I've been thinking over a vague design for a 3 way system which would hopefully produce very high quality sound for a moderate price. I should point out I don't intend on building this right now but thought I would start a discussion and gain some opinions 
For the low end I have in mind the Monacor SPH-275C woofer. This woofer seems to offer ideal parameters for TL loading and should have plenty of output for most living rooms. In a TL I would hope for strong output down to 20Hz. I wouldn't expect the bass units to be small but in a folded design it should be possible to make them a suitable height for the mid+treble unit to sit on top. This driver is around £70:
http://www.monacor.com/en/produktseite_suche.php?artid=3735&spr=EN&typ=full
For the mid I have in mind the Vifa (Peerless V-line) XG18 driver. As a woofer it reportedly lacks low end but most people say the mid is clear and extended on it. Sealed, TL or open baffle are all loading options for this but I am inclined to go with the TL again; for aesthetic reasons i'm not so keen on open baffle. Aperiodic is also interesting however and It could potentially offer the clearest mids. A nice point is aesthetically it should match the woofer quite nicely . This driver is around £50:
http://www.tymphany.com/datasheet/printview.php?id=108
For the treble unit the Morel MDT33 is interesting. It is recieved well by virtually all and seems to fit the bill. Reportedly smooth sound is appealing, I hate that accentuated sibilance you get with many drivers. This driver is around £70:
http://www.morelhifi.com/products/pdf/mdt 33.pdf
There is a fair variation in sensitivity between these drivers which could cause a problem, but then I never planned on using a passive crossover. The crossover would be based on this design:
http://sound.westhost.com/project09.htm
with points at around 250Hz and 2.5Khz.
For the bass amplifier I was thinking about this:
http://sound.westhost.com/project101.htm
having used it in my sub project I can say it offers the tightest most powerful output I've come accross. Offers a suitable level of power for the woofers.
The mids and highs are undecided but a chip-amp could be viable for one or both, or the P101 could be used for these too (with lower supplies if desired). The Vifa driver crossed with a steep slope at 250Hz should handle more power than listed so every driver in this system should easily handle over 100W and in total offer a respectable output.
Any thoughts?
For the low end I have in mind the Monacor SPH-275C woofer. This woofer seems to offer ideal parameters for TL loading and should have plenty of output for most living rooms. In a TL I would hope for strong output down to 20Hz. I wouldn't expect the bass units to be small but in a folded design it should be possible to make them a suitable height for the mid+treble unit to sit on top. This driver is around £70:
http://www.monacor.com/en/produktseite_suche.php?artid=3735&spr=EN&typ=full
For the mid I have in mind the Vifa (Peerless V-line) XG18 driver. As a woofer it reportedly lacks low end but most people say the mid is clear and extended on it. Sealed, TL or open baffle are all loading options for this but I am inclined to go with the TL again; for aesthetic reasons i'm not so keen on open baffle. Aperiodic is also interesting however and It could potentially offer the clearest mids. A nice point is aesthetically it should match the woofer quite nicely
. This driver is around £50:http://www.tymphany.com/datasheet/printview.php?id=108
For the treble unit the Morel MDT33 is interesting. It is recieved well by virtually all and seems to fit the bill. Reportedly smooth sound is appealing, I hate that accentuated sibilance you get with many drivers. This driver is around £70:
http://www.morelhifi.com/products/pdf/mdt 33.pdf
There is a fair variation in sensitivity between these drivers which could cause a problem, but then I never planned on using a passive crossover. The crossover would be based on this design:
http://sound.westhost.com/project09.htm
with points at around 250Hz and 2.5Khz.
For the bass amplifier I was thinking about this:
http://sound.westhost.com/project101.htm
having used it in my sub project I can say it offers the tightest most powerful output I've come accross. Offers a suitable level of power for the woofers.
The mids and highs are undecided but a chip-amp could be viable for one or both, or the P101 could be used for these too (with lower supplies if desired). The Vifa driver crossed with a steep slope at 250Hz should handle more power than listed so every driver in this system should easily handle over 100W and in total offer a respectable output.
