DIY Modular Hifi Rack

[Garnett]

I'm thinking of building my own Hifi rack. Having been inspired by TNT Audio's design for the "Flexi Table": http://www.tnt-audio.com/clinica/flexye.html and another thread I read recently but cannot find now about someone's ikea "lack" table-based idea, I thought I might combine the ideas, and make something unique.

Please can you look at the design below. It shows one leg of a table which will have one leg at the rear and 2 at the front (three in total for ease of balance - I know TNT recommend not having a leg running at the centre of the back of the units because it gets in the way of the cabling, but I prefer it for aesthetic reasons). I have sourced all the materials and this can be put together for £40 plus some postage costs I do not know yet. I expect the total build to be fairly short of £60.

Does this seem like a good design? Are there any aspects I've neglected? One problem I envisage and cannot counter yet is one of balance. When I tried the suash ball technique, I noticed the weight in my cd and amp are not evenly distributed (why should it be?) and so the units tilted significantly. Any suggestions? I do not know how firm the damping sheeting is (it is incredibly heavy - each layer will weigh nearly 2 kg).

Any advice/comments/words of warning would be greatly appreciated





Thanks.

Garnett

looks fine, adjustability will be built in if you use threaded connectors, i guess, you might also want a nit and washer inside (between mdf and bitumen) as well as outside, and use rubber and plastic washers, to isolate shelf from rods a little more.

[Garnett]

Thanks analoguekid. I hadn't realised the connectors would allow me to adjust the height of each leg but that is, as you point out, a very useful feature.

On further reflection, and on finding out the damping material is thin - 2.5 mm (or 5mm if I want the heavier stuff), I've had to revise the design:

I've countersunk the bolts so that the damping material runs flush over the lower MDF board, so it should work more effectively. I should be able to sacrifice the capacity to add another bolt and washer as suggested, because the damping material now runs uninterupted over the whole layer - this should halt the transmission of any vibration between each pair of MDF boards.

I've decided to thicken the MDF solely so that the machine screws have more directional support since it is important they act as firm legs for each layer. For this reason the upper of each pair needn't be any thicker than in the original diagram, but this is a matter of pence and will eventually be decided on aesthetics.

[Dexter]

Hi Garnett,

Here's the link to the IKEA LACK rack you mentioned above:

http://www.musiikin.com/lackmod.htm



Looking at what you've shown us thus far, I have a couple of questions for you.

First, precisely what equipment are you going to be putting on your rack?

Second, when you refer to the threaded roofing bolts or countersunk machine screws, what diameter are you planning to use and are they to operate as the legs themselves or are they at the center of something more substantial?

Cheers, Dex

[Garnett]

Dexter wrote

Looking at what you've shown us thus far, I have a couple of questions for you.

First, precisely what equipment are you going to be putting on your rack?

Second, when you refer to the threaded roofing bolts or countersunk machine screws, what diameter are you planning to use and are they to operate as the legs themselves or are they at the center of something more substantial?

Hi Dexter,

Thanks for that link.

In answer to your questions:

I'm going to be putting a tuner, a CD Playe, and an Amp on this rack.

I intend to use M6 (6mm) screws (because most of the commercial spikes available have m6 threads).

I was going to use the screws on their own. I thought they should be strong enough. Do you think this will be a problem?

Let me know if you have any thoughts.

[Garnett]

Here's another picture to make the whole thing easier to envisage. I intend to use different length machine screws for eac level so that each component on the rack fits snugly into its layer.

At the moment I'm also wondering about reverting to the thicker top layer of MDF so that I can drill holes in it to recess the spike receptors, to improve looks, and also decrease the chance of a heavy bump knocking one layer off the one below it.

Garnet PM tecnobear, he made a fine modular rack from ikea lack tables, had adjustable spike and looked like a fine modular design, there is a threadf somewhere on here about it, but I can't find it, it's similar to what dexter mentions, but with out butchering the shelves.

[Dexter]

analoguekid wrote:

it's similar to what dexter mentions, but with out butchering the shelves.

Yes, I've used LACKs before but never bothered to tear into them.

A few further thoughts, Garnett.

As far as I know, there are two schools of thought regarding the handling of resonant energy with equipment supports:

1) Coupling (low mass & spikey)

2) De-coupling (high mass & squishy)

By attempting to combine both it seems that you are trying to achieve conflicting goals here. On the one hand, you are trying to couple your equipment and shelving to ground with the spiking arrangement. But on the other hand, you are also attempting to de-couple via the damping and rather substantial shelving sandwiches (in at least 2 of your examples).

I don't believe that they will be effective if employed together.The damping will effectively hinder the spikes from doing their job properly, that it to channel resonant energy to ground. Rather, it will tend to hold and then ultimately rebound that energy in an uncontrolled manner. And for massy, we're talking granite here and lots more squishy damping stuff.

And in either case, I'm not terribly confident that those 6mm screws would be sufficient and not be prone to torquing under load.

So I think you need to decide upon one direction; I chose light & spikey with my DIY-modded rack and have gotten tremendous results.







Best, Dex

[Kiang]

I look at things from a different perspective but perhaps with the same design goal.

CD players(source components ) are inherently vibration producing devices while amps are required shielding from vibration. Both need de-coupling but the source in addition needs vibration absorption.

http://www.auricles.com/new_page_15.htm

this looks at solutions for amps



http://www.auricles.com/new_page_47.htm

http://www.auricles.com/new_page_16.htm

These two above are my solutionsfor source components.

