In for a penny in for a pound
You really need to ask?????
😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀
Of course I'm building it!!
And costs can be kept down depending on what you the builder can do. Remember a medium sized drill press, a fly cutter and some patience in layout and clamping will cut the holes in the square tubing. If all you need done out of your shop is the pole piece then you are laughing.
Mark
You really need to ask?????
😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀 😀
Of course I'm building it!!
And costs can be kept down depending on what you the builder can do. Remember a medium sized drill press, a fly cutter and some patience in layout and clamping will cut the holes in the square tubing. If all you need done out of your shop is the pole piece then you are laughing.
Mark
I'm wanting to build one, sure!RHosch said:
But then again, I have this fun little fantasy that's WAAAAY outside the scope of this application.. 😀
I thought it would massively fun to be able to enter car audio sound quality competitions, with absolutely no commercial products in the car at all...
...home-made ESL panels for both doors, and the dashboard top surface (R, center, L)...
...home-made tube amplifiers, or better yet, gainclones, to power them...
...and something like this, or my home-made servo project, in the trunk as a subwoofer...
As far as unleashing this in my house, I'd be seriously concerned about cracked drywall and neighbors calling the police...
For some odd reason, I don't have the same concerns, inside my car.
But, just for the heck of it, grasp this:
If we are talking about 120dB at 20 hz... when you factor in a vehicle's "cabin gain" (room gain), you end up somewhere in the (brace yourself) 168dB region (assuming 12dB/octave gain, starting at 160hz).
OK, maybe I should change that to "SPL competitions". 😀
Maybe I'd just need a "mini" version. 😀
Alright, sorry to get off-topic there...
I think Dan W. and Bill F. know that I've got a little collection of interesting subwoofer drivers going at my house... either visualy interesting, or unique technology... I think this would make a fine addition! 😎
I think that's a reasonable target figure (same in £'s) based on my preliminary estimates. I think all steel can be had for sub £200, possibly much less if this square tubing works out. 32 N45 24mm OD x 10mm approx £80. Can someone put an estimate on baffle, spider, former?
Then there's small amounts for mdf perhaps for support, screws, glue, enamelled wire etc. I personally will attempt to wind my own former, and assume to use some friendly contacts to do any metal machining.
I think the half grand figure is a good starting budget.
OK, then we need estimates from someone on diaphram, spider, and VC costs. We have to settle the question of one or two spiders and diaphrams before making that estimate. We also need to commit to a material system for the diaphram. A composite sandwich seems like a good option, but if we go with two spaced diaphrams an aluminum sandwich would be easier to work with.
Costs
Diaphram is going to cost however you want it to.
Foam cored composite?? Or whatever. That one is up to the builder unless one of us wants to create a partial kit.
Voice coil??? Custom to say the least!! I haven't ordered a coil in six years so my contacts are gone. I'll try my hand at winding or having one wound. It depends on the availability of CCA wire.
The spider is the fly in the ointment. It might have to be custom. It must be a large diameter to allow for the required X-max. Plus it must be stiff at rest and have a progressive compliance. Not an off the shelf product.
As for the cool coil springs on the Parthenon I'm guessing that they are more for centering and excursion control than for alignment.
Either we go for a guided DDD approach with the spring coils or try to get one big spider!
Mark
Diaphram is going to cost however you want it to.
Foam cored composite?? Or whatever. That one is up to the builder unless one of us wants to create a partial kit.
Voice coil??? Custom to say the least!! I haven't ordered a coil in six years so my contacts are gone. I'll try my hand at winding or having one wound. It depends on the availability of CCA wire.
The spider is the fly in the ointment. It might have to be custom. It must be a large diameter to allow for the required X-max. Plus it must be stiff at rest and have a progressive compliance. Not an off the shelf product.
As for the cool coil springs on the Parthenon I'm guessing that they are more for centering and excursion control than for alignment.
Either we go for a guided DDD approach with the spring coils or try to get one big spider!
