开云体育

Still working on my 577 (see my other post regarding power supply restoration). Voltages are all good now (after replacing all electrolytic caps and tantalums on the power supply and collector supply boards).
I read the other posts regarding retracing of curves but can't seem to get anywhere. In my case, it seems to be a lot worse that what I saw with the other posts so I think my problem must be different (especially since I already replaced the caps that were mentioned in one other post).
Here are some pictures I took with a 2N3055 at various step settings: /g/TekScopes/album?id=79121

I am going to dig deeper into the collector supply board but appreciate any thoughts/feedback.

Thank you,
Wolfgang

On Fri, 23 Nov 2018 15:17:15 -0800, you wrote:

Still working on my 577 (see my other post regarding power supply restoration). Voltages are all good now (after replacing all electrolytic caps and tantalums on the power supply and collector supply boards).

That part is good.

I read the other posts regarding retracing of curves but can't seem to get anywhere. In my case, it seems to be a lot worse that what I saw with the other posts so I think my problem must be different (especially since I already replaced the caps that were mentioned in one other post).
Here are some pictures I took with a 2N3055 at various step settings: /g/TekScopes/album?id=79121

That is a bit on the nasty side. I'm more concerned not with the
doubling of traces (which could be argued), but the lack of blanking
showing through.

Is there a blanking waveform? Is there scope blanking for traces? If
so, what's the waveform look like and what should be generating it?

Harvey

I am going to dig deeper into the collector supply board but appreciate any thoughts/feedback.

Thank you,
Wolfgang

Here are some pictures I took with a 2N3055 at various step settings:
/g/TekScopes/album?id=79121

I'm not at all surprised - a 2N3055 is a power transistor with a very large junction area. So it has a
very high capacitance.

I'd have to look it up, but one of Tek's curve tracer manuals sets out a procedure to match the
looping with removing the transistor and putting a capacitor instead. When the capacitor looping
matches the transistor looping, the capacitor value is the same as Ccb of the transistor.

Craig

For whatever it might be worth, I tried six variations of 2N3055 on a 577 that has had only enough repairs to keep it functioning.
The 3055's varied with a 3055H and a 3055HV mixed in with epitaxial and non-epitaxial 3055's, including an actual Tektronix 151-0140-?
All of the traces looked like what you would expect, and nothing like what Wolfgang is seeing.
The gain of the 3055's varied widely, with the 3055HV having the lowest gain as you might expect.
All of the above means absolutely nothing in this current discussion, but any discussion about my favorite curve tracer model needs to be followed and looked at.
tom jobe...

I have to wonder what were your settings for H Voltage, and
V Current, Supply voltage and load impedance, and your step size.

You can curve trace a 3055 at amps of current, or you can trace
it at microamps of current. Capacitance is much more evident at
lower currents.

-Chuck Harris

OBTW, there is no z-axis blanking on these curve tracers.

Wolfgang Schraml wrote:

On Fri, Nov 23, 2018 at 05:11 PM, Chuck Harris wrote:

I have to wonder what were your settings for H Voltage, and
V Current, Supply voltage and load impedance, and your step size.

You can curve trace a 3055 at amps of current, or you can trace
it at microamps of current. Capacitance is much more evident at
lower currents.

-Chuck Harris

OBTW, there is no z-axis blanking on these curve tracers.

Wolfgang Schraml wrote:

Still working on my 577 (see my other post regarding power supply

restoration). Voltages are all good now (after replacing all electrolytic caps
and tantalums on the power supply and collector supply boards).

I read the other posts regarding retracing of curves but can't seem to get

anywhere. In my case, it seems to be a lot worse that what I saw with the
other posts so I think my problem must be different (especially since I
already replaced the caps that were mentioned in one other post).

Here are some pictures I took with a 2N3055 at various step settings:

/g/TekScopes/album?id=79121

I am going to dig deeper into the collector supply board but appreciate any

thoughts/feedback.

Thank you,
Wolfgang

good point ... didn't write down the settings. I think I used pretty low current for the step setting. Let me run another series of traces with documented settings. Maybe I am chasing a problem that doesn't exist ...

Thank you,
Wolfgang

OK. I uploaded another series of pictures. Same folder: /g/TekScopes/album?id=79121

I used a different 2N3055 than in the previous pictures.

Settings:
- Vertical: 50mA/Div
- Horizontal: 5 Collector Volt/Div
- Step/Offset: 0.1mA
- Max Peaks Volts: 25V
- Variable Collector %: 100%
- Series Resistor: 120 Ohm

I took a picture for every step increase. What's interesting is that "odd" numbers of steps seem to have less of the retracing issues than "even" numbers. The flickering of the display is also more pronounced with even numbers. I ran all samples with the fastest step rate.
Another questions since I have no experience with this instrument - is it "normal" that I get two traces even with just a step count of "1" ("number of steps" knob fully ccw)?

