On Thu, 17 Nov 2016 20:31:16 -0200, you wrote:

...

Sadly, it didn't last too much and in the next few minutes, even the modest
brightness improvement apparently vanished away .

My conclusion is that in this CRT, the getter is probably already saturated
and not capable of absorbing any material that I eventually manage to strip
away from the cathode...and it ends up deposited back in the cathode.

Still, as I mentioned at the beginning of all this story, it's not THAT BAD.
Especially talking about brightness... It's usable...
And taking in account it's a storage CRT, the variable persistence mode can
(still) make-up for the weakened traces at high sweep rates with delayed
time base.

The only thing that really annoys me is that the trace doesn't focus all
that well... I can get it quite sharp on vertical oriented traces, and
almost as much on the horizontal traces, but never the best in both
directions... no matter how much I fiddle with Focus and Astigmatism
controls...

The 464 is the only other analog storage oscilloscope I have used and
it was neither bright nor sharp. The one I had did not display double
peaking but it did perform like you describe compared to the sharpness
and brightness of an analog non-storage oscilloscope so it seems to me
that your CRT rejuvenation was a success if the double peaking has
been reduced.

I thought the 7834 I got a couple years ago was bright and sharp until
I got a 7904. Now the 7834 looks dim but still useable under normal
room lighting which of course it should.

I have had one fail out of many scopes. Qservice has a new replacement.

On 11/18/2016 6:52 PM, 'gmail' guidozonderspam@gmail.com [TekScopes] wrote:

Hi all, Albert, thanks for the input.

Had to figure out where i stopped last time i worked on the unit.
Turned out that CR515 was completely missing... I collected BA280 and
MBD101 SMD diodes in the past. Tried both connected to the 2ns output
and the BA280 looked most promissing, so i put that one in. Then after
some tweaking i now have 1ns out :)

It is only about 120mV

How are you measuring this? with the 7T11?

but i have no idea if the very good looking cable is really that good
(RG142). It is also not completely stable, the whole sine jumps up and
down a bit. Not sure if it is the TG501 or the sampling unit, but at
least there is an output.

Make sure the out put is terminated in 50 ohms, preferrably at the
output of the TG501 with one of those pass through BNC
terminations..lack of termination can result in all kinds of weird
ringing at this frequency (it is easy to forget that 1ns is really a
1GHz sine in this case. Use the trigger out from the TG501 to trigger
the scope and see if that helps

The offset out voltage of the 7T11 shows a lot of peaks (540mV
peak-peak at 28kHz, so 35.7us apart). Zoom in and you see a dying out
23MHz sine for some 1.2us. My initial thought would be that the 7704A
switching power supply causes this. It is all old stuff :) But it also
seems to matter which TM i use for the TG502. One gives a more "jumpy"
result than number two. Time for a recap.

The TG503 2ns is a whopping 1.5V peak to peak. Not the nicest sine but
the higest amplitude to get most out of 1ns. 5ns is 1V and looks ok.
The 2465 sees also 1V with 5ns, 0.8V with 2ns and a few mV at 1ns. Not
much, but it is a sine at one GHz :) Triggering of channel two, x10
mag on.

Again make sure you are terminated in 50 ohms this will clean things up
if your not. Also you should do all the tuning of the 5/2/1ns sections
terminated this can effect some of the tuning and the tuning interacts
so you have to go through it more than once

I don't have any other source or scope in the voodoo region, but it is
good to have at least something to work with if needed. From what i
read getting a trace on the 7S11/7T11 is already an achievement on its
own.

An RF voltmeter (Boonton or HP) is good at this frequency especially
for tuning the caps in the RF filters in the 5-2-1 sections

Cheers
Guido

--
Dave
Manuals@ArtekManuals.com
www.ArtekManuals.com

On Fri, 18 Nov 2016 at 18:18 thtraynor@att.net [TekScopes] <
TekScopes@yahoogroups.com> wrote:

I just traced a no turn-on problem with my 2465 to the main AC power
switch, S350. One of the two poles is not connecting. Is this a common
problem? Are these switches readily available, or should I just jump out
the open pole and button it up?

Hey Tom,

my 2465 had the same problem, and I found a replacement readily enough on
Digikey. I could even fit the plastic cover from the old one over the
replacement. Sadly I can't give you a part number, this is a few years ago.

Siggi

Hi all, Albert, thanks for the input.

