uTracer (Micro Curve Tracer)

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

Post Reply
Matt D
Posts: 170
Joined: Fri Apr 22, 2011 6:21 pm

Re: uTracer (Micro Curve Tracer)

Post by Matt D »

I had some time with my friend's Amplitrex yesterday and tested some new JJ ECC81's in the ways described above. I let the tubes cool completely between each test.

The first test involved turning the heaters on for 1 minute and measuring both sides immediately. For the second test I turned the heaters on for 3 minutes and then tested each side immediately. In the third test I turned the heaters on for 1 minute and allowed the tube to flow current (10.5 mA at 250V on the plate I believe) for 2 minutes per side.

Tube A:

IP:
1) 9.2 / 8.6 mA
2) 9.4 / 9.0 mA
3) 9.4 / 9.4 mA

GM:
1) 5100 / 5060
2) 5130 / 5200
3) 5100 / 5290

RA:
1) 10K / 10K
2) 10K / 8.3K
3) 10K / 10K

mu:
1) 51 / 50.6
2) 52 / 43.1
3) 51 / 52.9

Tube B:

IP:
1) 8.2 / 8.6 mA
2) 8.8 / 8.4 mA
3) 8.8 / 8.8 mA

Gm:
1) 5000 / 4810
2) 5070 / 4900
3) 5000 / 5080

RA:
1) 8.3K / 10K
2) 10K / 8.3K
3) 10K / 10K

mu:
1) 41.5 / 48.1
2) 50.7 / 40.6
3) 50 / 50.8

Anecdotally, later when I was testing a used 6BQ7A out of an old radio I noted that over the course of 3-4 minutes the plate current rose from 6.2 / 4.6 mA up to 8.6 / 5.1 mA

The Amplitrex certainly has it's own flaws and inaccuracies and the results above are actually just a bit more consistent than I'm used to seeing, but there are some significant changes with increased heater time and current flow before measuring. Hope this helps add to the discussion and thanks to Martin and everyone who has contributed to this thread.
User avatar
martin manning
Posts: 13080
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: uTracer (Micro Curve Tracer)

Post by martin manning »

Interesting results, thanks Matt. What was the test point for the ECC81? I'm guessing 250V Va and -2V Vg?

Seems like the drift is greatest on section 2 on both tubes, and the values are low on test 2. It'd be interesting to see what the measured current does over time using resistors, as I did above, to see if any of that variation is in the Amplitrex itself. That could be done with and without plate current flowing, too.
John_P_WI
Posts: 1456
Joined: Wed Aug 15, 2007 4:29 pm
Location: Wisconsin

Re: uTracer (Micro Curve Tracer)

Post by John_P_WI »

After a day and a half of building and testing, the pcb works! With a lot of thanks due to Martin for his contribution to the manual.

I did have one little snafu - I broke a 1K resistor (R29) when piggy backing a small cap with it. Grabbed another from my pedal parts and away I went. All testing went well until calibrating Ia and Is as Is wasn't the diagonal line as expected - After digging around and doing a few measurements against the schematic and transistors I discovered that the resistor I grabbed was actually 10k! Damn those 4 band vs 5 band colors :roll: Put me an hour behind!

FYI, the notebook I use on the bench is an old Lenovo running XP and so far haven't had any problems with an old Belkin USB to RS 232 converter.

It certainly was an easy and stress free build,except for my one mistake. Next is the enclosure.

Now I just need to figure out how to use this as there are sure a lot of options... :oops:
User avatar
martin manning
Posts: 13080
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: uTracer (Micro Curve Tracer)

Post by martin manning »

Great, glad to hear it's up and running! I had a resistor incident myself, accidentally swapping a 12k1 for a 121k. The test as-you-go process caught that, so it wasn't much of a problem. To be clear, the construction manual is all Ronald, and it is very good indeed. I helped out with the user manual, proofing and editing.

