@profiguy, might it be feasible to use a vented enclosure with small ports for your 8NDL51(s), for the sake of cooling? If you had one or more small ports above the midpoint of the motor, and one or more small ports below the motor, you would get a bit of chimney effect cooling as the hot air inside the subenclosure exited the upper port(s), drawing cooler outside air into the lower port(s). You might have to adjust the highpass filter and/or crossover frequency somewhat if you were relying on the enclosure to prevent over-excursion.
I used this technique in a custom studio monitor build. The ports for the 12" midwoofers are more for cooling than for boosting the low end.
I used this technique in a custom studio monitor build. The ports for the 12" midwoofers are more for cooling than for boosting the low end.
In the case of B&C drivers, the Xvar number is more a practical limit than the specified Xmax, because the former is based on real distortion measurements.
The 8MBX51 have 1mm more Xvar than the 8NDL51, have the same electrical power handling, but the MBX have higher efficiency (or Voltage sensitivity for the same impedance), so you don't need to push as much power to the driver for the same SPL, so the power compression would be lower. Btw the 8MBX51 is the newer driver in the lineup, so the parameters is probably not by accident.
There is no 4 Ohm version of the 8MBX51 but since the efficiency is higher, the 8 Ohm version have about the same Voltage sensitivity as the 4 Ohm version of the 8NDL51.
Last edited:
I agree with @audiokinesis regarding possible solutions to cooling. The lack of airflow to the midbass driver enclosure presents the main limitations in powerhandling. An extended shelf reflex tuning, moving the second impedance peak down past the equivalent sealed box resonance peak would be effective in providing sufficient cooling while not affecting the linearity or impedance characteristics at the originally specified sealed box tuning. A supplemental over-excursion safeguard would be needed.
An alternative to a lower reflex tuning would be the use of liquid cooling. I’ve previously used direct pelletier cooling with the neo motor equipped B&C drivers, but even more effective is indirect liquid cooling. A copper coil wrapped around the 8NDL51 would be quite effective at raising power handling, but there was still a derating in total power handling, with cooling via motor radiation being not quite as effective as cool air being delivered directly to the VC gap. This is all considering using a 4 ohm version of driver. A more effective and economical solution is the use of 2 pieces of 8NDL51 8 ohm in parallel, theoretically doubling free air power handling. In practical sense, 200 W continuous roughly translates into 119 dB output, factoring in acoustical gain of 3 dB on top of 93 dB sensitivity of a single 8NDL51, provided both drivers placed in closest practical proximity to each other. Adjusting midbass HP to avoid over excursion would be a secondary measure.
The larger VC 8NDL64 would be a better fit as a single driver if its lower xmax is watched closely.
An alternative to a lower reflex tuning would be the use of liquid cooling. I’ve previously used direct pelletier cooling with the neo motor equipped B&C drivers, but even more effective is indirect liquid cooling. A copper coil wrapped around the 8NDL51 would be quite effective at raising power handling, but there was still a derating in total power handling, with cooling via motor radiation being not quite as effective as cool air being delivered directly to the VC gap. This is all considering using a 4 ohm version of driver. A more effective and economical solution is the use of 2 pieces of 8NDL51 8 ohm in parallel, theoretically doubling free air power handling. In practical sense, 200 W continuous roughly translates into 119 dB output, factoring in acoustical gain of 3 dB on top of 93 dB sensitivity of a single 8NDL51, provided both drivers placed in closest practical proximity to each other. Adjusting midbass HP to avoid over excursion would be a secondary measure.
The larger VC 8NDL64 would be a better fit as a single driver if its lower xmax is watched closely.
Too bad its not easy to aquire SF6 gas to fill the midbass chamber. SF6 is a great thermal conductor and would also lower Fs due to increased mass, but that would significantly reduce driver base sensitivity due to the extra added moving mass. The potential problem with using other cooling strategies is condensation forming inside the motor. This can be significant and is the main limiting factor how much delta in temperature is reliably achievable. This is all headed in the direction of overthinking a practical solution to a problem which may or may not be an issue under most use scenarios.
Okay, I've looked into it. What doesn't fit in the requirements is the high Xmax. For example, the 8NDL51 will do ~1,7mm excursion at 300 Hz with 600W. With that power, any high excursion driver will burn up the VC ends because they never meet the pole plate to dissipate the heat. That means, looking for high Xmax will exclude all drivers that are suitable for 300-900 Hz.