Had to figure out where i stopped last time i worked on the unit. Turned out that CR515 was completely missing... I collected BA280 and MBD101 SMD diodes in the past. Tried both connected to the 2ns output and the BA280 looked most promissing, so i put that one in. Then after some tweaking i now have 1ns out :)

It is only about 120mV but i have no idea if the very good looking cable is really that good (RG142). It is also not completely stable, the whole sine jumps up and down a bit. Not sure if it is the TG501 or the sampling unit, but at least there is an output.

The offset out voltage of the 7T11 shows a lot of peaks (540mV peak-peak at 28kHz, so 35.7us apart). Zoom in and you see a dying out 23MHz sine for some 1.2us. My initial thought would be that the 7704A switching power supply causes this. It is all old stuff :) But it also seems to matter which TM i use for the TG502. One gives a more "jumpy" result than number two. Time for a recap.

The TG503 2ns is a whopping 1.5V peak to peak. Not the nicest sine but the higest amplitude to get most out of 1ns. 5ns is 1V and looks ok.
The 2465 sees also 1V with 5ns, 0.8V with 2ns and a few mV at 1ns. Not much, but it is a sine at one GHz :) Triggering of channel two, x10 mag on.

I don't have any other source or scope in the voodoo region, but it is good to have at least something to work with if needed. From what i read getting a trace on the 7S11/7T11 is already an achievement on its own.

Cheers
Guido

I'm guessing a 545B based on the two oval serial number plates above the
screen. Here's a screen cap so you can see what I mean:

http://i.imgur.com/7zsymWJ.png

And here's a 545B:

http://www.thevalvepage.com/testeq/tek/545b/front.jpg

-Bob


On Thu, Nov 17, 2016 at 11:36 PM, goodappl@comcast.net [TekScopes] <
TekScopes@yahoogroups.com> wrote:

I recently heard about a new movie to be released in early January, 2017,
called "Hidden Figures". It's based on the true story of 3 African American
women who were engineers/scientists in the then new 60's NASA space
program. About halfway through the official 2.3 minute trailer (
https://www.youtube.com/watch?v=RK8xHq6dfAo ) there is an excellent shot
of a vintage TEK scope - looks like probably a 535 or 545 - on a cart,
followed by a close-up of the screen clearly displaying a pulse waveform.
Was thinking it would be a good movie to see, even before I came across
that cool scene!
Tom

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]

I just traced a no turn-on problem with my 2465 to the main AC power switch, S350. One of the two poles is not connecting. Is this a common problem? Are these switches readily available, or should I just jump out the open pole and button it up?

Thanks,

Tom

On Fri, 18 Nov 2016 12:16:07 -0500, you wrote:

At 09:57 AM 11/18/2016, Merchison Burke wrote:

And also re-lubing the switches and detents which were originally lubed
as suggested in the Tektronix article "Tektronix products get dirty too"
published in TekScope, Volume 8, Number 4, 1976 .

I do not see that issue online anywhere, and it is not at TekWiki. If anyone can point to a scan online, or is willing to scan it for TekWiki, I would appreciate it.

Dale H. Cook, GR / HP Collector, Roanoke/Lynchburg, VA
http://plymouthcolony.net/starcity/radios/index.html

I do not have that issue either.

But TEKscope 1972 Volume 4 Number 4 July has an article at the end
titled Washing Your Tektronix Instrument.

http://w140.com/tekwiki/images/f/f9/Tekscope_1972_V4_N4_Jul_1972.pdf

In case this gets missed, like I said in my other post, in retrospect
I think I could have done the removal safely with solder wick and my
good temperature controlled iron with a big low temperature tip. On
the other hand, the hacked together vacuum desoldering tool has served
me well since then for replacing or adding DIP sockets in other
Tektronix instruments so the investment was worth it anyway.

Had the non-volatile SRAMs I bought not worked in place of the
original Dallas memories, I was considering reverting the board to the
earlier design which used the external battery and low power SRAMs.

On Thu, 17 Nov 2016 22:52:43 -0700, you wrote:

...

It also depends on the board, it sounds like that junked PCB has good
holes drilled for the chips, that makes things much easier.  If the
through holes are just barely large enough for the pin especially due
to the plating-through, it becomes very painful to desolder (capillary
action sucks the solder back into the hole...).

The junked personal computer motherboard that I practiced on with my
hacked vacuum desoldering tool was *much* more difficult to work with
for the reason you identified and more. The plated through holes were
just large enough for the pins and the massive copper power planes
made it difficult to apply enough heat.

Removing the Dallas non-volatile memory was trivial by comparison and
in retrospect, I think I could have done it easily and safely with
just solder wick and my temperature controlled iron.

Incidentally, I have a dual Pentium Pro motherboard with a Benchmarq
RTC chip not unlike the Dallas Semiconductor chips (probably a
clone...).  This board has the holes drilled so small that it's
virtually impossible to suck the solder out of the through holes from
RTC chip.  Its battery is dead too, hence the desire to replace it.  I
tried to cut the top off the chip and solder another battery, but found
it leaks so much current that any new batteries would need to be
replaced at least once a month.

Probably time to pitch this board.

I had several P55 type motherboards that died once the embedded
lithium cells in the non-volatile SRAM/clock modules discharged. They
would not even start the POST (power on self test) presumably because
of the dead battery. Replacing the SRAM/clock modules was just too
difficult to do without wrecking the boards.

Some correction to my last post,
The original IBM PC/AT battery backup circuit worked with 6V battery (and
external 6V lithium battery pack, about the size of an AA battery).
The IBM PC/AT Technical Reference can be found here:

http://www.minuszerodegrees.net/manuals/IBM_5170_Technical_Reference_1502243_MAR84.pdf
Schematics of the battery backup circuit is at page 1-78 System board.

My memory should have tricked me because the factory I worked at, was
already using a modified version of this circuit, but to use one 3V lithium
battery, that I don't remember the details anymore.
Since it was a local, Brazilian IBM/PC clone factory, I don't think I will
ever find it on the net.

But, basically, the function which the original Dallas chip performed is
exactly that... And emulate it using a plain CMOS SRAM chip and one (or a
couple) of 3V lithium battery + a few discrete components is easily done.


2016-11-18 15:28 GMT-02:00 Fabio Trevisan <fabio.tr3visan@gmail.com>:

I was working on the Personal Computer manufacturing business back when
those first Dallas chips showed up on the market.
They appeared as a more practical solution to replace the, back then in
use, battery backup circuit that was used on all IBM/ATs and their clones,
which used for the first time then, an RTC (Real Time Clock) chip, which
was one that contained a Clock / Calendar circuit + 68 bytes of Static CMOS
Ram used to store the parameters that ended up being known to everybody, by
the name of the memory that was used to hold them... the "CMOS settings".
In its original version, from IBM AT, this chip was fed by a small 2
transistor circuit, which would feed the CMOS RTC chip, taking supply from
2 sources... from the Computer's 5V supply, and from a 3V lithium battery.
The circuit would detect when the 5V supply would just start to fall below
4V (capitalizing on the fact that TTL logic needs quite a strict 5V
+/-0.25V supply), then disconnect the CMOS chip from the 5V supply and
connect it to the 3V from the battery.
The original circuit used bipolar transistors, and would rely on that the
CMOS chip would still hold its function with some 0.2V below 3V (while 3V
was the minimum by the datasheet)

I`m going to try to find that original schematic somewhere and post it to
Photos area.


2016-11-18 14:41 GMT-02:00 szigszabolcs@yahoo.com [TekScopes] <
TekScopes@yahoogroups.com>:

Hi,

I did recently replaced these SRAMs in my 2440. It wasn't that of a scary
job. They had 1993 datecode and one of them already failed.

First I removed the whole CPU board, it's simple and makes it much
comfortable to work on it. I used solder sucker to clear the holes for
every pin. Make sure you heat the pins well, so that the solder melts all
the way trough the hole. If necessary, add more solder first!
Then I used solder wick to remove the rest of the solder. Then, one by
one on every pin, I used a small screwdriver, while heating the pin to move
it away from the wall of the hole, and make sure that it can move freely
and is not stuck to the PCB. This wasn't a quick job, but this ensures that
you won't accidentally pull out the plating from the hole. Once done, the
RAM practically falls out by itself.

All went without any problem. I put in high quality sockets and new RAMs.
(Amazing I paid nearly the same amount for the new RAMs than for the whole
scope - it had had a failed PSU, so got it cheap.)

Then did a full calibration according to the service manual, but it looks
like simply doing a self cal and an ext cal is sufficient, unless other
parts were changed in the scope.

All in all, it took maybe 2-3 hours to fix, and hopefully it'll keep
working for the next 20 years.

I'm planning to cut open the old RAMs to see if I can connect a new
battery to it. I have successfully done it with a similar RAM in an old Sun
workstation, where I was unable to source replacement.

Szabolcs

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]

I was working on the Personal Computer manufacturing business back when
those first Dallas chips showed up on the market.
They appeared as a more practical solution to replace the, back then in
use, battery backup circuit that was used on all IBM/ATs and their clones,
which used for the first time then, an RTC (Real Time Clock) chip, which
was one that contained a Clock / Calendar circuit + 68 bytes of Static CMOS
Ram used to store the parameters that ended up being known to everybody, by
the name of the memory that was used to hold them... the "CMOS settings".
In its original version, from IBM AT, this chip was fed by a small 2
transistor circuit, which would feed the CMOS RTC chip, taking supply from
2 sources... from the Computer's 5V supply, and from a 3V lithium battery.
The circuit would detect when the 5V supply would just start to fall below
4V (capitalizing on the fact that TTL logic needs quite a strict 5V
+/-0.25V supply), then disconnect the CMOS chip from the 5V supply and
connect it to the 3V from the battery.
The original circuit used bipolar transistors, and would rely on that the
CMOS chip would still hold its function with some 0.2V below 3V (while 3V
was the minimum by the datasheet)

I`m going to try to find that original schematic somewhere and post it to
Photos area.


2016-11-18 14:41 GMT-02:00 szigszabolcs@yahoo.com [TekScopes] <
TekScopes@yahoogroups.com>:

Hi,

I did recently replaced these SRAMs in my 2440. It wasn't that of a scary
job. They had 1993 datecode and one of them already failed.

First I removed the whole CPU board, it's simple and makes it much
comfortable to work on it. I used solder sucker to clear the holes for
every pin. Make sure you heat the pins well, so that the solder melts all
the way trough the hole. If necessary, add more solder first!
Then I used solder wick to remove the rest of the solder. Then, one by one
on every pin, I used a small screwdriver, while heating the pin to move it
away from the wall of the hole, and make sure that it can move freely and
is not stuck to the PCB. This wasn't a quick job, but this ensures that you
won't accidentally pull out the plating from the hole. Once done, the RAM
practically falls out by itself.

All went without any problem. I put in high quality sockets and new RAMs.
(Amazing I paid nearly the same amount for the new RAMs than for the whole
scope - it had had a failed PSU, so got it cheap.)

Then did a full calibration according to the service manual, but it looks
like simply doing a self cal and an ext cal is sufficient, unless other
parts were changed in the scope.

All in all, it took maybe 2-3 hours to fix, and hopefully it'll keep
working for the next 20 years.

I'm planning to cut open the old RAMs to see if I can connect a new
battery to it. I have successfully done it with a similar RAM in an old Sun
workstation, where I was unable to source replacement.

Szabolcs

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]

At 09:57 AM 11/18/2016, Merchison Burke wrote:

And also re-lubing the switches and detents which were originally lubed
as suggested in the Tektronix article "Tektronix products get dirty too"
published in TekScope, Volume 8, Number 4, 1976 .

I do not see that issue online anywhere, and it is not at TekWiki. If anyone can point to a scan online, or is willing to scan it for TekWiki, I would appreciate it.

Dale H. Cook, GR / HP Collector, Roanoke/Lynchburg, VA
http://plymouthcolony.net/starcity/radios/index.html

Hi,

I did recently replaced these SRAMs in my 2440. It wasn't that of a scary job. They had 1993 datecode and one of them already failed.

First I removed the whole CPU board, it's simple and makes it much comfortable to work on it. I used solder sucker to clear the holes for every pin. Make sure you heat the pins well, so that the solder melts all the way trough the hole. If necessary, add more solder first!
Then I used solder wick to remove the rest of the solder. Then, one by one on every pin, I used a small screwdriver, while heating the pin to move it away from the wall of the hole, and make sure that it can move freely and is not stuck to the PCB. This wasn't a quick job, but this ensures that you won't accidentally pull out the plating from the hole. Once done, the RAM practically falls out by itself.

All went without any problem. I put in high quality sockets and new RAMs. (Amazing I paid nearly the same amount for the new RAMs than for the whole scope - it had had a failed PSU, so got it cheap.)

Then did a full calibration according to the service manual, but it looks like simply doing a self cal and an ext cal is sufficient, unless other parts were changed in the scope.

All in all, it took maybe 2-3 hours to fix, and hopefully it'll keep working for the next 20 years.

I'm planning to cut open the old RAMs to see if I can connect a new battery to it. I have successfully done it with a similar RAM in an old Sun workstation, where I was unable to source replacement.

Szabolcs

Somebody posted this one Monday, sorry, don't remember who. It tells of
washing the scope, etc.

http://w140.com/tekwiki/images/f/f9/Tekscope_1972_V4_N4_Jul_1972.pdf

Vince.


On Fri, November 18, 2016 10:36 am, Mike mdinolfo@erols.com [TekScopes]
wrote:

Merchison:

Is this article available on-line? I looked at the Tekscopes magazines
that are linked from www.w140.com, and while there are quite a few
Tekscopes magazines/editions linked there, the Vol 8 Number 4 1976
magazine is not.

Thanks!
Mike Dinolfo

On 11/18/2016 09:57 AM, Merchison Burke merchison@yahoo.co.uk
[TekScopes] wrote:

And also re-lubing the switches and detents which were originally lubed
as suggested in the Tektronix article "Tektronix products get dirty too"
published in TekScope, Volume 8, Number 4, 1976 .

On 2016-Nov-18 7:56 AM, Barry n4buq@knology.net [TekScopes] wrote:

If you washed the insides too, you might want to consider a final

rinse with distilled water on the circuits, switches, etc.

Regards,
Barry - N4BUQ

----- Original Message -----

From: "Don Lewis dlewis11193@yahoo.com [TekScopes]"

<TekScopes@yahoogroups.com>

To: TekScopes@yahoogroups.com
Sent: Friday, November 18, 2016 6:25:06 AM
Subject: [TekScopes] TEK 465M Saved


I found an old Tek 465M, 100MHz scope in a thrift shop for $35.00. I

thought

it was overpriced for the shape it was in, ..so dirty it was hard to

see the

front panel markings. I took the plastic cover off, doused it with

detergent

and soaped it thoroughly with a pressure hose in the back yard. It

now looks

brand new, ...one would have thought it was never used by the looks

of it.

I'll let it dry for a month and see what we've got. Will report back
how
the start-up goes. DonN5CID

------------------------------------

------------------------------------


------------------------------------

Yahoo Groups Links

--
Michigan VHF Corp. http://www.nobucks.net/ http://www.CDupe.com/
http://www.foggymist.com The Foggy Mist Emporium

Merchison:

Is this article available on-line? I looked at the Tekscopes magazines that are linked from www.w140.com, and while there are quite a few Tekscopes magazines/editions linked there, the Vol 8 Number 4 1976 magazine is not.

Thanks!
Mike Dinolfo

Here's a link to the second before it appears <
https://youtu.be/RK8xHq6dfAo?t=69> ... for two seconds or so.

On Fri, 18 Nov 2016 at 10:08 'Dennis Tillman' dennis@ridesoft.com
[TekScopes] <TekScopes@yahoogroups.com> wrote:

Look fast, its gone in an instant.
Dennis Tillman W7PF

I've been following this TM504 thread, and in my experience with 3 of these units, an alarm goes off in my head if any plugins mis-function. There are a lot of different plugin design styles, each using different resources in the mainframe. But those plugins that rely on the mainframe NPN/PNP mounted power transistors are very likely to be damaged by the presence of a shorted pass BJT. When a plugin stops working, I first check that the pass BJTs are ok.
I had one TM504 with a shorted NPN in one slot. When I moved a DM501 over into that slot I released great quantities of smoke. The DM501 uses the NPN with a LM723 to make 5v. When the NPN is shorted, (and they always seem to fail that way!), you get ~33v VCC and barbecued TTL. I gave up trying to fix the DM501 and got another 502.
Moral: the TM50x are not passive boxes. Make sure they are not the source of the problem, or you may propagate faults.

BTW, the original Motorola BJTs in the mainframe are now unobtanium. The MJE3055/2955 NPN/PNP pairs and their variants are an upgrade replacement. 10A, 60V, 90W 2Mhz ft TO126/225 style package.

Sent from my iPad

Look fast, its gone in an instant.
Dennis Tillman W7PF

And also re-lubing the switches and detents which were originally lubed
as suggested in the Tektronix article "Tektronix products get dirty too"
published in TekScope, Volume 8, Number 4, 1976 .