take two enclosures facing each other and screw a single driver into both of them at the same time so all of the output of the driver on both sides is going only into the enclosures themselves.
you can rotate the enclosures around a bit so their front baffles are exposed and can radiate sound into the room.
now play your music. you will hear what your enclosure is contributing to the mix.
you can rotate the enclosures around a bit so their front baffles are exposed and can radiate sound into the room.
now play your music. you will hear what your enclosure is contributing to the mix.
I'm sorry, but I think it is incorrect to say that any measurement is "more useful" than what you HEAR.
Measurements are not significant compared to how a speaker sounds to the listener, right? The main purpose of a speaker is for listening to, or for being measured?
Sure, measurements are useful for designing speaker systems, and for marketing, but, I think how the speaker sounds trumps any measurement.
In the end, listening tests (measuring with our ears) is the most important type of measurement when completing a speaker design, I think successful speaker companies and DIY enthusiasts have all learned this, or will eventually.
EDIT: Though, I do think that cabinet design is extremely significant and vastly overlooked by both most speaker companies and DIY folks. And, that an accelerometer (or other means of measuring enclosure resonances/transmission) should be part of the tool set of anyone hoping to design anything more than a mediocre speaker system.
Last edited:
I have to agree with Brett here. What you propose will give indeed some form of "residual" signal, but you simply don't know what the residual consists of. Assuming the two speakers are wired in reverse phase, and facing each other, the signal you still hear doesn't come from the cabinets alone. Any mismatch in the driver's transfer characteristics, crossover electronics, cabinet construction and such, will manifest itself as an additional "leaked" signal with a crooked frequency response, hard to discern from cabinet vibrations by casual listening.
It all depends on what you consider useful. If you want to know whether a system sounds right to you, by all means listen. But if you want to spot a potential issue and deal with it, by all means measure. Because your intention seems to be to assess cabinet vibrations, the casual test you describe just isn't going to nail them.
It all depends on what you consider useful. If you want to know whether a system sounds right to you, by all means listen. But if you want to spot a potential issue and deal with it, by all means measure. Because your intention seems to be to assess cabinet vibrations, the casual test you describe just isn't going to nail them.
Maybe I misunderstood the OP's intent when he said: "take two enclosures facing each other and screw a single driver" - what I thought when I read that is that you use a total of ONE driver for both cabinets. Doing it that way would avoid the problems that you mentioned, so long as there is a good seal between the two facing speakers.
I would add: use some form of gasket between the two speakers, also, rotating the cabinets by 90 degrees would allow more of the front panels to be exposed, which seems important to me as the front baffle is generally the most important one since it faces the listener and resonanace/transmission there would most likely be more audible, though, I suppose the combined output of two larger side panels could be more significant.
One thing I am certain of - this is an interesting and useful experiment...
Last edited:
This is correct. I don't want to 'hear' the enclosure at all, and measuring it is the way to get the most accurate results. Keep trying different designs until you get no significant response from the enclosure itself.
yes that's what i meant. the front baffle is important because it is most directly energized by driver's vibration.
since cone moving mass is usually an order of magnitude higher than airload mass most vibrational energy will not come from sound but from the driver itself and transfer directly to the baffle.
accelerometer will provide a different kind of information. neither one is "better" than another. both is better than either one alone. accelerometer may help you zoom in on a trouble spot that requires bracing but it won't tell you what your cabinet sounds like.
Any energy imparted physically to a baffle glued to the rest of the enclosure will transfer effectively to the rest of the enclosure.
It appears we are talking about very different things, because I don't want the enclosure to have any sound at all.
to have no sound from enclosure at all is great in theory but not practical.
in practice the resources put to eliminating enclosure contribution completely would be better put to use elsewhere.
for example i think most people over-brace and under-stuff speakers. then they have a completely dead cabinet that howls like a wolf through the port.
