of inductor should I use for my 0.5 driver? also, NO resistor in parallel with inductor correct? As we want to totally filter the HF on the rear?
Finally, Can I supplement the drivers HF response with a super-tweeter?
i.e. a piezoelectric or something? And where would the best place to mount it?
Thanks for all the info!!
p.s. vent on front or back?? I can get -3db at 45Hz with WINISD. Should I make the vent dimensions longer to drop off the response sooner to account for room gain??? |
|
|
| Colin |
Other than the HF loss (and that only happens when you're above or below the beam of the array), I don't think there is anything else to criticise about the JX53 array. Bear in mind they are designed to be quite directional and have been designed from the start with array applications.
Regarding the JX92, adding a tweeter may extend the HF but will probably muck up the stereo imaging to some extent (phase shifts, non-time aligned tweeter). The attraction of this driver is being able to use it as a pure point source.
Have a look at some of the Konus Essence reviews for comments on the sound of the driver used on its own.
Colin |
|
|
| planet10 |
| quote: | Originally posted by navin
this might be a culmination of a lot of ideas.
think jx53+125.
jx 53 above 300Hz, jx 125 below 300hz. 6db PLLXO.
jx 53 driven by EL83 PPP (parallel push pull).
jx 125 driven by 30W SS amp (made in India group project)
use amplifier gain to compensate for BSC
next level.
2-4 x jx 53 in line array open baffle.
2 x jx125 in push push box
|
EL83? Do you mean EL84 -- i've a couple Class A PP in the works.
I think either of these would be quite stunning, 300 Hz 1st order XO for a single JX53 might limit dynamics somewhat.
| quote: | Originally posted by AudioGeek
so do I have room boundary issues putting that 0.5 driver on the back??? |
A little bit, the speaker can't be right up against a wall (hardly ever a good idea anyway). With it used 0.5 it is less of an issue than run FR.
dave |
|
|
| AudioGeek |
so....
should I run the rear driver full range or roll of the HF????
oooorrrr.......
just thought of this... how about a notch filter?? keep the low end to account for the baffle step, roll of say 3K-10K and have 10K and up full.
That way I can keep them a little closer to the wall, and also have a bit of top end sparkle with the diffusion of the rear off of the walls.
How about it??? thoughts???
Thanks!!!! |
|
|
| planet10 |
| quote: | Originally posted by AudioGeek
should I run the rear driver full range or roll of the HF????
just thought of this... how about a notch filter?? keep the low end to account for the baffle step, roll of say 3K-10K and have 10K and up full.
That way I can keep them a little closer to the wall, and also have a bit of top end sparkle with the diffusion of the rear off of the walls. |
Try it all 3-ways... it will be a room dependant thing... my room loves bipoles (and dipoles)... i am using a broad "notch" on my curretn $20 specials for both BSC and a little more top (8" FRs), i don't see why something more radical couldn't work (as long as it doesnt cause an impedance problem with the amp)
dave |
|
|
| ozynigma |
| quote: | Originally posted by planet10
As you roll off the 0.5 driver its phase rotates thru 90 degrees. As it rotates it gets more & more out of phase with the driver that is not rolled off.
When you mount a 2nd driver on the back of the box it will perfectly compensate for baffle step loas. As long as the 0.5 rolloff is above the bafflestep frequency, the phase roll is hidden in the shadow of the box (ie this part of the sound only radiates backwards -- by the time it gets back off the back walls, it is so delayed and so mixed with the room, that the phase roll has little if any effect).
Further, by mounting on the back, it is convienient to mount the drivers back-to-back (push-push) and, if rigidly coupled, actuvely cancels driver vibration and improves downward dynamic range.
of course you can also choose to not roll off the back driver at all and have a bi-pole (works well with FRs)
dave |
I am working on some multimedia satellites with 2 x 4" drivers and a tweeter. Front baffle will be slanted back 15 degrees.
I was going to go 2.5 way with both drivers at the front in an MTM configuration and roll the bottom driver off for BSC.
For a rear facing driver how important is symmetry with the front driver?
