开云体育

On Mon, May 14, 2018 at 01:09 pm, Kurt Rosenfeld wrote:

Albert,

Good to know that GPIB doesn't work when it has hung. What's our best
hypothesis for why it hangs?

-Kurt

As you said it's probably "something" with the Front panel interrupt handling. E2131 points to that in the 11400 series you said, and also the CRT text hints in this direction: Executive - Front panel - Control - Interrupt. The signal & control path is via the A14 I/O board; this board issues (or passes?) an interrupt to the A17 Exe board Interrupt Controller.
Strange enough the start-up messages seem to be variable. At present I get E2131, 3 faults, with above mentioned text. Earlier it were 5 faults with text Executive - Front panel - Keys - Open. Then I noticed that the Measure button was stuck. I remedied that and got 2 faults. After several measurements at were those 3 faults. When I deliberately push the Measure button during start-up then I get 5 faults but with the Control text and not that Keys text.
Also yesterday the Display intensity changed when I rotated one of the knobs, but earlier and today there was no response at all to those knobs.
The problem could be as simple as bad contacts in a ribbon cable connector. I checked the 40-wire cable between A10 Front Panel Control and A14 I/O, all wires showed continuity. I also jiggled the other connectors at the A10 board.

W.r.t. GPIB, all settings are internal in the CSA and moreover might be destroyed by the NVRAM chips replacement. I tried to make contact at primary addresses 0 to 6 or so, the same way I do it with the 7854 and the 2232. At IDN? and serial poll I only get errors in return.

Albert

On Mon, May 14, 2018 at 01:40 pm, Raymond Domp Frank wrote:

I have two 468's they do exactly the same thing at the fastest sweeps. Takes

over 2 divisions before I have a trace.

You may check adjustment 5 in "Display and Z-Axis Adjustment", Vol. 1 page
4-33 in my Service Manual. Adjustment won't interfere with 'scope calibration.

Raymond

One thing after another... I just turned my 468 on and noticed some trigger flakiness also so while im in there I will look at the Z-axis adjustment also. Thanks for the tip!

I try to draw the line on whether or not I can lift it without assistance :-)

Paul

<grin>

The 175 is an insane unit. It is wired internally with welding cable and has an exceptionally
serious power transformer. All the switching is done using major-league relays. It has a 200A peak
collector current and 12A base current. It also has very, very large current sensing resistors.

When I first got it, I made up an umbilical to connect it to the 575 and fired it up. Just to test
it out, I wired in a random TO220 package transistor. There was a loud crack is it exploded into hot
shrapnel without very much effort at all. The tab burned into the carpet. "What is this mark on the
carpet? It looks like a burn" "No idea, dear...."

On Mon, May 14, 2018 at 01:33 pm, Dave Voorhis wrote:

On 14 May 2018, at 16:59, lop pol via Groups.Io
<the_infinite_penguin@...> wrote:

On Sun, May 13, 2018 at 10:55 pm, lop pol wrote:

lop pol
11:07 am #147825

When powering on the scope from cold the readout is high by about half a
graticule. After 3-5 minutes of warm up the readout drifts to the center of
the top graticule. Is this normal for this to happen.

...

And I'm pretty sure R2527 (thermal balance) is bad. Just touching it causes
the bottom readout to distort and jump to the middle of the screen. I

swapped

the readout board just to make sure and the issue still persisted. Tomorrow

I

will swap out the thermal balance resistor and go from there.

Swapping out the variable thermal resistor seemed to work.

My 7623 does the same warm-up readout drift. Where is R2527 on the 7633?

I can’t find a “thermal balance" on the 7623 schematic.

On the 7633 its on the top of the vertical amp board. /g/TekScopes/photo/49110/1?p=Name,,,20,1,0,0

I have two 468's they do exactly the same thing at the fastest sweeps. Takes over 2 divisions before I have a trace.

You may check adjustment 5 in "Display and Z-Axis Adjustment", Vol. 1 page 4-33 in my Service Manual. Adjustment won't interfere with 'scope calibration.

Raymond

On 14 May 2018, at 16:59, lop pol via Groups.Io <the_infinite_penguin@...> wrote:

On Sun, May 13, 2018 at 10:55 pm, lop pol wrote:

lop pol
11:07 am #147825

When powering on the scope from cold the readout is high by about half a
graticule. After 3-5 minutes of warm up the readout drifts to the center of
the top graticule. Is this normal for this to happen.

