[Onslo]

I only recently learned of OB speakers after a hiatus from DIY of several decades. I am experimenting with some drivers I have in stock. Those being
HiVi D10G +  Peerless Tymphony 164 nomex (2pr side) + Fountek Neos (cannot be OB)
My thought is to do a MTMW in an OB with these drivers. What are you folks expert thoughts. Get the saw out or find some other drivers and put these in a box?

Many thanks for any suggestions and ideas.
Regards Onslo

Product Specifications
Nominal Diameter10"Power Handling (RMS)150 Watts
Impedance8 ohmsFrequency Response25 - 2,000 Hz
Sensitivity88 dB 2.83V/1mVoice Coil Diameter4"
Thiele-Small Parameters
Resonant Frequency (Fs)22 HzDC Resistance (Re)6.5 ohms
Mechanical Q (Qms)2.41Electromagnetic Q (Qes)0.54
Total Q (Qts)0.41Compliance Equivalent Volume (Vas)4.88 ft.³
Maximum Linear Excursion (Xmax)8.3 mm
Materials of Construction
Cone MaterialPaper / Kevlar
Mounting Information
Baffle Cutout Diameter10"
HiVi D10G 10" Kevlar/Paper Woofer
BrandHiViModelD10G
Part Number297-430UPC844632021282
Product CategoryWoofersUnit of MeasureEA
Weight8.15 lbs.

Nomex 164
Specifications
Transducer Size         6.5 in   –
DC Resistance         6.37 Ω   ±5.0%
Resonant Frequency         45.82 Hz   15%
Half Space Sensitivity @ 2.83V         89.04 dB   ±1.01
Sensitivity @ 1W/1m         88.6 dB   ±1.01
Rated Noise Power (IEC 268-5 18.1)         75 W   –
Effective Piston Area         143.1 cm2   –
Moving Mass         15.9 g   –
Motor Force Factor         8.47 T•m   –
Voice Coil Inner Diameter         32.41 mm   –
Gap Height         6 mm   –
Maximum Linear Excursion         5.3 mm   –
Ferrofluid Type            –
Minimum Impedance          7.18 Ω   ±7.5%
Voice Coil Inductance         0.41 mH   –
Mechanical Q Factor         3.19   –
Electrical Q Factor         0.41   –
Total Q Factor         0.36   –
Ratio         126.93 fs/Qts   –
Test Spectrum Bandwidth (1300-20000 Hz)         40Hz - 3kHz 12 dB/Oct   –
Energy Bandwidth Product ((1/Qes)•fs)            –
Suspension Compliance         756.8 um/N   –
Effective Cone Diameter         13.5 cm   –
Motor Efficiency Factor         11.3 (T•m2)/Ω   –
Voice Coil Former Material         ASV GSV   –
Transducer Mass         1.56 Kg   –
Equivalent Volume         21.78 L

[Onslo]

 Done some early prototyping. Not yet allowed in house...

[JohnR]

That was quick! Do some measurements and see how you go. You will probably find yourself looking for bass reinforcement.

[richidoo]

Looks good! 

Do you plan on using active crossover, or passive? John is right, your challenge now is boosting that woofer enough to compensate for the major rolloff in the bass that is inherent to OB speakers.

If active, this is a nice choice, sounds very good with built-in ICE amps. One of these per speaker gives you active xo between bass and mid/tweet, with mid/tweet using a passive xo, 12dB LR @2k + pad on ribbon. DSP corrects baffle step diffraction and dipole cancellation for flat frequency response probably down to 40Hz at maybe 95dB?

The woofer isn't designed for OB (Qts is too low) so that kinda forces you into using an active crossover to let the bass amp make up for the woofer's incompatibility. Active allows you to BOOST the woofer to be as loud as the mid, while passive requires you to CUT the mid to be as quiet as the woofer. To CUT the mid enough to be flat FR at 50Hz you would need a lot of cut, so your overall system sensitivity would be in the mid 70s if you have balanced bass level. This is why passive OB speakers really need a woofer that is designed for OB use, with high Qts, like this one which gets louder as the frequency drops, so it self-compendsates for baffle step and dipole losses, and it can keep up with the mids without any passive or active EQ.  But with active amp, you can apply an EQ curve to the bass and all is well.

[Onslo]

Thanks for the responses. My intention with this was to spend very little money so the drivers and crossover bits and bobs I had to hand were used.
You are exactly correct the ribbon needed backing way off with an lpad. The bass is pretty shy too. I am waiting on a mike to allow me to measure.

