positioning drivers on baffle...

Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic

   ...irie all!
right...i'm still thinking about these top end cabinets i'm building..... i've just got a couple of questions on positioning drivers on a baffle to reduce interference effects.....
...this is just for the high frequency cabs.... 2x phl audio 6.5" (playing 500Htz to 3.5KHtz) and 1x p.audio bmd450 tweeter on horn(playing from 3.5KHtz).....initially i was thinking that i'd put the compression driver at the top of the cab and the 2 phl drivers vertically below it......this was to reduce lobing in the horizontal plane......i've read on diyaudio that to reduce the effect of interference tweeters should be no more than 1 wavelength at crossover frequency away from the drivers below them.....velocity of sound in air iz about 345m/s therefore if i'm crossing over to the tweeters at 3.5KHtz then they should be less than 10cm from the centre of the phl's.....in which case they'd need to be as close as possible really.....has anyone else heard this rule and is it worth following? in which case to reduce the interference and get both phl drivers as close to the tweeter as possible i'd put the tweeter in between the phl drivers MTM in the vertical plane (M = phl 6.5" and T=tweeter).
any thoughts? ....am i talking bollocks?
what about baffle edge effects? as far as i can decide these effects are small to none existent when dealing with these high frequencies as the driver doesn't begin to radiate into 4pi space due to high directivity involved with high frequencies....am i correct?
...also.....i seem to remember reading something that said don't put the tweeter directly in the middle of a baffle.....as it can have strange effects......is this bollocks?? coz my designs are heading towards a tweeter in the centre...
...oh yeh...i'm going for a 15degree trapezoidal cabinet to get wider dispersion with 2 stacked together.....
any advice appreciated!

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version Translate Topic
dubsuction Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 18 August 2004 Location: United Kingdom Status: Offline Points: 269	Post Options Post Reply Quote dubsuction Report Post Thanks(0) Quote Reply Topic: positioning drivers on baffle... Posted: 19 October 2004 at 1:13pm
	...irie all! right...i'm still thinking about these top end cabinets i'm building..... i've just got a couple of questions on positioning drivers on a baffle to reduce interference effects..... ...this is just for the high frequency cabs.... 2x phl audio 6.5" (playing 500Htz to 3.5KHtz) and 1x p.audio bmd450 tweeter on horn(playing from 3.5KHtz).....initially i was thinking that i'd put the compression driver at the top of the cab and the 2 phl drivers vertically below it......this was to reduce lobing in the horizontal plane......i've read on diyaudio that to reduce the effect of interference tweeters should be no more than 1 wavelength at crossover frequency away from the drivers below them.....velocity of sound in air iz about 345m/s therefore if i'm crossing over to the tweeters at 3.5KHtz then they should be less than 10cm from the centre of the phl's.....in which case they'd need to be as close as possible really.....has anyone else heard this rule and is it worth following? in which case to reduce the interference and get both phl drivers as close to the tweeter as possible i'd put the tweeter in between the phl drivers MTM in the vertical plane (M = phl 6.5" and T=tweeter). any thoughts? ....am i talking bollocks? what about baffle edge effects? as far as i can decide these effects are small to none existent when dealing with these high frequencies as the driver doesn't begin to radiate into 4pi space due to high directivity involved with high frequencies....am i correct? ...also.....i seem to remember reading something that said don't put the tweeter directly in the middle of a baffle.....as it can have strange effects......is this bollocks?? coz my designs are heading towards a tweeter in the centre... ...oh yeh...i'm going for a 15degree trapezoidal cabinet to get wider dispersion with 2 stacked together..... any advice appreciated!
	peace+respect-always HIGH PRESSURE SOUND SYSTEM www.highpressuresound.co.uk

