A pointed discussion

July 27, 2017
 by Paul McGowan

In yesterday’s post, I shared some common wisdom about spikes that turns out to be wrong. Thanks to my attentive and generous readers, I stand corrected.

The original thought process behind spiking speakers was not to isolate, as I had assumed, but the opposite—anchor the cabinets to the floor and transfer some of the excess cabinet energy into the floor. We don’t want the speaker cabinets to sing.

If the speaker sits atop carpet (which typically has a squishy pad underneath), it is effectively isolated from the firm anchoring possibilities of a cement floor. And it can be tippy.

All that aside, the use of spikes is still a hotly debated subject. Stereophile’s Art Dudley, for instance, suggests selling them to the scrap dealer in his Universal Tweaks #2.

Richard Murison, of Bit Perfect and Copper fame, posits: “I think the core point about spikes is that they simplify the mechanical interface between speaker and floor. It’s not the only solution to that problem, but it’s probably the simplest option – particularly with a rigid (i.e., concrete) floor. The problem gets much thornier, from a theoretical perspective, when the floor itself cannot be assumed to be rigid (such as a suspended floor) where the amount of damping in the speaker-floor coupling mechanism enters the equation.”

Like anything in audio, opinions abound. I have heard spikes and cones help, especially on equipment and turntables. Speakers always sound different when spiked, not necessarily always better. But that’s my experience. Your mileage may vary. Certainly, there’s no need to spike the giant 1.2 ton IRS V in Music Room One.

The original point I was attempting to make, that of keeping room vibrations alive to heighten the sense of reality, remains in the forefront of my thinking.

Subscribe to Paul's Posts

32 comments on “A pointed discussion”

  1. Thanks, Paul, for supporting my view of physics of spikes and my personal experiences with tweaking of stereo components. However I still haven’t got an explanation how Arnie Nuddell adds a delay to the tweeter in the middle of the tweeter column of the IRS V thus that the signal of the top and bottom tweeter arrive at the same time at your ears. And absolute weight is not synonymous with rigidity. The towers might be fixed at the bottom. But even shyscrapers are flexible and their top floor swings some meters. 🙂 Even the heavy but smaller Eiffel Tower.

    1. He does not delay them. I wish there were a way to do that but unless you have access to the signal in the first place and are willing to modify that signal I am unaware of any way to do that. Arnie likes the line source and it seems to work extremely well. Maybe Mark (Soundminded) can help us understand. I wish I knew more.

      1. Easy problem. You want all of the sound to arrive at a the same point at the same time? Choose the point in space you want, calculate the difference in arrival times based on the distances and the speed of sound, use a separate amplifier and time delay circuit for each driver, the closer the driver to the point you’ve chosen, the more delay you use, and then then you can have all of them arrive at the same chosen point at the same time. You can even compensate for the slight differences in loudness too. Another method is to arrange the drivers along a the curve of the arc of a circle instead of a straight line. This will focus the sound at one point like the radii of a circle, each one equidistant from the chosen point. The second method is cheaper. If you use the first method and arrange them along a straight line you might want to progressively tilt each one so that they point in the same direction at the focal point. This will result in the same FR from each one at that point.

        BTW, I just woke up, I haven’t even had my morning coffee yet, and this took all of one second to think of. My brain is still foggy. I almost said parabola instead of circle. If I gave it more thought, I could probably come up with some other schemes. If I can turn a problem into a geometric problem, I can usually solve it. that’s the way my mind likes to think about things. Will these ideas work? Yes. will the results be worth the effort? I dunno? I don’t care. It doesn’t really interest me. Good luck if you try it. Let me know if how it works out.

        1. Thanks, Mark but I don’t think it’s that easy. For example, how do you delay the signal? It comes in analog. Sure, there are old analog delay lines but in my experience, they muddle sound. If you’re willing to digitize everything then it’s of course easy.

          The real challenge is creating a passive line source with delays as he is asking.

