i'm 100% new to DIY audio (speaker wise), and i've decided on a reasonably big project to start with
i've built floor speakers, each consisting of a 1" tweeter, two 5" midranges and one 8" for lowrange
the limited width i set for the enclosure meant the 8" would be facing the side. this may not be ideal, but it means that by swapping the two speakers around in the room, i can either have the 8"s facing into the room, or at a wall.
i don't know what would be an acceptable XO frequency for the midranges - lowrange. i don't wanna screw up imaging by making the XO too high, but i also don't wanna waste the 8"s by barely using them
i have a 12" sub i already use separately (but will play with the new speakers), and can vary its crossover between 20-100Hz.
so ultimately my question - what sort of frequency would i be looking at to not stuff up imaging?
i've built floor speakers, each consisting of a 1" tweeter, two 5" midranges and one 8" for lowrange
the limited width i set for the enclosure meant the 8" would be facing the side. this may not be ideal, but it means that by swapping the two speakers around in the room, i can either have the 8"s facing into the room, or at a wall.
i don't know what would be an acceptable XO frequency for the midranges - lowrange. i don't wanna screw up imaging by making the XO too high, but i also don't wanna waste the 8"s by barely using them
i have a 12" sub i already use separately (but will play with the new speakers), and can vary its crossover between 20-100Hz.
so ultimately my question - what sort of frequency would i be looking at to not stuff up imaging?
Hi Phreeky.
To give you a proper answer we really need to know more about the drivers and the configuration of the cab.
Unfortunately a successful Xover implementation is much more than just picking frequencies at random, many factors have to be taken into account. I won't blind you with tech terms, but you may want to read up a little, and do some research on loudspeaker testing.
Good luck😉
To give you a proper answer we really need to know more about the drivers and the configuration of the cab.
Unfortunately a successful Xover implementation is much more than just picking frequencies at random, many factors have to be taken into account. I won't blind you with tech terms, but you may want to read up a little, and do some research on loudspeaker testing.
Good luck😉
well i'm not after a complete crossover, i'm going to try a few, but would like some kind of starting point to help minimize cost of replacing components.
i mainly just want to know at what sort of frequency it is generally accepted that direction is not important/cannot be heard. i do realise that regardless of the frequency, some direction will be heard regardless (due to some vibrations etc)
shoot technical terms at me, it's ok. i've done quite a bit of reading and understand the concepts quite well.
Not sure if these links will work, but drivers are:
tweeter: http://www1.jaycar.com.au/productVi...&pageNumber=&priceMin=&priceMax=&SUBCATID=418
midrange: http://www1.jaycar.com.au/productVi...&pageNumber=&priceMin=&priceMax=&SUBCATID=383
bass: http://www1.jaycar.com.au/productVi...&pageNumber=&priceMin=&priceMax=&SUBCATID=383
Currently looking at a 3way @ 200Hz & 3500Hz, 1st order. i got a mate to draw up the circuit in microcap and simulate it, and the response and phasing looks fine etc. but looking at the frequency response, i have no REAL idea what, for example, 6dB difference 'sounds' like.
just hoping for a little advice
i mainly just want to know at what sort of frequency it is generally accepted that direction is not important/cannot be heard. i do realise that regardless of the frequency, some direction will be heard regardless (due to some vibrations etc)
shoot technical terms at me, it's ok. i've done quite a bit of reading and understand the concepts quite well.
Not sure if these links will work, but drivers are:
tweeter: http://www1.jaycar.com.au/productVi...&pageNumber=&priceMin=&priceMax=&SUBCATID=418
midrange: http://www1.jaycar.com.au/productVi...&pageNumber=&priceMin=&priceMax=&SUBCATID=383
bass: http://www1.jaycar.com.au/productVi...&pageNumber=&priceMin=&priceMax=&SUBCATID=383
Currently looking at a 3way @ 200Hz & 3500Hz, 1st order. i got a mate to draw up the circuit in microcap and simulate it, and the response and phasing looks fine etc. but looking at the frequency response, i have no REAL idea what, for example, 6dB difference 'sounds' like.
just hoping for a little advice
Ah, I know these drivers pretty well....
firstly , you will have to run the pair of 5" in series of course, then everything is basically an 8 ohm load all the way down. I would go with an mtm + w arrangement, making the cabs about 1200mm high.