Given the original intent and that I have no digital source I am inclined to avoid the mini DSP route although I see them as great value and easy way of fixing lots of issues, although needing another power amp.

My thinking is to go MTMW with the next iteration hoping the extra peerless 164 will waste a bit less of the tweeter and maybe pop the D10G in a box. Otherwise it is lose the D10G, and get those SB 12s. The listening tests I have done were with a subwoofer reinforcing from 150hz. I didn't have components to roll the D10G off at that point but doing so would probably give it a bit better power handling.

What are folks thoughts on the Nomex 164 in OB? Maybe I could do a simple 2 or 2.5 way OB with them, straight to the sub.

[richidoo]

If you parallel 2 of those midrange drivers (88+6) you will get 94db. That will lessen the need to pad the ribbon, but will make it even more difficult (impossible?) for the D10 woofer to keep up while using a passive crossovers.

Putting the D10G into a box will only affect the bass response. The nominal (midrange) sensitivity isn't changed with a box, so you would still need to pad on mid and/or tweet.

You wouldn't need any other amp with the minidsp plate amp. It has 2 amps onboard each plate. One powers the woofer directly, the other amp powers both the mid and tweeter. DSP on the M/T amp provides high pass filter for the crossover to the bass. And the mid and tweeter have a passive crossover between them. It might be the lowest cost path for you to get up and running with the drivers you have now. But I understand your desire to keep the digital out.

[gab]

Study the OBL11 and OBL15 woofer crossover pages for some things you might try with that low Qts woofer.

http://www.troelsgravesen.dk/OBL-15.htm

[Onslo]

Thanks for the helpful advice.

If you were to use two 164 mid bass units per speaker would you suggest a 2.5 way or a MTM? How do you reach that decision? Is it very different for OB than other enclosures?

Last question... There are several designs around using these drivers in MTM and 2.5 way with several different tweeters, how different is the crossover likely to be in an OB setup than in a bass reflex. Excuse my ignorance but I assume if the bass driver is allowed to run down to its natural bottom then the difference would be minimal due to the major difference would be higher order roll off at the bottom in OB. unless you traded some sensitivity in the rest of the range to try to balance better with that lower output down low.

Thanks heaps for any information that increases my knowledge.

[richidoo]

If you were to use two 164 mid bass units per speaker would you suggest a 2.5 way or a MTM? How do you reach that decision? Is it very different for OB than other enclosures?

Some discussions of your driver that might be interesting to you:
http://www.troelsgravesen.dk/Peerless_HDS830875.htm
http://www.troelsgravesen.dk/NOMEX-164-mkII.htm

Both 2.5 and MTM have pros and cons. The biggest con you will hear for both arrangements concerns the vertical lobing effects caused by the drivers' separation being larger than the wavelengths they play. Ideally, the driver distance would be always smaller than 1/4 wavelength separation, but that is very impractical so most MTMs have larger distance separation than that. Larger vertical separation of drivers approaching 1 wavelength and larger causes interference and audible lobing - meaning the tone will noticeably change with your listening height. For many people this is not a concern as they don't bob up and down while they listen. But how do you ever know if you're sitting at the "right" height to get flat FR? The lobes can be only inches small at the higher frequencies. The benefit to these dual mid arrangements is higher sensitivity and greater dynamic range, and for midbass MTMs, the floor bounce interference is spread out across larger freq band. The acoustic impedance doesn't significantly change, so lower frequency extension via the cones is not really a benefit, but 2 woofers can power a larger reflex port, so you can get lower extension from the port. In the case of a D'Appolito Array, which is a specific MTM design formula, you would also gain vertical directivity control to attenuate reflections from floor and ceiling. So there are some good reasons to double midrange, and some technical reasons to avoid it. My current (commercial) speakers have dual midbass with large separation and dual ATM tweeters between them, and it sounds pretty darn good to me! A friend's Selah Audio Referimento has MTM with smaller separation and smaller drivers with active linear phase crossover and that was one of the best speakers I've ever heard. So the MTM fear on the internet is more dogma than a real risk, imo. But of course it's the whole speaker design taken together as a whole that matters. I'm sure there are plenty of true golden ear audiophiles who would be annoyed by the sound of the most refined MTM in some small way. But there are still plenty of highest of high end speaker mfgs still producing MTM designs. The most common reason to use dual mid is to improve dynamic range, so you can play louder with less midrange distortion, because driver distortion increases with cone excursion. Most modern drivers have pretty good dynamic range though, so unless you need to play very loud in a large room you may not need that much SPL to warrant dual mids.

Last question... There are several designs around using these drivers in MTM and 2.5 way with several different tweeters, how different is the crossover likely to be in an OB setup than in a bass reflex.

OB differs from box speaker in that it has an additional 6dB/oct rolloff with falling frequency. The crossover would have to deal with this in OB. Adding 6dB/oct is a huge difference and that would be required if you compared the same low Q driver with box and without.  But fortunately, there are drivers designed for OB which have higher Q, which means these have their own EQ built in, so they increase their volume to offset the dipole cancellation and give almost flat FR into bass freqs without filter EQ. For passive speaker this is pretty much essential to preserve sensitivity of the mid/tweeters. The cost of this added sensitivity in OB woofers is loss of cone control. The woofer increases it's sensitivity by losing control, the motor and suspension are deliberately too weak to control the cone, which allows it to flail around more, exceeding the displacement of a low Q speaker for same input signal, and this is what makes it play louder at lower freq. There are schemes like servo feedback that allow harnessing the sensitivity gain of low Q driver without losing the cone control.

Excuse my ignorance but I assume if the bass driver is allowed to run down to its natural bottom then the difference would be minimal due to the major difference would be higher order roll off at the bottom in OB. unless you traded some sensitivity in the rest of the range to try to balance better with that lower output down low.

The dipole cancellation causes the speaker to start rolling off at a much higher frequency than if the same driver was in a box. Iirc, the rolloff begins when 1/2 wavelength equals the baffle width. Maybe someone can clarify the frequency at which it starts attenuating. So with your 18" baffle the dipole cancellation would start at 375Hz, though the F3 would be lower.  A cone in a box has no dipole cancellation loss, so it can play flat lower. It's natural low frequency limit is only affected by it's cone diameter and excursion (velocity,) which affect it's grip on the air (acoustic impedance.) Of course, a driver in OB also has this impedance limit, so OB rolloff steepens at the freq that would affect a box speaker of same diameter cone. That rolloff is 12dB/oct, iirc. Reflex has perfect impedance (air:air) so it can play even lower, limited only by the box volume and swept volume of the cone.

http://www.diysubwoofers.org/dipole/

[Onslo]

What a terrific forum. Thanks a heap Richidoo for a thorough and educative answer. I will follow up with reading your links and put make some more learnings before I actually make more sawdust.

It is a rather addictive interest. I now know I have to stump up for a microphone because without a reference to listen to one never really knows how good ones creation sounds.

[Onslo]

Actually another question (s)... The lobing you speak of makes sense. I wonder though in an OB is this reduced considerably over a box? I expect the reflected contribution would be reducing the lobe effect. If I am understanding it correctly.

The TS parameters of the nomex 164 look suitable for OB am I correct? Any examples of anyone using these drivers in OB?

Thanks Onslo

[richidoo]

I now know I have to stump up for a microphone because without a reference to listen to one never really knows how good ones creation sounds.

Mic is useful, but let your ears be the judge.

The lobing you speak of makes sense. I wonder though in an OB is this reduced considerably over a box? I expect the reflected contribution would be reducing the lobe effect. If I am understanding it correctly.

Yeah, same thing, but in 'room acoustics' parlance they call it comb filtering. The experts tell us not to worry about comb filtering, "the brain will filter it out," they say.... but lobing is unacceptable?   Technically lobing is the same in box or OB, as it's measured from the front of the speaker. Your tweeter is not dipole, so it would not contribute to lobing in rear radiation, but two mids would have lobing effects to the rear like you say.

[/quote]The TS parameters of the nomex 164 look suitable for OB am I correct? [/quote]

Not really, because its Qts = 0.36 and ideal OB Q is 0.707. Putting this driver into a box of the correct volume will change the Q to .7, and that will achieve the "maximally flat response" we are usually after for a midrange.

In order to achieve maximally flat response in an open baffle speaker with no help from a box to raise the Q, a driver designed for OB must have its own Qts be 0.7 without a box. This allows it to play louder, lower, maintaining flat FR as low as possible, without a box. A .36 driver without a box will rolloff at higher frequency. But if you apply EQ to force it to the same loudness as a .7 driver, the .36 should, in general, have less distortion because of its stiffer suspension, stronger motor and better self damping. ymmv. That's where the DSP comes in.