E: Why not use a 10" midrange driver?
E: Why not use a 10" midrange driver?
@ICG I've looked at a few 10s, specifically with 3" VCs. They don't have the clean midrange of the 8NDL51. PRV makes an 8MB700FT-NDY with 2.5" VC and decent sensitivity, but a very light / lively cone. It will take 250 - 300 W continuous but only in a ported enclosure. The few JBLs I looked at were promising, but expensive for the really good ones.
Do you think it would stay in the box? For long?
I have bottle of it bought for exactly this. Haven’t tried it yet.
Barry.
It's impossible to use SF6 to cool the speakers. Since it's much denser than air, the speed of sound is a lot slower (125m/s), reflections in the basket are now in its reproduction range. Resonances will be lowered so upper end resonances will drop in frequency too which makes a non perfect behaving driver a nightmare. Not the biggest part of the problems though, it has a higher pressure than air because of the high mass. That means - even at normal air pressure - it pushes the cone outward and gives a severe DC offset. Also, the spring constant is different which makes the compilance extremely asymmetrical. The final knock out is if some of it escapes, it will pool in the lower part of the enclosure, the part of the cone still in it will be slowed down while the part above will not, which will result in a severe tumbling motion and deformation of the cone.
The best way to cool such a driver is to mount a heat pipe to the front pole plate and connect it to a heatsink with the fins on the outside.
@1audiohack Yes, as long as the edges are all caulked, inside walls coated with urethane or a glue/water mix and gaskets are used for the woofer. The driver needs to have an coated or treated cone and surround. SF6 is a huge molecule and won't easily leak. It will expand a tiny bit with minor temp changes but its not too severe. At worst it could cause a small amount of driver cone offset. Putting the gas directly in contact with the driver will aid cooling, but it may expand a little more with high VC temps.
Putting SF6 in a large bladder inside the speaker enclosure, occupying most of the internal space works best and will stay put for a very long time, even years as long as the bladder material is stable and durable. I've used small weather balloons to contain it. Trick is to have it sit there in uncompressed atmospheric pressure, same as the surrounding atmosphere. SF6 is an adiabatic (ideal) behaving gas. That means compressing it retains its kinetic energy at a linear rate and releases it back, as opposed to air.
In the same size airspace compared to a regular sealed air filled cabinet, I"ve seen Fs drop by 15 to 20% (sometimes more) with a partial mix of SF6 and nitrogen. There have been a few speakers (Corwin Vega) which have used this method to extend low bass response and in the case of some electrostatic speakers, it was employed to increase the bias voltage without arcing. Just be aware its a bad greenhouse gas with a rating of about 1500 times higher than CO2. Where did you buy it and how much was it?
Putting SF6 in a large bladder inside the speaker enclosure, occupying most of the internal space works best and will stay put for a very long time, even years as long as the bladder material is stable and durable. I've used small weather balloons to contain it. Trick is to have it sit there in uncompressed atmospheric pressure, same as the surrounding atmosphere. SF6 is an adiabatic (ideal) behaving gas. That means compressing it retains its kinetic energy at a linear rate and releases it back, as opposed to air.
In the same size airspace compared to a regular sealed air filled cabinet, I"ve seen Fs drop by 15 to 20% (sometimes more) with a partial mix of SF6 and nitrogen. There have been a few speakers (Corwin Vega) which have used this method to extend low bass response and in the case of some electrostatic speakers, it was employed to increase the bias voltage without arcing. Just be aware its a bad greenhouse gas with a rating of about 1500 times higher than CO2. Where did you buy it and how much was it?
@ICG You're correct that using SF6 in a midbass enclosure will lower certain resonance modes. In a subwoofer however, it won't encounter those higher frequencies. Thats why putting the gas inside a bladder is better, as its shielded from the higher frequencies and it won't mess with TSPs, which may shift undesirably in the case of a speaker reproducing higher frequency ranges. In my case in a sealed midbass enclosure, it wouldn't work well because of these reasons. It will however absorb heat better than air, hence cooling would be more effective.