If I rearface the second driver will it matter that the rear baffle is vertical and the front is slanted?
Does the rear facing driver have to be at the same height as the front driver?
Or what if I slanted the rear baffle mirror image to make a triangular bi - pole configuration?
Howard |
|
|
| AudioGeek |
so how do I do a broad and deep notch filter?
My baffle is about 12in wide, so probably start at 1Khz -> 10kHz for the notch.
4-6db down or even more.
Thanks!! |
|
|
| planet10 |
| quote: | Originally posted by AudioGeek
so how do I do a broad and deep notch filter? |
Essentially a variation on the BSC filter... check Martin King's site
| quote: |
My baffle is about 12in wide, so probably start at 1Khz -> 10kHz for the notch.
|
12" is an estimated 380 Hz BS -3dB.
dave |
|
|
| AudioGeek |
essentially use the 380Hz baffle step....
probably 8 ohms or more paralleled with a 1.75mH choke.
then where does the cap go?? across the choke and resistor?
for a 10Khz filter... what value should I use?
please forgive my basic questions.
Thanks!!!! |
|
|
| AudioGeek |
I want to build a HF diffuser on the back.... something to reflect the HF sound a bit equally everywhere.
How would I do this???
Something like this:
|< shape on the back? with the point close to the center of the driver and tapered out to the edge of the driver.
hope this makes sense. |
|
|
| navin |
| quote: | Originally posted by planet10
EL83? Do you mean EL84 -- i've a couple Class A PP in the works....I think either of these would be quite stunning, 300 Hz 1st order XO for a single JX53 might limit dynamics somewhat.
|
a single jx53 with a single jx125 will represent a point source more closely than a jx125 PP mated to a jx53 4-in-line. however the jx125PP + jx53 4-in-line would have considerably more SPL capability.
the reason i chose 300Hz is because most music has 50% of it's energy above 300hz and 50% below. Also I wanted as much of the vocal range to be managed by one driver. However maybe a XO at about 400hz might also suit me since my proposed baffle would be about 8.25" wide (210mm) and have a "lip" of about 1" (25mm) to simulate a baffle of about 10.25" (260mm).
the -3db step for baffle step would be about 440Hz.
BTW are the JX53 and JX125 that good. what about the Fostex FE83 and FW168N? Should that combo or any other also be considered? |
|
|
| planet10 |
| quote: | Originally posted by navin
BTW are the JX53 and JX125 that good. what about the Fostex FE83 and FW168N? Should that combo or any other also be considered? |
I haven't unleased my JX150s yet, so don't talk from personal experience, but i have reports that even FR the JX125 & 150 will outdo a JX92 -- no mean feat.
The FW168HP would be very interesting... with something smaller than the FE83. I don't have an opinion on the FW168N, but it looks good...
dave |
|
|
| navin |
| quote: | Originally posted by planet10
I haven't unleased my JX150s yet, so don't talk from personal experience, but i have reports that even FR the JX125 & 150 will outdo a JX92 -- no mean feat.The FW168HP would be very interesting... with something smaller than the FE83. I don't have an opinion on the FW168N, but it looks good...
dave |
i could not access any data on the 168HP from the fostex site. i guess the question is....
which combo is better...
Jx 53 with a JX125
or Fe83 (does fostex make a widerange that is smaller?) with a 168HP?
if either the jx53 of fe83 requires some hf help (which i doubt given that there are enough out there that are happy with a jx 92's hf) does adding a rear facing piezo make any sense. |
|
|
| Nardis |
Although the mid is great, I found the top end of the JX92 lacked a bit of openness, especially on orchestral strings & cymbals.
I am now using them with Fountek JP3 ribbons crossed over 1st order at about 6K and the result is much better. The Aurum Cantus ribbons/JX92 combination I've heard sounds better still - but at a price.
There was an interesting contribution on the JX92 impedance from member 6010 a couple of months back and I'm going to try his Zobel and redo the xover. |
|
|
| Colin |
Adding an additional HF to either the JX53 or the JX92 is a personal thing. I find the HF of the JX53 is very good and the lack of additional drivers and crossovers over 500Hz makes for a very natural, coherent sound, more like an ESL than a traditional moving coil speaker and totally removed from the sound of a dome tweeter. It's worth getting the 53 and 125/150 drivers as close as possible, though, as the first order crossover recommended by Ted Jordan means there is a considerable overlap.
