DIY Espek Bases - How do upgraded speaker bases really work?

[Ceilidh]

Hi Everybody!

I've 2 theoretical questions, and 2 practical queries. Here are the theoretical questions:

Q1 (Theory): How exactly does an upgraded speaker base (such as the polymer/granite composite bases optional for many past Linn floorstanders) actually improve performance? Is it simply by adding mass to the speaker so that it doesn't shake around while reproducing bass notes? Or is there a more subtle interaction going on?

Q2 (Theory): Should a speaker base "ring", such that it does not absorb high frequency vibration and is able to vibrate at such frequencies? Or should it be somewhat damping, so that it does not reflect vibrations back into the speaker? (Note the word "somewhat": I understand from Mr. Lejonklou's past Topica posts that heavy damping (e.g., via loose sand or lead shot) kills the treble performance -- but to what extent do the Linn-approved polymer/granite bases damp vibration?)

...And, here are the practical questions, which will explain why I'm asking the above: :D



Background to Practical Questions: Linn Espek speakers are said to benefit greatly from the optional composite (polymer/granite) bases, and as these bases are fairly rare on the used market, I'm going to try a Do-It-Yourself (DIY) approximation for my father's Espeks. Here's the plan:

a. The stock Espek bases are hollow molded plastic. They can be filled with something heavy.

b. The something heavy will be 1mm steel shot (yes, I know the real upgraded bases use some sort of powdered or granulated granite, but I suspect the manufacturer employs a combination of high molding pressures + a range of grain sizes + vigorous industrial mixing to get the density up and the porosity down -- I can't do that, and so I'm stuck with a porosity of ~40%; steel shot is a lot denser than granulated granite, and it's very inexpensive on eBay).

c. The steel shot will be bound together with some sort of resin (which will also fill all the pores). Choices here are boat-building epoxy (extremely strong, fairly rigid -- the end product will "ring" somewhat), vs. polyester/vinyl ester resin (weaker but more rigid -- more "ringing"), vs. polyurethane (essentially a high-grade rubber, which is weaker and will somewhat damp high-frequency vibrations).

Q1 (Practical): Does anyone know how much the official Espek upgraded bases weigh?

Q2 (Practical): How rigid/brittle should the base be? I'd like to use epoxy (for a variety of practical reasons), but if need be I can maximize stiffness (by using Vinyl ester), or conversely I can minimize it (via polyurethane). Any thoughts on what sonic effects there might be, with varying material rigidity/brittleness? Opinions gratefully accepted! :D

Thanks very much in advance! One way or another, I'll let you know how this DIY project works out. Cheers!

-C

[SaltyDog]

Lead shot will be much better at minimizing vibration than steel. Steel wiil transmit very well. If you have seen the 5 steel ball pendulums in series this will give you a good idea of how well steel will transmit vibration. Lead is used extensively in industrial applications to reduce or eliminate vibration. Lead shot works better than solid.

I don't know but think that lead shot at the base - isolated from the cabinet - would be different than lead ON the cabinet.

Anything touching the speaker will affect the tune of the speaker.

I think the purpose of the base is to isolate the speaker from the floor and vibration.

[Ceilidh]

Thanks SaltyDog, :D

Agreed that lead shot would offer considerably more damping (and it would also be cheaper, and easier to attain high weights and densities). My concern is threefold:

1) Somewhere in the maddeningly user-unfriendly Topica forum, there are posts made years ago by (I think) Mr. Lejonklou, in which he mounted bookshelf speakers (Katans or Kans maybe?) on a base filled with loose sand or lead shot; he reported the result as surprisingly dull and lifeless.

2) On theoretical grounds, Mr. Lejonklou's findings make sense: Linn voiced and tuned the Espek's treble performance with a certain amount of cabinet vibration in mind. If I damp out that vibration with lead shot or sand, then I'll be changing the voicing. And since we're very happy with the Espek's treble and midrange performance, I'd rather not do that!

3) And on a practical note: I've got the optional polymer bases on my Ninkas, and they're a very interesting material. They don't "ring" like solid steel would, nor do they respond to rapping in the manner of a stone slab. But conversely, they also lack the deadening response of rubber, sand, lead, or loose shot. In very gross terms, they seem to have the acoustic properties of MDF while being much heavier -- which I suppose is what one might expect for granite particles embedded in a polymer binder....

All in all, I'm guessing that Linn was shooting for some moderately damped resonance in these optional bases. And after a couple of days of reflection (and thank you for the feedback, SaltyDog!), I think I'll go ahead with the steel shot & epoxy route: the steel will hopefully yield the weight needed to damp cabinet motions with low bass notes (the reports on the Espek bases tend to point to "tighter" and more focused bass performance), and the epoxy will provide some moderate damping at treble frequencies without making things completely dead....

We'll see how it goes. :D

-C

[SaltyDog]

Ahh I see you are saying that the base you are planning will be filled with steel shot AND resin mixed to form the base.

I haven't seen Espeks. On my keltiks and kabers the bases (which have always been on them) are attached to 4 feet. I don't see how they affect the resonance of the cabinet. The setup does allow the use of spikes and makes their adjustment easier.

Are the espeks attached in some other way? Do they have spikes without the bases?

Good luck.

[lejonklou]

Interesting project, Ceilidh.

I don't know the exact properties of the granite/polymer bases, but I can tell you that Linn had several companies manufacturing them for them. As quality and delivery time seemed to be a problem, they changed manufacturer several times. For the Kabers and Keilidhs, there were 2 different kinds of bases (made by different companies) and while one used less material and had a cross section like an inverted U, the other (much harder to find) version was completely filled and heavier.

I compared these two on a pair of Keilidhs and the completely filled and heavier one was clearly better.