JBL uses aluminum enclosures on their midbass drivers designed for horn loading. I see many of these designs from other manufacturers who design high power midbass speakers with built in enclosures. Another option is using a heat pipe with freon in it, as Panasonic did on some of their 80s and 90s amplifiers. Pelletier coolers are still the most effective at removing heat from the motor. We used modified CPU cooler setups in some large FLHs to keep the woofers alive. It increased power handling by more than 25 percent if copper coils were used
The actual precise GWP of SF6 is estimated between 1000 and 25000 depending on its life cycle. We used to use it in magnesium casting, which released alot of it. The manufacturer quoted the specs to us, which differ from your source of info. In any rate, its a bad substance to release into the environment, regardless of how exact its GWP is.
JBL uses aluminum enclosures on their midbass drivers designed for horn loading. I see many of these designs from other manufacturers who design high power midbass speakers with built in enclosures. Another option is using a heat pipe with freon in it, as Panasonic did on some of their 80s and 90s amplifiers. Pelletier coolers are still the most effective at removing heat from the motor. We used modified CPU cooler setups in some large FLHs to keep the woofers alive. It increased power handling by more than 25 percent if copper coils were used
The actual precise GWP of SF6 is estimated between 1000 and 25000 depending on its life cycle. We used to use it in magnesium casting, which released alot of it. The manufacturer quoted the specs to us, which differ from your source of info. In any rate, its a bad substance to release into the environment, regardless of how exact its GWP is.
Last edited:
Anyways, getting back to my driver choice, I've done more number crunching on using dual 8 ohm 8NDL51s in parallel. I also brainstormed using these mids with the chassis exposed to the outside. I'll need to measure the actual FR with the driver radiating backwards and see if there is any noise coming from the motor radiating out. The actual FR may still be very linear up to 600 hz or so.
Unusual specs, but the Beyma 8MC800Nd seems to be a contender for max SPL (from an 8" 300-900Hz)
https://www.beyma.com/en/products/c/low-mid-frequency/108MC50ND8/altavoz-8mc500nd-8-oh/
8 ohm unfortunately, and begs for an amp with 750W@ 8 ohm for max output
https://www.beyma.com/en/products/c/low-mid-frequency/108MC50ND8/altavoz-8mc500nd-8-oh/
8 ohm unfortunately, and begs for an amp with 750W@ 8 ohm for max output
@mark100 Nice driver, but base efficiency is only 94dB at 1W and looks to be only available in 8 ohms. The VC is only 2", so its the same dilemma as the 8NDL51.
@abstract. The issues with blocking off the rear vent on the 8NDL51 is they're not vented in the traditional sense being a neo motor and spider is vented in the rear of the chassis already. There are also holes right behind the VC gap, so the driver is already being cooled the way you suggest.
There is always the option of using very quiet PC fans pulling air through constantly, but I don't think that will make a difference in a small sealed enclosure without fresh air access.
There is also critical dampening using a restricted port to make it act sealed without completely blocking airflow. Overexcursion can be a danger with this. I may still visit the idea of inward facing chassis mounting depending on how clean the backwards radiated response is. According to basic math, running 2 of the 8NDL51 im 8ohms each sealed will be possible using some heatsinking. I already have the drivers. Just need to try a mockup enclosure. The car audio guys use these drivers all the time and push them very hard in small sealed boxes, so that in itself is promising.
@abstract. The issues with blocking off the rear vent on the 8NDL51 is they're not vented in the traditional sense being a neo motor and spider is vented in the rear of the chassis already. There are also holes right behind the VC gap, so the driver is already being cooled the way you suggest.
There is always the option of using very quiet PC fans pulling air through constantly, but I don't think that will make a difference in a small sealed enclosure without fresh air access.
There is also critical dampening using a restricted port to make it act sealed without completely blocking airflow. Overexcursion can be a danger with this. I may still visit the idea of inward facing chassis mounting depending on how clean the backwards radiated response is. According to basic math, running 2 of the 8NDL51 im 8ohms each sealed will be possible using some heatsinking. I already have the drivers. Just need to try a mockup enclosure. The car audio guys use these drivers all the time and push them very hard in small sealed boxes, so that in itself is promising.
Yes, the Beyma 8MC800Nd is only available 8 ohms. VC is 2.5".
It is quite a jump up from the 8NDL51 in terms of power handing / SPL.
Very quick and nasty sims, 20L closed box....steep BW 8th orders at 300Hz and 900Hz.
Both at their max power handling (halfway between continuous and program)
Beyma on the left, B&C the right.