An earlier version of the JX125 was used in a Townsend design, crossed over to a ribbon tweeter at around 5kHz. It got good comments on the stereo imagery though I suspect that was mostly the contribution of running the JX125 so high. As an experiment, I tried running the JX125 full range but things got a bit hard at the top end (they aren't really designed as a full range).
Re Jordans vs Fostex - there are comments on the sound of these at the zHorn site. A recent posting by the designer on the Konus site says he prefers the sound of the JX92 in the Essence-style enclosure.
Regarding using two JX92s in one enclosure, the VTL on the Jordan website would make a good starting point for this. The easiest way to roll off the HF of the rear driver (assuming you need to - it would depend on room and reflections) would be to put some sound absorbent material over the driver and let it run full range (no crossover to worry about). Another route would be to mount the second driver on the top, firing upwards. There have been a few designs using this approach and reviews seem split on whether it results in a more natural soundstage or simply messes up the imaging.
Alternatively, the Essence/ VTL design using one driver is supposed to work quite well against a wall.
Colin |
|
|
| planet10 |
| quote: | Originally posted by navin
i could not access any data on the 168HP from the fostex site. i guess the question is.... |
try Madisound
| quote: |
which combo is better... |
probably both.... i'd expect them both to sound really good, but quite different.
dave |
|
|
| navin |
| quote: | Originally posted by Colin
...I find the HF of the JX53 is very good...It's worth getting the 53 and 125/150 drivers as close as possible....I tried running the JX125 full range but things got a bit hard at the top end (they aren't really designed as a full range).
Re Jordans vs Fostex - there are comments on the sound of these at the zHorn site.
Alternatively, the Essence/ VTL design using one driver is supposed to work quite well against a wall.
Colin |
I would love to see links to the zHorn and Essence sites. |
|
|
| pooge |
I've been considering using a Jordan line array with 5 drivers instead of 4, in an MTM arrangement. The top and bottom pairs would be connected the same as a 4 driver line array, and crossed over at high frequencies to a single driver in the middle of the array. I thought that this would help increase the power response at high frequencies, and get some sparkle back, while still keeping the imaging qualities of the line array. Has anyone tried this?
I figured that if I use 5 drivers, I could also experiment with connecting them up in an alternative array I read about. I'll be damned if I can remember the name of the array or the author of the article. I'll have to spend some time digging it up. Basically, the 5 driver array was connected up in various phase combinations which, in sum total, maintained a fairly constant dispersion pattern throughout its band. I think it did so vertically, also. This is the problem with a line array. At high frequencies, the vertical pattern collapses. I don't know about recent history, but in the distant past, when Transendental Audio distributed Jordans in the late seventies, they recommended rolling off a pair of in the array at high frequencies. I figured that using 5, and having a single Jordan at high frequencies, would be better, eliminating combing as well as increasing the power output in the upper treble range. |
|
|
| planet10 |
| quote: | Originally posted by pooge
The top and bottom pairs would be connected the same as a 4 driver line array, and crossed over at high frequencies to a single driver in the middle of the array. I thought that this would help increase |
I've got some drivers i'm doing this with as well (not Jordans). Wiring correctly it allows for a very consistent impedance. XO point should be near the baffle step frequency. Bi-amping here makes a lot of sense too (1st order PLLXO would work well.
dave |
|
|
| pooge |
| I don't think the baffle step comes into play, here (with the Jordans). XO would be pretty high. to be determined by the spacing between the upper and lower pair. I think I would use a first order series XO here. The impedances should be the same on the 4 drivers series/parallel combo and the single tweeter. They should blend perfectly, being the same drivers. |
|
|
| planet10 |
| quote: | Originally posted by pooge
I don't think the baffle step comes into play, here (with the Jordans). XO would be pretty high |
Of course baffle step would come into play, determined by the width of the cabinet.