Mark
As regards the sandwich type diaphragm... it doesn't need to be flat, to be a sandwich... and if we don't care about mounting depth (I don't think we do), we can easily fabricate a curved sandwich diaphragm.
Vacuum bagging is easy and relatively inexpensive.
Not to bring up a familiar link... but if you check out my servo project page and scroll down to my comments on "Update: 03-29-2003", you can see the vinyl foam core that I used in my vane assembly...
And if you scroll down to "Update: 4-16-2003", I walk through the vacuum bagging process.
It's quite easy, really.. and the cool thing is, you don't need to make a flat shape, it can be contoured any way, and the vacuum will draw it into that shape... much like a vacuum forming table does with sheet acrylic.
If we did want to do a flat panel though, take a look at my panel on that page though also... since I did fabricate a flat panel sandwich diaphragm there. I cut channels into mine, that we could consider for our diaphragm as well, IMO. It stiffens the end product, much in the same way that the shape of an I-beam gives it strength.
What do you think?
Vacuum bagging is easy and relatively inexpensive.
Not to bring up a familiar link... but if you check out my servo project page and scroll down to my comments on "Update: 03-29-2003", you can see the vinyl foam core that I used in my vane assembly...
And if you scroll down to "Update: 4-16-2003", I walk through the vacuum bagging process.
It's quite easy, really.. and the cool thing is, you don't need to make a flat shape, it can be contoured any way, and the vacuum will draw it into that shape... much like a vacuum forming table does with sheet acrylic.
If we did want to do a flat panel though, take a look at my panel on that page though also... since I did fabricate a flat panel sandwich diaphragm there. I cut channels into mine, that we could consider for our diaphragm as well, IMO. It stiffens the end product, much in the same way that the shape of an I-beam gives it strength.
What do you think?
I think two diaphrams would greatly help in VC alignment, regardless of whether we went with a DDD approach or not. Without a wide spaced dual spider, a central guid rod will be a must.
With the length of the motor, a flat diaphram would really help cut down on length. A curved diaphram will have to have significant depth over a 36" or so diameter/width to add enough rigidity to be worth the effort. That's one reason I suggested the dual diaphram... same or better stiffnes/weight ratio as a deep cone or curved diaphram, but less overall room required (thickness of one extra panel + 6" for excursion clearance) compared to the depth of a traditional cone. It's the same principle of going with a non-flat diaphram... just not a traditional way of accomplishing that. It's essentially making a really deap diaphram structure, and placing the motor assembly in the middle of the diaphram sandwich. Perfect symmetry for a dipole. 🙂
Regardless, I think you are right in that the diaphram design can be left up to each individual builder. We still need some reasonable weight estimates though to go further with motor design. I'll aim for a target of 36" wide diaphram and see what it takes to get modes >100Hz and deflections that seem reasonable under the anticipated accelerations.
Carbon or Glass fiber? Makes a big difference.
With the length of the motor, a flat diaphram would really help cut down on length. A curved diaphram will have to have significant depth over a 36" or so diameter/width to add enough rigidity to be worth the effort. That's one reason I suggested the dual diaphram... same or better stiffnes/weight ratio as a deep cone or curved diaphram, but less overall room required (thickness of one extra panel + 6" for excursion clearance) compared to the depth of a traditional cone. It's the same principle of going with a non-flat diaphram... just not a traditional way of accomplishing that. It's essentially making a really deap diaphram structure, and placing the motor assembly in the middle of the diaphram sandwich. Perfect symmetry for a dipole. 🙂
Regardless, I think you are right in that the diaphram design can be left up to each individual builder. We still need some reasonable weight estimates though to go further with motor design. I'll aim for a target of 36" wide diaphram and see what it takes to get modes >100Hz and deflections that seem reasonable under the anticipated accelerations.
Carbon or Glass fiber? Makes a big difference.
Been Vacuum Bagging for over 10 years Bud
Stiffness to weight ratio of the two competitors?
Hyperbolic curve is a must. Without it we will be introducing allot of doppler related distorsion which if I remember right is odd order. Not wanted.
And yep I plan on punching some holes so the two sides can meet too.
Mark
Stiffness to weight ratio of the two competitors?
Hyperbolic curve is a must. Without it we will be introducing allot of doppler related distorsion which if I remember right is odd order. Not wanted.
And yep I plan on punching some holes so the two sides can meet too.
Mark
I'm not sure I understand your question. Which two "competitors" were you referring to? Single/dual... flat/curved?... carbon/glass?
How does a flat diaphram with equal stiffness to a curved one introduce doppler distortion? Isn't that a function of the maximum diaphram velocity and the upper end of the frequency range being asked of it?
The more I think about it, the more concerned I become about a single diaphram, for the following reasons (and yes, I know Adire used a single diaphram in the demo version, but that was in combination with a guide rod):
(1) Cantilevered diaphram weight. With the diaphram weight hanging off one end of the former piece, there will be a significant tendancy for the voice coil to rock. Even wide spaced spiders might not be sufficient to control the motion. I'm afraid that we'll wind up with either a central guide rod solution (which might not be so bad), or significantly wider gaps to allow for this rocking and misalignment (bad).
(2) High moment on diaphram/former connection. With only the central portion of the diaphram to make connection with, a very secure attachment method will have to be developed. This especially concerns me with such a large diaphram. The force caused by accelerating air should be symmetrical (top/bottom and left/right), but any irregularities in the diaphram construction will cause non-symmetric forces during operation. Not only will that contribute to the already problematic VC rocking, but it will also put a considerable moment on the former/diaphram connection, which will tend to cause the diaphram to pitch one way or another... further increasing VC rocking. If the VC doesn't scrape, or the spider give out, then the connection might.
Perhaps it will be just fine... but all in all I think there are better ways to skin a cat. If there were two diaphrams, all that would be needed is essentially a pinned connection at the center of each. The diaphram/diaphram stiffening connections will easily control diaphram pitching problems, as the number of connections through which moments are passed can be increased by an order of magnitude (and placed where their resistance to torque are the highest). Mirror the diaphrams and balance the static weight, spread the spiders for higher VC control, and connect the two diaphrams to combat former connection issues and diaphram asymmetry problems.
Bah, I don't think my technical explanations are clear. Maybe I need to make some sketches to illustrate my point.
How does a flat diaphram with equal stiffness to a curved one introduce doppler distortion? Isn't that a function of the maximum diaphram velocity and the upper end of the frequency range being asked of it?
The more I think about it, the more concerned I become about a single diaphram, for the following reasons (and yes, I know Adire used a single diaphram in the demo version, but that was in combination with a guide rod):
(1) Cantilevered diaphram weight. With the diaphram weight hanging off one end of the former piece, there will be a significant tendancy for the voice coil to rock. Even wide spaced spiders might not be sufficient to control the motion. I'm afraid that we'll wind up with either a central guide rod solution (which might not be so bad), or significantly wider gaps to allow for this rocking and misalignment (bad).
(2) High moment on diaphram/former connection. With only the central portion of the diaphram to make connection with, a very secure attachment method will have to be developed. This especially concerns me with such a large diaphram. The force caused by accelerating air should be symmetrical (top/bottom and left/right), but any irregularities in the diaphram construction will cause non-symmetric forces during operation. Not only will that contribute to the already problematic VC rocking, but it will also put a considerable moment on the former/diaphram connection, which will tend to cause the diaphram to pitch one way or another... further increasing VC rocking. If the VC doesn't scrape, or the spider give out, then the connection might.
Perhaps it will be just fine... but all in all I think there are better ways to skin a cat. If there were two diaphrams, all that would be needed is essentially a pinned connection at the center of each. The diaphram/diaphram stiffening connections will easily control diaphram pitching problems, as the number of connections through which moments are passed can be increased by an order of magnitude (and placed where their resistance to torque are the highest). Mirror the diaphrams and balance the static weight, spread the spiders for higher VC control, and connect the two diaphrams to combat former connection issues and diaphram asymmetry problems.
