The Trouble with Tube Rolling
Dick Olsher (January 2021)
I recently rediscovered a letter sent to me by the late David Manley in the early 1990s in which he made his feelings quite clear regarding tube rolling. This was in the way of a follow up to our previous discussions on this subject and was intended as a genuine effort to offer additional insights rather than just attack my standard operating procedure of substituting NOS types and occasionally near equivalents. He hoped that I would see fit to publish or quote some of his feelings and understand that what he has to say is in the best interests of both the consumer and tube electronics manufacturing industry. And that’s exactly what I intend to do. What follows are excerpts from David’s letter. Keep in mind that all of this dates back some 30 years.
“I have been doing quite a lot of investigative talking to other tube electronics manufacturers of which Bill Conrad CJ, Jean Paul of Jadis, and Roger Cox [at] Fender tell me they feel exactly as I do and are prepared to write letters to that effect. Audio Research’s views are well known to me (very rigid and un-bending as regards warranty wok). They would not accept the KT90 even though it exceeded the handling requirements of the 6550 (NOT GE “A” 42 watts, but the EGC-Sylvania rated at 35 watts plate dissipation) they were using. Unfair though it seemed to me at the time I did agree that any manufacturer has the right to disclaim a part until and if they actually approve it.”
“So let me speak now strictly from my personal position. I DEPLORE ‘tube-rolling’ and have been so impassioned about that in the press that Roger Modjeski sought to lampoon me and make me the subject of mockery by handing out buttons marked ‘NTRS’ (Nimble Tube Audio Society). You will not be slow to recognize that he had a commercial axe to grind: he sells, promotes, markets, makes money out of neurotic suckers who fall victim to a ‘selected’ (possibly Yugo, Tesla, Chinese, etc.) 12AX7 being sold for $50 retail.”
“I regard changing tubes from those supplied and CLEARLY STATED so in the manual as Modification!... I do not agree that a buyer should be encouraged to start experimenting with a newly acquired piece of electronics, encouraged to believe that the designer/manufacturer (well me anyway!) did not offer his best shot in the original parts he supplied.
“As regards New Old Stock” I would have no problem with a genuinely superior part (i.e., KT66 Gold Lion or Sylvania 6L6GC STR 387 in the case of the 175’s) but it is the word GENUINE that bothers me.
FACT: From Richardson (in all respects supposedly a blue-chip $150 million international outfit) I have had certifiable “Pulls” (very used!) sold as new in their boxes, freshly re-painted as to number and brand.
FACT: Also from Richardson, ordered 5751’s and got “non-specific” 12AX7’s stenciled “5751.”
FACT: When Mullard closed down and ran out of their specific tubes (KT88, 66, 77 and others) they were quite happy to go on selling IN GOLD LION BOXES tubes they bought from China.
FACT: When ECG-Sylvania closed down here in the USA they likewise went on supplying… dealers any quality (Yugo, E. German, Chinese, etc.) of similar part number in their familiar black and yellow ECG boxes.
“I have no confidence in the man in the street being able to get the “real thing” or be able to tell the difference from a junk tube by looking at it.”
“The way I have to go about it (aside from designing and causing a tube to get made for me like the KT90) is to get samples in quantity (after asking for such and such a spec) and then evaluate the actual BUILD-QUALITY of a given type by torture-testing (to find true break-down voltages) and by breaking (carefully) the glass envelopes and measuring thickness and gaps of grid spacing wire, plate material, quality and distance of lead-out wires to pin outs etc.”
David’s concerns are still valid after all these years. Do not attempt any tube substitutions without first checking with the factory. This is especially true in the case of power tubes when well-meaning tube-rolling may result in damage and a voided warranty.