First post so please be gentle 😀 and apologies as it is a little long. Please bare with me.
Been into Hifi for a while working through a number of systems -
1980's Sevenoaks pick n mix - Rotel CD, Sansui amp, Mission 737
Proper System (sold 🙁) - Technics 777 CD, Quad 44/405-2, Kef 104/2 (AKA the Wardrobes by Significant Other)
Current AV - Denon BT2500, 3808, B&W FPM6/4/M1
Current Office - PC, Fostex A3 Dac, 2x Rotel RB850 (bridged), B&W CDM1SE
Also have built a Kit car and restored classics so am reasonably OK with tools etc.
Anyway, looking for a future project and have had a few ideas that I would like to see if I am completely barking mad or not - All comments gratefully received.
Idea -
Looking to build some floor standing monitor speakers i.e. flat response. I always liked the Kef 104 mid range and the sound of both my CDM1SE and my Son's Yamaha H7's, which are pretty neutral and really show off good recordings and show up bad ones.
Doing a bit of lateral thinking and too many hours on the web I thing the following setup maybe / possibly work at a sensible cost (approx £500 per pair)
Drivers -
Monacor DT-107 1" soft dome tweeter (80W)
Monacor SPH-145HQ 5.5" Mid (50W)
Monacor SPH-8M 8" Woofer (100W)
If you get the frequency/dB plots off the Moncore website and overlay them it looks like they all have a pretty flat response and with cross overs at approx 300hz and 3Khz the composite curve looks pretty good (I know this does not take into account enclose effects and is rough and ready but a start). I can also get the drivers for less than £150 per speaker.
In order to keep the inherent flat response I am looking at a sealed box design for the mid and woofer and reducing any internal reflections or box flexure. Using the Monacor data and the MH-audio closed box calculator for a total Q of 0.65, the Mid volume should be about 7.5l and the woofer about 29l. Comments on these volumes please and use of total Q = 0.65.
I have looked at various speaker designs and like the organic shapes and reading about the nautilus shell / helix / spiral speaker performance I think this is what I will be looking into.
For the Woofer I am assuming the box shape is less important so will be something like a water melon shape i.e. no sharp edges and for the mid a nautilus shell logarithmic spiral shape. I can do designs on CAD and scale to get the internal volumes correct (see above). The tweeter only needs a cavity to put it in, no volume, correct?.
There are many ways of manufacturing this but I will be looking to mould the shapes into a fibreglass half shell (think jelly mould) using the CAD model to provide profiles that are stacked up in wood to make the melon/shell shapes. The fibreglass shell will be backed filled with flexible polyurethane foam to provide some rigidity and also damping (self expanding, similar to memory foam matresses without the memory), these will then be bonded together and encased in a more traditional looking 1" MDF enclosure, leaving space at the bottom to mount some active components.
I also like the idea of a quasi active set up so as I already have the rotel power amps I am looking to do the following -
1x RB850 running in stereo for the Tweeter and Mid (could expand to two bridged in future)
1x RB bridged running the Woofer
(need to get 2 more Rotel Amps). RB850 is rated at stereo 50w RMS into 8 Ohms, or bridged 150w RMS into 8 Ohms.
Both amps will be fed by a 3 way mono active crossover like the Behringer CX2310 (£70) so I can set the individual driver gain and low/high cross over points.
You still with me?
In summary - Semi active, floor standing full range (ish) monitor speaker -
Behringer CX2310 active crossover (x2, one per speaker)
Rotel RB power amps - initially 2 per speaker, 1 stereo for Tweeter/Mid and 1 bridged for Woofer
Sealed box design per driver - Woofer in elipsoid shape and Mid in Log Spiral shape made from fibreglass shell, back filled by flexible damping foam and in 1" MDF enclosure.
So, as already said, am I mad, is this feasible, does it have a chance of providing what I am after and thoughts on the Monacor drivers (don't seem to be used in UK much) or potential other alternatives for similar price (Dayton Audio RS Paper range ?).
Thanks.