...

And I'm pretty sure R2527 (thermal balance) is bad. Just touching it causes
the bottom readout to distort and jump to the middle of the screen. I swapped
the readout board just to make sure and the issue still persisted. Tomorrow I
will swap out the thermal balance resistor and go from there.

Swapping out the variable thermal resistor seemed to work.

My 7623 does the same warm-up readout drift. Where is R2527 on the 7633?

I can’t find a “thermal balance" on the 7623 schematic.

Hi David,

A little over a year ago I completed my Vacuum Tube Curve Tracer adapter which worked with ALL Tek curve tracers. After seeing the beautiful job Glydeck did on his simple triode curve tracer adapter for a 576 I decided I must make one myself if I ever got a 576.


Then one day a 576 fell into my lap for $25 and I set to work. My goal gradually evolved past Glydeck's. I wanted to test all kinds of tubes, not just a few triodes. With great advice from George Lydecker (Glydeck) I came up with a low cost solution for an easy to make adapter that, in conjunction with an inexpensive tube tester, would test any tube on any Tek curve tracer.

My goal was to make something anyone with a curve tracer could build at the absolute lowest cost that would test the greatest number of tubes. It had to work on any Tek curve tracer without modification (575, 576, 577, 7CT1N, 5CT1N) and perform most of the tests the original Tek 570 tube tester could do. Those were pretty aggressive goals but I met most of them and wrote about how I did it in a very detailed 29 page paper complete with parts lists and schematics. It may help guide you in your design. You can download a copy from:


To my amazement I almost immediately got requests from 75+ members of Tekscopes for the PC Board I made for the VTCT Adapter. I gave a talk on the adapter at SeaPac last June and I sold 10 more there. I have sold about 105 of them all total by now including a few completely assembled units and a few kits that include all the parts.

In between everything else I am doing I'm investigating what it would take to design a new version that would be capable of testing tubes under the exact conditions in a Williamson amplifier and its derivatives that many audiophiles will tell you is the ultimate in tube amplifiers.


Dennis Tillman W7PF

Hello Albert,

- You used internal triggering. Yes, but not triggering by the S4 (with I previously thought and couldn't understand)
but by the S-52 rear pretrigger.

Bingo! The idea was to show the S-52's signal into the S-4 so having the S-52 in the same mainframe with the other stuff made that simple.

- The S11/S4 is set for Smooth. That explains the smooth curve.

Yes and not really: Yes, the 7S11 in the 7854/7T11A/7S11/S4 setup was - inadvertently - set to smooth, and no, what you're seeing is the averaged digital result of 100 samples. You can see it from the readout info and in
the photograph "7854_7T11A_S-4 CRT.jpg". In my experience, smooth won't result in as smooth a trace as that by far.

In the 7904/7S12/S-6, the setting was High Resolution to get the smooth trace.

Raymond

Albert,

Good to know that GPIB doesn't work when it has hung. What's our best hypothesis for why it hangs?

-Kurt

On Mon, May 14, 2018 at 12:13 pm, Fabio Trevisan wrote:

When using "A sweep" mode only, especially disappointing on the last sweep
speed (50ns/div), the trace fades-in very slowly, and takes almost 2 divisions
to reach full intensity.

I have two 468's they do exactly the same thing at the fastest sweeps. Takes over 2 divisions before I have a trace.

I would be interested in buying 2 extenders (parts kits) if they become available again. Happy to commit with a deposit.

On Sun, May 13, 2018 at 12:58 pm, Albert Otten wrote:

On Sun, May 13, 2018 at 10:33 am, Kurt Rosenfeld wrote:

Since it sounds like there is a problem with your front panel, I recommend
running the diagnostics via RS-232 instead of via the front panel.

I doubt that communication via RS-232 or GPIB would be possible now that the
system "hangs" after the error messages at start-up.

Hi Kurt,
I tried GPIB to no avail.
Albert

On Mon, May 14, 2018 at 10:47 am, Harvey White wrote:

On Mon, 14 May 2018 06:11:27 -0700, you wrote:

I have acquired a TM506 with 7 plug ins PG506,TG501,DC503A,PG502,SG503,SG504

and DM502. All the units seem to work except the PG502. Looking for some
guidance as to the likely failure points of this module from the much more
experienced members. The TM506 had a fan that had just exploded and was
locked up. I replaced the fan and found that the PG502 was not powering on,
the power light on the panel did not come on. I found that It had a blown
F645 1A sub-miniature fuse on the board, I replaced that fuse and now when I
plug it in and turn on the power, it blows the 7A fuse on the board of the
TM506. Funny thing is that the little sub miniature that I replaced does not
blow when this happens. I immediately stopped inserting the module, in fear
that it might cause some other damage to the TM506. Anyone have any ideas? I
do not want to keep poking around without getting some ideas from the forum.
Thank You in advance!

I remember that TEK does not recommend hot insertion of modules in
equipment. Part of the problem is that if the connectors are not
aligned properly from the very start, adjacent pins can short at the
very beginning of the process, yet be correctly aligned when the
plugin is fully and properly inserted.

With no plugins, if the TM506 main fuse blows, then there's a nice
short somewhere. Obviously, if the fuse doesn't blow with no plugins,
and then blows once a plugin is inserted and power is applied, I'd be
looking at that plugin. A good set of extenders can be a very nice
addition.

As far as the PG502 is concerned, I'd check the capacitors down line
from the fuse. I'd also check the major capacitors in the PG502 just
as a matter of course, if you have an ESR checker.

Harvey

Harvey,

You are correct, TEK does not recommend "Hot swapping". I do not do hot swaps. The short is definitely in the PG502 as it blew that fuse on the TM506 as soon as power came on. With a full complement of Six functional plugins, (Excluding the PG502) there is no problem with the fuse. I am confident that the TM506 is good to go. Remove PG502- replace fuse-all is well. Others have also focused on the caps as well, so that will be a good place to start. Appreciate your insight and feedback.

On Mon, May 14, 2018 at 08:28:10PM +0100, Craig Sawyers wrote:

For the record, I have a 576, 577 and a 7ct1n (which is the one that seems to get used the most

since

I'm mostly checking small signal transistors).

I have all those (although the uTracer, though working, is awaiting a hardware implementation)

Plus a 575mod122C, and another 575 plus 175 high current adaptor (which weighs even more than the
575!)

I have a bit of a thing about curve tracers ;-)

Craig

I try to draw the line on whether or not I can lift it without assistance :-)

Paul

--
Paul Amaranth, GCIH | Rochester MI, USA
Aurora Group, Inc. | Security, Systems & Software
paul@... | Unix & Windows

Based is probably a little to strong of a suggestion.

Definitely siblings, though.

Both were developed in the same time frame,
both were designed to host a fiber optic readout system.
Both are an early adoption of a mix of semiconductor
and IC circuitry. Both fit in about the same footprint.
Both had flat, rectangular, internally ruled graticules,
on rectangular ceramic funnel section CRTs. Both used
epoxy potted EHT transformers in the anode section.
Both have convection cooled linear regulators plating the
back of the instrument.

Put a 7454 and a 576 side by side, and the familial
resemblance is hard to dismiss.

If Gilbert hadn't developed his superior on screen readout
system, The 7000 series would have been a carbon copy of
what you see in the 576.

-Chuck Harris


zenith5106 wrote:

If you're interested primarily in tubes, you should check out the uTracer which is a very clever
implementation using pulsed power supplies and a computer for the analysis/display. Google

uTracer

and you'll get a zillion hits but the home URL is

www.dos4ever.com/uTracer3/uTracer3.html

I do very little with tubes, but I was very impressed with the implementation.
He has a kit available too, which makes construction easy.

Wow, I just checked the web site and there's now a Linux version of the GUI.
The latest version goes up to 400V as well. Extremely impressive. I'm tempted to get a kit just

for the

heck of it.

For the record, I have a 576, 577 and a 7ct1n (which is the one that seems to get used the most

since

I'm mostly checking small signal transistors).

I have all those (although the uTracer, though working, is awaiting a hardware implementation)

Plus a 575mod122C, and another 575 plus 175 high current adaptor (which weighs even more than the
575!)