Regarding lead it's true that I've had no success with that material so far. Seems to make things over-damped, which causes a numbness in the music. I'd go for steel shots if those two are the options available.

Best of luck and please do report on your progress!

[Ceilidh]

Thank you, Mr. Lejonklou!

Sounds like mass without gross damping is the goal; I appreciate the info on the solid vs. inverted-U bases.

(Also, just to let you know: there's an option beyond steel or lead shot that I'm not pursuing -- fine garnet grains are also readily available (and they're even less expensive than lead). These would more closely approximate the characteristics of granite. However, they're so light, and so much of the stock Espek base volume is taken up by plastic ribbing, that using garnet would only add ~8lbs to the weight of the base. So I guess I'll go with steel for the added mass (and if the rumored Majik 140 bases are steel, perhaps that's another reassuring note?).)

I'll let you know how it goes! :D

[Ceilidh]

Hi Everyone,

The DIY (Do-It-Yourself) Espek bases have been fabricated and installed, and they are a complete and unequivocal success -- a surprising success, in fact, and they constitute one of the biggest single improvements ever experienced on my parents' system.

As the following report is very long, here's a Table of Contents. :D All best, everybody!

-C

TABLE OF CONTENTS

1) BACKGROUND TO DIY PROJECT
2) DESIGN AND CONSTRUCTION
3) PERFORMANCE
4) TECHNICAL ANALYSIS


1. BACKGROUND TO DIY PROJECT

A while ago, we replaced the Aktiv (w/ Chakra x100) Ninkas on my parents' system with a pair of passive Espeks (driven by a single passive Klout). The results were mixed: in terms of Tune Dem, the performance seemed roughly comparable, and whilst the Aktiv Ninkas seemed to have more "detail" (e.g., you could hear violin pizzicatos more clearly), the passive Espeks had a tonal quality that somehow seemed more pleasing to the ear. There was a problem, however, in the bass: whereas bass lines with the Aktiv Ninkas were always clear, restrained, and in proportion to the rest of the music, with the passive Espeks there was a persistent bass "muddiness" that proved difficult to control. This muddiness was partly an outright boominess (which careful speaker placement could greatly control), and partly a waffly, murky "thudiness" (to coin an awkward word), whereby deep bass notes that should have sounded pure came out instead as nondescript low "thuds". It was the latter problem (the muddy "thud") that eventually defeated us, and we wound up using the preamp's bass management to divert deep bass away from the Espeks and into the subwoofer (a Sizmik); only then could music with a lot of bass energy be really enjoyable.

According to various Linn-centered forums, an upgraded Espek speaker base would perhaps correct the "thud" problem: all posts on the subject have extolled how these bases, constructed of some polymer+granite mixture, would "tighten" the deep bass (and mid-bass as well). I have a variant of these bases on my Ninkas and can attest to their (small, but discernible) effect on those speakers, and many have claimed the performance improvement to be far greater on the Espek than on the Ninka. (To quote "TSwam" on Topica (who sounds a lot like a respected contributor to the Lejonklou Forum! :D), Espek owners should "run, don't walk, to the dealer" to purchase the upgraded bases.....) So if our muddy "thud" is the opposite of "tight", then an upgraded Espek base would seem to be the solution....

The problem, however, is that upgraded Espek bases are very difficult to get hold of. They're out of production, most dealers seem to have sold off their stock, and they don't often show up on the used market. After hunting around for a good long while, my father and I decided to try a Do It Yourself project....


2. DESIGN AND CONSTRUCTION

What makes a DIY project possible is the unusual nature of the stock Espek base: in contrast to the Ninka base, which is a solid piece of MDF, the ordinary Espek base is a hollow piece of injection-molded plastic, with the hollow section opening downwards towards the floor. As such the base can be filled with a polymer-based concrete that's hidden by the (aesthetically-pleasing) plastic shell.

Since the "official" upgraded Espek base is said to be mixture of granite and polymer, the first thought was to fill the stock base with crushed granite in a resin binder. But on technical reflection (see section #4, below), I thought it best to try 1mm steel shot + epoxy resin. This combination hardens into an extremely dense, heavy, and rigid substance that is moderately damped against "ringing" vibration: in the "knuckle-rap" test, it sounds very similar to the granite & polymer material used in my Ninka bases, and as discussed in a previous post it will neither "ring" nor unduly damp out high-frequencies.

In terms of construction: the mixing & casting process is a right mess, and in the end I developed a healthy respect for the official optional base. That is, the polymer & granite base is probably priced about right for what goes into it, and if anyone actually has access to the real upgraded base, he/she should certainly go for it in lieu of a DIY item. The problem is not material cost (with judicious bargain hunting, the materials can be gathered for under $100 US), but rather the labor: whilst it is easy enough to make a slurry of steel pellets embedded in a resin matrix, such is not the desired end goal. Such a slurry, once hardened, would be fairly dense -- but its rigidity and general mechanical properties would be that of epoxy, not steel. In contrast, the desired material is a "close-packed" assemblage of steel grains, with each grain glued to neighboring grains by a microscopic film of epoxy. Such an assemblage has the rigidity of steel (or fairly close to it), and the epoxy filling the voids between grains serves primarily as a vibration damper. Producing such an assemblage, however, is pretty finicky and labor-intensive, and it literally takes several hours.

In the end, after a bit of experimenting, grinding with a Dremel tool, and generally making a sticky epoxy-based mess all over the workbench, we wound up with the sought-after close-packed steel & epoxy filled bases: each base weighs 22.5 lbs (~10 kg, in contrast to the stock 2.5 lb (~1.1 kg) value), with rigidity values close to that of solid steel, and acoustic properties (judging from knuckle/screw driver/ wooden stick/ etc. rapping) roughly similar to that of the official granite & polymer bases.