Been into Hifi for a while working through a number of systems -
1980's Sevenoaks pick n mix - Rotel CD, Sansui amp, Mission 737
Proper System (sold 🙁) - Technics 777 CD, Quad 44/405-2, Kef 104/2 (AKA the Wardrobes by Significant Other)
Current AV - Denon BT2500, 3808, B&W FPM6/4/M1
Current Office - PC, Fostex A3 Dac, 2x Rotel RB850 (bridged), B&W CDM1SE
Also have built a Kit car and restored classics so am reasonably OK with tools etc.
Anyway, looking for a future project and have had a few ideas that I would like to see if I am completely barking mad or not - All comments gratefully received.
Idea -
Looking to build some floor standing monitor speakers i.e. flat response. I always liked the Kef 104 mid range and the sound of both my CDM1SE and my Son's Yamaha H7's, which are pretty neutral and really show off good recordings and show up bad ones.
Doing a bit of lateral thinking and too many hours on the web I thing the following setup maybe / possibly work at a sensible cost (approx £500 per pair)
Drivers -
Monacor DT-107 1" soft dome tweeter (80W)
Monacor SPH-145HQ 5.5" Mid (50W)
Monacor SPH-8M 8" Woofer (100W)
If you get the frequency/dB plots off the Moncore website and overlay them it looks like they all have a pretty flat response and with cross overs at approx 300hz and 3Khz the composite curve looks pretty good (I know this does not take into account enclose effects and is rough and ready but a start). I can also get the drivers for less than £150 per speaker.
In order to keep the inherent flat response I am looking at a sealed box design for the mid and woofer and reducing any internal reflections or box flexure. Using the Monacor data and the MH-audio closed box calculator for a total Q of 0.65, the Mid volume should be about 7.5l and the woofer about 29l. Comments on these volumes please and use of total Q = 0.65.
I have looked at various speaker designs and like the organic shapes and reading about the nautilus shell / helix / spiral speaker performance I think this is what I will be looking into.
For the Woofer I am assuming the box shape is less important so will be something like a water melon shape i.e. no sharp edges and for the mid a nautilus shell logarithmic spiral shape. I can do designs on CAD and scale to get the internal volumes correct (see above). The tweeter only needs a cavity to put it in, no volume, correct?.
There are many ways of manufacturing this but I will be looking to mould the shapes into a fibreglass half shell (think jelly mould) using the CAD model to provide profiles that are stacked up in wood to make the melon/shell shapes. The fibreglass shell will be backed filled with flexible polyurethane foam to provide some rigidity and also damping (self expanding, similar to memory foam matresses without the memory), these will then be bonded together and encased in a more traditional looking 1" MDF enclosure, leaving space at the bottom to mount some active components.
I also like the idea of a quasi active set up so as I already have the rotel power amps I am looking to do the following -
1x RB850 running in stereo for the Tweeter and Mid (could expand to two bridged in future)
1x RB bridged running the Woofer
(need to get 2 more Rotel Amps). RB850 is rated at stereo 50w RMS into 8 Ohms, or bridged 150w RMS into 8 Ohms.
Both amps will be fed by a 3 way mono active crossover like the Behringer CX2310 (£70) so I can set the individual driver gain and low/high cross over points.
You still with me?
In summary - Semi active, floor standing full range (ish) monitor speaker -
Behringer CX2310 active crossover (x2, one per speaker)
Rotel RB power amps - initially 2 per speaker, 1 stereo for Tweeter/Mid and 1 bridged for Woofer
Sealed box design per driver - Woofer in elipsoid shape and Mid in Log Spiral shape made from fibreglass shell, back filled by flexible damping foam and in 1" MDF enclosure.
So, as already said, am I mad, is this feasible, does it have a chance of providing what I am after and thoughts on the Monacor drivers (don't seem to be used in UK much) or potential other alternatives for similar price (Dayton Audio RS Paper range ?).
Thanks.
I wonder if you are focussing your box ideas on the internal shape, over getting the sound out into the room. At any rate your ideas seem fairly conventional, if not a little traditional, but nothing crazy.