Bah, I don't think my technical explanations are clear. Maybe I need to make some sketches to illustrate my point.
Choices of words at the end of a long day at work NUTS
The "choices" were glass fiber and carbon fiber.
I agree in principle about the two diaphrams. They can produce a greater spl. And they will provide greater balance when the driver is in motion. But to do so they must be stiff. Hence the curved idea.
The term doppler distortion is not the best. It describes only a ( poorly at that ) byproduct of cone flexture. The problem lies in when a cone wants to stay in motion when we don't want it to. It will keep on singing it's own tune because it remains moving when there is no signal or while there is a signal. The later is when we are getting into resonance modes. Not what we want. A curved wall diaphram is known to exibit greater stiffness than a flat one. ( no kidding that's why we have curved loudspeaker diaphrams )
RH your post on the advantages of two diaphrams are on the money. So why not. Two diaphrams. Thirty six inches in diameter. A curve to be what, tractrix so that we all could refer to a known shape. ( any other ideas? ) Thicknes as per original ideas ( in a nut shell 6mm of foam and then sandwiched in between skin of ........ Enliighten me on the benefits of our two most common fibers.
A small thought on the project. We are going where no sane people go! We are studying drivers that are hard to make. Have a limited market and are protected by patents. ( pending or issued ) The DDD ( I hope I have it right ) motor idea is smart. Our application of the dual diaphram is also well grounded in good engineering ( pat on the back RH ) The dual spider or even the concept of dual symetrical suspension is also briliant.
So in the works is an uber driver. Over six inches of throw and argueably the only thing short of a coffin sized box of 15" drivers that will create concert levels of any instrument that you can think of.
WOW😀
I tell people about this thread from time to time. Old clients eyes glow and the odd one asks some good questions, stimulates the thinking process. It's truly a pleasure to be a part of this little adventure. And my hats off to those that are contributing so much of their thoughts and time.
Lets get this baby built
Mark
The "choices" were glass fiber and carbon fiber.
I agree in principle about the two diaphrams. They can produce a greater spl. And they will provide greater balance when the driver is in motion. But to do so they must be stiff. Hence the curved idea.
The term doppler distortion is not the best. It describes only a ( poorly at that ) byproduct of cone flexture. The problem lies in when a cone wants to stay in motion when we don't want it to. It will keep on singing it's own tune because it remains moving when there is no signal or while there is a signal. The later is when we are getting into resonance modes. Not what we want. A curved wall diaphram is known to exibit greater stiffness than a flat one. ( no kidding that's why we have curved loudspeaker diaphrams )
RH your post on the advantages of two diaphrams are on the money. So why not. Two diaphrams. Thirty six inches in diameter. A curve to be what, tractrix so that we all could refer to a known shape. ( any other ideas? ) Thicknes as per original ideas ( in a nut shell 6mm of foam and then sandwiched in between skin of ........ Enliighten me on the benefits of our two most common fibers.
A small thought on the project. We are going where no sane people go! We are studying drivers that are hard to make. Have a limited market and are protected by patents. ( pending or issued ) The DDD ( I hope I have it right ) motor idea is smart. Our application of the dual diaphram is also well grounded in good engineering ( pat on the back RH ) The dual spider or even the concept of dual symetrical suspension is also briliant.
So in the works is an uber driver. Over six inches of throw and argueably the only thing short of a coffin sized box of 15" drivers that will create concert levels of any instrument that you can think of.
WOW😀
I tell people about this thread from time to time. Old clients eyes glow and the odd one asks some good questions, stimulates the thinking process. It's truly a pleasure to be a part of this little adventure. And my hats off to those that are contributing so much of their thoughts and time.
Lets get this baby built
Mark
Could someone explain what DD and XBL^2 technology is, and importantly what we can't copy from them. Is our motor in its current form using (and thus potentially breaking patents in) both technologies?
Do both apply regardless of the number of plates/gaps?
Do both apply regardless of the number of plates/gaps?
