Date   

Re: Tek 475: square wave leading corners too rounded?

 

On Sun, 17 Apr 2016 00:46:02 +0100, you wrote:

On 16/04/2016 05:48, David davidwhess@gmail.com [TekScopes] wrote:

I scrounged a BNC-T and 50 ohm terminator and improved my Arduino-based
square-wave generator to emit higher frequencies with less jitter.

For reference, the Gratten GA1042CAL with 250khz input via the BNC-T &
50ohm terminator: https://goo.gl/photos/Jd7Mn88dpd3ZhkRn6

The 475 on Channel 1 at .5 volts/div:
https://goo.gl/photos/xgeeqQ8j7RKHNq1c8

The 475 on Channel 2 at .5 volts/div:
https://goo.gl/photos/P6G4vp6kSqA4qf6Z7

The tilt on Channel 2 varies with attenuator setting. At .2 volts/div
it looks more like Channel 1.

Am I right in thinking this suggests vertical attenuator problems in
both channels, likely either due to dirty/oxidised leaf switch contacts
and/or cold solder joints in the attenuator blocks?
Yes, there is something going on with the attenuators but I doubt the
problem is with the switches or solder joints. An open circuit
usually causes a very high pass filter characteristic but at 250 kHz
that might be difficult to see. It would be helpful to lower the
square wave frequency a lot, like to 1 kHz, so that the DC response is
more evident.

Besides using a lower frequency square wave, the same square wave test
needs to be done at 5mV/div where none of the attenuator sections are
in the circuit. That will verify if the problem is only in the
attenuators. Knocking the 5 volt output from your Arduino down to 20
millivolts can be done with a couple of resistors to make a low
impedance attenuator at the Arduino.
I have done so. The squarewave frequency is now 1.25khz. The crudeness
of my ad-hoc Arduino-based squarewave generator is such that it isn't
(yet) easy to dial up an exact desired frequency; hopefully it's close
enough.

For reference, this is the output of the frequency generator on the
Gratten GA1042CAL: https://goo.gl/photos/ehgMShwBNJADPqfw7
Good.

This is Channel 1 on the 475, AC coupled:
https://goo.gl/photos/KBMdeGE8tjfVs72u5

This is Channel 2 on the 475, AC coupled:
https://goo.gl/photos/bQ35QgFkUjpwYd9W6
This is Channel 1 on the 475, DC coupled:
https://goo.gl/photos/mfHQWHazjLktXQKp8

This is Channel 2 on the 475, DC coupled:
https://goo.gl/photos/dapEDZRzWcyNZ7F68
Ok, so the tilt we see here is just the AC coupling which is normal.

All are shown with BW limit turned on (set to 20mhz on the 475; on the
Gratten it's just on/off), including on the Gratten, to limit noise
emitted by the Arduino that otherwise makes the traces hairy.

Note that the slow rise of the leading edge of the squarewave plateau is
still present. This is the 475 on Channel 2 at 5 microseconds per
division, demonstrating roughly 20 to 25 microseconds to reach the
nominal plateau: https://goo.gl/photos/oxUYm3ETvwcVJBxBA

I believe that slow rise accounts for the obviously rounded leading edge
of the square wave when displayed at .2 milliseconds per division.
And this is a problem.

Viewing the output of the Channel 2 vertical amplifer on the Gratten --
via the BNC connector on the back panel of the 475 -- demonstrates what
appears to be the same slow rise, though it's perhaps not as obvious due
to the graininess of the Gratten display. I presume this localises the
problem to the vertical amplifier rather than the vertical deflection
circuitry?
Unfortunately unlike some other designs, the vertical amplifier
circuitry in this case includes the vertical channel switch which is
what produces the channel 2 vertical output. That might explain why
we see the same "dribble up" on channel 1.

I think on the 465 the high impedance x10 and x100 attenuators are in
little hybrid modules which are installed into collet socket pins. You
can pull them out and reseat them to check that the socket pins are
making a good connection.

I did have one of these hybrid modules with a bad connection inside
where the pin attached. I touched it up with silver solder and it
worked as good as new.
On the 475, the attenuators are hybrid plug-in modules.

I'm not sure what to tackle next. Any suggestions?
I think you did rule out a problem with the vertical CRT deflection
amplifier.

Double check that the "dribble up" occurs on both channels. If so, I
think either they were both miscalibrated in the same way or there is
something going on with vertical channel switch U370 shown on
schematic 3.

I would be a little suspicious of C403. Can you reseat U370? Can you
replace U350? The 7474 IC is pretty common in parts bins.

Also double check that the "dribble up" is not an artifact from your
pulse generator using your Gratten. Examine the falling edge closely
to see if it displays the same thing using both oscilloscopes.

Does the "dribble up" change as the vertical position of the trace is
moved?

Note that "dribble up" refers to an actual transmission line
phenomenon at high frequencies which should not be present at 20 MHz
(or 200 MHz) and below. Your measurements using your Gratten rule out
this problem with the test coaxial cable you are using.


Re: PG-502 blowing fuses - PS repair

 

On 16 Apr 2016 03:03:14 -0700, you wrote:

But doesn't U640 require +12V to be there in the first place ?
It does because +12 volts to pin 12 powers the control circuits in
U640 which is a 723 voltage regulator. But +12 volts is independent
of what the +15 volt regulator is doing. It only depends on fuse
F600.

And when I provide the +5V I'll have to enable the +33.5 V branch with the consequence that +15V is going up also which in turn can have other consequences e.g. on the thyristor circuit.
The +15 volt regulator can be deactivated by removing Q600.

That's why I was asking to start the repair "outside" the frame by which I didn't mean an extender but *really" outside.and measure certain branches by supplying the required voltages externally.
What I do is use one of my power supply mainframes, a TM501 but the
others work for this, with the covers removed so I can access the
sides of the plug-in which also has its covers removed.


2430 fix repair or parts

Cuc Tu
 

I think this one is not a basket case since it will power on, pass POST, and ran the self cal. OK.


It was not used for a few years, but worked fine. It would first lock during POST, then it would complete POST with different failures (WP, CCD, PA), finally it passed all self tests and completed a self cal, but the display goes off after a minute of being on. I have only given it a handful of power cycles to get through this. It will not sweep and just shows a static, full scale repeating sine wave burst. When the sweep time is lowered, I can see a sine wave at each end with the center flat (in case that is characteristic of something).


I have opened the case and found the Lithium battery, which measured 3.67V. It's quite clean inside and no corrosion. That is as far as I went into it. Next I will get to the PS and I plan to just recap it. I have some references to a 2440 with a parts list to check against.


Just wondering if there are any specific things to look for beyond the usual leaky/bulging caps? I did not find the U800 chip often discussed. Does this model use that chip?


Thanks for any guidance here.


Re: Tek 2467A X/Y issues

Siggi
 

Hey Jeff,

this, I think, is mostly working as intended. Possibly the CRT adjustment
procedure would allow the X/Y display to show a fraction brighter. I just
tried this on my 2467, and I get the same exact thing.
What you need to do is to (carefully) crank up the intensity after you
toggle to X/Y mode.
This is happening because the 2467s jealously guard against CRT burn-in,
and the CPU/MPU is simply turning down the beam current/intensity when you
click over to X/Y mode.

There's no way this has to do with the front panel switches, and whoever
indicated this might be the case has sold you a packet. The 2400-series is
all drive-by-wire, meaning that most all of the scope is controlled by the
MPU, which takes input from the front panel switches.
Incidentally, my 2467 had terrible glitching on the front panel switches,
but they cleaned up nicely by teasing some IPA in there and giving them a
good workout. I may also have given them a squirt or three of DeOxit D5 - I
don't remember.

BTW: Jeff's video link is here <https://youtu.be/b1BeuiPimmg?t=144>.

Siggi

On Mon, 18 Apr 2016 at 22:52 Huzzunga.Din@gmail.com [TekScopes] <
TekScopes@yahoogroups.com> wrote:



I have gone through all the boards with a cap upgrade, and there is a 100%
improvement in the scopes lowered intensity & sharper focus.


What started all this was the scopes inability to present, and then hold
an X/Y display. The scope would hold the 45 degree display for about 1/4 of
a second. It is about the same. I think I imagined it longer in the video.
The X/Y jumpers are in the outlined X Y locations.



I cleaned the Hybrid IC, and I am down to thinking the Sec/Div control.
While I had the scope apart for the A 1/ A 5 boards I separated the control
panel modules from the panel face. I took the Sec/Div control switch apart
noticed the inline switch that rides the Sec/Div cam like, had 3 terminals.
One terminal leg was cut off, and only the other 2 soldered in the board.
Likely an OFF/ON/OFF switch.



I have the Service Manual via download, waiting the arrival of the real
thing. It is a lot better than nothing, but not fully functional as I am
used to in these type of docs.


Because I was able to see the Sec/Div control to drive this in-line
switch, it seems mechanically that is okay, the tan plastic moves the
inline switch. This brings up a couple of questions in my mind.

1. Is it normal to see a 3 position switch with a lead cut of used in
original installations?

2. Is it normal to see a 3 position position switch with a lead cut off in
repair situations?


I think this switch is the heart of my X/Y problem. It is not holding ON
position, so my X/Y falls off. Makes sense to me, but I'm just a self
taught solder slinger.

If you want to see it happen, or have a laugh, there is a YouTube video on
my channel Huzzunga.


One more symptom I was unable to duplicate is when the switch is flipped,
and I am driving to stop the trace to display X/Y, after it does the
momentary X/Y thing sometimes turning the control back CW will flash all
the panel LED lights as opposed to begin driving the trace faster again.
If that symptom rings a bell for someone let me know.


I would rather buy an in-line switch, and replace it first, rather than
purchase a 70 something dollar used panel from eBay.


The symptoms again;
The scope displays X/Y momentarily, and sometimes after attempting an X/Y
display all the control panel LED lights will flash as the Sec/Div control
is turned Clock Wise.
If you have a part number on the Off/On/Off switch let me know. If it
sin't controls it starts getting harder to find. Or an IC.
Thank you for your time
Jeff


Re: 5441 High Voltage

Nathan Johnson
 

I'll let you know on that, pending some 300ma fuses to test, not something I seem to be able to get locally.
Nathan KK4REY

Sent using CloudMagic Email [https://cloudmagic.com/k/d/mailapp?ct=pi&;cv=7.4.15&pv=9.1&source=email_footer_2]

On Tue, Apr 19, 2016 at 16:14, kurt.harlem@gmail.com [TekScopes] <TekScopes@yahoogroups.com> wrote:
If you lift the base of Q410, does the fuse still blow?

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


Re: 5441 High Voltage

Kurt Rosenfeld
 

If you lift the base of Q410, does the fuse still blow?


Just Joined

Cuc Tu
 

Hi All,


I'm the proud owner of a 2430 and 496. Both were my grandfather's and I've had them since his passing about 7 years ago, but we both used them when we had occasion to do so. The 2430 now has a problem after not being used for a couple years. I'll make a new topic just for this attempted repair. Also, the 496 has an intermittent phase lock error. Me, I'm jut a hobbyist technician with some experience in RF and microwave.


5441 High Voltage

Nathan Johnson
 

Working on a 5441, blown HV fuse. Transistor is good, so I'm suspecting one
of the HV diodes(CR412) and the multiplier(U410).
There really isn't a good way to test these diodes is there? Pretty much
need to replace and try that? Anybody have a modern 12kv/10ma diode they
like?
For the multiplier, if it is toast, I should see a measurable short from
either the input or output to ground?
Thanks for any suggestions!
Nathan KK4REY
Sent using CloudMagic Email
[https://cloudmagic.com/k/d/mailapp?ct=pi&;cv=7.4.15&pv=9.1&source=email_footer_2]


Does anybody have the Tek hardware catalog that covers knobs?

 

Like for semiconductors, there was an internal Tek part number catalog for knobs, which I really need to try and correctly number the knobs on our site. knobs have no number on them, and many look similar, but have minor differences, making it very hard to identify some of these parts correctly. it is currently driving me crazy as I try and clean up our stock after the floor repairs and restoration (9 months of hell).

I have many other Tek internal parts catalogs, but not this one. I have also never seen the Tek CRT catalog, if it exists. I would really like to see that too.

any help greatly appreciated,
all the best,
walter (walter2 -at- sphere.bc.ca)
sphere research corp.


Re: FS: AM503 current probe amps

Artek Manuals <manuals@...>
 

On 4/19/2016 4:29 PM, jerry@hanler.com [TekScopes] wrote:
No, without the probe. I was told they worked but without a probe I
can't test them. They look to be in good shape.


I kicked around buying a probe but they are expensive.


Thanks Jerry

At $30 I susupected as much

I will pass

Dave
manuals@artekmanuals.com
--
Dave
Manuals@ArtekManuals.com
www.Arte
kManuals.com


Re: FS: AM503 current probe amps

Jerry
 

No, without the probe. I was told they worked but without a probe I can't test them. They look to be in good shape.


I kicked around buying a probe but they are expensive.


Re: FS: AM503 current probe amps

Artek Manuals <manuals@...>
 

Jerry

Does that include the probe and do they work?

Dave
manuals@artekmanuals.com

On 4/19/2016 3:34 PM, jerry@hanler.com [TekScopes] wrote:
I have two AM503 current probe amps that were included in a frame I
bought. They look to be in decent shape. One has a broken pull-tab and
the second has a cracked plastic bevel corner. I would like $30 per plus
shipping. I've been kicking around the idea of gutting them and using
them as modules for other projects as I don't have a need for them. Send
a PM if interested.


Thanks


Jerry


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


2567B auto-shutdown and auto dimming

bogroca
 

Hi guys,


I have two 2467Bs that I would like to restore. After re-caping the PS they seem to work, but:


First (SN 050183), has a very fast (1-2 seconds) auto-dimming regardless on how bright the manual control is set. This is w/o any signal on inputs. As soon as signal is present (triggered or not), then manual brightness control holds. Is it possible to do something about this or is the CRT going, going, gone?


Second (SN 053818), does not go into auto shutdown at all (went in EXER09 and set it up). Any idea?


Best regards,
Bogdan


FS: AM503 current probe amps

Jerry
 

I have two AM503 current probe amps that were included in a frame I bought. They look to be in decent shape. One has a broken pull-tab and the second has a cracked plastic bevel corner. I would like $30 per plus shipping. I've been kicking around the idea of gutting them and using them as modules for other projects as I don't have a need for them. Send a PM if interested.


Thanks


Jerry


Re: Tx line sampling

Ed Breya
 

When all of this is said and done, what is the measurement goal? Ed


Re: 5k plugins

 

Hi,


Sounds like your are talking about the Pulse Instruments PI-451. It makes a great prototype base. I may have the model number wrong. My plan is to use one to house and power a cheapo curve tracer kit and feed its X-Y output to an MR or SC501 next slot.


Jerry Massengale

-----Original Message-----
From: Doxemf doxemf@aol.com [TekScopes] <TekScopes@yahoogroups.com>
To: TekScopes <TekScopes@yahoogroups.com>
Sent: Mon, Apr 18, 2016 8:53 pm
Subject: Re: [TekScopes] 5k plugins






They are physically compatible but the ID slots cut in the edge connector are not so as to prevent the accidental wrong mainframe use. No big deal. Just pull the blocking inserts out of the connectors or cut new slots. There double contact and plated thru connections to watch out for. If you haven't seen it, there is a 500 series backplane app manual on tekwiki. I have a custom PI somewhere that has the edge conn terminated in wirewrap terminals for custom builds. It has some odd protoboard mounted but easy to strip out. Don't remember what series it is and will check.
Bill, KB3DKS

-----Original Message-----
From: jerry massengale j_massengale@aol.com [TekScopes] <TekScopes@yahoogroups.com>

Hi,

Is there a problem with using a 5K plugin module to build a TM500 plugin prototype? If so , what?

Jerry Massengale

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


Re: 7854 RAM card upgrade only - any interest?

Mark Wendt
 

On 04/19/2016 07:57 AM, Nathan Johnson jdownj@gmail.com [TekScopes] wrote:
The draft that I'm working on is looking like a total parts cost of ~70USD.
This is based on what is believed to be the last revision of the combined
RAM/ROM card. The PCB is the primary cost, so there is little to be gained
by making a RAM or ROM only card. I am going SMT for a good part of this,
but all SOP/PLCC that can be drag soldered, nothing crazy. My goal is to be
able to offer assembled/tested units for $100.
Nathan KK4REY
Nathan,

Excellent! You guys can count me in then too, for sure.

Mark


Re: 7854 RAM card upgrade only - any interest?

Nathan Johnson
 

The draft that I'm working on is looking like a total parts cost of ~70USD.
This is based on what is believed to be the last revision of the combined
RAM/ROM card. The PCB is the primary cost, so there is little to be gained
by making a RAM or ROM only card. I am going SMT for a good part of this,
but all SOP/PLCC that can be drag soldered, nothing crazy. My goal is to be
able to offer assembled/tested units for $100.
Nathan KK4REY

Sent using CloudMagic Email
[https://cloudmagic.com/k/d/mailapp?ct=pi&;cv=7.4.15&pv=9.1&source=email_footer_2]

On Tue, Apr 19, 2016 at 05:52, Mark wendt.mark@gmail.com [TekScopes]
<TekScopes@yahoogroups.com> wrote:
On 04/18/2016 04:06 PM, test2cal@yahoo.com [TekScopes] wrote:
> Brian, has the buyer of the 7854 diagnostic card replied yet to your
communication through Solano?
>
> If we don't get the diagnostic ROM image, we should look at a cost
reduced solution.
>
>
>
> I have identified 8 possible buyers of the card through expressed
interest or through having made substantive contributions to this thread.
We need most of these to follow through to spread the setup cost.
>
>
> I think we should separate the issues: Some users have already solved
the ROM rot problem. Some users don't have the ROM rot issue (non-Mostek
ROMS already installed by Tek; which is my case). The Diagnostic ROM
versions issue is not clear at present, and we have no images of the
contents. We may be reducing the market by increasing the cost to folks who
don't need these solutions.
>
>
> We should consider a RAM only card, as small as possible while still
allowing for the backup battery. The card might not even have to reach the
card guides or retainer if it is small (and light) enough. Installing or
removing the card might require removing the adjacent cards; not a major
problem. That would lower the cost (fewer square inches and fewer parts).
Users who still really want card retention could make their own from
plastic or metal if provision was made for bolt holes at the top of the
card.
>
>
> David, we need to know if earlier 7854s respond to the KBID signals the
same way as late versions. What would be the simplest test for this?
>
>
> I think we have offers to do the layout, to order the cards and parts
kits and to distribute them. We need to firm this up.
>
>
> Cliff

Cliff,

I may be interested in this too. Have we come up with a rough ballpark
figure on the cost yet?

Thanks,
Mark


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

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


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

Yahoo Groups Links


Re: 7854 RAM card upgrade only - any interest?

Mark Wendt
 

On 04/18/2016 04:06 PM, test2cal@yahoo.com [TekScopes] wrote:
Brian, has the buyer of the 7854 diagnostic card replied yet to your communication through Solano?

If we don't get the diagnostic ROM image, we should look at a cost reduced solution.


I have identified 8 possible buyers of the card through expressed interest or through having made substantive contributions to this thread. We need most of these to follow through to spread the setup cost.

I think we should separate the issues: Some users have already solved the ROM rot problem. Some users don't have the ROM rot issue (non-Mostek ROMS already installed by Tek; which is my case). The Diagnostic ROM versions issue is not clear at present, and we have no images of the contents. We may be reducing the market by increasing the cost to folks who don't need these solutions.

We should consider a RAM only card, as small as possible while still allowing for the backup battery. The card might not even have to reach the card guides or retainer if it is small (and light) enough. Installing or removing the card might require removing the adjacent cards; not a major problem. That would lower the cost (fewer square inches and fewer parts). Users who still really want card retention could make their own from plastic or metal if provision was made for bolt holes at the top of the card.

David, we need to know if earlier 7854s respond to the KBID signals the same way as late versions. What would be the simplest test for this?

I think we have offers to do the layout, to order the cards and parts kits and to distribute them. We need to firm this up.

Cliff
Cliff,

I may be interested in this too. Have we come up with a rough ballpark figure on the cost yet?

Thanks,
Mark


Tx line sampling

mosaicmerc
 

I found this to be very useful regarding how to go about calculating the figures of merit from sampling a Tx line.
http://www.amanogawa.com/archive/docs/F-tutorial.pdf http://www.amanogawa.com/archive/docs/F-tutorial.pdf

Also, it appears that for best accuracy a wideband, variable, precision attenuator may be best to keep detector non linearities out of the picture.

In theory, terminating the Tx line signal V+ with an open or short will produce maxima |2V| and minima |0| standing waves which can be measured for amplitude and wavelength.

Matched impedance termination will produce a flat DC sample voltage which may need to be chopped and AC processed to removed any DC offset drifts due to electronics to produce the best precision.

A signal from a DUT can then be analyzed for amplitude, frequency and VSWR/impedance matching.

Since the VSWR and matched impedance DCV amplitudes are static relative to the TX line, sampling along the length of the line can produce a fairly jitter free representation of the waveform.

Thus any signal with a reasonable SNR can be processed via a mobile digital sampling head and then uploaded to a PC for charting or out to a DSO for viewing and interactive measurements or processing.

As I have several (6) old HP & Brother inkjet printers on hand, assembling a mobile head with optical stepping precision should not be difficult. I should be able to do bidirectional USB control and data gathering with the system.

Limitation is no frequency sweep like a VNA to leverage the shifting of minima delivering phase info.

But cost effectiveness may be good with a single sampling head and perhaps a good representation of 2 cm to 30cm or more wavelength signals in several seconds. By offloading all the logarithmic processing to a PC or android device cost drops further. But a DSP capable uC is just a few dollars these days and it adds portability with a TFT screen or simply a wifi IoT uplink to an android device.

Please poke any holes in this concept....weaknesses can be weeded out early.

I'd hope for perhaps a +/- 0.25dB uncertainty as a goal when its done.

Ancel

---In TekScopes@yahoogroups.com, <mosaicmerc@...> wrote :

Yes, it's a revisit of an old principle.
It might lead to some interesting results using modern PCB SMT assembly
methods and flatness calibration using digital sampling correction
coefficients.

52301 - 52320 of 179451