I actually have a 3-way TL using the 10", 5" unshielded, and response tweeter. It uses a linked series x-o and sounds pretty good. Email me if interested in the circuit, 2 coils, 2 caps and a couple of resistors.
firstly , you will have to run the pair of 5" in series of course, then everything is basically an 8 ohm load all the way down. I would go with an mtm + w arrangement, making the cabs about 1200mm high.
I actually have a 3-way TL using the 10", 5" unshielded, and response tweeter. It uses a linked series x-o and sounds pretty good. Email me if interested in the circuit, 2 coils, 2 caps and a couple of resistors.
very funny u say what u did.
cabinets are built. yes, mids in series. strangly enough cabinets are exactly 1200mm tall, 160mm wide and 340mm deep (all externally)
internally there is a seperator around 700mm up. there is also another at the top making the combined mids volume around 6.4L (double the recommended for a single driver). the 8" has around 30L, although i will reduce this VERY mildly for cosmetic reasons.
thanx very much for any advice
cabinets are built. yes, mids in series. strangly enough cabinets are exactly 1200mm tall, 160mm wide and 340mm deep (all externally)
internally there is a seperator around 700mm up. there is also another at the top making the combined mids volume around 6.4L (double the recommended for a single driver). the 8" has around 30L, although i will reduce this VERY mildly for cosmetic reasons.
thanx very much for any advice
first -- take a look at some of the crossover designs which are on the website for Audax, some of the designs which are Zalytron's website -- at least this will give you a starting point.
a good way of addressing this problem generally is by multi-amping and using an adjustable active crossover to determine what sounds best -- (barring an analyzer etc.) you can make an active crossover adjustable by using molex 0.100" female headers as "jacks" so to speak -- to plug in different 1% resistors for setting the frequency -- Analogue Devices has a great inter-active filter designer on their website (see my post.) It does get pretty hairy when you are using more than one driver, however.
oftentimes (as in the case of Audax) you can get an idea of the driver's response curve from the manufacturer's PDF, plug it into Excel (it then becomes an estimate, since you have to eyeball the numbers) and back out from the combined frequency response the curve of the filter response.
btw, I can't emphasize enough the utility of this tool by ADI. It allows you to specify Low Pass, High Pass, Bandpass; Butterworth, Bessel, Chebyshev transfer functions, Sallen Key, MFB, State-Variable etc. topology; gain, and filter order up to #8.
a good way of addressing this problem generally is by multi-amping and using an adjustable active crossover to determine what sounds best -- (barring an analyzer etc.) you can make an active crossover adjustable by using molex 0.100" female headers as "jacks" so to speak -- to plug in different 1% resistors for setting the frequency -- Analogue Devices has a great inter-active filter designer on their website (see my post.) It does get pretty hairy when you are using more than one driver, however.
oftentimes (as in the case of Audax) you can get an idea of the driver's response curve from the manufacturer's PDF, plug it into Excel (it then becomes an estimate, since you have to eyeball the numbers) and back out from the combined frequency response the curve of the filter response.
btw, I can't emphasize enough the utility of this tool by ADI. It allows you to specify Low Pass, High Pass, Bandpass; Butterworth, Bessel, Chebyshev transfer functions, Sallen Key, MFB, State-Variable etc. topology; gain, and filter order up to #8.
cool thanx for that. i've looked around a lot on the net before i decided to post here, so had plenty of ideas. didn't want to get building it all to find it sounded like crap tho
i can use some software i have to basically perform as an active filter, using front and rear sound card outputs. i've been playing around with sine wave generators as well, although that obviously is no good for phasing problems, as the crossover in software has no phase changes. but it still helps with the frequencies.
