kipman725

the GAS COMP 12D1 is a 12" with a Qts of 0.2 and presumably high excursion capabilities: https://www.gascaraudio.com/produkter/comp12d1?lang=en#1462284074811-88c8730f-f38bd1d2-1d988827-f5a0 Also this was brought to my attention: https://www.ground-zero-audio.com/en/produkt/gznw-38neo-spl/ BL^2/Re = 1043 Regarding the MTX: using 16 of them you would not be able to get enough power from a domestic supply. I actually do have 16*15" drivers and the next upgrade for my system is to increase the efficiency by moving away from sealed boxes as I can't sensibly increase the amount of amplifier power. Another issue with running a lot of small boxes is that the cabling gets super annoying to setup and tear down.

I managed to perform some measurements using a sharp GP2Y0A51SK0Fdistance measurement sensor and a scope. The distance sensor samples at ~60Hz asynchronous to the signal gen so I used 16 average mode on the scope (upper trace is distance sensor). I drove the driver at 5 Hz with a cloud VTX1200. I have also taped white paper to the cone to get a better reflectivity. The distance from the sensor can be found using the following equation (from published curve): z = (V/3.76)^(-1/0.814). From this I found a difference of 1.5 cm between the minimum and maximum value which is ~ xmax (8mm one way). Obviously easy way to improve the results would be to get the sensor closer to the cone into its more sensitive region. I also noticed some asymmetry in the clipping when not driven so hard presumably from gravity. Quite interesting results could be obtained with some PC data logging and a current sensor as it would be possible to plot the displacement Vs coil current and also the back EMF.

Alpine do a quite interesting 12" (SWR-12D2) that claims Qts = 0.44 and a one way xmax of 20 mm: https://www.alpine.co.uk/fileadmin/user_upload/manuals//car_audio_manuals/SWR_10D2_4_12D2_4/OM_SWR-12D4_EN.pdf not too expensive but needs a bit more motor to be your dream driver.

BC218 is the first time I have seen a prosound sub sold on how big it is. (to get the directivity) https://hal.archives-ouvertes.fr/hal-00462380/document Consider figure 15. of Performances and Design of Ironless Loudspeaker Motor Structures. Reduce distance e to 0 resulting in maximal magnetic flux but over a very small depth. However compensate this by repeating the motor structure and using voice coils of depth J. Put hall effect sensors in the centre of the coils or measure linear displacement of the whole voice coil assembly to work out the current direction in each coil. Use acceleration feedback to control the coil current to make the whole system linear. A kind of brush less DC linear motor voice coil that would only be limited by suspension performance. Also no iron so reduced mass and higher peak flux density possible.

I also got into a discussion about optical methods of measuring excursion which some people are doing already for driver protection. The sensors suitable for higher frequencies are quite expensive but at a 60Hz sample rate there is a sharp sensor that looks quite easy to deal with: http://www.sharp-world.com/products/device/lineup/data/pdf/datasheet/gp2y0a51sk_e.pdf could output directly into a sound card and use the computer to calculate the peak to peak excursion.

Sure if your into folk music. I spent Friday night stood next to a quad stack of this: http://voidacoustics.com/docs/products/Stasys_X_V2.shtml Full body experience.

In terms of voltage I have used SiC modules rated at 1200V hard switched at 100khz with decent efficiency so it should be posible to make ~1000v rms amplifiers.

I actually got as far as phoning up a company that made piezo actuators once when I saw they they were offering actuators that could do mm movements. My thoughts where that they would be ideal for the distributed motor idea. As power is not required to hold position against zero presure they could potentially make very large surface area actuators (something like an air motion transformer) and be efficient at bass frequencies. Unfortunately I found out such actuators where about 8000euro per actuator, which put a stop to the idea of using 100's of them. It would be interesting to see a loss breakdown of something like an IPAL driver in a sealed box, I presume that most of the losses are 'copper' (yes I know that coil is alu on the B&C) losses in the coil. The other losses like suspension and air compression/expansion must be small otherwise they would get hot. This would imply that significant improvements in efficiency without radical change would only be possible via: 1. Lighter moving parts 2. more conductive coils (I come from the wireless power world.... everyone wants more conductive materials, its not going to happen, at least at sub MHz frequencies) 3. Better magnetic materials (sometimes there is slow improvement here, slightly improved neo grades etc...) I have just thought of one improvement for situations which arn't excursion limited. Simply double up the motor on both sides of the coil. Should give twice the power handling, twice the motor force and have slightly less moving mass gain than using 2x the number of drivers. It should also improve linearity a little, as one motor is going in the other goes out. Probably would mean the spiders could be less stiff as well as there isn't the pivot point that normally exists.