| quote: | | I think I would use a first order series XO here. The impedances should be the same on the 4 drivers series/parallel combo and the single tweeter. They should blend perfectly, being the same drivers. |
For a passive XO, i'd try series 1st too... "identical drivers" makes getting it work a lot easier. You need to use the baffle step to your advantage to compensate for the xtra 6 dB of sensitivity of the series/parallel set.
dave |
|
|
| pooge |
Oh, I got you now. I wasn't thinking of the sensitivity situation. However, regarding the baffle step, I think I would be crossing over to the tweeter well above the baffle step frequency to do what you suggest. I could pad differences in sensitivity. I want to get the 4-driver array well out of the power range to keep the single tweeter from overload.
One of the reasons I'm contemplating this arrangement is because I have drivers from different eras. My older ones are not as good at the top end as the newer ones. I though this would be a good way to use them all. They might not be identical at crossover, but would likely sound more alike than using other drivers. Using these in the alternative 5 driver circuit I mentioned may not be as idea as 5 identical drivers, though. |
|
|
| pooge |
| quote: | Originally posted by planet10
You need to use the baffle step to your advantage to compensate for the xtra 6 dB of sensitivity of the series/parallel set.
|
Are you sure about this? The series/parallel combination should make for the same sensitivity as a single driver, as they would be the same 8 ohm load. The mutual coupling at low frequencies might raise sensitivity there. Is this what you are talking about, vis-a-vis the baffle step? I'm rusty on the sensitivity of plural drivers, though. I'll have to dig up my info on this. |
|
|
| planet10 |
| quote: | Originally posted by pooge
I think I would be crossing over to the tweeter well above the baffle step frequency to do what you suggest. I could pad differences in sensitivity. I want to get the 4-driver array well out of the power range to keep the single tweeter from overload. |
The XO at the BS should be between the 4 drivers & the single driver... you don't want to be padding your "bass" drivers.
| quote: | Originally posted by pooge
Are you sure about this? The series/parallel combination should make for the same sensitivity as a single driver, as they would be the same 8 ohm load. |
everytime you double the number of drivers you gain 3 dB (ignoring differences, if any, of amplifier output into different impedances) so 4 drivers is 6 dB.
The confusion arises in that many, when talking about multiple driver sensitivity, will lump in an extra 3 dB for the (theroectical) increase in amp output (in actual fact this is rarely the case, and certainly not if you talk tube amps).
dave |
|
|
| AudioGeek |
anyone have any comments about my rear diffuser?????
can I build a mechanical device that will block midrange frequency but pass anything 400Hz and below and 10Khz and above??
Thanks!!@! |
|
|
| navin |
| quote: | Originally posted by planet10
I've got some drivers i'm doing this with as well (not Jordans). Wiring correctly it allows for a very consistent impedance. XO point should be near the baffle step frequency. Bi-amping here makes a lot of sense too (1st order PLLXO would work well.dave |
I loved the look of the essense. the box was a bit too wide but otherwise i like the slim look. I'd have to show it to my wife to get her approval though.
what i was thinking about was a similar looking box for the 125 with the front bafle extending on the top so teh JX53 could be open baffle.
Looking at the specs of the 125 however i noticed that it requires over 20 liters to work. EJJ actually recomends 25 liters (min 18 liters). This means I can only fit 1 JX125 in the box (intrenal dim. 24"Hx6"Wx12"D)
That is why I started investigating Fostex.
however i'd like to know what drivers you are using in the system described above. I like the idea of using 1st order PLLXO at baffle step. |
|
|
| planet10 |
| quote: | Originally posted by navin
however i'd like to know what drivers you are using in the system described above |
I bought a box of a dozen "famous maker" surplus 6" with a claimed 96.5 dB efficiency (i suspect Klipsch).
The JX125 TL i did is almost 100 litres.
dave |
|
|
| johninCR |
| quote: | Originally posted by pooge
I've been considering using a Jordan line array with 5 drivers instead of 4, in an MTM arrangement.....