Re the spider:
I still think that the 4 rolling flexible spring steel strips might work precisely enough. The advantages are:
almost no friction, long travel, probably will center former precisely(keep in mind that it is resisting forces normal (perpindicular) to the flexing direction. So the sideways forces are resisted by say a 4" wide piece of thin steel.
Problems are that there is no progressive spring action, but this can be supplied as the Parthenon does it or with long coil springs parallel to the former.
I just can't see how a conventional spider will work with these extensions. I'm certainly open to other appproaches, but the sliding rods are about the only other thing that might work I think, but they have their own possible problems of friction, noise, wear, precision, etc.
I still think that the 4 rolling flexible spring steel strips might work precisely enough. The advantages are:
almost no friction, long travel, probably will center former precisely(keep in mind that it is resisting forces normal (perpindicular) to the flexing direction. So the sideways forces are resisted by say a 4" wide piece of thin steel.
Problems are that there is no progressive spring action, but this can be supplied as the Parthenon does it or with long coil springs parallel to the former.
I just can't see how a conventional spider will work with these extensions. I'm certainly open to other appproaches, but the sliding rods are about the only other thing that might work I think, but they have their own possible problems of friction, noise, wear, precision, etc.
Vikash posted:
DDD (what it's called by Harman) was invented by HyperDynamics. It's the technique of using the return flux in a motor. You typically have two gaps and two voice coils. The flux in each gap is oriented a different way. Efficient use of flux, well balanced, but because of two gaps the flux in the motor as a whole tends to be a bit low.
XBL^2 is the technique of using one or more voice coils transversing two or more same-polarity magnetic gaps. The flux in each gap is in the same direction. This also yields a well balanced (flat) BL curve, and can be extended as far as you want to go. Also quite a bit cheaper to build, because you don't need more than one winding, or if you use multiple you don't have to counter-wind your voice coils (if using more than one)
XBL^2 can be combined with DDD, as in some of the earlier drawings. Use of the return flux (DDD) with multiple gaps (XBL^2).
Dan Wiggins
Adire Audio
DDD (what it's called by Harman) was invented by HyperDynamics. It's the technique of using the return flux in a motor. You typically have two gaps and two voice coils. The flux in each gap is oriented a different way. Efficient use of flux, well balanced, but because of two gaps the flux in the motor as a whole tends to be a bit low.
XBL^2 is the technique of using one or more voice coils transversing two or more same-polarity magnetic gaps. The flux in each gap is in the same direction. This also yields a well balanced (flat) BL curve, and can be extended as far as you want to go. Also quite a bit cheaper to build, because you don't need more than one winding, or if you use multiple you don't have to counter-wind your voice coils (if using more than one)
XBL^2 can be combined with DDD, as in some of the earlier drawings. Use of the return flux (DDD) with multiple gaps (XBL^2).
Dan Wiggins
Adire Audio
Couple of things...
- Keep an eye on mass and Fs. Remember that the higher the Fs for a given mass, the stiffer the suspension, which means the less reason you need a guide rod. The reason we had the guide rod wasn't because it's always needed, it was because the Mms was ~800 grams, AND the Fs was down in the 6 Hz range. REALLY soft suspension. So in this case mass is a good thing, because it makes the suspension stiffer and better able to support the diaphragm. Of course, it also kills efficiency...
- Doppler distortion is from cone velocity (like the sound of a siren when it passes) shifting the reproduced signal. It is, IIRC, all even order, and thus from a subjective standpoint quite benign in terms of IMD components.
Cone flex/breakup modes are completely unrelated to the spectrum (they can be excited by IMD or THD components, for example). "Cone cry" is a definite issue with a wideband driver, but for a simple sub - operating down below 100 Hz - it's not going to be too much of a problem. Rigidity for a decent lifespan will be more critical, but proper selection of material for the diaphragm and bracing/reinforcement on that unit will be paramount.
