I am trying to calibrate a hickok 533a. I purchased a 6L6 calibration tube from Sonny the radiolaguy. It was measured at 5500 mhos. I have also downloaded the instructions from BAMA. All the voltages are correct, and the tube measures 5500 correctly on the ENGLISH scale, but the 6000 and 15000 scales test the tube at 4200. According to the instructions, I have to adj. the resistance of the range spools.

Here's where I run stuck. How are these adjusted, and which ones do I adjust? I see 6 spools mounted on the MICROMHOS switch, and 2 more under P5 and P6.

Any help would be greatly appreciated.

EDIT:

When I adjust the bias dial so that the tube reads 3000 on the 6000 scale, then switch to the 3000 and 15,000 scales, all the scales read 3000.

The tube I purchased is for the 6000 series testers, it will not correctly calibrate a 539 or 580.

I don't think I would jump to changing the spools, the resistance of these does not change. You did say the voltages were correct.

The "english scale" is just a good bad scale, it does not read conductance.

I am always suspect of the so called 'calibration tubes' which are sold. The tube must have been tested on a calibrated 533A to work on your tube tester. Verify your tube before going any further.

You can always calibrate your tester using the method which does not require a calibration tube. The "calibration tube" is used to periodically check the calibration of your tester.

Your tester may be fine just the way it is.

Dave

Hi Dave, how do you do it without a calibration tube?

Instructions sent, check your e-mail

Dave

Hi Dave, thanks for the file, it does help some. I have been double checking my voltage measurements, and I found one that isn't quite right.

Quote from the instructions:

"Connect a high impedance AC voltmeter to pins 8 and 5 and measure the grid signal voltage. The AC grid signal voltage should be 2.5 volts plus or minus .05 volts. No calibration adjustment is available to change this."

This is 2.2 VAC. If this is low, will it affect the readings? Bias is out by maybe 0.1 volts at the most, all the rest are fine.

EDIT:

Would an incorrect bias fuse lamp cause this problem? It is supposed to have a #49 lamp (2V, .06A), but has a 6.3V .35A lamp instead. I'll have to get another lamp and try it out.

EDIT:

The lamp doesn't make a difference. I have discovered that the grid signal voltage is low due to 2 out of tolerance 500 ohm resistors. One of them measures 625 ohms!

Dave,

I have a Hickok 533. Could you please send me the instructions on how to calibrate without a "calibration tube"?

Thanks,
Geoff

Should be similar to the 600/600A Calibration, using a variable transformer and an isolation transformer as outlined in the calibration instructions. Works well, have also used it on the Hickok 539 series.

http://www.radiolaguy.com/PDF_Files/Hic ... ration.pdf

you are not using the tester correctly, you don't read 5500 on the English setting only good-bad scales. Don't go making adjustments till you sort things out. You said"When I adjust the bias dial so that the tube reads 3000 on the 6000 scale, then switch to the 3000 and 15,000 scales, all the scales read 3000." That's exactly what it should do. How does a new 6l6 test on the English setting?

PM me with your email and I will send it to you.

Dave

The Gm for all tubes varies with plate voltage and current, by as much as 300%. Now the plate voltage in a Hickok is 150V, so that takes care of one of the variables, but the plate current is dependent on grid bias, so a "calibrated" tube is only useful when you know your grid bias is the same as the machine it was calibrated on.

The Gm reading is also dependent on the grid signal level, if the level is high relative to the bias (and it almost always is).

I don't think a honest person would sell a calibrated tube.

Dave

The final check on many Hickok tube testers per their calibration instructions is to use a calibrated 6L6. The problem is that when using a calibrated tube, and also comparing different tube testers, one assumes one size fits all. If you know the operating parameters of a tube and have verified them in a test jig, then the tube should test similarly as a calibrated tube in a tester that uses the same parameters. Most "Hickok" testers are quite limited in their testing parameters, have a number of testing variations between models, and testing values between tube testers are often not interchangeable. The testing parameters and results are specific to that tester. I make my own calibration tubes/plate curves using a test jig with parameters specific to the calibrated operating conditions for the individual tube tester. I use a set of three calibration tube types for GM (0 to ~10,000) and plate current (0 to ~100 ma), warms my heart that they test similar between tube testers. Nice, but most testers only allow gross adjustments, if that.