I figured that if I use 5 drivers, I could also experiment with connecting them up in an alternative array I read about. I'll be damned if I can remember the name of the array or the author of the article. I'll have to spend some time digging it up. Basically, the 5 driver array was connected up in various phase combinations which, in sum total, maintained a fairly constant dispersion pattern throughout its band. I think it did so vertically, also. |
You are referring to a Bessel Array using amplitude weighting. Here's a link. http://joelist.free.fr/Bessel_Array_DavisDavis.pdf
I've tried this in my experiments with arrays and it works for it's intended purpose to make an array act like a point source for sound dispersion (ie. no more vertical plain of sound). The problem with it is that imaging totally disappeared. The sound seemed to come from everywhere between the speakers. I believe this was caused by the out of phase wiring of some of the speakers. Maybe it resulted from me trying it with dipoles.
A good use may be an application where you don't want imaging, like for a bar, but then you'd give up the advantage in SPL vs distance that line arrays give you. |
|
|
| navin |
| quote: | Originally posted by planet10
The JX125 TL i did is almost 100 litres. |
ouch! 100L! no way I can get away with that. 25 liters was pushing it. |
|
|
| pooge |
| Not sure if the Bessel was the one I saw. I think I read it in the AES Journal in the last 10 years. The connections were similar, though. I'll try to locate it and report back. I did have reservations about it, though, for the reasons you gave. |
|
|
| Colin |
100L for the JX125 is a little large - recommended max for a selaed enclosure is 50L. I once asked Ted if the JX92 could be used as a bass for the JX53 (given it's minimum enclosure size) and he said using a JX125 in a smaller box (ie less than 25L) would be a better solution.
The Carolina Audio site has some variations on the VTL/Essence style enclosure for larger drivers at http://rthack.tripod.com/KOTA/Page_2.html
Colin
PS ref the rear-facing diffuser idea: if we're still talking about Jordan units, I'd think it would be working against them - they are specifically designed to be directional and for the axis of each driver to cross in front of the listener. |
|
|
| AudioGeek |
my intention was to diffuse or break up the very directional HF response.
Would it be better to just add a piezo tweeter x-over at >10Khz???
also in a bipole design... should I roll of the rear driver with an inductor or just use some sound absorbing material??
Thanks!!! |
|
|
| GM |
>anyone have any comments about my rear diffuser?????
can I build a mechanical device that will block midrange frequency but pass anything 400Hz and below and 10Khz and above??
====
Well, it will need to be fairly bulky to attenuate down to 400Hz, so the sides would have to be extra deep to hide it, which in turn would lower the BS, requiring an even bulkier filter, etc.. I prefer acoustic solutions for acoustic problems, but in this scenario, best I think to just use a big electrical notch filter or roll off the rear driver and add a piezo for some top end 'air'.
GM |
|
|
| AudioGeek |
I think I will go with a switchable inductor.
and maybe a piezo later on.... recommendations for a good piezo??
Thanks!@ |
|
|
| planet10 |
| quote: | Originally posted by Colin
100L for the JX125 is a little large - recommended max for a selaed enclosure is 50L. I once asked Ted if the JX92 could be used as a bass for the JX53 (given it's minimum enclosure size) and he said using a JX125 in a smaller box (ie less than 25L) would be a better solution. |
TLs tend to be bigger. I posted a drawing earlier. Here it is again with dimensions & notes
the design falls right out of Martin King's tables.
dave |
|
|
| pooge |
| quote: | Originally posted by planet10
The XO at the BS should be between the 4 drivers & the single driver... you don't want to be padding your "bass" drivers.
dave |
The line array would be used as midrange, not bass. The midrange may often be padded anyway, to match the bass drivers. I agree, though, that it would be more desirable not to pad. Part of the problem in using the identical driver for the tweeter vs. a higher sensitivity one, I guess. Anyhow, I don't see the tradeoff of using the single driver cross-over at such a low baffle-step frequency as being a good thing. It would not get the advantage of the 4-driver array of JX53s over a significant bandwidth, and would further not have an upper crossover far enough away from the lower crossover point. Bullock has reported on this problem with three-way crossovers. The midrange must have a wide enough bandwidth to avoid crossover interaction problems between the two crossover points.
| quote: | Originally posted by planet10
everytime you double the number of drivers you gain 3 dB (ignoring differences, if any, of amplifier output into different impedances) so 4 drivers is 6 dB.
The confusion arises in that many, when talking about multiple driver sensitivity, will lump in an extra 3 dB for the (theroectical) increase in amp output (in actual fact this is rarely the case, and certainly not if you talk tube amps).
dave |
Depends if you are combining in series or parallel. I'm not ignoring impedances. They are an issue with SS amps. Combine in series, and you get the same pressure response. Combine in parallel, you get 6db increase in pressure response. However, at low frequencies, a combination of drivers will have mutual coupling, increasing the radiation resistance. This is where I would prefer to offset the baffle step, because that is a decrease in radiation resistance. This would offset apples with apples. |
|
|
| pooge |
| quote: | Originally posted by planet10
You need to use the baffle step to your advantage to compensate for the xtra 6 dB of sensitivity of the series/parallel set.
dave |
There are other ways besides padding to match sensitivities. The 4-driver array could be connected all in series to match the sensitivity of a single driver, making them a 32 ohm load. To somewhat offset the ugliness of this arrangement, a resistor could be placed in parallel with each driver to help ensure a more consistant voltage sharing between the drivers, and to provide individual electrical damping for each driver that would be at least as good as for the individual drivers in a series/parallel combination. Maybe better, because it would be purely resistive. The bigger problem would probably matching the base unit, if parallel drivers are used there. |
|
|
| planet10 |
| quote: | Originally posted by pooge
Depends if you are combining in series or parallel |
Makes no difference whether you connect in series or parallel, that only affects the impedance (and the potential power available from the amp).
dave |
|
|
| planet10 |
| quote: | Originally posted by pooge
There are other ways besides padding to match sensitivities. The 4-driver array could be connected all in series to match the sensitivity of a single driver, making them a 32 ohm load. To somewhat offset the ugliness of this arrangement, a resistor could be placed in parallel with each driver to help ensure a more consistant voltage sharing between the drivers, and to provide individual electrical damping for each driver that would be at least as good as for the individual drivers in a series/parallel combination. Maybe better, because it would be purely resistive. The bigger problem would probably matching the base unit, if parallel drivers are used there. |
Sounds a mess to me... now you have to add a BSC filter too.
dave |
|
|
| pooge |
| quote: | Originally posted by planet10
Makes no difference whether you connect in series or parallel, that only affects the impedance (and the potential power available from the amp).
dave |
Not true. Speakers are voltage driven. Put two in parallel with the same voltage, you get double the output. Put two in series, and you get half the voltage to each driver, summing to one. |
|
|
| pooge |
| quote: | Originally posted by planet10
Sounds a mess to me... now you have to add a BSC filter too.
dave |
BCS will not correct the pressure response and the power response equally. If you space the drivers and select the baffle width accordingly, you can offset the baffle response loss with the increase in the power response of the drivers due to mutual coupling at low frequencies. The distance of the drivers to the floor, wall, or other boundary may also be useful. |
|
|
| planet10 |
| quote: | Originally posted by pooge
Put two in parallel with the same voltage, you get double the output |
But i said the same power -- you only get twice the output because of the halving of the impedance (and with most amps you would have to turn it up to maintain the same voltage (and 2x the power)).
Thruout this whle discussion we have made the assumption, ignoring differences in amp output due to impedance differences. We are talking 1 W/1m.
dave |
|
|
| planet10 |
| quote: | Originally posted by pooge
BCS will not correct the pressure response and the power response equally. If you space the drivers and select the baffle width accordingly, you can offset the baffle response loss with the increase in the power response of the drivers due to mutual coupling at low frequencies. The distance of the drivers to the floor, wall, or other boundary may also be useful. |
Mutual coupling is a red herring in this situation... if you don't deal with baffle step in one of the ways outlined here
http://www.diyaudio.com/wiki/index....le-step+Physics
you will still end up with weak bass.
dave |
|
|
| pooge |
| quote: | Originally posted by planet10
But i said the same power -- you only get twice the output because of the halving of the impedance (and with most amps you would have to turn it up to maintain the same voltage (and 2x the power)).