Any thoughts?
Hi Dr. EM,
Remember that the drivers constitute only part of the cost of DIYing a set of loudspeakers. There's all the other, relatively mundane stuff as well, such as:
-glue
-wiring
-MDF panels/ plywood/ alternative enclosure materials
-passive crossover components*
-rubber feet
-sheets of rubber for chassis gaskets
-various nuts and bolts**
-pillow stuffing/foam/etc
-sandpaper
-lacquer/ veneer/ protective fabric and so forth.
Depending on where your priorities are, you can cut costs on things that are related to appearance but have little impact on the sound: veneer, lacquer and some of the sanding.
*If you already have a multi-channel amp lying around somewhere, you may find that it's worthwhile investing some extra time in an active crossover. I don't buy into all that pseudo-science regarding posh, gold-plated inductors and whatnot. IMO a few hours laying out and soldering a prototyping board is time well spent
**In terms of longevity, the box will almost certainly last longer than the drivers. So even if you're not worried about aesthetics, build quality is still important, and that includes the type of screws used for mounting the drivers. I like to use grub bolts with nuts on both ends, so the design doesn't rely on woodscrews embedded into the baffle.
...So, if you can find ways to save money on consumables, and maybe factor in longevity and reusability as a "cost", you can use that use that as an excuse to spend more on drivers or more complicated box shapes?
Remember that the drivers constitute only part of the cost of DIYing a set of loudspeakers. There's all the other, relatively mundane stuff as well, such as:
-glue
-wiring
-MDF panels/ plywood/ alternative enclosure materials
-passive crossover components*
-rubber feet
-sheets of rubber for chassis gaskets
-various nuts and bolts**
-pillow stuffing/foam/etc
-sandpaper
-lacquer/ veneer/ protective fabric and so forth.
Depending on where your priorities are, you can cut costs on things that are related to appearance but have little impact on the sound: veneer, lacquer and some of the sanding.
*If you already have a multi-channel amp lying around somewhere, you may find that it's worthwhile investing some extra time in an active crossover. I don't buy into all that pseudo-science regarding posh, gold-plated inductors and whatnot. IMO a few hours laying out and soldering a prototyping board is time well spent
**In terms of longevity, the box will almost certainly last longer than the drivers. So even if you're not worried about aesthetics, build quality is still important, and that includes the type of screws used for mounting the drivers. I like to use grub bolts with nuts on both ends, so the design doesn't rely on woodscrews embedded into the baffle.
...So, if you can find ways to save money on consumables, and maybe factor in longevity and reusability as a "cost", you can use that use that as an excuse to spend more on drivers or more complicated box shapes?
Thanks for your input
Yeah, I've seen how all the odd little bits add up in projects many times from various (much smaller scale) electronics projects. The workings are cheap but then knobs and mounting of boards, cases etc add up.
I think to make this system for £1000 is just possible. That is quite a lot but compared to what you'd expect commercially for that money it isn't a bad deal (just a small pair of stand mounts can easily cost this).
I haven't considered finish too much and it would be a worry. It's something which requires real experience to do right it seems. I am trying to veneer a sub project now and it isn't exactly going well. Where there are two bits of veneer required to be layed next to each other the join is very obvious. Something as seemingly simple as cutting two straight edges is suprisingly difficult (I am using a steel ruler and x-acto style knife). That said the finish would be important on a project costing this much so it would be an additional cost to get right. The finish also applies to the 3 DIY amplifiers and crossover/attenuator/selector unit; buying in those nice ATI aluminium chassis is too costly.
For assembly I used solely glue on my sub project. Its so simple yet so ideal. No wood splitting, complete seal between joins, high strength. To test the join I glued to smaller scraps together; once set I snapped them apart with my hands and they only break apart because the top layer of one piece gets torn off. A full length joint will have plenty of stregth. Dowelling may help with extreme output subs where the joints are stressed more but that isn't an issue here. I use the nuts and bolts from MAS, very effective and simple to install . I think i'd use plywood if I ever did this, so thats a bit more than MDF, but it seems easier in a lot of ways.