3. PERFORMANCE

After letting the bases sit outside in the sun to cure for a full day (and to vent off the various noxious fumes that gave me a headache the first evening), we brought the experimental pieces back inside, bolted them in lieu of the stock bases to the family Espeks, and sat back to listen. Neither my Dad nor I were expecting particularly much: from my prior experience with upgraded Ninka bases, I was hoping for a similarly modest improvement in bass-clarity; whereas my father was simply there in his usual role of moral support for his audio-hobbyist son (as a conscripted heavy machine-gunner back in the Olde Countrie before coming to America, he spent a good part of his youth rattling a .50 cal on the firing range without earplugs or ear protection -- hence subtleties in stereo reproduction are generally beyond him, and he's supported tweaks to the home system primarily because Mom loved the improvements :D). The test CD was Enya's "Watermark", selected because it's about the only CD in the house with a strong deep bass....

[I'll have to stop here (errands to run), and will hopefully pick things up later this evening or tomorrow -- if anyone's actually reading this thread & post, please don't post any responses until I finish the report/analysis, as it'd be nice to have everything all together. Thank you, and have a good Sunday!]

- C

[Ceilidh]

3. PERFORMANCE (continued)

[Ok, back again. :D]

We've left things with the new 22.5 lb (10kg) steel + polymer + injection molded plastic bases bolted onto the bottom of the family Espeks, the speakers placed in exactly their original locations, a bass-rich Enya CD in the CD tray, and myself and my army-induced-hearing-deficient Dad seated on the couch. We hit "play", waited a minute for the usual Enya shimmering intro to pass by, and....

Oh my goodness.

The reason I've cited my Dad's army experience is to indicate he's by no means a "critical listener". He can't hear what most tweaks & upgrades do (e.g., he can't hear the differences between an LK140 vs. a Klout vs. a 2250 vs. a Chakra, nor between single K20 vs. biwired K400 vs. passive biamped K400, nor between various connectors, interconnects, grades of solder, nor between Classik Music vs. Classik Movie vs. Unidisk SC (again, the home system was built up for the sake of Mom, who could hear and enjoy every upgrade)). But when the first low notes came in, my father laughed and turned with a look of surprise in his face: there were suddenly low bass notes that we'd never heard before, bass notes that were rock solid and pure and weighty and in-the-pit-of-your-gut concussive. It was a revelation.

The improvement an Espek experiences with the optional bases (and I'm assuming the "official" bases must work even better than my steel & epoxy DIY project) is far, far greater than the improvement undergone by an optional base-shod Ninka. There's really no comparison. As mentioned earlier, I've installed the official granite & polymer bases on my own Ninkas up in Boston, and the improvement was "nice" -- but the Espek upgrade is in an entirely different league. After listening to a variety of tracks on a variety of CDs, here are the audible improvments:

a. The "thud" is gone. Low bass notes that used to sound like muffled drum strikes now take the correct form as pure bass tones. One of the checks here involves switching from Stereo to Stereo-Sub (with the latter diverting the low bass away from the Espeks and into the Sizmik): before the base upgrade, the bass tones would sound radically different; now they're similar enough as to make no difference.

b. As mentioned above, the "reach" of the bass is much lower than before: very low notes have appeared where none used to be (apparently) present.

c. The bass in general is much louder (but not in the "boomy", "thuddy" sense that was previously so annoying -- the sound is instead "rich").

d. The Tune Dem benefits (and this one really surprised me) extend far beyond the low bass (obviously, if we couldn't hear bass lines before, it's much easier to "follow the [low-bass] tune" now!), and indeed go all the way up to the mid-range -- even to low female voices (e.g., on a Kate Rusby track, there's a section where I could tell 3 voices were singing in close harmony, but I could only vaguely pick them apart; now it's quite easy to sing along with any of the individual singers). More than that, the music is better: one can much more easily hear the subtle inflections that make a singer "great".

e. In sum up, the Espeks are now much nicer speakers. Before, it was unclear if they (in passive form) were really an upgrade over the previous Aktiv Ninkas (and the "thud" problem was really vexing); now, the system is simply delightful to listen to. :D


4. TECHNICAL ANALYSIS

As the performance enhancement was so much greater than expected (and even greater than what Internet forum posts have suggested -- was there something anomalously wrong with our original Espek setup?), and as by training I'm a physicist and this whole DIY project was taken on as a fun experiment, here's a quick little stab at accounting (perhaps!!) for the improvements. The usual caveats apply here: I've neither professional nor academic experience in audio engineering; I don't have the measurement or modeling tools to really analyze what's going on; and I've not cranked through the math -- so what follows is a sort of technical guess from an interested hobbyist. :D

A. Weight (Mass) is of secondary importance.

A lot of internet chatter talks of how important is weight (or more accurately, mass) in "damping" vibrations and "stabilizing" a speaker -- the idea seems to be that a heavy upgraded speaker base keeps the speaker cabinet from moving around and "wasting" the energy that should be going into the speaker cones, and that the suppression of cabinet resonances "tightens" up the bass by eliminating extraneous harmonics.

Well....

The above line of argument has always troubled me -- not because of the general reasoning (which I wouldn't fault), but because of the magnitudes of the effects cited. To grossly over-simplify a multi-faceted phenomenon (i.e., here's an example of a place where I don't want to grind through a huge amount of math), let's just look at the resonant frequency of a "simple harmonic oscillator / SHO" (which is a good a starting place as any for this type of discussion): a vibrating speaker cabinet will behave much like an SHO (in various vibrational modes, for the mechanicists reading along), and if the cabinet is to siphon off energy that should be going into the drivers, then its resonant frequencies have to be close to the desired audio driver frequencies (if you're unfamiliar with SHO, click on the following link, and set the "damping constant" to zero and the "Natural frequency" to something close to 20; you'll see how much the cabinet (red) will move when its natural frequency approaches that of the driver (blue, sort of)):

http://www.upscale.utoronto.ca/GeneralI ... enSHM.html

(The above link is fun to play with: if you set the natural frequency to 32 (with the damping constant at zero), you can see how cabinet resonances can generate unwanted harmonics: as the blue "speaker cone" (sort of) moves at a regular frequency, the red "speaker cabinet" takes on a more complex motion that can be decoupled into higher and lower harmonics....)