- Price. I wish I could get my labor that cheap! Any machinists looking for a free summer intership? 😉 Also, if at all possible, I'd strongly suggest looking for 1010 grade steel at the minimum (1008 or 12L14 also work). It's a good step up from 1018. And of course the larger the motor/more complex, the higher the price. If it were me, I'd trade off some size/complexity and increase the quality of the steel used, especially since I'd have plenty of stroke with either approach.
- Tolerancing. Rather than use pre-built structural steel, I'd look at square flat panels and welding. Curvature can be a MAJOR problem here, especially if you want to maintain any kind of tolerance. Otherwise everything will have to be cut much wider to account for assembly. I spec as loose of a tolerance as I can, because tighter costs money. Even then, I had to go to 0.007" TIR across the plates to properly maintain the gap widths.
- Voice coils. Sticking with standard 3" or 4" units is highly recommended, because they can be purchased from a wide variety of sources. In fact, how about designing the motor geometry around a standard pro audio voice coil (like one of the big JBLs) so that it can be sourced from just about any reconer?
Tolerance in winds of the voice coils is a serious concern in production, and what may look "good enough" can have some pretty significant impact on the actual performance of the product. And with the performance mentioned in this thread (4-6" peak to peak) it can be accomplished with a standard 2" long voice coil.
Dan Wiggins
Adire Audio
- Keep an eye on mass and Fs. Remember that the higher the Fs for a given mass, the stiffer the suspension, which means the less reason you need a guide rod. The reason we had the guide rod wasn't because it's always needed, it was because the Mms was ~800 grams, AND the Fs was down in the 6 Hz range. REALLY soft suspension. So in this case mass is a good thing, because it makes the suspension stiffer and better able to support the diaphragm. Of course, it also kills efficiency...
- Doppler distortion is from cone velocity (like the sound of a siren when it passes) shifting the reproduced signal. It is, IIRC, all even order, and thus from a subjective standpoint quite benign in terms of IMD components.
Cone flex/breakup modes are completely unrelated to the spectrum (they can be excited by IMD or THD components, for example). "Cone cry" is a definite issue with a wideband driver, but for a simple sub - operating down below 100 Hz - it's not going to be too much of a problem. Rigidity for a decent lifespan will be more critical, but proper selection of material for the diaphragm and bracing/reinforcement on that unit will be paramount.
- Price. I wish I could get my labor that cheap! Any machinists looking for a free summer intership? 😉 Also, if at all possible, I'd strongly suggest looking for 1010 grade steel at the minimum (1008 or 12L14 also work). It's a good step up from 1018. And of course the larger the motor/more complex, the higher the price. If it were me, I'd trade off some size/complexity and increase the quality of the steel used, especially since I'd have plenty of stroke with either approach.
- Tolerancing. Rather than use pre-built structural steel, I'd look at square flat panels and welding. Curvature can be a MAJOR problem here, especially if you want to maintain any kind of tolerance. Otherwise everything will have to be cut much wider to account for assembly. I spec as loose of a tolerance as I can, because tighter costs money. Even then, I had to go to 0.007" TIR across the plates to properly maintain the gap widths.
- Voice coils. Sticking with standard 3" or 4" units is highly recommended, because they can be purchased from a wide variety of sources. In fact, how about designing the motor geometry around a standard pro audio voice coil (like one of the big JBLs) so that it can be sourced from just about any reconer?
Tolerance in winds of the voice coils is a serious concern in production, and what may look "good enough" can have some pretty significant impact on the actual performance of the product. And with the performance mentioned in this thread (4-6" peak to peak) it can be accomplished with a standard 2" long voice coil.
Dan Wiggins
Adire Audio
Thanks for the comments Dan.
That was one consideration that went into my flat-stacked plate design. Whether the plates are welded or machined for bolted assembly, it may be possible to finish bore the gaps as an assembly. Might not be a standard machine shop setup to do a barrel bore like that, but a possibility. Tolerance would progressively increase for each individual plate (or square tubing section...), but total cylindrical tolerance would be much better for the assembly.