Since you can save the complete test setup and representative test results, I have a file for each tube type (a .uts file) that includes the setup for both the single operating point Quick Test and a trace of Ia vs. Va-k at constant Vs for pentodes (in uTracer-speak "I(Va, Vg) with Vs, Vh constant"). For single triodes, that same measurement type is used, with the Vs being irrelevant. Dual triodes get the "I(Vg, Va=Vs) with Vh constant" to trace both sides at once, and the Quick Test form includes inputs and nominal results for both sections. For pentodes, you can include both the pentode and the triode connection Quick Test setups and nominal results.
JoeTele
Posts: 127
Joined: Sun Oct 27, 2013 2:20 pm
Location: Milwaukee, WI

Re: uTracer (Micro Curve Tracer)

Post by JoeTele »

I've been skimming this thread and find it fascinating. I don't want turn it into another discussion into the general merits of tube testing, this or that tester, etc., so could someone just tell me, even if without any real detail, whether this is really the holy grail of tube testing/data? In other words, if the tube is able to reach full voltage, not smoke, explode, etc, and produce on the curve tracer a respectable approximation of the curves on the data sheet, does that circumvent the need for the other various tests done by other devices (excluding those that might be done before putting it on the tracer, such as tests for shorts)?

Thanks as always!

Joe
Matt D
Posts: 170
Joined: Fri Apr 22, 2011 6:21 pm

Re: uTracer (Micro Curve Tracer)

Post by Matt D »

I don't know if there is a Holy Grail of tube testing. This looks great though:

http://www.roehrentest.de/EnglishInfo.html
User avatar
martin manning
Posts: 13080
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: uTracer (Micro Curve Tracer)

Post by martin manning »

JoeTele wrote:...if the tube is able to reach full voltage, not smoke, explode, etc, and produce on the curve tracer a respectable approximation of the curves on the data sheet, does that circumvent the need for the other various tests done by other devices (excluding those that might be done before putting it on the tracer, such as tests for shorts)?
There are some defects that wouldn't be obvious from the traced curves or a spot point measurement (like the uTracer's Quick Test). Some vintage testers had "Gas" and "Leakage" test functions aimed at detecting these. The leakage test is primarily concerned with the integrity of the heater-cathode insulation, and the gas test looks for grid current. The "Life" test on vintage Hickok testers reduces the filament voltage by 10%, with the pass criteria being not more than a 20% reduction in gm. The "Life" test would be easy to reproduce on the uTracer. The leakage test could be done using the anode or screen supply to measure current from k to h at rated Vh-k, with the heater on, and placing a large resistance in the grid circuit could be used to detect excess grid current.
JoeTele
Posts: 127
Joined: Sun Oct 27, 2013 2:20 pm
Location: Milwaukee, WI

Re: uTracer (Micro Curve Tracer)

Post by JoeTele »

The leakage test could be done using the anode or screen supply to measure current from k to h at rated Vh-k...
Would this be as simple as connecting an ammeter from the heater pin to the cathode pin?

Joe
User avatar
martin manning
Posts: 13080
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: uTracer (Micro Curve Tracer)

Post by martin manning »

No, to make the h-k leakage measurement a voltage potential has to be included in a series loop including the h-k pins.
v846
Posts: 117
Joined: Fri Jan 21, 2005 9:50 pm
Location: T-Ville, CT

Re: uTracer (Micro Curve Tracer)

Post by v846 »

Matt D wrote:I don't know if there is a Holy Grail of tube testing. This looks great though:

http://www.roehrentest.de/EnglishInfo.html
A friend has one, its pretty impressive.
JoeTele
Posts: 127
Joined: Sun Oct 27, 2013 2:20 pm
Location: Milwaukee, WI

Re: uTracer (Micro Curve Tracer)

Post by JoeTele »

No, to make the h-k leakage measurement a voltage potential has to be included in a series loop including the h-k pins.
If one wanted to do this outside the context of the u-tracer, could the heater windings come from the same PT secondary as the voltage potential in series with the h-k loop provided the center-tap of the heater windings was lifted, or otherwise isolated from the h-k loop? I ask because I'm working on a design for a home-brew tube/cap leakage tester.