The radiolaguy http://www.radiolaguy.com/info/tubeTesterRest.htm does seem to tailor the calibration tube to the tester, so for many individuals it is a reasonable check to feel comfortable their tester is in the ball park. Hickok military tube testers do not use a tube to calibrate GM and specify using a variable transformer, an isolation transformer and a 10K resistor, also recommend by Daniel Schoo. This specifically gives GM, and removes some of the other variable like plate, screen and filament voltages when using a 6L6. Joel Hatch’s “Transconductance Tube tester Standard” article in the February ’05 issue of AudioXpress is a simple fabricated standard and that sidesteps many of the issues around the usual "Hickok 6L6 standard tube" and it does not change between testers or with time. On my Hickock testers, using Joel's standard with some mods, the testers measure within 5% at it's calibrated GM of 1500, 3000 and 6000. All these methods work well, and in my experience give fairly close GM values. My issue with the 6L6 tube standard is that it only tests the GM parameters under one set of conditions, and you can be way off under other test conditions because you messed with the tester to get it right for the 6L6. All calibration voltages, especially the grid signal must meet specifications, otherwise you may be way off on the GM/ma readings. http://www.radiolaguy.com/PDF_Files/Ove ... ration.pdf.

Does anyone have a scan of that article that they can post? I'm sure it would be helpful to a lot of people on this site.

- Geoff

I have one of Joel's calibration tubes and the article, at least on paper. I'll see if it survived my last move.

He was going to make a revised version at one point, but I think he stopped the project. I should see if I can find his E-mail too.

Tube testers are great fun (I should know) but you can make too much of this "calibration" business. I calibrated my TV7B fifteen years ago and it works fine. If a bunch of new tubes started showing up as weak, I'd know something was wrong. In the meantime I don't worry about it. Yeah, I could keep a "standard" 6L6 on hand as a periodic check, but who cares? The Space Shuttle won't crash if I measure a tube wrong.

The Gm meter in a Hickok is measuring AC current, which for a known AC grid voltage, translates directly to mutual conductance (Ohm's Law). The 10k resistor/50V transformer provides a known AC current (5mA) for the meter to measure. That's about as basic as it gets. The 10k resistor is a pretty good imitation of a medium-power tube.

If Gm numbers are really important, you need a better tester than the average service-shop Hickok. Cardmatics work well for indirectly-heated tubes, and are relatively easy to come by. Except for setup time, and the difficulty of finding one, the AVO VCM163 could really spoil you. Likewise the full laboratory models (RCA, Weston, New London, or the Tek curve tracer), but in no way are those suitable for routine testing. Some modern-day instruments get around the setup problems for commonly-used audio tubes, and you could make your own. It's not (despite my earlier "space shuttle" reference) rocket science.

I agree Alan. I've read enough about tube testing to know that the test itself is not the final word on the acceptability of a tube. I'm just trying to find a standard way of doing the calibration without the 6L6 tube since the consensus is (from what I can tell) that there seems to be a bit of "snake oil" in the ones advertised as "calibration" tubes. I'd like to at least get my 533 to some sort basic standard to start from after going through and checking the resistors.

- Geoff

It's not so much snake oil as it is demanding of the same configuration. So, filament voltage, bias, signal level, plate voltage and load all have impact. So, if the reading's not up to snuff, which variable(s) to correct? Best to do it (right) from the start to finish. Then you certainly don't need a pricey tube to verify that things are still in calibration. Simply warm things up and test one of your own tubes. Save it to use if/when you want a sanity check.