[Garnett]

Dexter, thanks and wow. That is a gorgeous piece of design.

Your advice is much appreciated.

I can understand the theory behind De-coupling, but not coupling. The idea of having heavy masses that can "ride out" external vibrations makes sense to me. Having read a little about it it seems to me coupling is reliant of "fixing" or "coupling" everything together so either it all moves or none of it does. This is why three legs and sikes are good because it is "concentrating its mass to at least 3 or perhaps even 4 small spots". So it would appear mass is important in coupling as well.

The damping shouldn't rebound energy. It should be dissapated as heat when the damping material is squashed.

I am starting to agree about the screws as a potential weak spot.

I will let you know what I eventually decide...

[MartinC]

What I do now is to use a standard HiFi rack frame, but the shelves that sit on it are a bit like Garnett has suggested but using 12mm thick neoprene to seperate two layers of MDF forming the shelves. What this means is there is no solid contact path between what the piece of equipment sits on and the metal frame (just that through the neoprene). I wonder if rather than the multiple spiked level design you could just have a frame built from three long sections of threaded bolt like the Flexy design, but using the two part shelves like you've suggested, something like:


Noting the wider hole on the top half to prevent the top section touching the bolt. Exactly as drawn this may not be a totally rigid design, but you could either countersink a second nut into the top surface of the lower shelf section and clamp something like this:



or possibly have some additional cross piece just below each shelf that is rigidly clamped to.

I don't really see the point of the multiple spike design (spikes couple not isolate in case there is any confusion), which seems overly complicated for no real purpose, unless I'm missing something?

[MartinC]

Garnett wrote:

The damping shouldn't rebound energy. It should be dissapated as heat when the damping material is squashed.

It will reflect energy to some extent. That energy that is transmitted will largely by dissipated by conversion to heat.

When a sound wave meets a boundary between two materials, how much energy is reflected and how much transmitted will depend on the acoustic impedance of the two materials. The acoustic impedance is given by the product of the density and sound speed for the medium. The higher the impedance mismatch the more energy is reflected. I work with underwater acoustics where for instance there is virtually complete reflection at a water air interface (e.g. the sea surface) due to the large impedance mismatch.

The situation we're dealing with here is rather more complicated, but the above gives the basic idea of why some energy is reflected.

[Garnett]

Thanks Martin. After reading your post and Kiang's I think I understand.

Anything that causes its own vibrations (CD Players, Record Players etc) need to be coupled to something heavy to kill any vibrations they may be emitting.

Anything that is sensitive to vibrations needs to decoupled from external sources of vibration.

So if you have a heavy rigid frame decoupled from the floor, and couple the CD player to it, that should solve any problems, right?

[MartinC]

My HiFi sits on a suspended wooden floor that is far from stable, so my main concern is decoupling from the floor. I do assume though that the damped shelves idea will act as a sink for component vibrations to some extent. Although obviously CD players generate more vibration internally all will generate some (transformers etc), plus of course all will be exposed to the sound field emitted by your speakers...

[mosfet]

Anything that causes its own vibrations (CD Players, Record Players etc) need to be coupled to something heavy to kill any vibrations they may be emitting.

Mass or “something heavy” does nothing to remove vibrations. Vibration is kinetic energy traveling through a medium as a wave; as it is an energy it cannot be destroyed only converted to another form. Thus only those materials that convert kinetic energy to another form (usually to heat or sometimes light) are useful in this respect. Rubbers of various types, open cell PU foams (acoustic foams) and so on.

What mass does do is change the frequency of the waves (vibration) travelling through the medium rather than the commonly held myth of ‘sinking’ energy.

The best use of mass is as part of a mass-spring arrangement where the resonant frequency of the arrangement is 10Hz or lower. I’ve used the description of a jelly wobbling on a plate to try to illustrate this in the past in a less technical way.

Anything that is sensitive to vibrations needs to decoupled from external sources of vibration.

If you’re building your own rack Garnett, with objective of de-coupling kit, have a look at turntable suspensions for good design ideas. These commonly comprise the mass-spring arrangements referred to above.

I remember seeing one particular design (can’t remember who) where the platter was hung from vertical pillars by rubber strips. Possibly Roksan – someone else may know.

[rockmeister]

Mossy...just a thought, but you might find that there's a bit (or more) of lateral sway, since your roofing spikes are bedded into a wooden and bitumen structure and, being thin themselves, wont ne clamped rigidly at their top end (I imagine that one could, holding the spiked end with the shelf upside down, sway each leg a little side to side?). I doubt that this would matter sonically, but would you want to balance heavy electronics on it? One solution might be to insert a steel tube of say 5cm diameter into the shelf corners where the legs will go, fill these with concrete and bed the legs into that? More complex of course but rigid...and the concrete will effectively dampen out most vibrations from reaching the shelf at the top of the leg?
I'll do a diagram in a bit, but have just received instructions to walk the dog!...back in a mo.

[rockmeister]

like that?

[mosfet]

This is the design I was referring to meister. SME Model 30/2. This could be adapted to use for the top shelf of an equipment rack with a little thought.

[rockmeister]

I can understand the idea, but it's a fantastically inellegant solution ain't it (and I don't mean just aesthetically)...all that vibrating rubber!?

[MartinC]

mosfet wrote:

Mass or “something heavy” does nothing to remove vibrations....

Is there an argument of coupling something to a large mass limiting vibrations in the first place due to increased inertia?