DIY Curve Tracer

[jackinnj]

Ale: You've done such nice work on the instrument, why not get a silk-screen kit for the lettering.

[mogliaa]

Originally Posted by jazbo8

Ale,

Sorry to bother you again... What are the front panel controls highlighted in the attached picture? I can't see them on your schematic.

Thanks,
Jaz

Hey Jazz,
I'm on the road with no laptop.
Ia: is the pot for the dn2540 cascoded CCS I use for measuring transconductance and THD
G: grid output to calibrate steps using the scope
Grid zero: bias pot to set Grid starting point of Vgk curves
Switch on the bottom is power switch
Ia switch disconnects HT from CCS

All that part of the circuitry blew up with my last 600V test. Hopefully I will sit down to repair it when back
Cheers
Ale

[mogliaa]

Originally Posted by jackinnj

Ale: You've done such nice work on the instrument, why not get a silk-screen kit for the lettering.

You may be right, but haven't done that before, any idea what/where to get it from?
Thanks!

[jazbo8]

Originally Posted by mogliaa

All that part of the circuitry blew up with my last 600V test. Hopefully I will sit down to repair it when back
Cheers
Ale

Thanks for the clarification.

Too bad about your gm/THD test rig, I read about it on your blog, something to do with Millet's soundcard input, was it? I am still gathering parts for the build and going over the Tek 570 schematic, I'd love to incorporate the Plate-Screen-Grid switch for the vertical amp, but the switching and the extra power supplies are not "DIY friendly"... Those Tek engineers sure knew what they were doing!

Thanks,
Jaz

[mogliaa]

Thought about the addition of a simple short/leakage testing to my tester. Now that will have to open and do some repairs, I can easily drill a couple of testing connectors and use existing +/- 80V supply available.

I thought about the proposed leakage circuit in the Sussex tester, so came up with the simple modification in the attached.

With a couple of probes can use this to test anode/grid leakage, filament continuity and cathode leakage as well.

Only concern is whether gain is too much. MPSA42 has an hFE=90@Ic=1mA. So input sensitivity is about 5uA to light the russian neon bulb.
Looked around for grid current values thresholds used in testers but couldn't find any. Old testers use a simple neon bulb to identify leakage so I guess they are measuring 100uA or above in order to get a dim light on the neon bulb?

Suggestions?

Thanks
Ale

[v4lve lover]

see the Neuberger tube manual . http://www.jogis-roehrenbude.de/Roeh...400/RMP400.pdf

There is a part about grid current See Page 18b

"the value of the measured grid current can be used as a criteria, as a rule of thumb

Tubes whit a Bad vacuum (tested bad on the procedure in 18A)
preamp tubes whit less than 4 uA are usable
output tubes whit less than 10 uA are usable


tubes that are good and show little Gas on the gas test.
should have less than 0,6-1uA for preamp tubes and less than 1.5-2 for output tubes.

lot of useful data in that manual.

v4lve

[mogliaa]

Thanks v4lve!
So in principle 5uA is not bad at all for this purpose. This circuit should work fine. Can build it very quickly and report results.

Only point is that output valves with 5-10uA will turn on the lamp

I'm trying to avoid the addition of a grid sensing resistor and an op amp and the output to measure the grid current externally, no additional space inside the tracer am afraid.
Ale

[mogliaa]

Originally Posted by v4lve lover

see the Neuberger tube manual . http://www.jogis-roehrenbude.de/Roeh...400/RMP400.pdf

lot of useful data in that manual.

v4lve

Wish I could translate this, but unfortunately don't speak German!

[jazbo8]

I'm also thinking about incorporating a shorts test into the tracer, I like the B&K version, the detailed schematic is in the B&K 747 tester if you want to take a closer look, the basic setup is as follow:



Jaz

[jazbo8]

I ran a quick test on my adapter this morning, it worked BUT how do I get rid of the re-traces from the display? I played with the Phase and Sweep adjustments (per Ale's schematic), but they did not seem to have an effect... I also tried blanking the Z-axis by connecting the Y-sweep to it, which helped may be a little, but most of the re-traces remained... Did I overlook something obvious?

Jumbled curves:


TIA,
Jaz