Timber_MG Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 27 March 2004 Location: South Africa Status: Offline Points: 1454	Post Options Post Reply Quote Timber_MG Report Post Thanks(0) Quote Reply Posted: 19 October 2004 at 2:21pm
	OK, behold the power of M$ Excel. The FRD Consortium http://www.pvconsultants.com/audio/frdgroup.htm has many useful tools to visualize things not seen on many of the conventional modeling programs. For your lobing issue, check out the horizontal array proggie (3d viz). just remember that in PA you don't often exceed 10-20 degrees off axis (vertically) and it's often better to have less lobing going down (crossover also affects this, as it effectively sets the delay between the different drivers at crossover) The lobing in the horizontal plane should be fairly un-affected by your driver spacing vertically. However the vertical lobing does very well change, and this is where the age-old argument of MT vs MTM vs MMT comes in. On baffle edge diffraction, here is the link to the edge tool i told you about earlier. http://www.tolvan.com/edge/help.htm Your goal is to have a cab which has a 60 degree horizontal dispersion throughout it's frequency range. Now this isn't possible physically, as at low frequencies the waves "wrap around" the cabs. This isn't a problem with stacking, as a uniform wavefront is still emmited from the pair (in this case). However as the frequency increases, you want to have a polar response which narrows down to -6dB point at 30degrees off-axis for the system to be sensibly arrayable. Having the drivers closer to one another between the two cabs would reduce the ammount of (almost unavoidable but in the end bearable) interference (actually, move it to a frequency range where it's less noticable). The other approach is to have the cabs spaced a bit further appart, but neither have fixed guidelines as the specific driver arrangement/offset and polar plots affect behaviour. Don't get held up too much on this point (as you'll probably end up using a pre-made crossover anyways which will again affect the vertical lobing). If you were after maximum fidelity, you wouldn't be arraying cabs anyways (with some notable exceptions such as (pseudo) conical waveguides which do sum almost perfectly (eg. SPL td-1). Using more conventional ways can still deliver an acceptable end result, just keep the drivers as close to one another vertically and get a reasonable position on the baffle (edge) Martin

dubsuction Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 18 August 2004 Location: United Kingdom Status: Offline Points: 269	Post Options Post Reply Quote dubsuction Report Post Thanks(0) Quote Reply Posted: 19 October 2004 at 3:11pm
	cheers martin.... i've been toying with diffract.exe which is the simple predecessor to edge i think.....i'll have a look on the updated version now, the effects are quite pronounced eh!! ..it seems to be quite tricky to minimise the effects....i've been able to get a reasonble (ripply but not too far off flat) in the phl 500-3500Htz range by off setting slightly to one side and going for MTM arrangement...hmmmm thanks for the links man cheers
	peace+respect-always HIGH PRESSURE SOUND SYSTEM www.highpressuresound.co.uk

dubsuction Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 18 August 2004 Location: United Kingdom Status: Offline Points: 269	Post Options Post Reply Quote dubsuction Report Post Thanks(0) Quote Reply Posted: 19 October 2004 at 4:15pm
	yeh...the edge seems to agree, i modelled for two drivers and it seems that to avoid a dip of around 2dB at 3.5KHtz (cross over) the drivers need to be 3cm off centre on the baffle, with this configuration..MTM and drivers positioned 3cm horizontally off centre axis, there's only about 1dB variation between 500Htz and 3.5KHtz .....this seems to be the best i can do without making a massive cab... ..question... i've modelled the edge effects of a single cab with two drivers in but i assume that when stacked the front of the whole stack can act as one large baffle? this i imagine has some effect on the response too?! .....i'm just going to build em i reckon one more thing...... does having a protruding 'rim' around the edge of the cabinet (effectively recessing the baffle a bit) have much effect on the sound? i'm thinking in terms of protecting the drivers(i'll use grills aswell) and storage+ aesthetics.... cheers
	peace+respect-always HIGH PRESSURE SOUND SYSTEM www.highpressuresound.co.uk