I think the ultimate driver would be designed for a sealed box avoiding these port compression issues. To generated good SPL it would need a large surface area and excursion capability. To do this we would have to move beyond single voice coil motors and suspensions systems. I have had several ideas which have also been in part explored by others. Ideas for increased excursion: 1. No front surround instead the driver is a piston inside a tube with a piston ring style seal. 2. Motor is linear motor BLDC type using closed loop position control. I have attached a paper about a motor designed for vibration damping train carriages, perhaps a smaller version could work. Linear bearings replace the spider. 3. In the most advanced case no bearings would be needed as a magnetic bearing would be used. Ideas for increased surface area: 1. Distributed motor drivers, instead of pushing the cone in one place push it in many places allowing a thinner larger flat diaphragm. This way one face of the box could be almost all driver. If we look at something like the Maul it can do 134.6 dB @ 20Hz/1m and has a frontal area equivalent to a 40" driver to equal this with a sealed box where the whole of the front was a driver would require an xmax of 93 mm. However the moving mass of such a box would necessitate dual opposed operation in which case this would be halved. applsci-07-00152.pdf

I am fully in agreement Lilmike that this is the best test as after all xmax is only the way we generate the SPL. I just lack suitable facilities as I only have a small outside area at my house (5m x 1m) and get too many reflections/objects resonating that push peaks above the limits of my mic (130dB) at power levels that are below any visible power compression. So the idea was that by verifying the xmax I would have a better idea of the system capabilities. I should be able to perform power compression tests by late August as I'm doing an outdoor event that has electrical power and am planning to allow myself a leisurely setup time span to perform such tests. For context I live in the UK where houses are tiny and all the space is used by someone so only the wealthy/old tend to have large gardens.

The mini-dsp could also be for the main speaker crossover? I would recommend getting an old pro dsp and the non-dsp version of the NU3000 if its much cheaper than the DSP version. I use a symetrix symnet 8x8 it was only £100, such devices support limiters on their outputs. The usual procedure is to turn the sub amp gain to maximum and use attenuation on the DSP output to level match, then set limiters to the maximum voltage swing you want. Doing it this way also allows more advanced limiting to be used; for example short term excursion and longer term thermal limits. The power ratings on your drivers also suggest that an NU6000 would get you more output, you can use the money saved on DSP to upgrade

Thanks for the advice, I just wanted to check there wasn't something obvious I was missing... Using the wedge micrometer method (http://www.linkwitzlab.com/faq.htm) on my spare driver I got to about 20mm peak to peak before the noise was objectionable. However I suspect that this noise was at least in part the amplifier I was using clipping as the only one available to me right now is only 50W. I have already used 12 of the drivers in anger with 8 of the drivers (quad driver boxes) on one channel of the NU6000 to make a 4 ohm load and 4 of the drivers (dual driver boxes) on the other channel to make an 8 ohm load with bass heavy program material, often hitting the built in limiter, for about 20hrs without driver failure. So unless I up the amplifier power significantly I'm not too concerned about thermal failure when I change the boxes run them magnet out as the power available per driver is low. I have included some pictures of the driver (with vent mesh removed), the voice coil former looks to be aluminum and claims to be a "2.5-Inch four-layer dual voice coil". The ventilation seems to work pretty well with a lot of air exchange when the excursion is pushed. If I block off the magnet vent the driver makes far more noise. When I measure the top plate by pixel counting the mechanical diagram I get 13 mm thickness however this diagram is probably not showing this accurately. The driver is overhung type and the xmax is listed as 8mm voice coil overhang. A lot of these excursion questions could be answered by properly measuring my existing boxes in the style of the systems on this site, however I don't have access to a suitable outdoor environment with sufficient power. I could take one to the park but then there would be no power! indoor measurements of these boxes so far have been pretty useless. PLPW15D-2.pdf

Long time reader, first time caller. In lots of the reviews the excursion capabilities of the drivers are stated E.G on the 2269H: How is this measured? I have some low cost drivers (Pyle plpw15d) with a stated xmax of 8mm, however the other T/S parameters were not correctly stated by the manufacturer. They seem to start making bad noises far beyond 16 mm peak to peak, the top plate looks to be about 6-8 mm thick (hard to get calipers in). I would like to have a better idea of the actual excursion capabilities for system optimisation as I am using 12 of these drivers but would like to re-design my system to use 16 of these drivers and need to do some thinking about whether its worthwhile and how much amplification I should apply before diminishing returns. I have attached a picture of one of my current subs, its a quad driver opposed sealed about 250L. My current plans are to use 16 drivers in dual opposed slot loaded magnet out sealed boxes that are sized so that using the quasi continuous full output of a NU6000 (60Vrms) they can reach 8mm excursion at 20Hz when wired as series parallel 4ohm loads (30Vrms per box). This change is mainly because the quad boxes are too heavy, I lose output over time due to driver heating and I keep having to un-bend grills. The main objective of the system is maximum output at 20Hz which for 8mm excursion should be about 120dB/1m/2pi (assuming I could locate the drivers in the same point of physical space).