I have a bit of a thing about curve tracers ;-)

Craig

If you're interested primarily in tubes, you should check out the uTracer
which is a very clever implementation using pulsed power supplies and a
computer for the analysis/display. Google uTracer and you'll get a
zillion hits but the home URL is

www.dos4ever.com/uTracer3/uTracer3.html

I do very little with tubes, but I was very impressed with the implementation.
He has a kit available too, which makes construction easy.

Wow, I just checked the web site and there's now a Linux version of the GUI.
The latest version goes up to 400V as well. Extremely impressive. I'm
tempted to get a kit just for the heck of it.

For the record, I have a 576, 577 and a 7ct1n (which is the one that seems to
get used the most since I'm mostly checking small signal transistors).

Paul

--
Paul Amaranth, GCIH | Rochester MI, USA
Aurora Group, Inc. | Security, Systems & Software
paul@... | Unix & Windows

I just sold a 576 for $750 and I think, for a fully
working unit, it was a buyer bargain. The owner just wanted it gone....

My 576 cost me - precisely zero! Known to have the epoxy HT transformer problem. I rewound it from the
specification sheet, and potted it in wax. Works absolutely perfectly.

Craig

On Mon, May 14, 2018 at 10:58 am, Raymond Domp Frank wrote:

[Dennis:]
You didn't say what your signal source was in any of these photos. Was it
the S-52: If so how did you get it to trigger.
I was aware that the 2nd pulse was taller and I tried to trigger on it but I
could only trigger on both simultaneously which
made for a very confusing waveform and a lot of knob twiddling to no avail.

I used two different S-52's, please refer to the new photographs. In the 7T11A
setup, I used internal triggering. Only with the trigger amp. on X1 I was able
to filter out the one waveform, strangely enough. I remember the same thing
from my previous experiments.

Hello Raymond,
Two things you didn't mention explicitly but follow from the 7854 photos:
- You used internal triggering. Yes, but not triggering by the S4 (with I previously thought and couldn't understand) but by the S-52 rear pretrigger.
- The S11/S4 is set for Smooth. That explains the smooth curve.
I think Dennis tried to trigger from the S-52 TD output (or internal S-4). In sequential mode that would be problematic since you have to trigger on another edge than the one you want to display. That in turn would require a large Range setting at the 7T11A.

Albert

Hello guys,
On my final attempts doing an overall wrap-up to my 464, to eventually put it for sale, it always annoyed me the disappointing performance (to my taste at least) of the Z axis intensity at the fastest sweep speeds.
When using "A Intensified by B" modes on the three or four last sweep speeds, the transition from normal intensity to intensified is rather sluggish.
When using "A sweep" mode only, especially disappointing on the last sweep speed (50ns/div), the trace fades-in very slowly, and takes almost 2 divisions to reach full intensity.
I took some pictures and upload them here:
/g/TekScopes/album?id=49127
Can someone of the group, who has a 464 like mine, or even others of the same family, like 465, or 466, take a look at the waveform coming out of the Z-Axis amplifier and coming into the Z-Axis DC restorer (on the 464 this is TP1443) and let me know if this rather slow rising edge is normal, or is it some problem with my 464's Z-Axis amplifier?
On the pictures, you will see the trace of the 464 itself, with A sweep at 200ns/div, intensified by B sweep at 50ns/div, and the Z-axis amplifier output, showing the waveforms, first without intensified and then with intensified portion.
The transition of the 25V step is about 80~90nS (from 10% to 90%) and the transition of the 8V of the intensified step is about 100ns.
1. The wave-forms shown are taken already with the best possible adjustment of the frequency compensation trimmer capacitor... but even if I try to push it harder, this trimmer doesn't really help to improve the rising edge of the waveform, it only makes the falling edge a little bit faster, and adds an undershoot, which creates a dip in trace intensity just after the end of the intensified portion (which is no good either).
2. I already tried to remove the upper transistor of the amplifier (the active load, or current source if you like) and performance degrades noticeably, giving insight that the active load is playing its role.
I also tested the transistor off-circuit (on an AVR transistor tester) and it measures well.
3. Looking at the signal incoming to the amplifier, at the emitter of Q1424 is not conclusive, as this is a current-driven input so, I wasn't able to tell if the slow rising transition is due to the input signal or low performance of the amplifier.
I appreciate any comment.
KRgrds,
Fabio