6L6GC vs 7581A...

This confusion seemed to mostly come about, because GE sometimes labeled their 7581s as "7581A KT66" (in fact, I have some of those labeled that way). Those GE 7581s would DEFINITELY sub in for a KT66 (they had way higher ratings in every way), but the opposite is NOT categorically true at all!!

That is not the only difference the KT66 require much higher applied negative voltage to the control grid than 6L6GC or 7581....so if the amplifier has no bias controls or limited available negative voltage the tube will run away red plating or not bias if adjustable. This is an absolute problem in a Scott 296 for instance.

I'm not one to believe published tube specifications all that much I've found too many of them to be complete hogwash especially with output tubes....
 
Every tan/ beige base GE 7581 i have ever found tests bad , glad to hear some found some good ones .

The 7581 and 7581A are NOT the same, as far as allowable rating are concerned- and there's a reason that GE seemed to QUICKLY replace the 7581 with the 7581A!

The 7581 may well have been the same plate-set as the 6L6GC- but the 7581A seemed to (from physical comparison) have the same plates as the 7027A. Those definitely WOULD handle higher plate dissipation than the 6L6GC (just ask any one who has ever owned an Ampeg VT40- the 7027A and 7581A are the only GE tubes that will live in those, without mods!).

The other exception to the dissipation and voltage limit, seems to be the later model Philips/Sylvania 6L6GC STR tubes. Those seem almost indestructible. I suspect those also have 7027A-spec plate assemblies...

Regards,
Gordon.
 
I don't know Gordon I'm holding a GE 6L6GC and a 7581A side by side right now and see very little visual difference. Other then the actual way the two plate halves are fastened together at the wings and the glass envelop being maybe a 1/4 taller on the 7581A which is pretty common even in the same type from one run to the next. While there maybe a slightly different bias point in one circuit to the next for the two tube types I personally suspect they are dead nuts identical...

I'm not at all familiar with the Ampeg amps...so I cannot comment first hand on how they were electrically setup..... I just did a quick google and see they are cathode biased. Have you ever checked those amps to see what voltage is developed across the cathode resistor with both tube types installed? The difference of these tubes in that type of amp could be the difference of cathode just as easily the plates themselves.

Kind of the same as the difference between the original 7591 and that god awful poor rendition from EH.... it takes nearly + -10 control grid volts to bias the EH compared to NOS 7591's in the exact same amplifier. The JJ is just slightly better in that regard.. In a tube tester setup for 7591's the EH flat out fails brand new... change the pertinent settings to 6L6 and it passes every time. After all the years of that tube been in production it still will become hugely unmatched after a few hundred hours of use....what a POS.
 
Edit: Forgot 6L6 GC STR added it and more material at the end with links to GE publications.

I'm no expert but based on what I've read this is what I think happened. 6L6GC, 7581. 7027, 5881, are all 6L6 family tubes with the following official ratings and similar if not identical transfer characteristics:
6L6 19W I think that these are metal case G is for glass
5881 23 W This is an industrial type power derated 6L6GB
6L6GB 25 W
6L6GC 30 W
6L6GC STR 387 35W By Philips/Sylvania STR is Special Tube Request - See note one below
7581 30 W
7581A 35 W Phillips ECG made a JAN Military version see note 2 below
7027 35 W Pin out is slightly different and may not work in all 6L6 applications
807 is a 6L6 type with a different base and plate cap
There are a few others.

Back in the 60s and 70s production people were learning that common parts reduced production costs. To make a 19W, 25W, 30W, and 35W 6l6 costs more money to set up production for each type. Now, this is my guess, they said why not stuff 6l6 GC, 7581A, with 7027 guts and make them all the same. The lower rated parts will simply have more margin and last longer. Also, companies were pushing the parts right to their limits where they did not last long. Companies that did this would have longer lasting tubes. So, 7027 and 7581A are offically 35 W, some 6L6 GC are 30W and some are truly 35 W with 7027 guts but there is no designation to indicate 35W - you have to look at the guts. Edit, the 6L6 GC STR is a 35W tube so there is an indication. I knew that the STRs by PhilipsECG/Sylvania were 35 W when I wrote this, but not that the STR meant special tube request so the 35W guts were requested by one or more companies, probably Fender. The interesting thing is that Sylvania also did the 35W upgrade to other non STR types so perhaps the efficiency idea hit them at some point. There are modern STRs and I would not count on them being the same.
The reasons are my speculation, but I have read that the guts are positively the same as the 7027 for certain 6L6GC's and those tubes can run all day at 35W on a tester. There is even a tube rated for 19W with 35 W guts.
Also,
5XXX tubes are usually industrial quality tubes
7XXX are optimized for audio use, lower noise and microphonics as I understand it.
The rest are common grade.

KT66 is similar but changes were made that make the transfer characteristics and transconductance different, they require different biasing.

GE publication on the 6L6GC and the 7581, they run them at 80W to show them running red plate:
http://www.triodeel.com/ham1.gif

Continued with a picture of the 5-ply plate material (used in the 6L6 GC not the 6L6GB) that has copper in the core to conduct the heat:
http://www.triodeel.com/ham2.gif

Interesting 1950's paper introducing the 5881 with many features to make it more rugged. It had gold plated grid wire:
http://www.pearl-hifi.com/06_Lit_Archive/07_Misc_Downloads/TungSol_5881_Info.pdf

The Tung-Sol 6L6 GC STR is NOT a 35W tube, at least not rated as one:
http://www.tungsol.com/tungsol/specs/6l6gcstr-tung-sol.pdf

Note 1. - I found this in an ebay add so readers can decide if it is accurate or not:
The STR387 was a "Special Tube Request" by USA amplifier manufacturers seeking a heavy duty 6L6 power tube that would last longer at the higher voltages produced by audio and guitar amplifiers. STR tubes have heavier plates, taller and thicker glass bottles with a higher vacuum so these last up to 4x times longer than imported 6L6 tubes being made today...

Note 2. - There seems to be a military version of the 7581A according to Watford Valves:
https://www.watfordvalves.com/product_detail.asp?id=1317

From that web site, I don't think that all 7581A's are JAN Military:
"Philips ECG JAN / 6L6GC/STR387-7581A-PHILIPS
Philips electronics components group was the name given to the products marketed in America after Philips bought the American valve manufacturer Sylvania in the early 1980’s. Sylvania was a major manufacturer of valves and Philips bought them to get the lucrative American Government and military contracts. These valves are of the highest military grade which offer outstanding performance and were all made in the USA.
The 7581A is the high grade military version of the famous Sylvania 6L6GC STR 387. The 7581A was made to have a higher plate dissipation of 35 watts, where all other 6L6GC types have a 30 watt rating, Philips Sylvania uprated anode plate coating allowing better heat transfer. The valve also uses the highest grade nickel for the anode plate construction so the 7581A gives a long life."
 
Last edited:
As always, read the fine print on the data sheets. Are tube ratings "Design Center", or "Design Maximum", or "Absolute Maximum"?

First allows about 20% worst case margin, second about 10%, last allows NONE. Earlier tubes used the first rating system, later ones were less conservative due to competition. But the laws of physics still apply - run it hotter and it won't last as long.
 
I'm no expert but based on what I've read this is what I think happened. 6L6GC, 7581. 7027, 5881, are all 6L6 family tubes with the following official ratings and similar if not identical transfer characteristics:
6L6 19W I think that these are metal case G is for glass
5881 23 W This is an industrial type power derated 6L6G
6L6G 25 W
6L6GC 30 W
7581 30 W
7581A 35 W
7027 35 W Pin out is slightly different and may not work in all 6L6 applications
807 is a 6L6 type with a different base and plate cap
There are a few others.

Back in the 60s and 70s production people were learning that common parts reduced production costs. To make a 19W, 25W, 30W, and 35W 6l6 costs more money to set up production for each type. Now, this is my guess, they said why not stuff 6l6 GC, 7581A, with 7027 guts and make them all the same. The lower rated parts will simply have more margin and last longer. Also, companies were pushing the parts right to their limits where they did not last long. Companies that did this would have longer lasting tubes. So, 7027 and 7581A are offically 35 W, some 6L6 GC are 30W and some are truly 35 W with 7027 guts but there is no designation to indicate 35W - you have to look at the guts.
The reasons are my speculation, but I have read that the guts are positively the same and those tubes can run all day at 35W on a tester. There is even a tube rated for 19W with 35 W guts.
Also,
5XXX tubes are usually industrial quality tubes
7XXX are optimized for audio use, lower noise and microphonics as I understand it.
The rest are common grade.

KT66 is similar but changes were made that make the transfer characteristics and transconductance different, they require different biasing.

For the most part I agree!
 
As always, read the fine print on the data sheets. Are tube ratings "Design Center", or "Design Maximum", or "Absolute Maximum"?

First allows about 20% worst case margin, second about 10%, last allows NONE. Earlier tubes used the first rating system, later ones were less conservative due to competition. But the laws of physics still apply - run it hotter and it won't last as long.

What makes you think I am not aware of that?
They are all Absolute Maximum, as far as I know, that people should not design to but many do.
 
What makes you think I am not aware of that?
They are all Absolute Maximum, as far as I know, that people should not design to but many do.

Lots of people will read this thread, not all have read the fine print. Here are the values from some data sheets (Different brands may be rated differently)

6BQ5: 12W, design center (T-S)
7189: 13.2W, design maximum (Syl)
6L6G: 19W, design center (RCA)
5881: 23W, design center (Tung-Sol)
7027: 25W, design center (RCA)
6L6GC: 30W, design maximum (GE)
KT66: 25W, des. max, 30W abs. max. (M-O-V)
7581: 30W, design maximum (T-S)
7027A: 35W, design maximum (RCA)
7581A: 35W, design maximum (GE)
6550: 35W, design center (RCA)
6550: 42W, design maximum (T-S)
6550A: 42W, design maximum (GE)
KT88: 40W, des. max, 46W abs. max (M-O-V)
 
Kind of the same as the difference between the original 7591 and that god awful poor rendition from EH.... it takes nearly + -10 control grid volts to bias the EH compared to NOS 7591's in the exact same amplifier. The JJ is just slightly better in that regard.. In a tube tester setup for 7591's the EH flat out fails brand new... change the pertinent settings to 6L6 and it passes every time. After all the years of that tube been in production it still will become hugely unmatched after a few hundred hours of use....what a POS.
Sorry for the OT post.

Yes the EH 7591s all seem to test as "bad" on antique tube testers which only run them at about 100V or less. But the fact is that when these are run at the voltages which these were designed to run at, they actually seem to run pretty well in some amps.

It may well be that their bias point is way off. That would not surprise me. But somebody performed a fairly thorough test of the new vs the old tubes about a year ago or so. I want to say that it was Dave Gillespy, but I'm not 100% certain about this.

The gist of the test results was that when run at real world voltages, the EH tubes actually did OK. Not great mind you, but OK. They were about 20% lower power output than the Westinghouse tubes. But the JJs were about 40% less power than the Westinghouse tubes. And when run near their rated voltages, the EHs performed better all around than the JJs, despite testing like crap on a Hickok.

I don't make any claims to knowing how well the EH 7591s perform in other circuits. But inside of the MC225s at least, they seem to work well enough that I personally don't find that I miss the Westinhouses at all.

But that doesn't excuse their size, which is ridiculously large in comparison to the originals.
 
Last edited:
Back
Top Bottom