Joe
User avatar
martin manning
Posts: 13080
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: uTracer (Micro Curve Tracer)

Post by martin manning »

Testing a 6688A sharp cut-off, frame-grid pentode. This little tube has a very small grid bias voltage requirement: At typical Vg2 it takes only about -3V to reduce anode current to ~zero. Transconductance is high at 15.5 mA/V (@Ia=11.5 mA)... compare that to an EF86, where Gm is 2.0 (@Ia=2.0 mA). This tube did not want to cooperate straight away, as it showed signs of oscillation (red, magenta, and blue traces, first screen shot). Grounding g1 to the cathode stopped that, and I was able to get a true Vg1=0 trace (black trace, first screen shot). Next a ferrite was clipped to the (green) g1 patch cord (photo), and that worked for all Vg1's (second screen shot). Note the Vg1=0 line in the second screen shot is not as high as the grounded g1 trace due to the inability of the grid bias circuit to set 0V on g1 and grid current flowing through the cathode.
You do not have the required permissions to view the files attached to this post.
User avatar
JazzGuitarGimp
Posts: 2355
Joined: Mon Jul 23, 2012 4:54 pm
Location: Northern CA

Re: uTracer (Micro Curve Tracer)

Post by JazzGuitarGimp »

Matt D wrote: Sun Aug 21, 2016 4:16 am I don't know if there is a Holy Grail of tube testing. This looks great though:

http://www.roehrentest.de/EnglishInfo.html
Wow, yeah, that's a good looking piece of gear! Whole lotta sheet metal though. I wonder if it can plot posative grid voltage anode curves for power tubes.... i see the Va goes all the way up to 600V, which is nice.
Lou Rossi Designs
Printed Circuit Design & Layout,
and Schematic Capture
User avatar
martin manning
Posts: 13080
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: uTracer (Micro Curve Tracer)

Post by martin manning »

The uTracer must use a DC voltage for the heater, and I prefer to use an external power supply to insure that the correct voltage is set and for monitoring of the heater current. However, testing tubes with directly-heated cathodes will produce some confusing results when an external power supply is used due to the variation in cathode voltage from one end to the other. This is worth understanding, since vacuum rectifiers with directly heated cathodes are still fairly common (5U4 and 5Y3 e.g.), and their heater current demands generally exceed the capacity of the uTracer's internal supply.

Two test runs on a 5U4GB dual diode are plotted below, with the polarity of the external heater supply voltage reversed on the second run. The inset diagram shows the test setup and a physically correct schematic symbol for the tube. Most vacuum rectifier diagrams show a single filament, but there are actually two cathode/filaments in series, one for each diode, and they will have different potentials when a DC supply is used. This changes the average Va-k and therefore the anode current. Here the difference in average Va-k seems to be consistent with the observed effect on anode current. If an accurate result is desired, the traces with heater voltage reversed should be averaged, which should produce a result equivalent to the test performance on AC, where the average voltage on the filaments is the same.
You do not have the required permissions to view the files attached to this post.
User avatar
martin manning
Posts: 13080
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: uTracer (Micro Curve Tracer)

Post by martin manning »

Here's the other common configuration for high-power rectifiers, i.e. GZ34/5AR4, where the two heaters are connected in parallel. The cathodes are indirectly heated, but both share a common pin (8) with the heaters. In this case the polarity of the external DC heater supply does not matter since the cathodes and the cathode terminal of the uTracer are at the same potential either way. Should the uTracer's cathode terminal be connected to the wrong pin (2 vs. 8), an offset of plus or minus 5V is created by the heater voltage, depending upon the heater polarity.
You do not have the required permissions to view the files attached to this post.
Post Reply