Thanks for the comment, do you mean how the drivers interact with the front face of the speaker. My initial thought was to conventionally flush mount them with the fibreglass providing only the cavities to form the sealed boxes. However, I could build it so the drivers protrude significantly from the front face and so have effectively no baffle i.e. mount then on cylinders that stick out 1-2" from the front face of the speaker. The third option is the make the fibreglass very thick so it does not need the wood enclosure and so the speakers would again be in free air - similar to the B&W Nautilus speaker.
What effect would this have, any thoughts or clarificaiton ?
Thanks for the comment. So looks like the Sphere give the flattest response. Will tray and aim for that in the Woofer.
Have added a noddy sketch below to try and explain the three option for mounting the drivers flush with the front face, raised above or in "free air". If I get chance I will CAD up but may not be for a while.
Also here is an overlay of the frequency plots for the Monacor drivers for info. As said, not very scientific but may gives a quick idea how they may integrate.
Links to Monacor data sheets -
SPH-8M - MONACOR - SPH-8M
SPH-145HQ - MONACOR - SPH-145HQ
DT-107 - MONACOR - DT-107
Have added a noddy sketch below to try and explain the three option for mounting the drivers flush with the front face, raised above or in "free air". If I get chance I will CAD up but may not be for a while.
An externally hosted image should be here but it was not working when we last tested it.
Also here is an overlay of the frequency plots for the Monacor drivers for info. As said, not very scientific but may gives a quick idea how they may integrate.
An externally hosted image should be here but it was not working when we last tested it.
Links to Monacor data sheets -
SPH-8M - MONACOR - SPH-8M
SPH-145HQ - MONACOR - SPH-145HQ
DT-107 - MONACOR - DT-107
Last edited:
The sphere does indeed look ideal but that truncated pyramid looks like a reasonable alternative.
What is the difference between a flat baffle and (hypothetically) no baffle? If you embrace the idea you might find there is no problem that needs (or can) be fixed by eliminating the front in this kind of design. Anyway, that small mounting surface will be a half space baffle especially at higher frequencies, and doesn't sound like a good idea to leave it like that. You could provide easement at least.
A simple baffle diffraction simulation of large round overs for a T-M-W 3-way baffle illustrates low tweeter diffraction effects through the crossover, which should produce a consistent soundstage through the TM crossover. Below the tweeter crossover, both the M and W generate similar 1.8db slow baffle humps, which then correctly linearly drop with the baffle W*H dimentional effects. Textbook crossovers perform well with rounded baffles. A pair of side-side counter-force woofers can significangly reduce vibration. You can still taper(angle cut) the volume behind the midrange and use multi-hole bracing to reduce bass resonances. Many cabinet shops have a 3" quarter-round cutter. You can purchase 3" to 6" radius molded MDF quarter round edge material. Trans-Laminate construction is probably the best solution for high performance cabinets. CNC + 10 sheets of MDF + GLUE.
---Big round overs = Small diffraction effects = Smooth and Uniform soundstage =??= NEUTRAL?
Octavia: The tapered truncated pyramid cabinet constructed and measured in 1950 by Dr. Harry Olson, and marketed by Avalon usually sounds accurate and "interesting" in most rooms. Why? A simple baffle diffraction simulation illustrates how the T and M each produce low but variable triangle-bevel diffraction generated dither in their frequency band.
---Deep bevels = reduced but irregular diffraction effects = comfortable yet interesting soundstage =??= CAPTIVATING?
The Visaton website has a diy Avalon-style design with an 8" woofer called Sonja. You can free download BoxSim CAD with all the speaker models for Visaton drivers.
---Big round overs = Small diffraction effects = Smooth and Uniform soundstage =??= NEUTRAL?
Octavia: The tapered truncated pyramid cabinet constructed and measured in 1950 by Dr. Harry Olson, and marketed by Avalon usually sounds accurate and "interesting" in most rooms. Why? A simple baffle diffraction simulation illustrates how the T and M each produce low but variable triangle-bevel diffraction generated dither in their frequency band.
---Deep bevels = reduced but irregular diffraction effects = comfortable yet interesting soundstage =??= CAPTIVATING?
