Eico 666 test settings - suggestion for Type 45 (VT45)

[Brienne R.]

Hey folks, first post here at AK!

I've been trawling the net to find info on Eico 666 test settings, and found many great resources like the BAMA pages and of course Conrad Hoffman's extensive work with the Eico 667.

http://bama.edebris.com/manuals/eico/666/

http://conradhoffman.com/EicoTubeTesterSettings.xls

Unfortunately, one of the reasons I'm interested in searching for these setting is because there seem to be few resources for older, preoctal tubes. I found the old docs for those tubes, but the test settings seem to leave something to be desired.

I have some old vt45's and wanted to test them. Never used these old things, and wanted a good shot at their validity. So I cooked up my own settings... and then I found the old docs - woops! Anyway, their settings put them at the following:

Grid - 7 (3.2v)
Plate - 44 (43mA)
Levers - 23511 11111
V - 4
S - 2

The weird bit is that they call the plate voltage 90v by tying it to the Screen connection. Whatever, just a name right? The datasheet calls for 180v in typical Class A, so I figured I'd work back from those settings, but now I'm not so sure.

Any folks know if this is sufficient to test these old bottles truly?

 

[FlaCharlie]

Hey folks, first post here at AK!

I've been trawling the net to find info on Eico 666 test settings, and found many great resources like the BAMA pages and of course Conrad Hoffman's extensive work with the Eico 667.

http://bama.edebris.com/manuals/eico/666/

http://conradhoffman.com/EicoTubeTesterSettings.xls

Unfortunately, one of the reasons I'm interested in searching for these setting is because there seem to be few resources for older, preoctal tubes. I found the old docs for those tubes, but the test settings seem to leave something to be desired.

I have some old vt45's and wanted to test them. Never used these old things, and wanted a good shot at their validity. So I cooked up my own settings... and then I found the old docs - woops! Anyway, their settings put them at the following:

Grid - 7 (3.2v)
Plate - 44 (43mA)
Levers - 23511 11111
V - 4
S - 2

The weird bit is that they call the plate voltage 90v by tying it to the Screen connection. Whatever, just a name right? The datasheet calls for 180v in typical Class A, so I figured I'd work back from those settings, but now I'm not so sure.

Any folks know if this is sufficient to test these old bottles truly?

First, welcome to AK.
I've used my 666 to test 45s using the following settings, which seem to be the same as the ones you mentioned except that your's aren't complete.

Fil: 2.5v
Grid: 7
Plate: 44
Levers: 23511 11111
V: 4
S: 2
Leak Push: 2, 3
Merit Push: 2

Here's a 45 data sheet: https://tubedata.altanatubes.com.br/sheets/127/4/45.pdf

Very few testers test tubes at the full voltages and currents seen in the tube data sheets. The Eico 666 is fairly basic tester. They call it a Dynamic Conductance tester which is just a bit better (supposedly) than the most basic testers which are called Emissions testers.

BUT . . . the Eico data sheets are known to have numerous settings that are either questionable, just plain wrong, or that conflict with the settings shown on other Eico data sheets. So other than giving you a basic Good / Bad / Questionable reading, I wouldn't put too much stock in the readings you get. Unless a tube tests solidly in the Bad range, it might still be useable in an amp.

I just checked a 45 that tested at 80 on the Eico. Unless you are familiar with the 666 you would think that 80 is pretty decent. But the Good scale on the Eico runs from 80 to 140. So 80 is the minimum good reading.

To compare, I checked the same tube on my Hickok 600A, which is a mutual conductance tester which also can be set to test on a Good / Bad scale. The same tube tested about 1/4 of the way into the Good using the Good / Bad test. There are no % markings on the meter. That's a little higher than the Eico so I suspect the Eico reading for 45s would be a bit lower than it should be.

I got a transconductance reading of 1650 on the Hickok. An average new 45 would read 1850, so that's about 89% of a new reading.

Comparisons like this are totally unscientific of course.

 

[Brienne R.]

Thanks for the detailed comparison of results from the Eico to your Hickok, Charlie. That's handy to keep in the back of my mind, to be sure!

I've read extensively and studied much of the primary lit and the forums in which folks discuss validity of each tester, along with their relative use and limits. It seems to confirm that for many tubes the Eico 666 - though many settings can be rough on tubes, or just plain wrong - is generally not bad.

I still maintain that the only real way to test tubes is at load in a circuit-proper. I may only be 29, but after a decade of hearing the old timers (no srsly, the guys who are >70 at my local electronics B&M) harp on this stuff, I get it.

IIRC, Hoffman wrote (in his Eico 667 settings .xls file..?) that using a dynamic/transconductance tester is a pretty good way to match tubes. Lets be real - if I test two 45's and they are at 84 and 86, I could be confident in their ability to match well for a push-pull circuit. Tubes and their circuits may have a tolerance that is easily +/- 10-20%... As long as I have a tolerance within a few percent, I think that's acceptable. One at 70, and another at 92? Probably not...

(Ok here comes my sacrilege: in a guitar amp output - my primary deal - if I have 20% tolerance, I'm totally fine... whatever kills hum at least. 30% THD at full power? Right on! But sound REproduction? I'd definitely like no more than 5% and tight matching)

Foolproof? Hah! I mostly use these things to see if tubes are showing leakage, shorts, are gassy, or otherwise have some sort of wonky issues.

That said, I wonder about finding a way to test them at higher settings on the 666. I tried to dink around, but where I was getting some reasonable readings at first glance told me I was being way too hard on the tester and the tube. Back of the envelope calculated 20W from that 45... whoops!