So in theory, the idea of adding-mass-to-tighten-the-bass makes sense: if bolting a heavy base to a speaker cabinet can shift the resonant peak (of various vibrational modes) downwards to a non-audio frequency range, we can drastically reduce the coupling between cabinet and speaker driver (or in plainer English, we can keep the cabinet from moving around and siphoning off energy / generating spurious tones).

The problem, however, is that you need a lot of mass to shift the resonance peaks! My DIY Espek bases, being made mostly of steel, are probably heavier than the Linn granite & polymer bases -- but at 22.5 lbs, they add only 42% to the mass of the stock 47.4 lb Espek (with standard base). This may seem like a large amount -- but the resonant frequency of a simple harmonic oscillator varies with the square root of (spring constant / mass): that is, by adding 42% to the mass of the speaker, I can shift the resonant frequency downwards by ~19%. If you work out the numbers, you'll find that a 19% frequency shift corresponds to a little over one musical whole step -- e.g., the difference between middle-D and middle-C. That's not a huge change, and it's hard to see how it can cause the radical transformation seen in my parents' Espeks. :D

(For the more technically inclined: the above discussion overstates the importance of mass in a speaker-base. If we look at the vibrational modes most likely to affect speaker driver performance (e.g., fore & aft rocking, bending, etc.), we'll find that an upgraded base puts the mass far away from where it's most needed -- so the effect should be even less than discussed above. Similarly, we can work out how the amplitude of cabinet motions change due to impulse & momentum effects, but we'll come up with similarly modest "improvements"....)

Anyway, if we work the numbers, we can convince ourselves that adding mass to a speaker base can change and perhaps improve the performance -- but if the speaker already is fairly heavy, then it's hard to convince ourselves that the improvement will necessarily be huge. (And in the case of the Espek, it's hard to see how it will have much effect at all.)


B. A Ninka (and Skeet?) Digression

Here's a sidebar that might be of interest in Charlie1 (as well as being a further argument against the sheer weight of the upgraded base being important):

While I was back home fiddling with the Espek bases, I took the opportunity to set up a second system in my parents' house (the components have been sitting in their boxes, waiting for eBay prices to recover somewhat before we try selling them; as they're just sitting around, I figured I might as well set them up). This system consists of a Classik Music, an LK140, and the Ninkas that the Espeks replaced.

As I was in a bit of a rush (only so much time before having to return to Boston), I simply placed the Ninkas on the carpet with the spikes in a semi-retracted position, figuring I'd come back and Tune Dem / level / lock down / etc. everything on a later visit. The setup was fast, I put in the same Enya CD that we used later to test the Espeks -- and the bass performance coming out of the Ninkas was much greater than we had ever heard from them before. So what was the deal?

Well, (and Charlie1, this is the part you might find interesting): I did a little "rocking" test on the Ninkas. When the speaker cones move back and forth, they tend to rock the speaker cabinets. To get a rough idea of the resonant frequency of this rocking motion, one can simply push (bounce) the top of the cabinet back (a tiny amount!) and let go; the amplitude and frequency of the resulting motion will be indicative. And with the Ninkas essentially sitting on top of a thick carpet (instead of before, when the spikes went deep down and anchored into the underlying wood), the rocking was big and (most importantly) slow -- so slow that I could almost count the swings, and so slow as to be far below the audio band (note: "slow" means a low resonant frequency). So basically the impromptu Ninka setup was akin to what happens if you set the sliders in

http://www.upscale.utoronto.ca/GeneralI ... enSHM.html

to zero damping and minimum (0.10 hertz) natural frequency: the resonance of the cabinet rocking motion was so far below the audio band that even low bass notes couldn't move it around very much. Hence we suddenly had much better bass performance than we had downstairs in the living room, when the Ninka's spikes went through to the wood -- and the resonant rocking frequency was considerably higher.

And so (for Charlie1): any ideas why it might work better to have a Ninka "floating" on Skeets (on carpet) than to have its spikes anchored to the underlying floorboards? :D


C. Rigidity in the Espek Base

Perhaps some of you now see where this is going: after playing around with the suddenly-surprising Ninkas, I went down and tried rocking the Espeks -- which were anchored via spikes to the underlying floorboards. The resonant frequency was much, much higher than with the Ninkas -- far too fast to count, and of a pace that I've seen bass speaker drivers move. That is, the Espeks on their stock bases would rock back and forth at a resonant frequency that was either close to or perhaps squarely within the low-bass audio band -- and that's why the low bass was either non-existent or muddy (again, play with the physics animation link given above!).