Thank you again, Wolfgang

What you are seeing on your 577 with these settings is as good as anything I have ever seen on my 577. The different behavior with different amount of steps is very normal too. Depending on settings and how many traces you are showing, it can freeze the display and have no flickering of any kind. I believe that extra line you see is just the base zero line, because when you count the traces there are 11 of them when you have the 577 set to the maximum of 10 traces. The base line also comes from with 'dot' at the lower left of your screen, and I believe the 'dot' is the zero point for the display on both the vertical and horizontal axis.
I have seen the kind of traces you have been worried about, but did not understand about testing the transistor with very low current.
I think that Chuck Harris has once again saved us with his excellent advice.
Thank You Chuck!
tom jobe...

On Sat, Nov 24, 2018 at 03:56 AM, tom jobe wrote:

What you are seeing on your 577 with these settings is as good as
anything I have ever seen on my 577. The different behavior with
different amount of steps is very normal too. Depending on settings and
how many traces you are showing, it can freeze the display and have no
flickering of any kind. I believe that extra line you see is just the
base zero line, because when you count the traces there are 11 of them
when you have the 577 set to the maximum of 10 traces. The base line
also comes from with 'dot' at the lower left of your screen, and I
believe the 'dot' is the zero point for the display on both the vertical
and horizontal axis.
I have seen the kind of traces you have been worried about, but did not
understand about testing the transistor with very low current.
I think that Chuck Harris has once again saved us with his excellent advice.
Thank You Chuck!
tom jobe...

On 11/23/2018 8:37 PM, Wolfgang Schraml wrote:

OK. I uploaded another series of pictures. Same folder:

/g/TekScopes/album?id=79121

I used a different 2N3055 than in the previous pictures.

Settings:
- Vertical: 50mA/Div
- Horizontal: 5 Collector Volt/Div
- Step/Offset: 0.1mA
- Max Peaks Volts: 25V
- Variable Collector %: 100%
- Series Resistor: 120 Ohm

I took a picture for every step increase. What's interesting is that "odd"

numbers of steps seem to have less of the retracing issues than "even"
numbers. The flickering of the display is also more pronounced with even
numbers. I ran all samples with the fastest step rate.

Another questions since I have no experience with this instrument - is it

"normal" that I get two traces even with just a step count of "1" ("number of
steps" knob fully ccw)?

Thank you again, Wolfgang

Thank you everyone for the feedback. Like I said - have no experience at all with this amazing instrument and would have probably torn the whole thing apart again chasing an issue that doesn't exist.

I ran through the adjustment procedure last night and all except the "Polarity Positioning" when switching back from "Display Invert" to " Display Norm" (Adjustment Procedure, step 15 (f)) could be brought within spec.

The only remaining issue is the storage feature not working well - very weak overall and only works around the center of the screen. Not worrying about that right now, not planning to use storage.

Thank you again, Wolfgang

Thank you everyone for the feedback. Like I said - have no experience at all with this amazing
instrument and would have probably torn the whole thing apart again chasing an issue that doesn't
exist.

I ran through the adjustment procedure last night and all except the "Polarity Positioning" when
switching back from "Display Invert" to " Display Norm" (Adjustment Procedure, step 15 (f)) could
be
brought within spec.

The only remaining issue is the storage feature not working well - very weak overall and only works
around the center of the screen. Not worrying about that right now, not planning to use storage.

Thank you again, Wolfgang

See the note on page 10 of the operations manual regarding conditioning of the storage CRT if it has
been turned off for two weeks or more.

Craig

"See the note on page 10 of the operations manual regarding conditioning of the storage CRT if it has
been turned off for two weeks or more."

That's one of the first things I tried. Only leads to an approximately 1.5" diameter, washed out spot slightly to the left and up from screen center.

Wolfgang

Hi Wolfgang,

Please forgive the capital letters. I am not shouting but I do hope to get your attention to what the other TekScopes experts (Craig Sawyer, Chuck Harris, Tom Jobe) are telling you:
* THERE IS NOTHING WRONG WITH YOUR CURVE TRACER!!!!
* YOUR FINE PHOTOGRAPHS SHOW A PERFECTLY NORMAL SET OF 2N3055 CURVES ON A 577.
No matter how much you think your photos show your curve tracer is not working it is working perfectly.

Just in case that was not clear let me say it a little differently:
* YOUR 577 IS WORKING EXACTLY LIKE ALL OF THE 575s, 576s, AND 577s I HAVE EVER USED (I OWN 5 AND I HAVE USED MORE THAN 20).

There are several things that are causing this misunderstanding on your part:
* Several of the curve tracer controls interact with each other and/or with the transistor's actual parameters to create misleading (or annoying at best) visual artifacts that you are pointing out as performance defects of the curve tracer such the loops on the left side of each curve. Those are characteristics of the transistor itself. Others have pointed out how to prove this by substituting a small value capacitor for the transistor to see that it creates loops too.
* Your expectation of what ideal transistor curves look like conflicts with 1) how real transistors (with stray capacitances and other ugly details) actually perform; and 2) how real world instruments such as a 577 operate due to limitations of real world circuit designs, or constraints like costs, weight, size, heat, etc.
* Your incorrect expectations of what transistor curves SHOULD look like on a curve tracer are confusing you further and further.
* You don't have enough experience with Tek Curve Tracers to realize that yours is operating perfectly.
* You do not have enough hands-on experience with curve tracers to realize the subtleties of how they operate. For example I have 50+ years of experience with the 575/576/577 and I still haven’t used every capability of these sophisticated instruments.
* The designers of the 576/577 knew it would have to test devices that weren't invented yet. As a result these curve tracers have many more controls and capabilities that they would seem to need. Until you need these capabilities those controls only create more confusion if you set them wrong. This is part of your problem.
* There is no blanking on the 575/576/577 so you see the unusual traces between steps at the right side of each step in your photos. This is because you were using the FAST Step Rate. I could explain why this is so but it would be better if you learned why yourself (from the Theory of Operation" section in the Service Manual) so you would realize it is not a sign that there is something wrong.

Regardless of what "wrong" curves you get on your 577, if you provide a photo, we can explain why the curves look that way. It will either be due to:
1) The settings of the curve tracer (such as using the FAST Step Rate),
2) The way the curve tracer circuitry works (it doesn't use blanking),
3) A real world transistor is not the same as an ideal transistor (Miller Effect, leakage, etc).

If I succeeded in reassuring you that your curve tracer is working properly, CONGRATULATIONS!
On the other hand, if Craig, Chuck, Tom, and I haven't convinced you, then I suggest you sell the curve tracer. You will never be happy with it and you are going to waste a lot of time and money chasing a chimera.

My only request is that you think about how much time we have spent so far explaining to you that there is nothing wrong before you ask us for further reassurance. It won't be a good use of our time.

I hope you are able to stop worrying and begin to enjoy the remarkable instrument you have. You can download a copy of the Tek book "Semiconductor Device Measurements" (link below) to learn how to use your curve tracer to make all sorts of device measurements. If you look carefully you will see the steps have loops in some of the photographs in this book.


Dennis Tillman W7PF

Thank you Dennis.
I did accept that this fine instrument is working as specified. My last comment was regarding the storage option which I am not really worried about.

I apologize if I gave you the impression that I am not trusting the very helpful advice I received in the last couple of days. I truly appreciate all the help I received.

Thank you again,
Wolfgang

Hi Wolfgang,

Excellent!
By the way, my storage 577 has similar "quirks" to yours. I tried adjusting it according to the manual and I found the directions very confusing. I finally gave up. Now I am happy that it works well enough that I can use it when I need to and if I get a small cloud when I first turn it on pressing the erase button always makes it go away so it is not a problem.

There are several valuable things you can do with the storage 577 that the 576 cannot do so that is one reason I prefer the 577. One odd advantage I discovered to the storage is that since the 577 graticule is not illuminated it can sometimes be hard to capture it in a photograph, but if you turn the storage gun up a bit it creates a uniform green fog effect that back lights the graticule just enough that you can see it and your transistor curves together making it easy to get both in the photo.

Dennis

By the way, my storage 577 has similar "quirks" to yours. I tried adjusting it according to the
manual
and I found the directions very confusing. I finally gave up. Now I am happy that it works well
enough
that I can use it when I need to and if I get a small cloud when I first turn it on pressing the
erase
button always makes it go away so it is not a problem.

Oddly enough storage was/is perfect on my 577. That isn't the problem. A couple of years ago I had a
tant go short. To locate which board was the problem, I unplugged the power supply. Finding the
problem was then obvious and said capacitor was replaced.

Then I plugged one of the darned cluster of harmonica connectors one pin out, and the magic smoke came
out. I was so angry, disappointed with myself, and discouraged I parked it for a year. Then I set
about replacing everything that was clearly ex-silicon and a new clutch of ex-tants. Logic chips had
had some horrid voltage and most were dead. Op-amps were dead. There are two dual JFETS in there, that
were clearly dead - and much to my delight they are still currently available parts.

This saga was behind my word of caution to Wolfgang to check the connectors were plugged in correctly.

Now everything works apart from a remaining gremlin in the step generator or amplifier, so at least
I've put the magic smoke back in at last. It is again parked waiting for another burst of fixit
enthusiasm.

Craig