December 2015

Leave It On!
Roger Skoff writes on why his systems are never turned off.
Article By Roger Skoff

  Back one day in the early 1960s, I went to the Hollywood Boulevard studios of radio KFWB; told the people there that I was a student at UCLA and a Hi-Fi Crazy; and asked if there was ever such a thing as a guided tour of the station that I could sign-up for. To my great surprise, even though they told me that they didn't offer any such tour on a regular basis; they did introduce me to the station's Maintenance Engineer and ask him to show me around.

Although my parents had, in the past, taken me with them to be part of the studio audience for some of the radio and television shows broadcast by ABC, CBS, and NBC (remember that this was Hollywood in the 1950s and 1960s; that all of the networks had major studios in town, from which they regularly did "live" broadcasts; and that, just as now, because they wanted audience reaction, you could usually get free tickets to them just by asking), this was the first time that I had ever been in the actual "working" parts of a broadcast studio, and it was fascinating. Although the biggest broadcasters had already started the change to solid-state electronics, KFWB, as an independent local radio station, still had room after room of tube gear, all of it mounted in tall 19" relay racks, and all of it turned on!

I had already become a little familiar with professional sound and recording equipment by helping Jimmy Valentine (Valentine Sound Recorders, North Hollywood, California) as "junior engineer" (go-fer) on a couple of the Stan Kenton big band recordings, so – even though I didn't know what most of what I was looking at was, I was still blown-away by the thought of the sheer expense involved in buying that much stuff. I also found myself amazed that all that much of it could be needed to be in operation at the same time, so I asked the Engineer, and was told that most of it was, in fact, not in use at the moment, but was just "keeping warm" and ready until its operation was called for.

When I asked the obvious question – wouldn't it be a lot cheaper, not only in terms of the electricity for the equipment, but for the air-conditioning to cool the rooms it was housed-in, to just turn everything that wasn't being used off and then turn it on again a little before it was actually needed – he told me something that I'd like to pass along to you now:

"Tubes," he said (paraphrased, of course. It's been far too many years for me to actually remember his specific wording) "suffer most of their wear not so much from the hours that they're used, but from the eventual cracking and failure of their metal and other parts brought about by the repeated expansion and contraction cycles that result from the heating and cooling caused by repeatedly turning them on and off. Far better," he continued "to just turn them on and leave them on. That way they'll usually last for many years before needing to be replaced, and the cost for the electricity to just keep them on is usually less than the cost of replacing the tubes, so leaving them on is usually the cheaper option."

That was more than fifty years ago, so electricity was far cheaper than it is now. Tubes, however, were way cheaper then than they are now, so his statement still applies, and may be even more relevant.

I thought of this just a couple of days ago, when I went to set up a system in an unused upstairs room for the enjoyment of a visiting houseguest.  What I used initially was a Sony CD player and a pretty good solid-state integrated amp to drive a pair of middle-of-the-line Snell speakers that I had had sitting unused for many years. That sounded reasonable for a short while but, after just a few hours, started sounding awful, so I removed the grills and took a look and – lo and behold – the foam "surrounds" (why did people ever use those, anyway?) on the woofer cones of both speakers had died. This left the cones flapping uncontrolled except by their "spiders" (the sort of corrugated-looking part at the base of the cone that keeps the voice coils centered).

Obviously that was unacceptable, so I brought up a pair of very good prototype mini-monitors, which sounded great, except that they (as might be expected) lacked deep bass, so I decided to use a powered subwoofer – but couldn't, because the amplifier didn't have a suitable output to drive one. That led me to change over to, first, A Jeff Rowland "Consonance" preamp (specially modified by Rowland's service tech Ron Fleming with XLO connectors and internal wiring [yummy!] and then to drag up all 130 pounds (more than 59 kilos) of an XLO-modified Jeff Rowland Model 5 [one of a matched-pair usually used in bridged-balanced configuration], and that's what caused me to write this article:

The Rowland gear, when first turned on, didn’t sound good! Just as it always has, it took several days before it really reached its best performance, and that, finally, brings me to the subject of this article. With tube gear, it's obvious that it takes a "warm-up" period before it will operate. The electrons that tubes control to make the music come from heat, and until the tube's internal heating element has warmed-up enough to start the electron flow, nothing can happen. Tube Lovers (among which I also number myself for certain applications) know this, expect it, and accept it as part of their listening experience.

Solid-state, though, doesn't get its electrons from heat, but draws them from the capacitors in its power-supply section, from which they're available just about instantly upon flipping the "on" switch. It produces heat only as a by-product of operation – not as a prerequisite. So why does solid-state gear need to "warm-up"? The answer is simple, obvious, and applies to both solid-state and tube electronics: There are other things – capacitors, resistors, wire, diodes, rectifiers, transformers, etc. – in both solid–state and tube circuits, and they all need to warn up, too! Unless everything in a circuit is operating within its optimum temperature range, optimum performance won't be achieved.

That's why, for some applications, delicate circuits of either kind may be laid-out and tested many times with the same components and the same electrical connections and only the physical placement of the pieces changed: It's so that all points in the circuit will be at the same temperature during operation. It's also why many solid-state high end audio components don't even have a front panel on/off switch: Although some may have a reduced-power-consumption "standby" switch on the front panel, they’re intended to be left on all at all times, so that, after warming-up on first being turned-on, they will stay at optimum operating temperature at all times.

It's also why the current Tube Lover standard practice of turning their equipment on shortly before, and turning it off immediately after listening to it – "to conserve tube life" – may be counter-productive, and is almost certainly depriving them of some portion of the performance and enjoyment of the gear that they love so much and spent so much to buy:

For my own tube gear (VTL and Melody), as for all of my solid-state gear (Rowland, Krell, Klyne, Theta), I follow the advice of that long-ago radio station Maintenance Engineer (and a number, too, of the industry's top designers, whom I've asked specifically about this) and keep it all turned on all the time. Yeah, the electricity's gotten expensive, but the tubes are even more so and if keeping them always on will extend their life, as I have been told, good. And even if it won't, keeping them turned on all the time ensures that they will always sound their very best – which is why I bought them, in the first place.

After all, even though writing and talking about hi-fi is great fun, and so is going to shows and hi-fi club events, isn't the real purpose of all this to...

Enjoy the music.