The Visaton website has a diy Avalon-style design with an 8" woofer called Sonja. You can free download BoxSim CAD with all the speaker models for Visaton drivers.
Attachments
So, lots of googling and head scratching and finally a bit of cad work his evening.
First, thought I'd have a go at modelling a two spiral version - one spiral for the mid and one for the woofer. Drivers are roughly modelled from the Monacor data sheets, enclose is just for scale.
as you can see this produces a big enclosure (approx 1m x 0.5m x 0.45m) and the closed volumes of the spirals is no where near the predicted close box design figures - mid 3.65l vs 7.5l and woofer 14.14l vs 29l. This would also take a lot of wood and time to make.
So, back to my first plan, an ovoid woofer and spiral mid sealed enclosure. I increased the size slightly (10%) on the mid spiral and came up with this -
This is closer on the sealed volumes - mid is 5.5l and the woofer 23.3l. This gives a total Q for the mid of 0.723 and the woofer 0.695. so could go up in size a bit for the mid (5.8l give Q=0.707) and down a bit in size for the Woofer (22.l for Q=0.707).
This is also a more manageable size speaker - approx 0.75 x 0.45 x 0.45
Remembering that any space between the cabinet walls and the fibreglass sealed box shapes can be filled with expanding foam - either rigid or more like flexible.
As a back up I quite like the Jim Holtz Statement Monitor and Finalist designs with the vented woofer box and the open back mid. The rounded box in Linesource's (2nd) picture also caught my eye. So may look at a design uing the Finalist general design with the round 6" pipe open mid driver but incorporated into a curved box using flexiply (4x5mm layers).
First, thought I'd have a go at modelling a two spiral version - one spiral for the mid and one for the woofer. Drivers are roughly modelled from the Monacor data sheets, enclose is just for scale.
An externally hosted image should be here but it was not working when we last tested it.
as you can see this produces a big enclosure (approx 1m x 0.5m x 0.45m) and the closed volumes of the spirals is no where near the predicted close box design figures - mid 3.65l vs 7.5l and woofer 14.14l vs 29l. This would also take a lot of wood and time to make.
So, back to my first plan, an ovoid woofer and spiral mid sealed enclosure. I increased the size slightly (10%) on the mid spiral and came up with this -
An externally hosted image should be here but it was not working when we last tested it.
This is closer on the sealed volumes - mid is 5.5l and the woofer 23.3l. This gives a total Q for the mid of 0.723 and the woofer 0.695. so could go up in size a bit for the mid (5.8l give Q=0.707) and down a bit in size for the Woofer (22.l for Q=0.707).
This is also a more manageable size speaker - approx 0.75 x 0.45 x 0.45
Remembering that any space between the cabinet walls and the fibreglass sealed box shapes can be filled with expanding foam - either rigid or more like flexible.
As a back up I quite like the Jim Holtz Statement Monitor and Finalist designs with the vented woofer box and the open back mid. The rounded box in Linesource's (2nd) picture also caught my eye. So may look at a design uing the Finalist general design with the round 6" pipe open mid driver but incorporated into a curved box using flexiply (4x5mm layers).
Before you spend too much time "reinventing the wheel" - why not dig into the acoustic and production design of the B&W N 800 series? I've heard several model iterations of these since they first debuted in the local dealer's showroom. Cost of the latest generation aside, they're not quite my cuppa tea, but I'd attribute most of that to the implementation of drivers and XO design. As far as the enclosures themselves are concerned, there's a lot to like about them.
Study these carefully and make up your mind after that 😉...
You'll agree (I hope) that a flat cylinder wouldn't exactly be the best shape.
Did I not read these were hypothetical data and NOT physical based upon presumptions........
Earl Geddes may cast a view
.....................................
@Planet10.. duplicated by hand from actual measurements I'm presuming.
@Charlie, even if these were 'pre-computer baffle simulations' they contain enough information to be educational.
Flat from 100-4000, you could invert the non baffle affected response and add it to the result.
@Charlie, even if these were 'pre-computer baffle simulations' they contain enough information to be educational.
