Tube rolling ARC amps & Steelhead

A while back I decided to try tubes. I have had an ARC Ref 2 Mk II preamp for about 2 years, a pair of ARC Ref 300 Mk II monoblocks for about 1 year, and a Manley Steelhead for 3 months. They sounded great, but it never occurred to me until recently that they could sound better - I always assumed that the manufacturers knew best when they put the stock tubes in.

My speakers are Acoustat Spectra 66 full-range electrostats, and, of course, I wanted more bass. After some investigation, I replaced the stock Sovtek 6550's in the monoblocks (8 per channel) with CryoValve 6550C's (from TubeDepot). The improvement in bass was nothing short of dramatic, and I much preferred them to the stock tubes. Nonetheless, the midrange and treble seemed to be lagging behind. After some more investigation into tube replacement options, I replaced the stock 5AR4 rectifier in the Ref 2 Mk II with a Genalex Gold Lion 5AR4, let the amps warm up a couple of hours, and then listened. I was dumbfounded at the change in sound. There was no "good bass" or "good midrange", everything just fell together with remarkable imaging and coherence, and that was with CD's and the tuner! All from changing ONE tube! I then replaced the stock 6922's in the Manly Steelhead with Mullard E88C's (following the advice of another A-gon member), put on an album, and and watched the bar further raised beyond my wildest expectations. All the sound was in beautiful coherence, rock-stable imaging, perfect balance of frequencies, not a touch of harshness, enjoyable at every volume level, like Neil Young was sitting there in the room pouring out his heart to anyone who would listen. I had a friend with me who has heard my system many times before, and he independently remarked "I can't believe it, everything sounds PERFECT!"

My question is - is this a unique experience, or have others found NOS or other tubes to make such an significant difference? I can certainly see why manufacturers don't include scarce tubes with their products, but it still amazes me that the choice of tube can affect the sound so profoundly. I would be interested to hear if others have had similar experiences.

Other equipment:
VPI Scout / Benz Ruby
Levinson 390S
Luxman T12


1) no, and

2) no.

The topic of this (ancient) thread is the sound of different tubes, and respondents' enjoyment or lack thereof of various tubes in the specified equipment. The topic is not about objectively measuring anything. (Almost) everyone posting here clearly believes that they hear very significant differences in different tubes (I know I do), and I doubt that all of us are wrong. We enjoy sharing our impressions with other enthusiasts who feel similarly. No one posting here except you cares one whit about objectively measuring parameters that some people may or may not think are important. We have our systems because we enjoy them, not because we want to measure them.

The human auditory system has evolved over hundreds of millions of years, and it's pretty much a sure thing we will never completely understand it. Who knows what makes music sound "good" or "real?" We'll be asking and attempting to answer those questions for many years to come.

So please mmrkaic, and I say this sincerely. Sit back, chill, maybe get a drink, and just enjoy the music.

I rolled Steelhead with the Mullard gold pin E88CC and Tungsol 5687 and it sounds the best now.
I hope you guys know that in the Steelhead, the 5687s are used as cathode followers, both of them.  Only one of the two is used for the direct output, bypassing the volume control, when you use the Steelhead as a phono stage only.  If you use the Steelhead as a full function linestage, the signal then and only then passes through both of the 5687s (and the volume control), per channel.  Keep in mind that a cathode follower adds no gain to the signal.  Active devices that do not supply gain are much less able to cause distortion or to have a "sound" of their own, compared to active devices that supply gain.  The most likely way in which the 5687s could affect the sound of the Steelhead is if they are noisy.  But even then, noise in a CF tube is less obrusive than noise in a gain tube. Using a rare and valuable 6900 to substitute for the 5687 is not only a waste of a great tube but also may not be optimal for the 6900, because although the 6900 is a direct replacement for the 5687, its optimal operating points are quite different; the 6900 can tolerate and likes to see much more current compared to a 5687.  I don't know the operating points for the 5687s in the Steelhead (the schematic is a deep and dark secret), so I don't know how much current is drawn by each 5687, but likely not enough to get any extra benefit from the 6900's capability, even assuming one could hear it as a cathode follower. The 6922 is the single tube used for (phono) gain in the entire Steelhead.  I own a Steelhead and a gaggle of 6900s, too. 

The Steelhead is a great piece of gear. I found most gain in performance by replacing the coupling capacitors between the phono section and the volume control and between the volume control and the line output.  I use mine as a full function preamplifier, and it got a lot better after I replaced those parts. I'm still using the factory-supplied tubes, however.