Just a thought. Of course, the requirement for accurate alignment of VC to gap is still there, but that might well be handled through manual alignment after assembly. No need to rely on machining tolerance stackup through multiple pieces... just shim properly, and ream the alignment holes afterwards.
DanWiggins said:
That was one consideration that went into my flat-stacked plate design. Whether the plates are welded or machined for bolted assembly, it may be possible to finish bore the gaps as an assembly. Might not be a standard machine shop setup to do a barrel bore like that, but a possibility. Tolerance would progressively increase for each individual plate (or square tubing section...), but total cylindrical tolerance would be much better for the assembly.
Just a thought. Of course, the requirement for accurate alignment of VC to gap is still there, but that might well be handled through manual alignment after assembly. No need to rely on machining tolerance stackup through multiple pieces... just shim properly, and ream the alignment holes afterwards.
Of bits and pieces stuck together.
Dan thanks for some of the clarifications.
I don't completely agree that the even order distorsion is not that bad. I don't like it! And I can hear it and identify it every time I hear it.
Cone break up in the low bass is not generally a problem, but if and or when it happens it will effect the midrange.
And yep your doppler distortion explanation is correct. I tried to retract the goof up I posted. Thanks
Where we differ in ideas is why try to weld up steel plate when the only areas that we need to a specific tolerance is the holes through the square sections?
RH a bore through the stack of plates that you made in one of your renderings is possible and it will keep it's tolerance as a group. We need to model that motor setup to see if it will work.
The idea of building around an existing coil was sugested and I whole heartedly accept it. Dan what did you use??? ( approximately? ) How about a group buy???
Different places will sample if you are a legitamit business. The former will be fun too. I'm wondering if the whole assembly of voice coils and former will end up becoming a custom job.
Mark
Dan thanks for some of the clarifications.
I don't completely agree that the even order distorsion is not that bad. I don't like it! And I can hear it and identify it every time I hear it.
Cone break up in the low bass is not generally a problem, but if and or when it happens it will effect the midrange.
And yep your doppler distortion explanation is correct. I tried to retract the goof up I posted. Thanks
Where we differ in ideas is why try to weld up steel plate when the only areas that we need to a specific tolerance is the holes through the square sections?
RH a bore through the stack of plates that you made in one of your renderings is possible and it will keep it's tolerance as a group. We need to model that motor setup to see if it will work.
The idea of building around an existing coil was sugested and I whole heartedly accept it. Dan what did you use??? ( approximately? ) How about a group buy???
Different places will sample if you are a legitamit business. The former will be fun too. I'm wondering if the whole assembly of voice coils and former will end up becoming a custom job.
Mark
Lets figure out Mms
RH:
How much does a square cm of properly laidup carbon fiber weight??? Given your experience in the field I would prefer your ( or Magura ) thoughts on the idea. Thread count as well, double layered at what 45 degrees??? Any other ideas.
Once we have this we can figure out some other stuff. We need a standard voice coil. I'll do some searching around or if someone has please speakup. The former could be made from ..... hold on boys a big momma beer can ( I always knew that they had a second use ) No serious. Check them out. It could work. Clean off the varnish and if you want to anodise it or varnish it again heck paint it flat black. So we can easily weigh the can of choice once we have the happy voice coil worked out. So a searching I will go.
Mark
RH:
How much does a square cm of properly laidup carbon fiber weight??? Given your experience in the field I would prefer your ( or Magura ) thoughts on the idea. Thread count as well, double layered at what 45 degrees??? Any other ideas.
Once we have this we can figure out some other stuff. We need a standard voice coil. I'll do some searching around or if someone has please speakup. The former could be made from ..... hold on boys a big momma beer can ( I always knew that they had a second use ) No serious. Check them out. It could work. Clean off the varnish and if you want to anodise it or varnish it again heck paint it flat black. So we can easily weigh the can of choice once we have the happy voice coil worked out. So a searching I will go.