          1. Actually it is easy. Many digital delay units have analog in, analog out. The A/D and D/A converters are built in. That’s what I use. You’ll need one for each pair except the driver that is furthest away. You’ll also need a lot of small amplifiers instead of one big one. I’m not sure why you want to do this. if you figure about one millisecond per foot (it’s actually about 10% less using 1100 feet per second as the speed of sound) the worst case difference between the soonest arrival and latest arrival is probably only about one millisecond, 2 at most.

            The second method costs just about nothing except for the curved baffle to mount the drivers on. In that arrangement you can use just one amplifier for all of them. I’m still not sure what you hope to achieve. Sounds more like an experiment.

            1. Here’s how to test and time align the drivers. Note, time alignment does not necessarily mean phase coherence when sound is propagated from more than one central point, that is the geometric center of propagation.

              You’ll need a pulse generator, a microphone, and a dual channel storage scope or a single channel storage scope and a mixer. Set the microphone at the desired focal point. Test one driver at a time. Insert a single pulse into the driver and note it as time zero on the scope. Then note the arrival time of the pulse at the output of the microphone. The time difference will be the group delay of the driver, that is the time it takes for the pulse to be translated into a mechanical motion of the cone, the transit time between the speaker and the microphone, and the group delay of the microphone. When they are all aligned, this delay time, the difference between the electrical input to the driver and the corresponding electrical output of the microphone will be the same for all of them.

  2. There’s a couple of eco-warriors in the Bavarian foothills hand-making speaker stands out of FSC-certified trees with no wood or plastic involved. So definitely no spikes, and no levelling mechanism at all. That’s fine in Germany because in Germany the floors are perfectly flat. I’m not aware if there is a market for eco-audio in Trumpland, but they have proven very popular over here.
    They have an association with a remarkable audio and vinyl dealer near them that has a building in the countryside with no less than 7 listening rooms from small to very large.
    For such sensible people, they used to sell magic crystals and blocks. They do sell the (in)famous Cardas Myrtle blocks, little lumps of wood that if placed under speakers or anything else allegedly have astonishing audiophile properties. Something to do with the ability of Myrtle wood from Oregon to transmit vibrations better than any other species of tree known to man. As an alternative, there is a TonTrager eco-block, technically called a “magic cube”.
    Perhaps this is a good example of when good people set out to make a product that is both functional and will contribute to mankind’s continued existence, meet some audiophiles and 20 minutes later they are making something with “magic” in it. It’s a disease, I’m telling you.

  3. On a slightly more serious note, many floor standing speakers have feet or spikes in the base, and standmount speakers have bespoke narrow-profile stands. These are often inherently unstable. You have to tell the kids not to touch in case they topple over.

    There is a very popular type of metal open frame stand, made to measure for a wide range of speakers. The key feature is that the spikes are set into outriggers, so the contact area is much wider that the weight being supported. You have no idea how incredibly stable these are, they are lightweight, very rigid (square tube steel) and cheap.
    Makes the point that spikes in the base of speakers mostly fix them vertically, this arrangement also fixes them laterally. Given drive units move laterally in most speakers, lateral movement would seem to be more the issue.

  4. Sorry to complicate matters Paul but there is a whole class of speakers which are designed to dissipate energy rapidly by letting the enclosure sides vibrate at inconsequential low frequencies: those following the BBC 1970’s speaker design rules e.g. Harbeth speakers. These stand-mount speakers require care with the speaker-to-stand interface so that enclosure wall energy dissipation is not compromised. I’ve found that replacing the ‘traditional’ 4 x Blutack blobs as coupling between stand and speaker with 3 small metal discs produced a noticeably clearer, faster sound. I guess the Blutak blobs integrated the speaker and stand sufficiently to allow some slow energy dissipation to occur thro’ the stand that muddied the sound.