Now for the final piece of technical info: in the preceding post, I mentioned that we wanted the DIY bases to have more or less the rigidity of steel (which was a little messy to achieve). The reason for this desire was a hunch and a bit of eyeball engineering (i.e., no calculations or stress analysis, just a visual inspection of the stock Espek bases coupled with sad experience of other mechanical parts that failed to work properly in other applications.... :D): the stock Espek base is a trampoline. It's of heavy injection-molded plastic and appears well-engineered for strength and manufacturing -- but the rib pattern is not one of maximum rigidity. It'll flex. Minutely, but it'll flex. Moreover, the bolt pattern (the arrangement of mounting bolts vs. floor spikes) is one that does not brace the speaker cabinet very well against rocking motions, and the rocking motions that do occur will (because of fore/aft asymmetries in location and ribbing) will couple with vertical bouncing modes in a fairly complex manner. In short, the stock Espek bases appear (inadvertently) designed to allow the speaker cabinet to significantly rock back and forth when the (high-mounted) heavy bass driver tries to reproduce low notes (and perhaps even when the mid-range driver tries to reproduce higher notes), with the rocking motion directly coupling to more complex vibrational modes. So if the rocking resonance frequency enters the audio band, the result should be an acoustic mess -- and that appears to be what's happening.

Hence the desire for maximum rigidity. The optional Linn granite/polymer bases are heavy, yes, but they're also much, much stiffer than the stock bases (and Mr. Lejonklou, the beam rigidity of the bases will vary with the cube of the thickness (i.e., twice as thick will give you 8 times the stiffness) -- so the solid granite bases that you tried were probably much, much stiffer than the hollow/inverted-U bases that you found less effective). And to the extent that my DIY bases approach the rigidity of steel (reasonable under low-stress conditions, if the grains are in contact and connected by microscopic films of epoxy), then they too will be much stiffer than stock:

http://www.engineeringtoolbox.com/young ... d_417.html

The above table shows steel (see "Young's Modulus") to be ~50-100X inherently more stiff than most plastics. Add to that a ~10X increase in section area (because we've entirely filled all the voids with steel & epoxy), and we're roughly in the ballpark of ~500-1000X more rigidity. Using the sqrt(stiffness/mass) relation discussed above, we would expect the upgraded bases to shift the resonance peaks upwards by ~20-30X -- or about 4 or 5 musical octaves (contrast that with the downwards shift of one musical whole note, caused by the increased mass). This shift should put the resonance well up into the midrange (where cabinet-driver coupling should be less of an issue, due to the lower mass of the midrange driver); the drastically increased rigidity should also drastically reduce the amplitude of cabinet rocking at all frequencies.

In short, the rigidity of the upgraded bases (whether granite or DIY steel) should effectively eliminate base deflection as a major source of cabinet resonances in the Espek speaker, particularly those related to rocking of the cabinet in response to bass-driver motions. There will of course be remaining resonances, associated with bending of the underlying floorboards, of the cabinet walls and framing, of the mounting bolts and spikes, etc. -- but these are issues all speakers have to deal with, and the Espek will no longer have its injection-molded-plastic base trying to make it unusually boomy or muddy.

So that's what this hobbyist's eyeball analysis suggests. :D As a related sidebar, please note that the deeper a speaker cabinet is fore & aft and the stiffer its mounting plate/feet, the less base-related rocking will occur -- and interestingly the Akurate and Klimax floorstanders are quite deep, with metal feet....

Cheers everyone!

-C

P.S. (edit): I forgot to mention -- I tried the "rocking" test on the Espeks with the DIY bases installed, and sure enough the rocking was far smaller in amplitude and higher in frequency.... :D

[lejonklou]

FAT and extensive report! :D

Great to hear that your experiment was so successful. I wouldn't entirely rule out the possibility that your DIY bases are actually better than the stock granite bases.

This reminds me of when I compared the "thick" Keilidh/Kaber bases with the more common "hollow" bases (both from Linn but made by different manufacturers): I expected a small difference, but it was very big. Apart from musical differences, there was also a surprisingly large tonal change, all the way up to the treble. I was very surprised by this.

One question I have been thinking about is why Linn chose granite in their bases. Well, as I understand it, the concept of the Ku-stone dampers originated from another company (can't remember which, but they worked with acoustics). Linn picked it up and started using them inside their speakers, in the form of small square blocks that were glued to the back wall, often directly behind a drive unit.

Later they started making speaker stands of the same material. First in the shape of pillars to the stand of Index II. Soon after, bases with much more resin (the blocks had very little so they were porous and you could scrape off tiny pieces of granite).

I wonder if there was ever any comparison made between granite and (for example) steel shots. Maybe they just went with what had already been developed?

Back when the Katan was released and the original stands were quite wobbly (the first version of the bottom steel plate was not much bigger than the speaker!), I seriously contemplated creating an extraordinary alternative out of granite, resin and steel. But it turned out to be quite costly... and what if the first protos didn't sound right? Instead I just replaced the bottom plates with two 17 kg thick and rusty steel plates from the scrapyard. :D

If the Majik 109's are much better than the Katans, maybe they could be a candidate for a super stand?

[Charlie1]

And so (for Charlie1): any ideas why it might work better to have a Ninka "floating" on Skeets (on carpet) than to have its spikes anchored to the underlying floorboards?

Errrr - not really - although I was extremely happy to go along with your idea that its due to the rocking motion being much slower :D I did give mine a quick nudge and thought I could probably count the swings too, although I have no idea how they acted prior to the Skeets. I would imagine they would have made more of a thudding sound and come to a stop quicker.

A couple of things come to mind. Firstly, whilst I have found that the Skeets did improve the bass performance, it was the treble that grapped my attention most. Much more open and clear where before it was more diffuse and too faint in the mix. Do you think this also fits in with your findings?

Secondly, the Skeets also proved very beneficial under my rack where there is no rocking motion caused by speaker drivers, so I'm wondering if there are other factors at work.

[springwood64]

Am I right in thinking that the AV5140 is a precursor to the Espek, and would also similarly benefit from an upgraded base?

[Music Lover]

Yes, Espek is more or less a 5140mkII and you can upgrade a 5140 with Espek treble/mid/bass units. Still not going to be an Espek but it going to be better than stock 5140.


BUT...
Not sure if the bases are identical - anyone tried Espek bases on 5140?