Mark
Re: Lets figure out Mms
Composite mass is more easily handled in terms of volume. Assuming roughly ~65% fiber to resin volume (indicative of a vacuum assisted resin transfer system or a really good hand-layup and vacuum bagging), here are the values I typically use to begin a design task:
E-glass/vinylester: 1.88 gram/cm^3
Carbon/epoxy: 1.58 gram/cm^3
Those figures apply to balanced and unbalanced weaves, as well as unidirectional cloths and tapes. Cloth thicknesses for either glass or carbon range from around .010" to .025" per layer. Thinner and thicker cloths can be found, but aren't going to to be applicable for what we're doing.
I'd suggest something in the .015" to .020" range.
Still can't calculate the diaphram weight until you target a core thickness and stiffness goal.
mwmkravchenko said:
Composite mass is more easily handled in terms of volume. Assuming roughly ~65% fiber to resin volume (indicative of a vacuum assisted resin transfer system or a really good hand-layup and vacuum bagging), here are the values I typically use to begin a design task:
E-glass/vinylester: 1.88 gram/cm^3
Carbon/epoxy: 1.58 gram/cm^3
Those figures apply to balanced and unbalanced weaves, as well as unidirectional cloths and tapes. Cloth thicknesses for either glass or carbon range from around .010" to .025" per layer. Thinner and thicker cloths can be found, but aren't going to to be applicable for what we're doing.
I'd suggest something in the .015" to .020" range.
Still can't calculate the diaphram weight until you target a core thickness and stiffness goal.
Check out my previous post, regarding the kevlar/foam diaphragm that I built for my servo project. 😉
Vacuum-bagging can be used to produce ultra-lightweight kevlar and carbon fiber (and fiberglass) parts, as it wicks out the excess resin, resulting in not only ultra-light, but also ultra-strong parts.
The vacuum also can be used like vacuum-forming, shaping your part over a mold, with a force up to several hundred pounds per square inch. So a curved or cone diapragm could be made, with a shape to mold to.
Although my diaphragm was flat, you can see that I had cut grooves into the foam...
...and when the kevlar/foam/kevlar sandwich was set up and vacuum pulled as it cured, the kevlar was pulled together into these grooves, giving my flat diaphragm strength under the same principles that give an I-beam strength.
That particular diaphragm is about 36" x 11", and not counting the steel shaft at the center, weighs about 0.75 lb after being trimmed to shape.
Vacuum-bagging can be used to produce ultra-lightweight kevlar and carbon fiber (and fiberglass) parts, as it wicks out the excess resin, resulting in not only ultra-light, but also ultra-strong parts.
The vacuum also can be used like vacuum-forming, shaping your part over a mold, with a force up to several hundred pounds per square inch. So a curved or cone diapragm could be made, with a shape to mold to.
Although my diaphragm was flat, you can see that I had cut grooves into the foam...
...and when the kevlar/foam/kevlar sandwich was set up and vacuum pulled as it cured, the kevlar was pulled together into these grooves, giving my flat diaphragm strength under the same principles that give an I-beam strength.
That particular diaphragm is about 36" x 11", and not counting the steel shaft at the center, weighs about 0.75 lb after being trimmed to shape.
A composite sandwich doesn't need grooves to be stiff... the geometry of the skin/core/skin already follows the principle that gives an I-beam stiffness. Though connecting the two skins coming into contact can do good things for shear strength...
Vacuum bagging produces a maximum ~14.7 psi of force... atmospheric pressure.
The figures I gave above should be achievable for small parts where there are few problems in the bagging process. Still need to decide on glass or carbon (kevlar is light and strong, but not stiff... which is what we really need for a diaphram). Simply a matter of cost there... carbon is better but more expensive.
Vacuum bagging produces a maximum ~14.7 psi of force... atmospheric pressure.
The figures I gave above should be achievable for small parts where there are few problems in the bagging process. Still need to decide on glass or carbon (kevlar is light and strong, but not stiff... which is what we really need for a diaphram). Simply a matter of cost there... carbon is better but more expensive.
- Status
- Not open for further replies.