Timber_MG Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 27 March 2004 Location: South Africa Status: Offline Points: 1454	Post Options Post Reply Quote Timber_MG Report Post Thanks(0) Quote Reply Posted: 19 October 2004 at 6:55pm
	In PA the grilles are a must (esp with the sorta stuff you're doing). Better yet is a layer of thin acoustically transparent foam covering the grille (beer and speakers....). Don't futz about 1dB. Get the Top as wide as the 15 and the Mid-high section as per your modeling. And build, nobody is going to mutter about a 1dB peak at 2k5 and a 3 dB dip at 2k when they have levels of well above 100dB to contend with. to be very honest. I don't know to what extent the edge tool takes directivity into considderation, but i recon build them and see what happens. With such nice mids it would be quite an achievement to have the box ending up soundling like a bunch of piezos. For two cabs next to one another, the angle of 30degrees won't make that much of a difference (for this sim). Assume them to be square cabs and give the mids a sim in the quad config. As I don't know how much directivity is taken into account, the results may model to your liking or not. Regards Martin

dubsuction Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 18 August 2004 Location: United Kingdom Status: Offline Points: 269	Post Options Post Reply Quote dubsuction Report Post Thanks(0) Quote Reply Posted: 20 October 2004 at 8:56am
	cheers martin ...i figured it'd be good enough...hopefully will have another 2 horns done by the start of the week..... ...can't wait to unleash this lot!!
	peace+respect-always HIGH PRESSURE SOUND SYSTEM www.highpressuresound.co.uk

Timber_MG Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 27 March 2004 Location: South Africa Status: Offline Points: 1454	Post Options Post Reply Quote Timber_MG Report Post Thanks(0) Quote Reply Posted: 20 October 2004 at 10:27am
	Horns sound good. A bunch of three should start to make themselves...more noticable. I wouldn't bother taking the direct radiators along to the following gigs then...except for a direct comparison ;-) Martin

dubsuction Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 18 August 2004 Location: United Kingdom Status: Offline Points: 269	Post Options Post Reply Quote dubsuction Report Post Thanks(0) Quote Reply Posted: 20 October 2004 at 11:01am
	coooooooooooooooooooooooooooooooooooooooooooooooooooooool
	peace+respect-always HIGH PRESSURE SOUND SYSTEM www.highpressuresound.co.uk

Timber_MG Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 27 March 2004 Location: South Africa Status: Offline Points: 1454	Post Options Post Reply Quote Timber_MG Report Post Thanks(0) Quote Reply Posted: 20 October 2004 at 11:45am
	Just try to amp them equally though if you can. A multimeter used at 50Hz (becasue the cheap non-tru-RMS ones only work properly around 50) is good for this if you're using different amps and for setting up a limiter to check whether it's working properly and at what level it starts cutting in. Martin

tb_mike Members Profile Send Private Message Find Members Posts Add to Buddy List Old Croc Joined: 01 October 2004 Location: New Zealand Status: Offline Points: 2744	Post Options Post Reply Quote tb_mike Report Post Thanks(0) Quote Reply Posted: 21 October 2004 at 8:16am
	So our major issues with mids and tops are Crossover Selection and optimisation Edge diffraction Driver placement Baffle step(slightly) Ive only just started all this stuff! the articles part of FRD is real good im getting thru them :-)

dubsuction Members Profile Send Private Message Find Members Posts Add to Buddy List Registered User Joined: 18 August 2004 Location: United Kingdom Status: Offline Points: 269	Post Options Post Reply Quote dubsuction Report Post Thanks(0) Quote Reply Posted: 21 October 2004 at 9:38am
	i'm just getting to grips with this myself man! i've been looking at driver placement in order to reduce the baffle edge diffraction effect...by having a driver positioned off centre on a baffle the baffle edges are different distances away from the driver in all directions...which helps to minimise this effect... ...as i'm understanding it edge diffraction and the baffle step effect are one and the same...driver placement affects this.... another factor is interference between multiple drivers... ...tricky stuff! cheers
	peace+respect-always HIGH PRESSURE SOUND SYSTEM www.highpressuresound.co.uk

tb_mike Members Profile Send Private Message Find Members Posts Add to Buddy List Old Croc Joined: 01 October 2004 Location: New Zealand Status: Offline Points: 2744	Post Options Post Reply Quote tb_mike Report Post Thanks(0) Quote Reply Posted: 21 October 2004 at 9:49am
	yeh its hard to get an overall idea of these inter related causes and effects.I agree. But i bet with a little learning+ some neat programs= results in pretty good sound !