[springwood64]

My brother is a stonemason and could make some stone bases for the AV 5140s, using the existing bases as a template (if I can persuade him!). If I go by the Youngs Modulus (compression) stiffness of various stones I can estimate the relative thickness against a solid steel base (200 x 10exp9 N/m2)to achieve comparable stiffness to the Ceilidh bases (assuming that stiffness increases linearly with the cube of the thickness):

1. Limestone: 37x 10exp9 N/m2 => approx 1.75 x thickness of steel base
2. Marble: 71 x 10exp9 N/m2 => approx 1.41 x thickness of steel base
3. Granite: 53 x 10exp 9 N/m2 => approx 1.55 x thickness of steel base

I'm assuming that if I get a close match to the stiffness of steel that I will hear the sonic benefits of the Ceilidh bases. One complication is that these values can only be considered very rough guides as rock composition varies enormously. For example, slate varies from less than 1 GPa to over 35GPa. Hmm.

In addition, the Ceilidh bases are wrapped in the original plastic bases. I would guess that this has some sort of damping that affects the transmission of resonance from the cabinet to the base and vice versa, although I'm not sure how significant this will be compared to the benefit of eliminating the 'trampolining' that Ceilidh has identified.

Any thoughts?

[Ceilidh]

Mr. Lejonklou and Charlie1,

I owe each of you a response, and will write you one anon, but first are some replies to springwood64. :D

If I go by the Youngs Modulus (compression) stiffness of various stones I can estimate the relative thickness against a solid steel base (200 x 10exp9 N/m2)to achieve comparable stiffness to the Ceilidh bases (assuming that stiffness increases linearly with the cube of the thickness):

1. Limestone: 37x 10exp9 N/m2 => approx 1.75 x thickness of steel base
2. Marble: 71 x 10exp9 N/m2 => approx 1.41 x thickness of steel base
3. Granite: 53 x 10exp 9 N/m2 => approx 1.55 x thickness of steel base

springwood (Pete),

1) My beam theory is a bit rusty (it's been a while!), but the cube-of-thickness relationship comes from the Euler-Bernoulli beam equation, which best applies when the span of a beam is great relative to the beam's thickness. For shorter (thicker) beams, such as the Espek bases, intra-beam shearing effects can become important, and thus stiffness will be a bit less than you'd expect. Hence you'll want to go a bit thicker if your goal is to match the stiffness of steel (as for just how much thicker: I don't know! I've never studied Timoshenko beam theory!)

2) Conversely, my bases are not solid steel: they wound up with a porosity of ~25%, so you can reduce your target stiffness by that amount.

As a wild guess, I'll posit than effects #1 and #2 roughly cancel each other out, so your suggested thicknesses are "probably"(!) in the ballpark. :D

One complication is that these values can only be considered very rough guides as rock composition varies enormously. For example, slate varies from less than 1 GPa to over 35GPa. Hmm.

Much of what the building trade calls "marble" is not marble, so you might want to stay away from that. Sedimentary rocks like limestone, sandstone, or shale -- or even worse, a sheet-like low-grade metamorphic rock like slate -- will often display weakness between the bedding planes, which can drastically reduce the beam stiffness. So a nice monolithic igneous rock like granite or basalt is probably what you'll want.

In addition, the Ceilidh bases are wrapped in the original plastic bases. I would guess that this has some sort of damping that affects the transmission of resonance from the cabinet to the base and vice versa, although I'm not sure how significant this will be compared to the benefit of eliminating the 'trampolining' that Ceilidh has identified.

The plastic shell may have some damping effect, yes, but I think (just based on material properties and relative masses) a much bigger effect in my bases is the damping caused by all the epoxy filling the 25% porosity between the steel grains. My epoxy & steel bases are probably fairly damped compared with solid granite -- but as Mr. Lejonklou has pointed out above, this damping might not be important in the proposed application (certainly the metal bases on the Linn Klimax and Akurate speakers suggest that damping isn't always needed in a floorstanding speaker base). In any event, you could always apply acoustic damping sheets (the heavy, tar-like stuff) to the underside of the granite, should damping prove desirable.

My brother is a stonemason and could make some stone bases for the AV 5140s, using the existing bases as a template (if I can persuade him!).

If I had had the tools and the ability to pursue this option, I would have! (The epoxy & steel route I followed was one of expediency....) If you go ahead with this project, please let us know how it turns out!!

Cheers,

-C

(P.S. -- glad the Klout-cleaning worked out! (Though isn't it amazing how poorly a partially-cleaned Klout can sound?) :D )

[springwood64]

Ceilidh wrote:

Though isn't it amazing how poorly a partially-cleaned Klout can sound?

Very much so. My first reaction was that I had damaged the amp. However the improvement brought by the Klout cleaning is behind my desire to sort the bases out in the AV 5140s. The bass now is much more prevalent, to the extent that it can be annoying and distracting, even when listening in another room.

[Music Lover]

Ceilidh wrote:

Though isn't it amazing how poorly a partially-cleaned Klout can sound?

Very much so. My first reaction was that I had damaged the amp. However the improvement brought by the Klout cleaning is behind my desire to sort the bases out in the AV 5140s. The bass now is much more prevalent, to the extent that it can be annoying and distracting, even when listening in another room.

It's not about "partially-cleaned" - it's about doing it correct!
ONLY mechanical cleaning (remove-insert-remove-insert etc) of everything you can remove and dont forget the chassis and especially the earth points.
And tighten all screws...

This applies not only to klouts!
Also, IC's, power cords, rack - yes everything really.

[Ceilidh]

It's not about "partially-cleaned" - it's about doing it correct!
ONLY mechanical cleaning (remove-insert-remove-insert etc) of everything you can remove and dont forget the chassis and especially the earth points.
And tighten all screws...