    1. A most convincing presentation and argumentation, Max, at a first glance. But sadly the direction of the measured vibrations/ accelerations are not clearly shown in the video. I doubt they have the same direction as the membranes of the speaker shown. What is the eigenfrequency of the platform depending on the speaker’s weight? I assume there must be a kind of pulsating Doppler effect when the speaker is swinging due to the forces correlated to the membrane movement. Maybe this effect is not audible. However finally the proof is in the pudding.

      1. The seismograph is showing acceleration in three planes X, Y and Z. For the app, go to: Android. Playstore. Seismograph. Calvico. Double swipe to amplify. (Unfortunately, there is no really good Mac seismograph.)

        With the tablet in the orientation shown in the video, the top trace, blue, is Z, (forwards and backwards), the middle trace, pink X (side to side) and the bottom trace, green, Y (up and down). It is a valid scientific test.

        We have been isolating speakers on very soft springs since 1990 and have sold thousands and thousands of units and choose springs to suit the speaker to achieve a resonance between 2Hz and 3.5Hz. This requires a compression of 20-30mm from no load to full load. (springs are pre-compressed 15mm). Speaker weights range from 20 lb ( 10 kg) to 500lb (250kg). We have only had one return of a set of Podiums and that was from a Punk Rocker who missed the bass thump that travelled through his suspended floor into his feet. He admitted that it improved the sound everywhere else though. Yes, the proof of the pudding is in the eating. Check townshendaudio.com

        There is no noticeable Doppler effect on normal sound in life as we have only ONE eardrum in each ear, handling the whole frequency range.

        Also, the cabinet movement is really small due to the huge difference in mass between cones and cabinets, 1000:1 or more (-0.01dB). There is more movement between the front baffle and speaker basket due to the compliant nature of MDF/wood, creating flimsy “launch-pads”. This problem is worse with electrostatic speakers where the launch-pad is a flimsy mesh, hence the lack of punch from such speakers.

        It is a poor concept “to drain energy away” through footers, as that is amplifier power wasted which could go into sound. Further, that drained energy travels straight through the building structure and into your the neighbours house, causing severe distortion on the way.

        If the speaker is free to move in space, there are no bending forces between the cabinet and the floor and no resistance to motion (and no big resonance). Therefore there is no work done, so no energy is taken out of the system. (work is force times distance and this requires energy). If there is no force, there is no energy (lost). This is why you system sounds a little louder when isolated (as well as sounding much better).

        Seismic isolation works for all speakers, in all houses, on all floors every time!

        BTW, carpet, foam, tennis balls etc do a little, but usually sound worse.

        1. Many thanks, Max, for the most detailed explanations. I think the most relevant data is the resonance frequency varying between 2 and 3.5 Hz. Thus the “tuning” of the speaker-platform resonance frequency seems to be similar to the tuning of the (lateral & vertical) tonearm-cartridge resonance. 🙂 What about having tuned a stand-mount speaker. Is there a platform for the compact sized speaker? Or do you recommend the platform placed below the stand?

  5. OK Paul,

    So after the correction, are you saying that you’d rather have the speaker’s spikes in direct contact with the floor (say, no carpets then)? Interesting debate for me at this moment as I was toying with the idea of re-carpeting my room.

  6. Well I guess I am more simple-minded. Not literally but I have as noted in yesterday post about this subject I have my 93lb speakers on sub-dudes from Auralex which on located on a cement floor with carpet and padding. Maybe it’s me but I hear things differently than I did with the 803’s directly on the floor. Yet, again repeating myself some people believe cement floor is the best for speaker placement when it comes to sound. I have had it both ways and having the 803’s on the sub-dudes with my hearing…the ultimate tool, my ears…they sound better than being directly located on the floor. This also IMO gives a better flavor to the low end. The sub-dudes act as a platform and serve as complete isolation from the floor which IMHO ALL of the sound flows thru the air rather than having any inference with sound waves bouncing off a cement floor.