bi-amping would be great, only if i had the money to get multiple amps together. i can for testing, but that's it. money is tight, being a uni student and all, so the less wasted components the better.
amp module kits are looking good at this stage.
only other question i have at this stage is to do with sensitivity. on my mids, which are in series, they are rated at 87dB, tweeter at 94dB and woofer at 90dB. that isn't good (right?). easy enough to attenuate the tweeter, i'll do that down the track a little and put it on a separate board. but the midranges, i've heard the sensitivity changes when u put them in series....? i dunno, i should just do a search for that and i'll probably find it 🙂
i can use some software i have to basically perform as an active filter, using front and rear sound card outputs. i've been playing around with sine wave generators as well, although that obviously is no good for phasing problems, as the crossover in software has no phase changes. but it still helps with the frequencies.
bi-amping would be great, only if i had the money to get multiple amps together. i can for testing, but that's it. money is tight, being a uni student and all, so the less wasted components the better.
amp module kits are looking good at this stage.
only other question i have at this stage is to do with sensitivity. on my mids, which are in series, they are rated at 87dB, tweeter at 94dB and woofer at 90dB. that isn't good (right?). easy enough to attenuate the tweeter, i'll do that down the track a little and put it on a separate board. but the midranges, i've heard the sensitivity changes when u put them in series....? i dunno, i should just do a search for that and i'll probably find it 🙂
multi-amping
this is where the LM3875 or LM3886 come in handy -- one of the projects I have been working on uses the LM1876 (a smaller LM3875 with 2 channels) for biamping computer speakers.
this is where the LM3875 or LM3886 come in handy -- one of the projects I have been working on uses the LM1876 (a smaller LM3875 with 2 channels) for biamping computer speakers.
Hi Phreeky
Corrections accepted, but I think that you should be OK with the mid driver sensitivity. If you wire them in series, each will get half of a 2.83V input from the amp, which will mean that they should produce 84 dB each. Because there are two of them and the sound is correlated, this will add 6dB, so net sensitivity is 90 dB/2.83V/m.
This assumes that the driver sensitivity is 87 dB/2.83V/m, which seems very low for a 4 ohm mid driver. It may be worth checking with Jaycar whether the sensitivity is per 2.83V, or per watt, which would changes things.
200 Hz seems to be a common choice for a side firing woofer. When I was toying with the idea of one of these I found of a couple of designs on the net which used this frequency (NHT + someone's home project).
If you are using a sub too, you may have problems with the crossover points of the low/mid and sub/mid being to close together. Normally you would want the XO points at least a couple of octaves apart, and this would mean that your sub would come in at 50 Hz.
In some experiments that I've just finished on my 3 way (see my previous thread) I found that an XO frequency of 130Hz or lower was needed to preserve good imaging and focus, but there were other factors at play in my situation. I'll try to write up my findings soon.
Mick
Corrections accepted, but I think that you should be OK with the mid driver sensitivity. If you wire them in series, each will get half of a 2.83V input from the amp, which will mean that they should produce 84 dB each. Because there are two of them and the sound is correlated, this will add 6dB, so net sensitivity is 90 dB/2.83V/m.
This assumes that the driver sensitivity is 87 dB/2.83V/m, which seems very low for a 4 ohm mid driver. It may be worth checking with Jaycar whether the sensitivity is per 2.83V, or per watt, which would changes things.
200 Hz seems to be a common choice for a side firing woofer. When I was toying with the idea of one of these I found of a couple of designs on the net which used this frequency (NHT + someone's home project).
If you are using a sub too, you may have problems with the crossover points of the low/mid and sub/mid being to close together. Normally you would want the XO points at least a couple of octaves apart, and this would mean that your sub would come in at 50 Hz.
In some experiments that I've just finished on my 3 way (see my previous thread) I found that an XO frequency of 130Hz or lower was needed to preserve good imaging and focus, but there were other factors at play in my situation. I'll try to write up my findings soon.
Mick
- Status
- Not open for further replies.