Hello Music Lover,

I fully agree!

But just to clarify: on my own Klout, I used mechanical cleaning only (plus screw tightening, etc.) - in fact, I basically followed your (much appreciated!) instructions from the old Topica forum. :D But there was one connector I couldn't budge at first, and as a result one stereo channel wound up behaving differently from the other channel. The effect was pretty awful! When I went back in and finally got access to the recalcitrant connector, everything became wonderful (much better than pre-cleaning) -- but there was a time in the middle that seemed pretty grim, and I was really having some second thoughts about the whole endeavor. (But it worked out well!)

...the improvement brought by the Klout cleaning is behind my desire to sort the bases out in the AV 5140s. The bass now is much more prevalent, to the extent that it can be annoying and distracting, even when listening in another room.

Pete,

I hope your stone bases work out -- certainly your problem sounds similar to what we had with our Espeks, and our steel & epoxy bases pretty much cured it entirely (the bass is actually even stronger than it was before, but it's no longer a one-note boom-whuffle-thud, and the strong bass is now an asset rather than an annoying liability). Hopefully you'll experience a similar improvement -- but I'm wondering: aren't the 5140 bass drivers different from the ones on the Espek? I vaguely recall Topica posts claiming "tighter bass" from retrofitting the Espek drivers into the 5140, so perhaps that's (also) part of the problem?

Perhaps some of the more experienced folks here can comment on the 5140 vs. Espek drivers?

In any case, I've a mild suggestion: it never occurred to me to check (i.e., I'm not sure this will work!), but it's possible that the bolts securing the 5140/Espek bases to the speaker cabinets have the same bolt threading as the spikes that hold the bases off the floor. If that's the case, then you can do a little pre-experiment to see whether the masonry-base option is worth pursuing: if the bolt sizes & threadings do match, you can remove the plastic bases from your 5140s, screw the spikes directly into the bottom of the speaker cabinet, and then set the speakers up (sans bases) in their original positions. Doing so will eliminate any "trampoline" effect caused by the original plastic bases, and will give you an idea of what a stiffer base may lead towards. If you hear a distinct improvement, then a stiff & massive stone base might be worth pursuing -- but if you hear no improvement at all, then your problem probably lies elsewhere(!).

In a related vein (and Charlie1**, could you maybe chime in here? - thank you!): you're undoubtedly already aware how important speaker positioning can be in determining the musicality and the sound of your system -- but for the sake of other folks who might be reading this thread in the future, I should mention that we wound up shifting our Espeks in half- and even quarter-inch increments (~1 cm) when searching for the best position. A small movement in the wrong direction had a surprising impact on the strength (and annoyance) of the original bass performance on our Espeks, and I suspect we could make the bass annoying even now, with our steel bases, if we moved things to a different spot.

Hence, if you perform the remove-base-and-directly-spike-the-cabinet experiment suggested above, please be sure you restore your 5140s to exactly the same position(!).

(And if you have not already done so, you may wish to TuneDem your speaker position, now that your Klout is putting out a healthy bass!)

Looking forward to hearing what you uncover, :D

-C

** - Charlie1, I know you have Ninkas instead of Espeks, but you can do a far better job than I of describing what TuneDemed speaker positions can lead to (i.e., I've enjoyed your posts on the subject!).

[Charlie1]

Charlie1, I know you have Ninkas instead of Espeks, but you can do a far better job than I of describing what TuneDemed speaker positions can lead to (i.e., I've enjoyed your posts on the subject!).

Thank you C.
I personally find Tune Dem'ing speaker positions one of the easier to perform now that I'm a little more practiced. I think this is because the 'sound' doesn't change significantly between small positional changes, so I find it much eaiser to hear what's changing musically and not be so easily swayed by better 'sound'. Whilst Tune Dem'ing Skeets under my Ninka's recently, I found it particuluarly difficult and had to go and hide in the downstairs toilet just to get away from hearing the differences in 'sound'! (My wife has given up with me as you can imagine)

I should mention that my Ninka's were in a particularly silly position first time I performed a Tune Dem, so the improvement was probably more marked than most owners would experience. Basically, the top end opened out, soundstaging improved, much faster transients and most importantly, a LOT more tuneful and musically whole. It was easily the best value for money upgrade I've ever done and really is well worth the effort.

Here's a good thread to read if you fancy having a go Pete and there are some great tips from the more experienced Tune Dem'ers!
http://www.lejonklou.com/forum/viewtopi ... c&start=20

[springwood64]

Thanks for your suggestions Charlie1 and Ceilidh. It seems my plan should be:

1. TuneDem the speaker positions (this will be my first serious attempt at TuneDem comparisons, so fingers crossed ....)
2. Compare A/A/B the AV 5140s on bases vs off bases

I checked the base fittings last night, and the bolts are M8 thread as are the spikes. I did a v quick comparison of the speakers directly on the spikes instead of the bases, but I could not get them stable (I didn't try very hard) and could not hear much difference. With the speakers on the spikes alone they seemed to sway in the breeze, which can't be good. :)

[Ceilidh]

Hello Pete,

Agreed: "swaying in the breeze" doesn't sound too optimal! :D Is it because the spikes aren't penetrating through the carpet?

In any event, your plan sounds like a good one: Tune Dem to get position first, then play around with base / no base etc. (Note: you can also try stacking a whole lot of encyclopedias on top of the speakers -- not to improve the performance (that'd be a pretty surprising "tweak"!), but to see if your 5140s respond in any way (negatively or positively) to changes in mass and resonance frequencies (e.g., if 20 lbs of encyclopedias atop the speaker have neither positive nor negative effects on sound & performance, and if comparing base / no base similarly yields no discernible differences, then one becomes a little less optimistic about the benefits of a custom granite base....).)