    Again, another theory or application that is subjective since everyone has different rooms and dimensions and search for that music nirvana in a way that they hears it best.
    There in not one set of rules for 2channel.
    Speaking of rooms…since someone mention Paul’s. My room was constructed according to the Auralex company. Two sheet of drywall thru out the room. Walls and ceiling 1.5″ & 5/8″ suspended on metal track and in between sheet bloc with mineral fiber in between the joists.
    All the work, time and money and I have been told by a notible in the industry who’s name I will not mention tell me that my application is the worse. Drywall is despised by him as well as Auralex.
    His application uses wood veneer at $60-$80 per sheet.
    Having a different idea than your neighbor is one thing but to scorn a company for there ideas I believe is out of line.

    Always interesting Paul,

      1. Stillpoints & Nordost cones claim to drain vibrational energy from the cabinet & con vert it to heat using steel or ceramic ball bearings inside the device. I have Stillpoints under my speakers & they sound noticeably clearer. Neither of these devices will pierce carpet.

  7. “The original thought process behind spiking speakers was not to isolate, as I had assumed, but the opposite—anchor the cabinets to the floor and transfer some of the excess cabinet energy into the floor. We don’t want the speaker cabinets to sing.”

    How do you do this? Let me count the ways 🙂

    Villchur and Kloss built their speaker boxes out of 13 ply marine grade plywood. They also filled them with lots of fiberglass for other reasons. Some internally braced theirs. Wharfdale put one speaker box inside another and filled the space between them with sand. Bose put his speakers in a box where the speaker baffles had no parallel walls behind them to prevent standing waves. Dick Diamond built his boxes out of slabs of aluminum. Dick Sequerra built his in the shape of a triangle. Parts Express built their best subwoofer enclosure out of 3/4″ MDF and braced it internally. Peter Qvortrup built his box out of Russian white birch so it would sing the right song. Atkinson tests for this by rapping the enclosure with his knuckes to hear what the boxes sound like (I call that the ape method.) One Russian guy built his sub woofer enclosure out of marble. Another guy built one out of concrete. Personally I like the box inside a box with sand idea myself. How do you feel about putting an 8″ woofer inside a box that weighs 200 pounds?

    Will spikes increase the transfer of energy from the box to the floor? I don’t know but my guess is it won’t. Usually transfer of energy requires a large contact surface with close coupling over a wide surface. That is why power transistors have metal cases, larger surface areas, and use thermal coupling grease to transfer heat to heat sinks. But I don’t know. They certainly won’t keep the box from resonating just as a transistor heat sink won’t prevent a transistor from generating heat. You want an enclosure that won’t resonate much? Make it in the shape of a sphere. Spheres are pretty strong for their size and weight.

  8. An interesting thought came to mind. Suspend a speaker with good bass response from a single point so that it stays just clear of the floor. Once it is stationary play something with good amount of bass and observe the movement of the speaker if any. This should help answer whether spiking is necessary or not. And yes make sure that the connecting cable is not to stiff since a stiff cable will tend to anchor the speaker. Regards.

  9. I don’t get it this spike argument. Connecting a speaker to the floor (or wall) chokes the speaker. Yes.. you will appear to get a cleaner sound because its constraining the speakers ability to project lower frequencies. Cabinets will vibrate to some extent. That vibration is a part of the sound that particular speaker produces. Might as well listen to your speakers in a semi anechoic chamber if you lock them to a solid surface. Speaker cabinets need to breath to an extent. In a concert hall big bass drums are hung from straps to allow them to float. Lock that same bass drum to a rigid stand connected to the floor? And you will wonder where all the impact and depth went.

    If you do not let the audio cabinets breath to some extent, you will be denying yourself of some of the spectrum that the speaker can produce. That is why some reviewers listening nearfield desktop prefer placing their bookshelf speakers on some type of semisolid foam. IsoAcoustics also produces stands that do not choke the life out of the speakers.