Great idea..the only problem i can thinkn of is that in a vented system(ported qwtl tl whichever) you are also going to hear whatever spurious output there is from the vent also.
n ice idea though and a more tangiable result that acc. displacment and velocity resutls than youll get using a velocimeter or accelerometer
It is possible to reduce enclosure vibration level greatly, depending upon how far you are willing to go. There are practical limits, especially for a DIYer. Measurement will allow you to accurately target where and how you have cabinet vibration and test solutions to rectify it.
Bracing and poor port design and implementation are two separate issues. If your port has en extended decay or drone, then it is implemented incorrectly.
Stuffing does not do anything much at LF to attenuate the rear wave inside the enclosure.
BTW, I forgot to mention in my first reply, that using your two enclosure method there will still be some sound spill from the front of the cones so you will not just be listening to the enclosure, but the sum of these two elements.
Last edited:
On the contrary, it is not practical for manufacturers of mass produced speakers to implement extensive/effective (expensive) methods of cabinet damping & stiffening, whereas, a DIY builder is at liberty to implement any of a wide variety/combination of techniques to dampen and/or stiffen their enclosures.
Some examples being: marble or granite enclosures, multi-layer construction using materials with differing Qs, constrained layer damping, lead sheet, sand-filled walls, and, on the extreme end: a double walled enclosure with a vacuum between the walls, like a Thermos.
Note: OP clarified that he did, indeed, mean just one cone between both enclosures, and, from where, will this (significant?) sound escape?
Sure, many of the methods you mention in the second paragraph would be difficult for a manufacturer to implement and ship affordably.
However, I was more thinking of rigid enclosures of CF, various resin compounds or metal which would tax the abilities of many, if not most DIYers.
I missed that clarification. Still with many bracing and construction methods it's going to be difficult to secure the driver to both enclosures.
Now let me see if I have this right, you have two identical sealed enclosures joined together at their sides with one driver shared between them. If this is correct, then I would assume that when the cone moves in one direction it will produce a positive pressure in one enclosure and a negative pressure in the other. In this case I am referencing positive and negative pressures relative to static air pressure. If my understanding is correct, the 5 exposed panels of one enclosure will be radiating in the opposite polarity to the 5 panels of the other enclosure, which surely must produce a null at all frequencies when the listening position is equidistant to the enclosures, regardless of how much the panels move! Or is my understanding of the test procedure incorrect ?
Peter
Peter
It sounds like you do understand the test procedure (or, at least, you agree with me). In addition to the problem you point out, the way the front of the driver would be loaded is tremendously different than what it would be when operating normally. Resonances and their intensities would be vastly distant under the two sets of conditions. Also, the extra mass of the second enclosure and it's contribution to stiffness throw another monkey wrench in. The OP had a clever idea, but anything learned through actually employing this method will lead to the same place as using common sense... brace as much as logically possible, use stiff and massive materials, and consider constrained layering. I'll stick to one of my best investments, a mechanic's stethoscope.
A screw driver's hand to the ear and the blade on the job
Cheers
It works to a degree, can't argue with that. But let's do a little math...
Mechanic's stethoscope costs $10 dollars. The cost of a screwdriver is anywhere from $1 to more than $10, but let's get a cheap one to make the price less than half of the former, say $3. Factor in it's usability as a device for driving screws, plus the capability to stab intruders, minus the fact that you won't look as important as when you have a medical-looking device hanging from you neck (I try to wear mine everywhere I go), add in the screwdriver's durability, and finally subtract my beard from my face and into the belt system of my running car engine. The math suggests the stethoscope to be a slight value leader. Alternately, consider one of those gigantic (18"+) flathead drivers for listenning and for use as a lever to replace the bar you lost for your floorjack. Add tubing & earpieces and have something to talk about for weeks to come.
Your logic is oversimplified. According to your logic dipoles and all planar speakers should be silent. The only frequencies that will be cancelled by mechanism you describe are ones below resonance anyway. Resonance in a panel means wavelength shorter than twice panel size. Just trust me that wouldn't be a big issue.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.