Technical Bulletin 01, part 1
Quick procedure to test EML tubes with AT1000
There are two ways to use the AT1000.
1) PC mode. Makes the AT1000 do things the way you want it, and get a printed report
2) Stand alone mode. Displays a partial result. No report, and no information how the test was done. Most if all, it uses a wrong default set up for the tester. (You have to change that)
The following applies for both PC and Stand alone mode.
Use always AUTO BIAS. Doing so, will EXACTLY repeat the test as it was done in the factory. We use also an AT1000 for this. Any deviation of a used tube will be: Gm and grid voltage will be another. Tested this way, normally you can say, if Gm is still good, the tube is good.
What is not working: Test with fixed grid voltage. No manufacturer supports this. It can make good tubes look bad, or bad tubes look good. If you have a strong reason to do so anyway, you can let us know what that is, but please do not ask us to comment on the result, as most of the time it will be invalid anyway.
PC mode.
For newer versions, most but not all EML tubes are present in the data base.
* If a tube is missing, you can add it with the Amplitrex programm called "tube editor". In the editor, open tubedate.csv. Use file manager to find it on your PC. Once opened, you can add a tube of your choice. As a start, look at tubes with a similar socket, and copy the socket connections from there. Insert a "new tube" at the position where you want to have it. Then enter: Grid voltage, plate voltage, transconductance and heater voltage, and save the tubedata.csv file, and you are done.
As said, choose AUTO BIAS mode for testing. This is just a click in the menu.
Stand alone mode
This is a primitive method, but it works without PC, whatever reason you have for that. It uses the database inside the AT1000. You can add a new tube into that as well. Use the AT1000 handbook for this, this is a bit diffuclt to do, since in the end you need a PC anyway, and need to edit the tube.csv file anyway. Once the tube is added, you need to change the general setting of the AT1000 to AUTO BIAS. Otherwise, results make no sense. This is done, by holding the left arrow key pressed, while switching the mains on. Then look at the display, and repeatedly press the arrow key, until you see "AUTO BIAS". If so, press the enter key.
TEST RESULTS in either PC or stand alone mode
If the tester is set up correct, plate voltage and plate current will be the same as on the EML tube box. With stand alone mode, you can not verify this, so make sure you did that right, while added it to the tubedata.csv file.
After testing, if Gm and -Ug (grid voltage) is more or less the same, you can say the tube has no, or little use.
If the tube is used, initially -Ug will go down. This is slow, continous process, already beginning after a few hundred hours. This has nothing to do with end of life. The next level of use becomes visible when Gm begins to drop below it's original value. Also this is normal, and a gradual process. There is some consenus, that a value of 70% is an acceptable wear out limit, but there is no fixed number for this.
PC mode gives additional quality information, such as Plate resistance and Gain. If a tube gets more used, not only Gm gets lower, but then also plate resistance begins to rise. Even that means not end of life. At further use, gain begins to decrease, and then end of life is getting nearer.
We need to make a strong note here, that end of life in the first place is determined by deviation of the original test value (as on the box) and less by absolut values as such. To give an example, a factory new tube can have a Gm from 5...6.8. This is not a matter of weak or strong emission, it is only a matter of mechanical distance tolerance inside the tube. If a Gm=6.8 tube tests at only 4.8, that means most life time has been used up. Wheras a Gm=5 tube, testing at 4.8 is as good as new. This is why you can say not much about random tubes, when original test data is missing.