Thanks to BitPerfect user Robin Wukits for sending me this link of a beautiful duet between a Soprano and a Cornetto. No, I didn’t know what a Cornetto was either 🙂
When I took delivery of my new PS Audio BHK300 Signature mono block amplifiers, together with a PS Audio P10 Power Regenerator, it presented me with an immediate practical problem. Those three units replaced the single unit of my Classé CA-2300 power amplifier which was installed in the bottom shelf of my “SolidSteel” equipment rack. All three PS Audio units share the same chassis, one which makes them comparable in size (and weight) to the Classé unit. So there was room for only one of the new trio in the SolidSteel rack. It was determined that the P10 would go in the rack, while the BHK300s – which could maybe profit from being located nearer the loudspeakers – would have to find a place to sit on the floor.
This new aesthetic immediately raised a question in my mind. While the SolidSteel rack attempted to provide a solid mechanical ground for the P10 via its three spiked feet sitting in conical cups, sitting the BHK300s directly on the suspended wooden floor did not seem so smart. Nonetheless that would have to suffice, while I thought about how I could provide a better solution.
Forced to consider the situation from theoretical as well as practical considerations, I immediately wondered about the relative benefits of a mechanical ground vs an isolation platform. The idea of a solid mechanical ground is that any vibrations in the product will be efficiently coupled out – just like any residual electrical signals in the equipment chassis will be efficiently coupled out to electrical ground via the ground wire in its power cord. Such an approach – whether electrical or mechanical – requires that the ‘ground’ we are coupling to be a true ground. Now, if my house were built directly on granite bedrock, which I exposed to form the floor of my listening room, a mechanical grounding approach could be ideal. But my house isn’t like that. I have a suspended wooden floor to which my speakers are coupled via their own ‘mechanical ground’ connection. The speakers are therefore in all likelihood transmitting a substantial proportion of any mechanical energy generated within their cabinets into the wooden floor and energizing it. The sound waves propagating back and forth around the room also energize the suspended floor, as do people walking about in the house.
Therefore, if I sit my amplifiers on a support table design which provides a solid ‘mechanical ground’ coupling to the floor, it seems that all this will do is potentially couple vibration from the floor up into the chassis of the amplifier just as efficiently as it would in the other direction. If there were more vibrations in the amplifier than in the floor, then it might be ideal. But I don’t think that is likely to be the case here. So a ‘mechanical ground’ approach might actually cause more problems than it would solve.
The alternative, if the thinking is that the floor represents a source of vibrations from which the chassis of the BHK300s are to be protected, is an isolation system. This is simply a mechanical system between the BHK300s and the floor which absorbs any incoming vibrations. Those of you who still own a turntable will know exactly what I’m talking about. Any turntable worth its salt will contain its own built-in isolation system, although all but the most extreme designs will still benefit from sitting on some sort of external isolation table.
The core of an isolation system is a damped spring. If you sit something heavy on a theoretically perfect spring, and tap down on it to provoke a bounce, then it will continue to bounce away forever, at a frequency determined by the stiffness of the spring and the weight of the object. If you introduce any damping into the system this will cause the bouncing to die down. The greater the amount of damping the more rapidly it will die down.
Consider a car driving along a rutted road. The car is a heavy object sitting on a damped spring (i.e. its suspension). A car driving along a rutted road is very similar to the same car standing still on a vibrating road. The purpose of the car’s suspension can be thought of as trying to isolate the occupants of the vehicle from the vibrations of the road. In truth a car’s suspension designer has a lot more on his mind than your comfort, but lets ignore that (although if you imagine a 1970’s Cadillac you might not be too far off the mark). The mass-on-a-spring will typically have a resonant frequency. If you push down on the fender of your car and suddenly release it, that is the frequency at which it will bounce up and down. The suspension’s damping determines how quickly the bouncing dies out. A modern car is typically very well damped – your 1970’s Caddy less so.
An isolation system designed this way tends to pass frequencies lower than the resonance (natural bouncing) frequency, and absorb the higher frequencies. That way, your old Caddy can drive comfortably along even a Montreal highway, smoothing over all but the biggest bumps, which are transmitted into the cabin. Damping is necessary for two reasons. First, because damping is what actually absorbs the vibrations fed into the system, turning them into heat which is conducted away. Second, because the amount of damping determines how well the isolator attenuates the frequencies it is designed not to pass.
Leaving the old Caddy to one side, for my audio application I want to make sure that my BHK300s are isolated from all frequencies at 20Hz and above. In fact, the lower the better. In my mind, my ideal would be something like a 1Hz resonance, with a small amount of damping that would allow the 1Hz bounce to die out over something like 4-5 seconds. Armed with these design objectives I can sharpen my pencil, sit down, and work out spring rates, damping factors, masses and so forth. And if I was designing something in a professional context that’s how I would approach it. But that’s not what’s happening here.
