Hi Folks,

Can anyone explain why my EICO 667 tube charts have TWO SETS of settings for TWIN TRIODE preamp tubes like 12AX7, 5751, 12AT7, 12AY7, 12AU7's?

I thought that since they're equivalent triodes that just have different plates (pin 1 and pin 6), the rest of the settings should be the same.

Seems like you should just leave the settings the same for each triode and just switch between button 1 and button 6 for the MERIT test.

What am I not taking into account? Thanks so much for clarifying this.

The grids and cathodes are also on different pins.

Yes, but if you set up the bank of levers so that each triode is equivalent and the heaters are connected on pins 4 &5 in-between:

4-5-1 -2-2- 4-5-1 11111....with the remaining levers set to 1 because they're not being used....

Then the measurement of the current through Pins 1 & Pin 6 using the MERIT lever gives the indication of how strong the tube is....right?

I honestly have no idea what you're saying, but I'm inclined to say no, because the grids and cathodes are on different pins for the different sections of the tube.

Here's some information about the essentially-the-same Eico 666 tube tester:

http://www.tone-lizard.com/Eico666.html

I believe Eico takes the approach that all elements not in use for testing should be grounded. Hence the different settings for multi-element tubes. But the writer states he found errors in the set-up charts. I'm not sure if this applies to the 667.

Dave

It's interesting that about 3/4 of the way down on that TONE LIZARD page, the author makes about the same comment I am making about the 2 equivalent halves of a TWIN triode:


"What makes the mystery truly perplexing is the fact that many tube charts have you alter the voltages, bias, and meter shunt for the two halves of the 12AT7 and 12AX7! Why?"

And as far as the grid and cathode being separate pins.... yes that's true, but that is compensated for by the use of separate levers for each corresponding tube pin.

That's why the EICO meters have 13 pins.... one for each pin of up to a 12-pin tube (like a Compactron), PLUS the cap connector, if there is one.


So anybody wanna venture a guess?

I notice my Hickok has a slightly different setting for each side of a dual diode and I came to think that was because they always reference the same end of the filament.

I know if you test a new 5U4 using the different settings you will get the same reading from each diode.

I expect the two sections of a twin triode to be identical and for that matter the same with a dual diode like a 5U4.

It may be that in a tube tester you have a filament winding that because of the multiple taps the signals used to drive the tube under test are all referenced from one end of the transformer.

Thats my theory

Jim

Hmm Jimmc,

But if the Filaments are connected....all part of the same contiguous element, why must the settings change to accommodate that?

Seems like a lot of extra lever flipping for not much benefit? I dunno, I think I just can't visualize how each triode is different, since the 2 plate merit tests seem to take into account each half of the tube separately already.

Thanks for the possible explanation.....

In your 1st and 2nd posts, I thought you were talking about the switch settings, not the pot settings. I had just assumed they would be the same like they are with my TV-7D/U. Maybe an Eico guru can explain the different settings of the controls.

Dave

Hi eugenedunn,
You have asked one of several $64,000 dollar questions that I have been asking about the Eico 666/667 tube testers for quite some time.I have owned a 667 for about 25 years,and the one thing I have learned about them,is to take Eico's tube testing data with a huge grain of salt.

Norm Leal has said here,many times,that Eico came up with their testing data by merely setting someone down with a supply of tubes and let them "twiddle knobs".When they found a random setting that produced a reading of 100,that became the data for that tube.This is what you are not taking into account,in a nut-shell.

I agree with Norm 100%.There appears to be no rhyme or reason to their data.Most of the time,they don't follow the rules they expect the owner to follow(per the manual),when coming up with settings for un-listed tubes.I have found tubes that were supposed to have a maximum of 2 or 3 ma of plate current,that the 667 was testing at over 30,and still not reading %100!Very,very scary.

Amen.

Ok, it seems there are quite a few people that do use the charts on faith.....

but I really wonder if it's better to take a reference or "calibration" tube you know is pretty good, and using the principles stated in manual, just come up with the best fit settings.

You would have to use due diligence to develop the best Grid and Plate Voltage settings in order to pass all but the most marginal tubes. Then, as a reality check, take those marginal tubes and test them in your amplifier to see if they still perform at all.

Even most of the lame settings should be able to tell you if the tube is shorted-out or something really serious.

That's the path I'm starting to plot.

Still trying to figure out why they seem to make one heater pin go to ground lever setting 1 and the other one go to the heater lever setting 2. I mean, they're each the terminals of a single heater element, so shouldn't they get the same lever setting?

Or is it because of what one of our forum members just said that one of the pins must be referenced to ground? Is that it? (^_^)

One side of the heater supply is probably grounded so one of the heater pins would need to be grounded while the other needs to be connected to the heater supply voltage.

Dave

Thanks for the reassurance, Dave. 'Makes sense.

Now, I wonder what other $64,000 questions I can ask about this model tube tester?

Mike C.? What are some of your concerns?

First, the sections of many dual triodes which are supposedly identical do not necessarily test identically, even when new. It's common to find slight differences. Sometimes you'll find one with a section that tests perfectly good, and the other section tests completely dead. The two sections have to be tested completely independent of each other to prevent interactions between the two from causing false readings.

If you look at the data sheets for certain dual triodes, eg. 12AZ7, you'll find that the plate resistances, and sometimes the Mu's of the two sections are different. This would explain why tube testers use different pot settings in those cases. It's not altogether clear why this would be necessary in tubes that have ostensibly identical sections, if that is what Eico has done.

Secondarily, the shorts and leakage tests are usually performed from the cathode to the other elements in each section, which are grounded in turn. (This is why one of the heater pins is usually grounded as well, since heater to cathode shorts are the most common problems of this type). Again, false readings are to be avoided so the sections have to be tested separately. There's no point in testing for shorts or leakage from one section to the other, since such defects would indicate something really drastic, like somebody put the tube on the floor and stomped on it!

However -- I have a 12AU7 (or a 12AX7, I'm at work now) with leakage from one section to the other. There's no visual evidence of it, and the pins are far apart on the base, but it leaks just the same. Only one element in each section is involved, as I recall a cathode and a plate.

What you said about taking a known good tube(or several examples of NOS tubes),and making up your own settings is the way to go,or at least verify Eico's settings(they did get some right).

My only real concern is the biggy-you really can't take the results these testers give very seriously until you take the time to check Eico's data.I know back in the early days, I threw a lot of good tubes away.But,from my education here,I have learned not to take ANY tube tester's results at face value.

Another interesting trait that I have noticed on both of my 667's,is they will indicate a fairly high level of heater/cathode leakage on certain tubes,when the condition doesn't exist.Every single 6a8 I have tested on my 667s shows moderate heater/cathode leakage.Makes no difference whether it is NOS,or a crusty,half dead specimen from the landfill-results are always the same.I believe I asked Norm Leal about this quite a while back and he said he had never experienced that issue with 6a8's.I had other people test the tubes with different machines(other than Eicos) and they reported no H/K leakage.

IIn my opinion,the first step you should take,if you haven't already done so,is modify your tester so you can monitor plate current during the merit test.This will help you greatly in you quest to come up with good test settings.It's also fun to see just how badly Eico screwed up on certain tubes.

Hey Mike C.

When you said make a modification so you can monitor plate current, did you mean put in some kind of pin-jack or terminals that you can connect your voltmeter to, and monitor current that way?

How did you accomplish that?


Everything else you said......I concur.

Isn't there a chart in the manual showing meter current for every range and pot setting? If so, you're already metering the plate current.

Awww dang Alan,

Now you're gonna make me decipher those flippin' charts too?! (^_^)

But you're right, I do remember seeing a set of 4 charts in the photocopied version of the manual I have.....

Ok...yes, there are 5 charts in the manual I have. Charts 1 thru 4 correspond directly to the S Lever positions #1 thru #4. Chart 5 corresponds directly to both S Lever positions #5 & #6.

The charts have Plate Control readings from 0 to 100 on the x-axis, and the y-axis shows corresponding milliamps.

Then, you're supposed to take the actual meter reading X, and divide it by 100 to get a fraction (X/100), which you in turn, multiply by the chart milliamps to get the actual millamps coursing thru your tube.


I wuz just hoping that I could rig up some test points and put my meter on 'em......