• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

How important are date codes when matching tubes in a preamp?

Probably shouldn't add my two cents worth, but 5 years is a long time (53 to 58). how many 12AX7's would RCA have made in that time? 50,000? 200,000? 1 million? Regardless, RCA would have continually reviewed and revised their processes to give better yield, higher quality, lower materials cost, lower labor costs, better processing rates, increased tooling durability, improved or less expensive testing procedures, revised vacuum systems, adjusted weld temperatures, etc. etc. I would expect an RCA engineer of that time would be able to list dozens of 'changes' during that time. Would any of those changes cause the latter tube to be different enough from an earlier tube to have different performance? Who knows, but it seems likely that those tubes would have more 'differences' than two tubes with closer build dates.
 
Towards the end of the 70s I suspect people weren't paying much attention to fussy stuff like matching. It was obsolete garbage and warranted only stuffing another tube in to make it go a little longer by that point.

I've seen gear serviced in the tube era that had mismatched stuff in it. Some repairman's sticker on the back listing a fix from 1965, open it up and you get 1 new output tube and one original in a push-pull pair. Sometimes you'd get one channel with new tubes and one with originals. My Pilot 260 came to me with an original Pilot-branded 5U4GA and one RCA 5U4GB . It also had 4 different old production EL34 tubes. Similar construction, but all different codes. They actually match up reasonably well and sound fine.
 
GordonW said:
Oh, and a note to the OP- the RCA black-plate 12AX7s- I have had some issues with RCA 12AX7 tubes- even older ones- being slightly too noisy (hiss/hum) for a Dyna phono stage. The later 12AX7A was specified out with a heater that would reduce hum- but the earlier 12AX7 (no "A") did not necessarily have that same degree of low noise as the 12AX7A. The 12AX7A was better than the 12AX7- but even so, it could still have a slight amount of audible hum in a Dyna.

That's ostensibly why they also later made the 7025- it was a "select" version, with even lower noise. Those usually are quiet enough.

Regards,
Gordon.
Click to expand...


Thanks for the feedback and interesting discussion, everyone.
Gordon -- Did not know about this. More food for thought, I suppose.

Joe
 
6DZ7 said:
And your date code could be the same for Monday morning or Friday evening.... I have two GE 6BQ5 with same codes and one has the heater sticking up out of the top of its K tube like a Christmas tree. I like to look at it. And it plays good.
Click to expand...

they aren't even made in the same factory, but they are all made by Phillips companies. Electrically they jive fairly well but they're very much not a set.
 
petercapo said:
Perhaps another good reason to just go with current production and avoid all that. I haven't always had difficulty getting matched date codes and electrical match with new tubes.
Click to expand...

Yeah, when they're installed. But later, the story has been sometimes different, once they get some hours on them.

It's getting better recently- but I've seen innumerable cases over the years, of (especially) then-current production Russian tubes and Chinese tubes, where they would go in as a "matched" set (and they would even initially test pretty closely on a tester)- but after a few months, they would diverge DRASTICALLY.

This is one of the reasons that Kevin Hayes of VAC came up with the iQ Autobias system- it not only makes the idle current constant- but it stabilizes the underlying bias even while the tube is passing signal- which goes a long way toward stabilizing the transconductance and gain of the tube, in operation. In essence- MAKE the tubes dynamic operational parameters as consistent as possible across a wide range of tube condition- even for tubes that aren't statically matched, due to wear...

Fortunately, this has somewhat rectified itself, at least in terms of the common Russian tubes- they've been more consistent in the last few years, even after being in use for significant time. However, current Chinese tubes- that's a different matter. If you can find "matched sets" at all, it's still a crap shoot, in many cases.

If China can ever get a brand new or rebuilt Shuguang factory on-line- maybe things will get better, though. Here's hoping they do...

Regards,
Gordon.
 
The sad thing is- that before the Shuguang factory had the fire that shut them down- they were making great strides, in making things FAR more consistent and accurate. The Shuguang tubes of that time, were great sounding, held up well, and maintained balance decently well.

Once the factory went off-line, and the supply of new, good tubes stopped- all the substandard and off-spec ones that had previously been rejected, that were left sitting in warehouses, apparently were repackaged by someone in China and sold to unknowing or uncaring vendors worldwide. That's why the last of the Shuguang-made tubes that were available in the last few years, were of such variant and oftentimes substandard quality.

But, if Shuguang is able to rebuild- to get back to the level that they were at, just before the factory burned- then they could be pretty solid performers across the board...

Regards,
Gordon.
 
Yes, undoubtedly changes were made in processes over the years, but did they induce significant differences? Say anymore than monthly production run variances would have, like machine temp at turn on verses at end of shift, or mid production materials resupply? I agree that getting matched dates with new manufactured tubes is easy enough it should be part of the matching process.

I would like to apologize to Petercapo, I didn't need to defend my position so vigorously. Growing up my father was a tv and radio repairman for many years, and later I worked on vacuum tube test equipment for the military until it was phased out. In hifi audio there are undeniable nuances that are difficult to quantify to measurements. It's true at the end of the day that some tubes just sound better to me than others. I can't convey that to someone else though, they have different ears than I do. The ultimate production run variance I suppose.
 
Shug451 said:
...How important is matching the date codes if the tubes both test the same value? Thanks for any and all advice.
Click to expand...

Just for sake of the discussion, even if you could hear no audible difference, would using different date codes still be a lingering question in the back of your mind?

I submit that to use for guiding light.
 
whoaru99 said:
Just for sake of the discussion, even if you could hear no audible difference, would using different date codes still be a lingering question in the back of your mind?

I submit that to use for guiding light.
Click to expand...

Ha! That might be the best argument of all. "Hmmm, the system sounds fine, but are the different tube production dates causing it to not sound as good as it should?"
 
IMHO, "matching" in a tube tester (for signal tubes) and current draw (output tubes) seems most critical. But, lot and date matching is important for certain aesthetic reasons. To me, the electrical data match is most important. Even between manufacturers, lot, dates of production.

Unfortunately, within a few months --- electrical matching between tubes can go astray. A lot depends on how tightly matched all the other components and their implementations are, within the unit.
 
Last edited:
I only glanced over this thead so it might have ben mentioned and I missed it.

If you have copious amounts of negative feedback then it does not matter.
Otherwise:
Use the same manufacturer and use the same plate structure. Different manufacturers may label a tube the same but in fact they arwe different although they may work in the same circuit. An European ECC83 is not performaing in the same way as a 12AX7 from the USA. Just check what Mazda recommends for operating points for the ECC83 versus another ECC83/12AX7.

An ECC83 with a spiral filament (Sovtek LPS 12AX7) does not work the same as a short plate with non-spiral filament. There will be hardly, if any differences when the same manufacturer and the same structure is used, even if there is a few years between them.

If the same date code can easily differ 30% in measured performance and still be within specification.

To confuse things even more: when some manufacturers ran low they might relabel some of their own tubes or have some other manufacturers tubes relabelled. Or it might come from another subsidary. (e.g. an E84L might be relabelled as EL84 or a Mulalrd might have been relabelled as Philips. Always check the look of the innards.)

On another note: if you use frame grid tubes or tetrodes: if you bump them then their performance may change due to their structure and elements in their structure moving.

AM
 
You must log in or register to reply here.
Back
Top Bottom