All I wanted was a simple test bed to see whether any of this stuff actually had any audible effect in my system. What I came up with was a Typhoon 17” x 13” Butcher Block which would form the base of my isolation table, and happened to be the exact same dimensions as the BHK300 chassis. Plus it looks good, and has the practical convenience of four sturdy, built-in legs. As a platform, Butcher Block is a mechanically well-damped material, which is a plus. For my damped springs I decided to use a high-technology pneumatic approach. I bought a set of 12.5” inner tubes for Stroller/Pushchair wheels. The idea was to inflate the inner tube, lie it on its side on the Butcher Block, and sit the BHK300 directly on top of it.
My calculations suggested that those inner tubes could support the weight of the BHK300 without my needing to inflate them to anywhere near their maximum rated air pressure, so I felt confident that the weight of the monoblocks wouldn’t just burst them. As it happens, without the tyre to constrain their expansion, you cannot pump these things up anywhere near their rated pressure! All I could do was pump them up as much as I felt they would safely sustain, and see how I got on. The pressure was too low to register on my automobile tire pressure gauge, so I can’t tell you what the actual pressure was. But that’s fine, because my target pressure was also too low to register!
Since the weight in the BHK300s is not conveniently centred, you have to position the inner tubes slightly to the right of centre on the Butcher Block. To my surprise it proved easy to get it lined up so that the monoblocks sit nice and level. See the photograph. Also to my surprise – and great pleasure – this arrangement proved to have a resonant frequency of ~1Hz and a natural resonance which damps out in 4-5 seconds. This is exactly what I thought in advance might be my ideal setup. This is great news, because manhandling those 83lb monoliths every time you want to make a change is not my idea of fun.
Of course if the isolation method is right, the mechanical grounding method must be wrong, no? I had one unused inner tube, so I pumped it up and put it under the P10 on the bottom shelf of the SolidSteel rack. This inner tube turned out be a bit narrower than the other two, and provides the P10 (a mere 73lb lightweight) with a resonance frequency more like ~5Hz, and is slightly more damped than the monoblocks. If I decide that makes a difference I can always shell out another $10 for one of the wider tubes.
So, how does it all sound? Let’s have a listen.
At this point I am somewhat concerned at the possibility of losing credibility. The changes I am hearing are not subtle. No, not subtle at all. Have you ever changed a pair of interconnects or speaker cables? How about a USB cable or a power cord? There is no doubt in my mind that those components can make a real and valuable contribution to the performance of a high-end audio system, particularly in the area where I operate, where vastly diminishing returns are the order of the day. But for sure the changes engendered by such tweaks are definitively subtle. I’m sure many people might listen along as I audition a pair of interconnects and shrug their shoulders, whereas I might conclude that one of the sets is worth an investment of a thousand dollars. Still others, not content with merely shrugging their shoulders, will fire off a spate of spiteful invectives on every audio forum that they can make the time to sign on to. Subtle effects are what we have become used to dealing with when auditioning audio “tweaks”.
Well, that’s not what is happening here. The changes wrought by those $10 inner tubes are more on a par with swapping out a pair of good 20 year-old loudspeakers for a pair of modern high-performance units. The basis is deeper and fuller, with less overhang and more tuneful delineation of pitch. The stereo image is tighter, much deeper, and more holographic. A whole layer of grain that I didn’t even know was there has seemingly been stripped from the midrange. Vocals in particular seem more natural and more three-dimensional. I could go on, but I won’t because the specifics of what I am hearing may prove to be specific to my particular system. However, I don’t think that the general level of benefits are going to be all that system-specific. We are talking about mechanical isolation, and I don’t think my PS Audio components depart in any radical way from industry norms with regard to their mechanical standard of construction. I expect major noticeable improvements are going to be evident regardless of what system you are using.
You might well point out that the BHK300 monoblocks contain vacuum tubes, and that vacuum tubes are well known for being particularly microphonic, and you would be right. But on the other hand I listened for a short while with inner tubes only underneath the BHK300 monoblocks, and when I placed the third inner tube under the P10 PowerPlant – which contains no vacuum tubes – the magnitude of the change was just as large, if not larger, and was probably more impressive in terms of the qualitative improvements it brought. Whereas the inner tubes under the monoblocks brought immediate and indisputable benefits, it was not until the final tube was placed under the P10 that everything suddenly came together as a coherent whole.
This was originally conceived as a trial experiment. The idea was to see how it went, and decide where to go next. In truth I’m not sure where to go next. All I know is I am going to focus on enjoying the music until the inner tubes eventually burst or deflate or whatever it is they are going to do. I expect stability and longevity is going to limit their long-term practicality, but until it does I’m going to be enjoying it for what it is doing right now.
Today we announce the release of v3.1 of BitPerfect.
As usual, BitPerfect 3.1 is a free upgrade to all existing BitPerfect users.
Today Apple released update 12.4.1 for iTunes. I have been testing it, and not only does it work fine with BitPerfect, it also fixes whatever was causing the problem with “Minimize iTunes Interaction” mode that occurred with iTunes 12.4.0, so BitPerfect users who prefer that mode of operation can feel confident enough to try enabling it again. However, it will still be necessary to disable and re-enable BitPerfect after making the change in order for it to come into effect.