But do please try the Tune Dem positioning exercise first: the results can sometimes be surprising. In addition to the excellent thread already suggested by Charlie1, you can also have a look at:

http://www.lejonklou.com/forum/viewtopic.php?t=209

...and...

http://www.lejonklou.com/forum/viewtopic.php?t=248

The first thread contains suggestions from many of the forum regulars on how to proceed with Tune Deming speaker location; the second similarly has comments from the (expert) regulars specifically on the Espek (which is closely related to your 5140).

Good luck, and do please let us know how things go (thank you very much!). :D

-C

[springwood64]

Ceilidh and Charlie1 thanks v much for those links - I can see that this is going to be a fairly lengthy process. To start the tune-dem refinement of speaker position I placed the speakers wider apart and much closer to the wall (about 9"), which has already made a big improvement. I've not got to grips properly with tune-dem, but I can already hear big changes :D

I have also noticed that the bass problem is much reduced for the LP12, but still prevalent for my Arcam DVD player.

Having spoken to my brother regarding the stone bases and pointed him in the direction of this thread, I've piqued his interest and he has identified some suitable stone. I'll report the outcome if we proceed.

[Charlie1]

To start the tune-dem refinement of speaker position I placed the speakers wider apart and much closer to the wall (about 9"), which has already made a big improvement. I've not got to grips properly with tune-dem, but I can already hear big changes Very Happy

Good to read that you've already heard a benefit. It can get a bit tiring after a while, moving heavy speakers back and forth, but it's well worth it in the end. Besides, and as you're already doing, you don't need to finish it in one go. As you've probably read, it's best to be relaxed when doing a Tune Dem. If you're tired or in a rush then the results are likely to be confusing.

[springwood64]

I've been bedding in my new speaker bases and I've got to the point where I can report the results.

In the interim I swapped my AV 5140s with a pair of Espeks. These sound a bit sweeter than the AV 5140s but can still suffer from overblown bass despite experimenting with positioning over a couple of weeks.

So to the new bases. My brother had some 40mm thick 'grub' limestone which he had previously used as a base for his own speakers. I sent him comprehensive dimensions for the stock AV 5140/ Espek bases, with particular attention to the locations of the bolt holes that attach the bases to the cabinets. He cut the limestone to the exact outer dimensions of the Linn bases and drilled counter-sunk holes for the M8 bolts that attach the bases to the cabinets, as well as drilling 9mm holes for the M8 spikes.

I bought some knurled brass M6 threaded inserts and epoxy-ed them into the 9mm spike holes.

A couple of days before Christmas we fitted the new bases. It was a bit of a tense moment as I located the bolts through the drilled holes - my brother had found drilling the hard limestone very difficult and the measurements only allowed errors of up to 2mm in the locations of the bolt holes. To our relief the bases fitted perfectly.

Immediately it was obvious that the bases make a distinct difference. As I'd expected from Ceilidh's report, bass was much better defined. Bass notes had a definite beginning, middle and end, rather than oozing out and wandering around the room, and subsequently I have heard new bass tones I'd previously missed. But more noticeable was the much brighter tone of the speakers, making the Grace Jones LP sound distinctly harsh and splashy in places.

Over the next week or so I tweaked the set up by inserting sections of bike inner tube as washers between the limestone bases and the speaker cabinets, adjusting speaker positions and swapping cabling around. I also tried to shine some light on Ceilidh's original question by reducing the coupling between the cabinets and the bases, and between the bases and the floor.

So, what is the verdict? Well, although the bass has benefited, the big winner is the treble, with the impact on the mid range appearing to be pretty neutral. The speakers have become much more revealing of changes earlier on in the signal path and in changes to positioning. The early harshness was down to inferior cable to the treble (I'm tri-wired+bi-amped). I'm still not qualified to talk in terms of tune-dem, but it is clear I'm getting a lot more information and the overall result is much more enjoyable.

However, I'm not convinced that my experimentation produces results that are consistent with the rigidity of the base being the primary factor in this change. If I reinstate the old bases, then the old sound re-appears like a sonic shroud. If I attach the new bases loosely on their rubber washers, with or without spikes, I get most of the improvement. In this arrangement the speakers are free to wobble quite a bit. I do get the best results with bases fastened securely and spikes in place, but the fact that I get most of the improvement just by having the bases loosely attached implies to me that rigidity is not the most important factor. Any suggestions?

Ceilidh, presumably you could also test this by loosening the bases of your Espeks and/or removing the spikes?

If you own AV 5140s or Espeks and don't have the ku-stone bases, I strongly recommend upgrading the bases. It is the second best upgrade I have done (after getting the AV 5140s in the first place).

[springwood64]

I've had more time to experiment with coupling between the bases and the cabinets, as well as swapping cables (again), changing spikes and adjusting speaker positions (again). Consequently I now think that I was wrong to say that rigidity is not the primary factor.

I have ditched the rubber washers between the bases and the cabinets in favour of 35mm steel washers. This time I performed a controlled A,A,B helped by my wife. It was clear that the washers controlled the bass further - pretty well eliminating boom. So now the speakers are very tightly coupled to the bases, and I've found myself rummaging through my record collection to discover what my LPs really sound like :mrgreen: .

I think that my previous results were clouded because the bases were sitting on inadequate spikes, which allowed a remainder of lateral movement. I have replaced these with broad spikes that use 15mm knurled rings which lock them securely to the bottom of the base.

[lejonklou]

Thanks for the report, Pete! Interesting stuff and I'm releived to hear that rigidity/stability proved to be important.

Maybe I feel that way because it better matches my experiences. If the loudspeakers are "free to wobble quite a bit" as you previously wrote, I tend to suspect something is not right.