Flat from 100-4000, you could invert the non baffle affected response and add it to the result.
Yes. Olson & his crew did an amazing amount of measurements. A lot faster than calculations at the time.
dave
Olson used a small (2 inch dome) source in the exact geometric center of a shape that was ~24" across. Worst case in all measurements. Putting it slightly off center (or having the entire face as the radiating surface) smooths things out quite a bit.
I think the OP is too focused on what is behind the speakers (doesn't really matter much as long as it is damped) and not focused enough on the rest of the design task, which is much more important. Flush mounting, crossover, etc is way more important.
The desire to use fiberglass skills opens up a huge range of possible enclosure shapes.
..and best for learning. I could assume the goal is to smooth the diffraction with rounding, centring the driver and still getting a smooth response.
Thanks all for the comments. What I do know is ....
The more I read the less I know I know!
Anyway, the general wisdom looks like I should stick with something more conventional for a first build. So, been doing a bit of searching and thinking about my requirement. So.
I have a pair of B&W CDM1SE’s which I like but they are quite “polite” and my son has some Yamaha H7’s which are ultra revealing so I am looking for something of a hibrid of these two desktop monitors. Smoother mid like the B&W but with more bass punch and extension and a slightly brighter treble.
Doing a search I have seen the following kits that I quite like
3 way - Jim holt statement monitors and troels grave-son 3 way classic (stepped baffle)
2 way - klang and ton Nada and troels 2way SBA’s
I understand these are optimised (cabinet and crossover) for the particular drivers but have been playing around with box sim and have a couple of decent 25ish closed box solutions.
So, either looking at a closed box or front ported monitor with 8” plus tweeter or 8” plus 4” plus tweeter. I also saw the labyrinth and slot ported boxes that look interesting as well.
So, numpty questions, can I take one of the above boxes e,g, Nada, substitute cheaper drivers (looking at £150ish per speaker) and use box sim to optimise the active crossover (24db/octave) using one amp per driver? Will this get me close enough to then refine with wadding and foams?
Finally any suggestions of driver combination and box designs that would work and out perform my B&W’s and is it even possible to use an 8” in a 2 way system?
The Monacor sph175hq and dt107 looks good or sph22Ohq, sph145hq and dt107 for a 3 way.
Or for a higher end 2 way sph220hq plus sb29rdc.
All comments welcome.
The more I read the less I know I know!
Anyway, the general wisdom looks like I should stick with something more conventional for a first build. So, been doing a bit of searching and thinking about my requirement. So.
I have a pair of B&W CDM1SE’s which I like but they are quite “polite” and my son has some Yamaha H7’s which are ultra revealing so I am looking for something of a hibrid of these two desktop monitors. Smoother mid like the B&W but with more bass punch and extension and a slightly brighter treble.
Doing a search I have seen the following kits that I quite like
3 way - Jim holt statement monitors and troels grave-son 3 way classic (stepped baffle)
2 way - klang and ton Nada and troels 2way SBA’s
I understand these are optimised (cabinet and crossover) for the particular drivers but have been playing around with box sim and have a couple of decent 25ish closed box solutions.
So, either looking at a closed box or front ported monitor with 8” plus tweeter or 8” plus 4” plus tweeter. I also saw the labyrinth and slot ported boxes that look interesting as well.
So, numpty questions, can I take one of the above boxes e,g, Nada, substitute cheaper drivers (looking at £150ish per speaker) and use box sim to optimise the active crossover (24db/octave) using one amp per driver? Will this get me close enough to then refine with wadding and foams?
Finally any suggestions of driver combination and box designs that would work and out perform my B&W’s and is it even possible to use an 8” in a 2 way system?
The Monacor sph175hq and dt107 looks good or sph22Ohq, sph145hq and dt107 for a 3 way.
Or for a higher end 2 way sph220hq plus sb29rdc.
All comments welcome.
Part of this can come down to their response, which to some degree might be equalised into any reasonable speaker.. which means to some degree, your choice of speaker comes down to its capabilities, eg: how loud, extension, as well as size, cost.