Thruout this whle discussion we have made the assumption, ignoring differences in amp output due to impedance differences. We are talking 1 W/1m.
dave |
I never made such an assumption, because it is not worth making.
Amp current output changes are necessary here to maintain the same voltage into parallel drivers. I am talking acoustic pressure vs. voltage input for sensitivity comparisons at crossover. A decent amp would maintain the voltage into 4 ohms. In any event, as I advocated putting the 4 drivers into a 32 ohm connection (maybe sixteen, if 8 ohm resistors are put in parallel with each driver), an amp should not drop voltage vs. an 8 ohm single driver. |
|
|
| pooge |
Mutual coupling (mutual radiation resistance) is not a red herring. It has every bit an effect on radiation resistance, and therefore pressure and power output, as baffle loss. Mutual coupling has the same effect as a boundary, boosting low end to a higher level. In effect, the drivers act as added boundaries to one another. And like a boundary and baffle step, it is frequency dependent. Changes in radiation resistance alter POWER output. See Roy Allison's patent for a clear teaching of these effects. You can read it many times before appreciating the full implications.
As baffle loss is caused by a lowering of radiation resistance, it is ideally corrected by an adjustment to the radiation resistance, not by electrical input adjustments. Adjusting on-axis frequency (pressure) response by electrical input (as indicated in your link) is not ideal because it will boost the power response too much, above the value it was above the baffle loss frequency. This is why some people report that electrical baffle-loss compensation sounds too heavy. Power response is boosted too high as a result. Also, unless the electrical response is rolled off lower down, it is added to room gain. Messy.
If you want to compensate for the baffle loss, better to offset it with radiation resistance than electrical input. |
|
|
| planet10 |
| quote: | Originally posted by pooge
If you want to compensate for the baffle loss, better to offset it with radiation resistance than electrical input. |
Why my preferred BSC is a bi-pole.
dave |
|
|
| planet10 |
| quote: | Originally posted by pooge
A decent amp would maintain the voltage into 4 ohms |
By that i assume you mean that a "decent" amp will double its power into 4 ohms vrs 8 ohms, then there aren't very many "decent" amps.... certainly a clear majority of amps sold or made today do not met this criterion.
There are many, many really good amps that don't meet this criterion.
dave |
|
|
| navin |
I like the Carolina Audio jordans a lot. The thin JX92 tower seems perfect fro my need except if I was building a florr stander i would prefer a little more extension at both ends of the freq. spectrum.
I have 2 options.
(1) wall mounted speaker like B&W VM1 operating from 100hz+ or (2) a floor stander operating from 40Hz+
each of these should be able to deliver 95db at their F3 (100hz or 40Hz) with about 10W rms (my amps will be capable of about 28-30Wrms but I'd like to have some headroom)
the speaker choice (wall/floor) depends on how the furniture in our apt gets arrranged by my wife and her friend. considering that the builder has not even finished the building yet we have time.
I am trying to have both options ready so that i can begin to collect parts as soon as my speaker location is chosen.
in either case I would have to supplement the bass using a sub. |
|
|
| AudioGeek |
I'm going to try the bipole jordans.
I threw together a small bipole with some RS drivers in a cardboard box tuned with a toilet paper roll.
:))
suprisingly very nice sound. actually very effortless.
SOO...
question is:
What type of wood should I use for the enclosure??
A) MDF with Veneer?
B) Plywood with Veneer?
C) Real wood?
Thanks!!! |
|
|
| planet10 |
| quote: | Originally posted by AudioGeek
What type of wood should I use for the enclosure??
B) Plywood with Veneer? |
Void free...
dave |
|
|
| Colin |
Dave - thanks for putting me right on the 100litre enclosure. Liked the design and may query you on it further when I plot my next JX125 enclosures...
Colin |
|
|
| AudioGeek |
| any good places for purchase? |
|
|
|