There are odd one off costs too like tools etc. I already have a basic scope, jigsaw, pillar drill, soldering iron etc, but a router would be essential. Also equiptment to make PCB's.
How does the VAS correspond to the size of a TL speaker incidentally. Defining the size of the bass unit would be a good start, i'm worried it may end up like a wardrobe
Yeah, I've seen how all the odd little bits add up in projects many times from various (much smaller scale) electronics projects. The workings are cheap but then knobs and mounting of boards, cases etc add up.
I think to make this system for £1000 is just possible. That is quite a lot but compared to what you'd expect commercially for that money it isn't a bad deal (just a small pair of stand mounts can easily cost this).
I haven't considered finish too much and it would be a worry. It's something which requires real experience to do right it seems. I am trying to veneer a sub project now and it isn't exactly going well. Where there are two bits of veneer required to be layed next to each other the join is very obvious. Something as seemingly simple as cutting two straight edges is suprisingly difficult (I am using a steel ruler and x-acto style knife). That said the finish would be important on a project costing this much so it would be an additional cost to get right. The finish also applies to the 3 DIY amplifiers and crossover/attenuator/selector unit; buying in those nice ATI aluminium chassis is too costly.
For assembly I used solely glue on my sub project. Its so simple yet so ideal. No wood splitting, complete seal between joins, high strength. To test the join I glued to smaller scraps together; once set I snapped them apart with my hands and they only break apart because the top layer of one piece gets torn off. A full length joint will have plenty of stregth. Dowelling may help with extreme output subs where the joints are stressed more but that isn't an issue here. I use the nuts and bolts from MAS, very effective and simple to install
. I think i'd use plywood if I ever did this, so thats a bit more than MDF, but it seems easier in a lot of ways. There are odd one off costs too like tools etc. I already have a basic scope, jigsaw, pillar drill, soldering iron etc, but a router would be essential. Also equiptment to make PCB's.
How does the VAS correspond to the size of a TL speaker incidentally. Defining the size of the bass unit would be a good start, i'm worried it may end up like a wardrobe
Dr.EM said:I but then I never planned on using a passive crossover. The crossover would be based on this design:
http://sound.westhost.com/project09.htm
with points at around 250Hz and 2.5Khz.
Hmmm........
Active crossovers are nothing like the universal cure all unless you
know exactly what you are doing and design a bespoke active c/o.
http://www.rjbaudio.com/Audiofiles/FRDtools.html
http://www.troelsgravesen.dk/Diy_Loudspeaker_Projects.htm
http://www.humblehomemadehifi.com/
http://www.rjbaudio.com/projects.html
http://www.zaphaudio.com/
All the above are worth a thorough perusal for inspiration and information.
FWIW do not use a driver that you cannot cross check its
performance if you do not need to, pick a known quantity.
Personally I'd say @ ~ £1k a bi-amped passive/active hybrid is your best bet.
The advantages of the above are described here :
http://www.musicanddesign.com/HybridDesign.html
/sreten.
I think the key to building a good loudspeaker is to first do the obvious stuff that has been mentioned here, like selecting drivers and box type, then do a bit of simulations based on whatever data there is on the drivers, then building a box, maybe at first a quick and dirty prototype box for measurement.
The comes the important part, that DIYers often neglect. Learn how and what to measure, and start an iterative process with measurement and crossover design.
The crossover is a great place to compensate for all sorts of quirks that the rest of the system has. Using off-shelf crossovers is IMO the most common mistake done in DIY projects.
I am not talking about expensive types of coils or capacitors, but just selecting the right values for the components and an appropriate design for the crossover.
The comes the important part, that DIYers often neglect. Learn how and what to measure, and start an iterative process with measurement and crossover design.
The crossover is a great place to compensate for all sorts of quirks that the rest of the system has. Using off-shelf crossovers is IMO the most common mistake done in DIY projects.
I am not talking about expensive types of coils or capacitors, but just selecting the right values for the components and an appropriate design for the crossover.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.