    You must either choose to resonate or choke. To resonate allows for a more organic sound. Too much cabinet vibration will smear the sound. Too little will shrink the sound. You want absolutely real? Look at a bass amp used by a musician. Look at the amps used by guitar players. They are not locked to the floor. Do that? And, you will choke the sound to some extent. That is why drummers many times will seek tom mounts that allow the toms to semi float when mounted to a bass drum. Lock em tight? It will choke the tone of both the toms and bass drum. I know. I witnessed to is and was heartbroken when I bought a new set that locked everything tight to each other. No more jiggle when playing… and the beautiful tone was gone.

    To some extent, audio speakers are still a music reproducer in themselves. Possibly, passive radiators help minimize certain cabinet resonances by giving the vibrations a place to go in a countering direction.

  10. Imho, The most substantial effect of spikes and speakers I have found is that it seems to ” tighten the bass response ” i’m not an engineer or acoustician like you guys but my take on it was that it prevents micro movement of the speaker which enables the speaker to have a tighter response. I have found that some speakers sound better without spikes for example my old Maggis seem to sound better with respect to midbass without spikes. Now that was the 3.6 is the 3.7i Version were better with spikes.
    But in most cases with conventional speakers spiking seems to be a good thing. But again I never thought of it as a coupling
    Or de-coupling scenario as with Electronics. More like a stabilization affect

  11. I make vibration cancelling subwoofers from Tymphany Linear Array Transducers which have dual opposing motors. Problem solved.

    Unless you are going for the “they are here” illusion where the music shakes the room. Ask any cellist “To spike or not to spike?” and they will reply, “Move that rug so I can dent your floor!”. This principle is made more consistent by Paul Galbraith’s custom 8 string guitar, which has a big spike coupling to an acoustic amplifier for solid bass notes. Note that added false floor resonator for the recording session:



    Opinions are more divided for bass clarinet, contra-alto clarinet, bass sax, bassoon, contrabassoon, contrabass flute and contrabass recorder. Some spike, some use rubber tips and some use neckstraps. (I have recorded all of the above). There are also singers who insist on a bare sprung wooden floor.

    The most interesting case was a novel patented piano leg. This dropped my jaw to the floor by turning a beat, cheap Yamaha 6′ practice piano into a nuanced beauty worthy of the finest halls. The room full of musicians and music faculty was likewise dumbfounded. The design sprung the body of the piano vertically, and with greater compliance, laterally. The vibrations in stringed instruments resonate by traveling end to end because they are plucked, struck or scratched (horse tail bows). Isolating these vibrations from the floor made the piano SING.

    1. Careful, these guys state “Power Amplifiers collect massive amounts of radio frequencies every millisecond and spend MOST of their power supply storage amplifying this airborne RFI. Amplifiers heat up due to RFI amplification”. Just one example of the bs on this site.
      Not our theme.

  12. So you folks intend to remove excess energy by transferring it into the floor one way or another. Sort of like “draining the swamp.” 🙂

  13. I gave decoupling my Vandersteen Treo CTs a shot a few months back just to see what the sonic benefits/drawbacks might be. I rested the speaker spikes into small round decoupling bases. The decouplers rest on carpet over a concrete floor. The clarity and smoothness to the sound was immediately obvious. The speaker vibrations were no longer finding their way to the spiked equipment rack and thus my equipment. Furthermore vibration from one speaker was not traveling up the spikes of the other speaker anymore.

    Uh oh! I have two 2wq subwoofers as well. Now it was time to decouple them. I purchased much larger round decouplers to put under the audio points I use on the subs. The instant perception was of less but tighter deep bass. I turned up the output level of each sub. to compensate and it sounds fantastic now. A great side benefit for the rest of the family is that the sub bass no longer rattles the rest of the house!! My wife loves that almost as much as our Westie who no longer thinks big orchestral bass drum thwacks are thunder!

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram