Need help building open baffle/bass reflex hybrid

Bensturgeon, feel free to post your questions here. There is no need to PM, we're all here to help :p No need to feel bad, I was once a total noob too!

Anyway to answer your questions in your original post:

1- With your amp it won't matter if you wire them in parallel or series. Your amplifier will work either way. I would personally wire them in parallel for a 4 Ohm load. As long as your amplifier is 4 ohm stable, you won't have problems. (You can tell this by looking up your amplifier, and if it gives a rated power input into a 4 ohm load, you're fine).

2- As long as you have a way to control your volume, you're fine. (More Volume ((with the same speakers)) = more power) . Also, If you use the TC9FD, as the Nola Brio Trio uses, as long as you stay under 10W you'll be fine. Worst case scenario is that you fry your speaker drivers.. which the TC9FD's are $12/piece.. so no worries.

3- I would highly recommend the TC9FD over almost any other fullrange driver. It's paper just like you asked (I also have a paper bias)! :] Xrk conducted blind tests on this website with several dozen fullrange drivers, and the TC9FD came out among the best 5 drivers, regardless of price. The TC9FD has a much better (flatter) on-axis response than the PS95, and also a much better impulse response. Don't be fooled by the price, this driver is designed and developed by one of the best speaker driver manufacturers to ever exist.

http://www.diyaudio.com/forums/full...ind-comparison-2in-4in-drivers-round-4-a.html

4- As long as you run these speakers under 5 watts/channel (aka be careful with the volume knob), there is no need for the crossover imho. Given the sensitivity of this setup at ~90dB 1 Watt/ 1 meter.. I highly doubt you'll need to exceed 5 watts/channel anyway.

The average listening level is 83dB at your listening position (seat/couch), which will probably be around 8 feet away from the speakers. Sound decreases 3dB for every doubling of distance, so with 1 watt the sound intensity at 8 feet away will be around 86dB. At the same distance with 2 watts, you'll reach around 89dB and with 4 watts 92dB. Be careful with the volume knob! :rolleyes: Dangerous Decibels How Loud is Too Loud?

5- The foam board construction was an experiment that was pioneered by Xrk and adopted by many (including myself) for rapid prototyping. Although, there is nothing wrong with using them long-term. I just use foam board that you get at Walmart/Target/Craftstores/Superstores. It's just foam sandwiched between 2 paper surfaces. It is usually sold by the office/school supplies and is often bought by high school students for projects.

Dangerous Decibels How Loud is Too Loud?
 
I just did makeshift holders for my pair of TC9s and placed them atop the TABAQs.

I like it! Most of the time...

For simple music, voice and guitar, or voice with piano, it's really great.

For jazz quartets, or small ensemble classical... very nice as well.

I even liked it for big choirs!

The feeling? Well, with only the TABAQs, the space is projected to you. And they do a pretty good job.

Adding the open baffle TC9s on top, I was there. Not projected space anymore, more like ... I'm there.

But, it didn't agree with my Dire Straits recordings... not at all. It sounded like they were playing in a very reverberant cave.

I had good success with most of my well recorded, High Rez files. But, with joint stereo mp3s, it was blurry, and down right annoying with some of them.

If I were to go DIY, I'd probably power them with a separate amp, that way I could control the amount of "there" I want, or need, for different kinds of music.

I liked it enough that I'll put them on top of my W8-1772 for fun as well!

PS... sorry, no REW sweeps as there were people around and couldn't get the time to do so.
 
Great to hear first impressions! I gotta order some of these drivers to try this out.

I've never used one, but couldn't you just use an L-pad attenuator for the top driver? Hell, I think you could even wire in different single resistors to experiment with instead of an L-pad.
 
Yes, an L-pad would work. A resistor would be more trouble, as I would need to swap the resistor depending on the music!

But, using an extra 2 channels and a stereo amp (I actually have an 8-channel interface with an 8-channel amp), I get to EQ and add hipass as I want, trying different settings separately from the mains.
 
Please post your impressions on high pass! I'm very curious if there is any real benefit. I hate to ask any favors or seem pushy, but if you could experiment with simple 1st order high passing that would be great :]. I'm all about simplicity..

Being able to produce an optimum result (if hi-passing proves beneficial) with a single capacitor would be awesome! :D
 
Last edited:
I didn't highpass the TC9 this time around. It was playing full range, and sounded fine. No problem with over excursion, as I was far from rock concert levels... Still, got to be a little careful since my amp can give 220W per channel.

The only cap I have laying around is a cheap 3.3uf... that gives me a 1st order f3 at about 6000Hz.
 
I found this website. It's a bit hard to understand what the author is trying to convey, but interesting.


Cardioid bass

Thanks for the link. It is hard to follow so I looked at other sources as well. Now it's my turn to "might be wrong" but I don't believe this is a cordioid system as I understand them. A monopole speaker has omni bass and an out of phase dipole is used to control the rear wave radiation usually with dsp. As the monpole becomes directional there is no need for this control. The dipole driver of the Brio cannot accomplish this since it's not producing any bass and the monopole is front firing (unlike the lxmini) so it cannot provide cordioid at the xover since it is directional at his freq. The lxmini can be placed close to walls because of the tube the full range driver is mounted in front of. According the Linkwitz it acts as a diffuser and attenuates high freq (~7.5khz?) to limit reflections off the front wall.
Feel free to chop up my analysis and bare my ignorance for the world to see!
 
You Guys have been super helpful thus far.. Thanks Much!!

More questions:

1. What gauge wire is best to use for internal wiring?
2. what part would I need to terminate the speaker wires going to the drivers to make them slip on an off easily while I am experimenting (I don't want to mess with Solder until I know my design works.. They are used in car audio a lot, I just don't what they are called)
3. How do I calculate the the size capacitor i would need to to filter out low frequencies in the upper driver. I'm thinking about experimenting with different crossover points. Maybe starting by filtering out everything below 200 hz to start . And are these just wired in line to the + terminal of the top drivers??

Thanks again for your help and insights!!
 
Jimbro, honestly, I have no idea on how this would work in real life as far as dispersion pattern. I believe that dipolar and monopolar radiation do sum to cardioid- but only when both sources are radiating at the same frequencies.. I believe the lx mini's rear waveguide/attenuator is used to make the cardioid response specifically because there is a crossover between the two drivers.

I believe that even in this design with out a crossover, bass will roll of sooner and faster in the dipolar driver than the monopolar driver. From my understanding maximum constructive/deconstructive interference happens when output between waves are more similar. So in theory if, I'm correct, cardioid response will increase as output differences from the monopole and dipole driver approaches zero. This should mean increasing cardioid response with frequency in this setup.

There is one large advantage to this system though- the fact that we are using two identical drivers with identical directivity (ignoring mono/dipole). I would have to assume this means that sense directivity is in essence as "matched" as it can possibly be (identical drivers) that a cardioid sum would be optimized.

Again this is all speculation based on the minimal knowledge I could scrap together on google. I can be totally wrong. We already have one impression though that these speakers are doing something fundamentally different as far as sound signature.

Only building these and giving them a listen will really tell. Thankfully because of Xrk's work we can use fantastic sounding drivers cheaply. :D

Measurements will tell us what's going on with the polar response. It's definitely different from the normal though. Part dipole, part monopole, tiny part linesource.
 
Last edited:
So...if the Nola speaker is 5.5 x 5.5 x 12, with half that open baffle....thats ...nowhere near the optimal .25 cu ft. Is It? Why did they go that small? Do i want to go that way or .25 cu ft??

Generally speaking, a larger enclosure will allow us to get a lower frequency response than a smaller enclosure. The Nola's use a subwoofer for the low end response, so I believe the designer made them smaller for marketing/visual/WAF (wife acceptance factor) purposes.

Ideally we want any 2 channel speaker system to get to atleast 80hz @ -3dB. Below 80hz the vast majority of people can not localize sound, so this allows us to use a subwoofer without "hearing" where it is placed in the room.

If you want to learn how to model low frequency response if different enclosures yourself, download WinISD and use this guide - WinISD Pro Tutorial and Download (A detailed guide on how to use WinISD Pro) - Home Theater Forum and Systems - HomeTheaterShack.com
 
You Guys have been super helpful thus far.. Thanks Much!!

More questions:

1. What gauge wire is best to use for internal wiring?
2. what part would I need to terminate the speaker wires going to the drivers to make them slip on an off easily while I am experimenting (I don't want to mess with Solder until I know my design works.. They are used in car audio a lot, I just don't what they are called)
3. How do I calculate the the size capacitor i would need to to filter out low frequencies in the upper driver. I'm thinking about experimenting with different crossover points. Maybe starting by filtering out everything below 200 hz to start . And are these just wired in line to the + terminal of the top drivers??

Thanks again for your help and insights!!

1. The biggest influence speaker wire has on speaker systems is resistivity. The general rule of thumb is, that the sum of all speaker wires to connect amplifier to a speaker, should not exceed 5% of the speakers impedance. So lets use 3 ohms as a good safety net. Generally speaking I would prefer 18 AWG- which has a resistivity of 0.006385 ohms / foot. You can order a long spool of magnet wire really cheap online. Try to get softer annealed wire.

Note: This Wikipedia article explains differences between inductance, capacitance, and resistance in cables. You CAN purposefully make wire that "violates" the rules to make them "transparent" to make your speaker wires sound different. That's exactly how $18,000+ speaker wires that companies sell do it. This doesn't mean a cable is better. Stick to the transparent guidelines at first and DIY other cables for fun in the future if you want to tru them out as "tone controls". Please don't spend $2,000+ on a few dollars worth of copper and plastic.

https://en.m.wikipedia.org/wiki/Speaker_wire

2. https://www.madisoundspeakerstore.com/quick-connects/

3. Someone will have to help you here. Crossover theory isn't my strong suit. I believe the capacitor is in series. Again, the top dipolar driver has more power handing than the driver in the ported enclosure- atleast in the enclosures I modeled. Really no reason to use a cap imho.
 
Last edited:
So...if the Nola speaker is 5.5 x 5.5 x 12, with half that open baffle....thats ...nowhere near the optimal .25 cu ft. Is It? Why did they go that small? Do i want to go that way or .25 cu ft??

Studying the Nola I got some rough dimensions. The ported enclosure is around 166.4"^3. The outside of the enclosure is just a 5.5 x 5.5 x 5.5 cube essentially, with the "top" of the enclosure slanted for looks and to somewhat reduce standing waves.

Optimum response in a 166.4"^3 enclosure is with a port tuning of 81hz. F3 is 70hz. You will want to use a 1" vent with a length of 2.6". Xmax is achieved at 2.3W. Port velocity at xmax is 17m/s peak, which shouldn't chuff.

Again, if my modeling is correct, the open baffle (top) driver will not hit xmax until ~4W of input. This is almost double the power of the TC9FD into the ported enclosure. There is no reason to hi-pass the open baffle driver, if the reason is to protect it from over-excursion. The ported TC9FD will hit Xmax way before the top driver does.

Of course, the enclosure interior volume is less than 166.4"^3 because the walls of the speaker enclosure have thickness. This isn't a problem imho (especially with foam core). Polyfill added into an enclosure has been documented to increase apparent enclosure volume by 40%. The rough rule of thumb is start with 1lb/1ft^3 of polyfill (will usually add ~30% more apparent enclosure volume), and add or subtract by ear. This enclosure is roughly 0.096 ft^3, which means you should start with 1.6 ounces of polyfill, if my math is right. Depending on your actual internal volume, you will need more or less.

I would suggest mounting the port in an up-firing position. Obviously, you're going to have to place it behind the speaker driver magnet (if you look down through the port, you shouldn't see the speaker magnet). This way the polyfill stays inside the enclosure better. ;)

Edit: If you create an enclosure to your liking in foam board, you can actually fiberglass it... with a lot of work. Here's a short video.
 
Last edited:
You could be right, Xaborus. I do know that the Brio and the Lxmini are monopole in the bass. According to Linkwitz the Lxmini is only cordioid in the xover region. The Brio? Guess we'll see. Actually I'm not that sure why it's that desirable to have a cordioid pattern. They would still need to be 2ft or so away from walls so I guess you gain a little placement flexibility. It wouldn't sound like a typical open baffle would it? What is to be gained?
 
You could be right, Xaborus. I do know that the Brio and the Lxmini are monopole in the bass. According to Linkwitz the Lxmini is only cordioid in the xover region. The Brio? Guess we'll see. Actually I'm not that sure why it's that desirable to have a cordioid pattern. They would still need to be 2ft or so away from walls so I guess you gain a little placement flexibility. It wouldn't sound like a typical open baffle would it? What is to be gained?

Actually, reduction of front-wall reflections is the main benefit (when the null faces the front wall). In linkwitz's words, you're increasing the direct-to-reflected sound ratio. I'm almost certain he came up with the concept because his listening room is mostly glass, and the omnidirectional pluto's had too low of a direct-to-reflected sound ratio without pulling them very near-field (away from the rear wall). Just like dipoles, line-sources, and waveguides, it's a form of creating directivity.

Dipoles will optimally reduce reflections, in a small angle, to the roof, sidewalls, and floor. Although, they will send lots of energy into the rear wall. That's the reason you have to "pull them out". They also have the hard-to-overcome bass response issue. Monopoles are omnidirectional anywhere below beaming, so have reflections everywhere, and induce the most room modes. Cardoid's ideally reduce rear wall interaction.

I think in theory a true cardioid should also decrease room-mode interaction, just like dipoles do.

When you think about it, most people fail to provide adequate clearance from the back wall for stereo speakers. So a cardioid response is actually a great radiation pattern for the general consumer and WAF. :p


Really, the big question is though.. if these do sum to a cardioid response, where is the null?..

Alas, if I'm completely wrong and wasted my time with the cardioid concept, it's still a novel speaker design which I haven't seen before. Sometimes being "different" is all a product needs to be perceived as good.
 
Last edited: