Topics

Tektronix 2245a Flickering Trace

William Schuler
 

That makes sense. 99% sure I did not short straight to ground, although you
never know... The 130 VAC supplies the multiplier, and is inverted ac,
Which I can’t measure straight with my multimeter because it is not true
RMS. Probably time to pony up for a Fluke.

I measured the voltage drop with a 1 Meg resistor, and came up with .150
VAC, so within the realm.

Thanks for the guidance, Bill

On Sun, Jan 12, 2020 at 2:42 PM Harvey White <madyn@...> wrote:

er... no....

you never measure current across anything, at least, not with an
ammeter. You always measure voltage drop across a known resistor to get
current (after a bit of computation). The ammeter function in a meter
just has built in resistors, requiring you to interrupt the circuit.

What I was suggesting is this:

Any measurement affects the circuit being measured. So with a 1X scope
probe hooked to a 1 meg input, you effectively are putting a 1 meg
resistor (the scope) from that point to ground. If the circuit you're
measuring has high enough resistance at points, then measuring that
point will obviously (as in you can see it happen when you do this)
change the circuit behavior. A rule of thumb is try to measure a
circuit with a meter that has at least 10x the impedance of the circuit,
more can be better. 10 x probes have 10 megs to ground, and less
capacitance, therefore upsetting the circuit a lot less.

So measuring the 130 volts to ground with an oscilloscope 1x probe puts
1 meg from there to ground. that's all of 130 microamps extra load,
which says that either something is REALLY marginal, or you may have hit
a nearby point in the circuit where 1 meg to ground *does* make a
difference.

The way that some scope probes are made has a large cylindrical ground
area behind the pin. It's quite easy to have that ground hit something
not intended to be grounded. Tektronix made tip protectors that would
cover that area and prevent accidents. All you had exposed was the
probe tip. With a little work, though, that can still short stuff out
(don't ask).

And 130 vac? Thought that was the DC voltage from the DC supply.

First rule of troubleshooting: check the power supply voltages. Then
check the point you're measuring. Then go check the supply voltages again.

I had a Fluke voltage standard that would never go above 250 volts where
it was supposed to go over 1000. Tried a number of things, then looked
at the main power supply voltage. LOTS of ripple, bad capacitor. The
meter was showing the average, not the peak. Replace the capacitor, and
it worked just fine, thank you.

Harvey


On 1/12/2020 4:18 PM, William Schuler wrote:
Hey Harvey, just to make sure I understand correctly: I shunt a one Meg
on
the resister from 130 VAC to ground, and measure the amperage across the
resister? I tried that, and got 0 amps.

On Sat, Jan 11, 2020 at 5:50 PM Harvey White <madyn@...>
wrote:

I'd say so.

First: check all the PS voltages, you could have blown a fuse.

Second: check the voltages in the HV circuit. 1 meg to ground on a 130
volt test point is 130 microamps. If that blows something, then you had
a VERY marginal circuit/situation to start. Possibly more likely that
you had the scope on 50 ohms input (if you had one), or had a 50 ohm
probe (X10 = 500 ohms). Not saying that you did, but these would be what
you'd be dealing with *if* you had. Possibly, did you short the 130
volts to ground?

Third: although you might want to do this second or as part of "first",
check the 130 volts.

Suggestion... Put a piece of heat shrink (or find the tip protector)
for your probe so that all you have as a connector is a tiny piece of
metal. Secondly, use an X10 probe just because. Switchable probes are
nicest on low frequency and low impedance circuits, exactly where you'd
think X1 probes are good. I'm assuming that the measuring scope is ok.

Harvey



On 1/11/2020 7:11 PM, William Schuler wrote:
Well I fear the worst... While self diagnosing power supply ripple, I
touched 130 V test point, saw a bright flash on screen, then
everything
went black. Unit powers up, LEDs on front light up, relays click, but
there
is no display on the screen. The beam find button does nothing. Bummer!
I
realized I didn’t have probe on 10 X. Is there any hope you guys?

On Sat, Jan 4, 2020 at 11:27 AM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <
guitardad1967@...
wrote:

That makes me wonder – can you use a mostly functional scope to
diagnose
itself?
I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf)
is
about "Front-Panel Diagnosis" which largely interrogates the
instrument
about its own faults. You can - to a point - use a scope to look at
its
own
power supply ripple and the like, and *some* of its internal state,
like
some portion of the sweep ramp(s). I guess it can be hard to know what
to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis
capability.
You
can get the service manual from here (
http://w140.com/tekwiki/wiki/2245
),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional
trigger
signal which is the line. You can also think of AUTO trigger mode and
the
holdoff as trigger signals. Then there's X/Y mode, which allows you to
introduce a calibrated signal to the horizontal circuitry. In your
case, it
might be interesting to inject DC on the X, and e.g. a sine or
triangle
wave on the Y. It might also be interesting to play with line
triggering to
see whether the glitching is line-synchronous, which could happen if
e.g.
you have excessive line ripple on a supply ... time passes ... these
scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi







Harvey White
 

er... no....

you never measure current across anything, at least, not with an ammeter.  You always measure voltage drop across a known resistor to get current (after a bit of computation).  The ammeter function in a meter just has built in resistors, requiring you to interrupt the circuit.

What I was suggesting is this:

Any measurement affects the circuit being measured.  So with a 1X scope probe hooked to a 1 meg input, you effectively are putting a 1 meg resistor (the scope) from that point to ground.  If the circuit you're measuring has high enough resistance at points, then measuring that point will obviously (as in you can see it happen when you do this) change the circuit behavior.  A rule of thumb is try to measure a circuit with a meter that has at least 10x the impedance of the circuit, more can be better.  10 x probes have 10 megs to ground, and less capacitance, therefore upsetting the circuit a lot less.

So measuring the 130 volts to ground with an oscilloscope 1x probe puts 1 meg from there to ground.  that's all of 130 microamps extra load, which says that either something is REALLY marginal, or you may have hit a nearby point in the circuit where 1 meg to ground *does* make a difference.

The way that some scope probes are made has a large cylindrical ground area behind the pin.  It's quite easy to have that ground hit something not intended to be grounded.   Tektronix made tip protectors that would cover that area and prevent accidents.  All you had exposed was the probe tip.  With a little work, though, that can still short stuff out (don't ask).

And 130 vac?  Thought that was the DC voltage from the DC supply.

First rule of troubleshooting:  check the power supply voltages. Then check the point you're measuring.  Then go check the supply voltages again.

I had a Fluke voltage standard that would never go above 250 volts where it was supposed to go over 1000.  Tried a number of things, then looked at the main power supply voltage.  LOTS of ripple, bad capacitor.  The meter was showing the average, not the peak.  Replace the capacitor, and it worked just fine, thank you.

Harvey

On 1/12/2020 4:18 PM, William Schuler wrote:
Hey Harvey, just to make sure I understand correctly: I shunt a one Meg on
the resister from 130 VAC to ground, and measure the amperage across the
resister? I tried that, and got 0 amps.

On Sat, Jan 11, 2020 at 5:50 PM Harvey White <madyn@...> wrote:

I'd say so.

First: check all the PS voltages, you could have blown a fuse.

Second: check the voltages in the HV circuit. 1 meg to ground on a 130
volt test point is 130 microamps. If that blows something, then you had
a VERY marginal circuit/situation to start. Possibly more likely that
you had the scope on 50 ohms input (if you had one), or had a 50 ohm
probe (X10 = 500 ohms). Not saying that you did, but these would be what
you'd be dealing with *if* you had. Possibly, did you short the 130
volts to ground?

Third: although you might want to do this second or as part of "first",
check the 130 volts.

Suggestion... Put a piece of heat shrink (or find the tip protector)
for your probe so that all you have as a connector is a tiny piece of
metal. Secondly, use an X10 probe just because. Switchable probes are
nicest on low frequency and low impedance circuits, exactly where you'd
think X1 probes are good. I'm assuming that the measuring scope is ok.

Harvey



On 1/11/2020 7:11 PM, William Schuler wrote:
Well I fear the worst... While self diagnosing power supply ripple, I
touched 130 V test point, saw a bright flash on screen, then everything
went black. Unit powers up, LEDs on front light up, relays click, but
there
is no display on the screen. The beam find button does nothing. Bummer!
I
realized I didn’t have probe on 10 X. Is there any hope you guys?

On Sat, Jan 4, 2020 at 11:27 AM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <guitardad1967@...
wrote:

That makes me wonder – can you use a mostly functional scope to
diagnose
itself?
I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf) is
about "Front-Panel Diagnosis" which largely interrogates the instrument
about its own faults. You can - to a point - use a scope to look at its
own
power supply ripple and the like, and *some* of its internal state, like
some portion of the sweep ramp(s). I guess it can be hard to know what
to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis capability.
You
can get the service manual from here (http://w140.com/tekwiki/wiki/2245
),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional trigger
signal which is the line. You can also think of AUTO trigger mode and
the
holdoff as trigger signals. Then there's X/Y mode, which allows you to
introduce a calibrated signal to the horizontal circuitry. In your
case, it
might be interesting to inject DC on the X, and e.g. a sine or triangle
wave on the Y. It might also be interesting to play with line
triggering to
see whether the glitching is line-synchronous, which could happen if
e.g.
you have excessive line ripple on a supply ... time passes ... these
scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi




William Schuler
 

I have the power supply open now. Starting to wonder if my problem might be
the multiplier...
Looks like it has a darkened area. I don’t have a means of testing with 2.8
KV. Is there a way to pull this from the circuit and test it?

-Bill



William Schuler
 

Hey Harvey, just to make sure I understand correctly: I shunt a one Meg on
the resister from 130 VAC to ground, and measure the amperage across the
resister? I tried that, and got 0 amps.

On Sat, Jan 11, 2020 at 5:50 PM Harvey White <madyn@...> wrote:

I'd say so.

First: check all the PS voltages, you could have blown a fuse.

Second: check the voltages in the HV circuit. 1 meg to ground on a 130
volt test point is 130 microamps. If that blows something, then you had
a VERY marginal circuit/situation to start. Possibly more likely that
you had the scope on 50 ohms input (if you had one), or had a 50 ohm
probe (X10 = 500 ohms). Not saying that you did, but these would be what
you'd be dealing with *if* you had. Possibly, did you short the 130
volts to ground?

Third: although you might want to do this second or as part of "first",
check the 130 volts.

Suggestion... Put a piece of heat shrink (or find the tip protector)
for your probe so that all you have as a connector is a tiny piece of
metal. Secondly, use an X10 probe just because. Switchable probes are
nicest on low frequency and low impedance circuits, exactly where you'd
think X1 probes are good. I'm assuming that the measuring scope is ok.

Harvey



On 1/11/2020 7:11 PM, William Schuler wrote:
Well I fear the worst... While self diagnosing power supply ripple, I
touched 130 V test point, saw a bright flash on screen, then everything
went black. Unit powers up, LEDs on front light up, relays click, but
there
is no display on the screen. The beam find button does nothing. Bummer!
I
realized I didn’t have probe on 10 X. Is there any hope you guys?

On Sat, Jan 4, 2020 at 11:27 AM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <guitardad1967@...
wrote:

That makes me wonder – can you use a mostly functional scope to
diagnose
itself?
I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf) is
about "Front-Panel Diagnosis" which largely interrogates the instrument
about its own faults. You can - to a point - use a scope to look at its
own
power supply ripple and the like, and *some* of its internal state, like
some portion of the sweep ramp(s). I guess it can be hard to know what
to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis capability.
You
can get the service manual from here (http://w140.com/tekwiki/wiki/2245
),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional trigger
signal which is the line. You can also think of AUTO trigger mode and
the
holdoff as trigger signals. Then there's X/Y mode, which allows you to
introduce a calibrated signal to the horizontal circuitry. In your
case, it
might be interesting to inject DC on the X, and e.g. a sine or triangle
wave on the Y. It might also be interesting to play with line
triggering to
see whether the glitching is line-synchronous, which could happen if
e.g.
you have excessive line ripple on a supply ... time passes ... these
scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi






Harvey White
 

aha.

The probe represents 10 megs to ground (or 1 meg).  Putting the probe on the circuit loaded that point by the 1 meg or 10 megs. That suggests that something that ought to be going from there to ground either isn't, or the voltage at that point is higher than it needed to be, and the scope probe loaded it down to the point where it was ok.

What happens if you put the scope probe back there?

Harvey

On 1/11/2020 7:51 PM, William Schuler wrote:
Tip of probe. Saw a bright, vertical trace, and when I removed my probe
screen was blank.

On Sat, Jan 11, 2020 at 5:24 PM Siggi <siggi@...> wrote:

Hmmm, what did you touch the test point with? Tip of the probe? If so, it
shouldn't be a problem: Maximum Input Voltage 400 V (DC + peak AC) or 800 V
(p-p ac at ≤ 10 kHz).
If you touched it with the GND clip you may have shorted something. Did you
measure the supplies post-gaffe?

On Sat, Jan 11, 2020 at 7:11 PM William Schuler <guitardad1967@...>
wrote:

Well I fear the worst... While self diagnosing power supply ripple, I
touched 130 V test point, saw a bright flash on screen, then everything
went black. Unit powers up, LEDs on front light up, relays click, but
there
is no display on the screen. The beam find button does nothing. Bummer!
I
realized I didn’t have probe on 10 X. Is there any hope you guys?

On Sat, Jan 4, 2020 at 11:27 AM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <
guitardad1967@...>
wrote:

That makes me wonder – can you use a mostly functional scope to
diagnose
itself?
I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf) is
about "Front-Panel Diagnosis" which largely interrogates the instrument
about its own faults. You can - to a point - use a scope to look at its
own
power supply ripple and the like, and *some* of its internal state,
like
some portion of the sweep ramp(s). I guess it can be hard to know what
to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis capability.
You
can get the service manual from here (
http://w140.com/tekwiki/wiki/2245
),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional trigger
signal which is the line. You can also think of AUTO trigger mode and
the
holdoff as trigger signals. Then there's X/Y mode, which allows you to
introduce a calibrated signal to the horizontal circuitry. In your
case,
it
might be interesting to inject DC on the X, and e.g. a sine or triangle
wave on the Y. It might also be interesting to play with line
triggering
to
see whether the glitching is line-synchronous, which could happen if
e.g.
you have excessive line ripple on a supply ... time passes ... these
scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi





William Schuler
 

All the rail voltages checked out OK. I don’t know how to measure the 130
VAC inverted power. With unit off,that tests as shorted. Is that normal?

On Sat, Jan 11, 2020 at 5:54 PM William Schuler via Groups.Io
<guitardad1967=gmail.com@groups.io> wrote:

Tip of probe. Saw a bright, vertical trace, and when I removed my probe
screen was blank.

On Sat, Jan 11, 2020 at 5:24 PM Siggi <siggi@...> wrote:

Hmmm, what did you touch the test point with? Tip of the probe? If so, it
shouldn't be a problem: Maximum Input Voltage 400 V (DC + peak AC) or
800 V
(p-p ac at ≤ 10 kHz).
If you touched it with the GND clip you may have shorted something. Did
you
measure the supplies post-gaffe?

On Sat, Jan 11, 2020 at 7:11 PM William Schuler <guitardad1967@...

wrote:

Well I fear the worst... While self diagnosing power supply ripple, I
touched 130 V test point, saw a bright flash on screen, then
everything
went black. Unit powers up, LEDs on front light up, relays click, but
there
is no display on the screen. The beam find button does nothing. Bummer!
I
realized I didn’t have probe on 10 X. Is there any hope you guys?

On Sat, Jan 4, 2020 at 11:27 AM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <
guitardad1967@...>
wrote:

That makes me wonder – can you use a mostly functional scope to
diagnose
itself?

I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf)
is
about "Front-Panel Diagnosis" which largely interrogates the
instrument
about its own faults. You can - to a point - use a scope to look at
its
own
power supply ripple and the like, and *some* of its internal state,
like
some portion of the sweep ramp(s). I guess it can be hard to know
what
to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis
capability.
You
can get the service manual from here (
http://w140.com/tekwiki/wiki/2245
),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional
trigger
signal which is the line. You can also think of AUTO trigger mode and
the
holdoff as trigger signals. Then there's X/Y mode, which allows you
to
introduce a calibrated signal to the horizontal circuitry. In your
case,
it
might be interesting to inject DC on the X, and e.g. a sine or
triangle
wave on the Y. It might also be interesting to play with line
triggering
to
see whether the glitching is line-synchronous, which could happen if
e.g.
you have excessive line ripple on a supply ... time passes ... these
scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in
the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi









William Schuler
 

Tip of probe. Saw a bright, vertical trace, and when I removed my probe
screen was blank.

On Sat, Jan 11, 2020 at 5:24 PM Siggi <siggi@...> wrote:

Hmmm, what did you touch the test point with? Tip of the probe? If so, it
shouldn't be a problem: Maximum Input Voltage 400 V (DC + peak AC) or 800 V
(p-p ac at ≤ 10 kHz).
If you touched it with the GND clip you may have shorted something. Did you
measure the supplies post-gaffe?

On Sat, Jan 11, 2020 at 7:11 PM William Schuler <guitardad1967@...>
wrote:

Well I fear the worst... While self diagnosing power supply ripple, I
touched 130 V test point, saw a bright flash on screen, then everything
went black. Unit powers up, LEDs on front light up, relays click, but
there
is no display on the screen. The beam find button does nothing. Bummer!
I
realized I didn’t have probe on 10 X. Is there any hope you guys?

On Sat, Jan 4, 2020 at 11:27 AM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <
guitardad1967@...>
wrote:

That makes me wonder – can you use a mostly functional scope to
diagnose
itself?

I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf) is
about "Front-Panel Diagnosis" which largely interrogates the instrument
about its own faults. You can - to a point - use a scope to look at its
own
power supply ripple and the like, and *some* of its internal state,
like
some portion of the sweep ramp(s). I guess it can be hard to know what
to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis capability.
You
can get the service manual from here (
http://w140.com/tekwiki/wiki/2245
),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional trigger
signal which is the line. You can also think of AUTO trigger mode and
the
holdoff as trigger signals. Then there's X/Y mode, which allows you to
introduce a calibrated signal to the horizontal circuitry. In your
case,
it
might be interesting to inject DC on the X, and e.g. a sine or triangle
wave on the Y. It might also be interesting to play with line
triggering
to
see whether the glitching is line-synchronous, which could happen if
e.g.
you have excessive line ripple on a supply ... time passes ... these
scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi







Harvey White
 

I'd say so.

First:  check all the PS voltages, you could have blown a fuse.

Second:  check the voltages in the HV circuit.  1 meg to ground on a 130 volt test point is 130 microamps.  If that blows something, then you had a VERY marginal circuit/situation to start.  Possibly more likely that you had the scope on 50 ohms input (if you had one), or had a 50 ohm probe (X10 = 500 ohms). Not saying that you did, but these would be what you'd be dealing with *if* you had.   Possibly, did you short the 130 volts to ground?

Third: although you might want to do this second or as part of "first", check the 130 volts.

Suggestion...  Put a piece of heat shrink (or find the tip protector) for your probe so that all you have as a connector is a tiny piece of metal.  Secondly, use an X10 probe just because. Switchable probes are nicest on low frequency and low impedance circuits, exactly where you'd think X1 probes are good.  I'm assuming that the measuring scope is ok.

Harvey

On 1/11/2020 7:11 PM, William Schuler wrote:
Well I fear the worst... While self diagnosing power supply ripple, I
touched 130 V test point, saw a bright flash on screen, then everything
went black. Unit powers up, LEDs on front light up, relays click, but there
is no display on the screen. The beam find button does nothing. Bummer! I
realized I didn’t have probe on 10 X. Is there any hope you guys?

On Sat, Jan 4, 2020 at 11:27 AM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <guitardad1967@...>
wrote:

That makes me wonder – can you use a mostly functional scope to diagnose
itself?
I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf) is
about "Front-Panel Diagnosis" which largely interrogates the instrument
about its own faults. You can - to a point - use a scope to look at its own
power supply ripple and the like, and *some* of its internal state, like
some portion of the sweep ramp(s). I guess it can be hard to know what to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis capability. You
can get the service manual from here (http://w140.com/tekwiki/wiki/2245),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional trigger
signal which is the line. You can also think of AUTO trigger mode and the
holdoff as trigger signals. Then there's X/Y mode, which allows you to
introduce a calibrated signal to the horizontal circuitry. In your case, it
might be interesting to inject DC on the X, and e.g. a sine or triangle
wave on the Y. It might also be interesting to play with line triggering to
see whether the glitching is line-synchronous, which could happen if e.g.
you have excessive line ripple on a supply ... time passes ... these scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi



Siggi
 

Hmmm, what did you touch the test point with? Tip of the probe? If so, it
shouldn't be a problem: Maximum Input Voltage 400 V (DC + peak AC) or 800 V
(p-p ac at ≤ 10 kHz).
If you touched it with the GND clip you may have shorted something. Did you
measure the supplies post-gaffe?

On Sat, Jan 11, 2020 at 7:11 PM William Schuler <guitardad1967@...>
wrote:

Well I fear the worst... While self diagnosing power supply ripple, I
touched 130 V test point, saw a bright flash on screen, then everything
went black. Unit powers up, LEDs on front light up, relays click, but there
is no display on the screen. The beam find button does nothing. Bummer! I
realized I didn’t have probe on 10 X. Is there any hope you guys?

On Sat, Jan 4, 2020 at 11:27 AM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <guitardad1967@...>
wrote:

That makes me wonder – can you use a mostly functional scope to
diagnose
itself?

I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf) is
about "Front-Panel Diagnosis" which largely interrogates the instrument
about its own faults. You can - to a point - use a scope to look at its
own
power supply ripple and the like, and *some* of its internal state, like
some portion of the sweep ramp(s). I guess it can be hard to know what to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis capability.
You
can get the service manual from here (http://w140.com/tekwiki/wiki/2245
),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional trigger
signal which is the line. You can also think of AUTO trigger mode and the
holdoff as trigger signals. Then there's X/Y mode, which allows you to
introduce a calibrated signal to the horizontal circuitry. In your case,
it
might be interesting to inject DC on the X, and e.g. a sine or triangle
wave on the Y. It might also be interesting to play with line triggering
to
see whether the glitching is line-synchronous, which could happen if e.g.
you have excessive line ripple on a supply ... time passes ... these
scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi





William Schuler
 

Well I fear the worst... While self diagnosing power supply ripple, I
touched 130 V test point, saw a bright flash on screen, then everything
went black. Unit powers up, LEDs on front light up, relays click, but there
is no display on the screen. The beam find button does nothing. Bummer! I
realized I didn’t have probe on 10 X. Is there any hope you guys?

On Sat, Jan 4, 2020 at 11:27 AM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <guitardad1967@...>
wrote:

That makes me wonder – can you use a mostly functional scope to diagnose
itself?

I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf) is
about "Front-Panel Diagnosis" which largely interrogates the instrument
about its own faults. You can - to a point - use a scope to look at its own
power supply ripple and the like, and *some* of its internal state, like
some portion of the sweep ramp(s). I guess it can be hard to know what to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis capability. You
can get the service manual from here (http://w140.com/tekwiki/wiki/2245),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional trigger
signal which is the line. You can also think of AUTO trigger mode and the
holdoff as trigger signals. Then there's X/Y mode, which allows you to
introduce a calibrated signal to the horizontal circuitry. In your case, it
might be interesting to inject DC on the X, and e.g. a sine or triangle
wave on the Y. It might also be interesting to play with line triggering to
see whether the glitching is line-synchronous, which could happen if e.g.
you have excessive line ripple on a supply ... time passes ... these scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi



Tony Fleming
 

First of all Harvey, Happy New Year 2020 and wishes of lots of health and
great ELECTRONS in your life!
If it wasn't for you, my scope would not work, so I trust you and wish you
are my neighbor!
Anyway, thanks for helping everyone here, including me. ( tonysfun.com and
click on MY VIDEOS to see how micro-knowledge functions in my very small
brain! ... ha ha ha)

Lastly, I wish you all HAPPY NEW YEAR 2020 and POSITIVE ELECTRON'S in your
life and scopes!

Tony

On Sun, Jan 5, 2020 at 10:35 PM Harvey White <madyn@...> wrote:

There are some tests you can do with a partially working scope, but you
are going to be very limited.

1) assuming horizontal sweep problems, you use an external ramp or the
"other" sweep if you have a dual sweep model.

2) if you have a horizontal amplifier problem, using the scope in XY
mode, even with the calibrator, can give you some idea of whether or not
it's the sweep or the amplifier.

3) if you have a dual channel scope, you can use a working channel to
debug the separate part of the vertical amplifier

These generally end up being go/no-go tests and are somewhat crude.

Function generators can make a very crude horizontal sweep. If they're
gated, then a gated ramp might work as a sweep.

Very crude, but better than nothing.

Generally, unless you have a scope specifically designed for self
diagnosis, you're better off with another, completely working, scope.

Even then, a scope designed for self diagnosis makes certain assumptions
about what's working, especially once you get past the "is the CPU
working...."

Harvey


On 1/4/2020 4:29 PM, Panos wrote:
I have read enough service manuals for the oscilloscopes, and till today
I didn't find any of them to mention, that someone can use the same faulty
scope (partially or totally) to repair his self.
But as always we can hope for something like that. After all, don't
they say that hope dies always last? :-)

In my opinion if you have time to play then do it by that way. But if
you want really to repair it, then proceed it with another oscilloscope.





battyhugh
 

Telonics 2003

I was hoping that the seller (of the manual) would link me with the buyer - no luck. So does some kind person had the manual for the earlier model - (or does anyone still have a 2003 manual loitering in a cupboard?? (or the old 3 pin plug?).

I'd hate to scrap it.

Cheers

hugh

William Schuler
 

Thanks Harvey, going through some family things right now, but hopefully I
can get to looking at the scope this weekend.

On Sun, Jan 5, 2020 at 9:35 PM Harvey White <madyn@...> wrote:

There are some tests you can do with a partially working scope, but you
are going to be very limited.

1) assuming horizontal sweep problems, you use an external ramp or the
"other" sweep if you have a dual sweep model.

2) if you have a horizontal amplifier problem, using the scope in XY
mode, even with the calibrator, can give you some idea of whether or not
it's the sweep or the amplifier.

3) if you have a dual channel scope, you can use a working channel to
debug the separate part of the vertical amplifier

These generally end up being go/no-go tests and are somewhat crude.

Function generators can make a very crude horizontal sweep. If they're
gated, then a gated ramp might work as a sweep.

Very crude, but better than nothing.

Generally, unless you have a scope specifically designed for self
diagnosis, you're better off with another, completely working, scope.

Even then, a scope designed for self diagnosis makes certain assumptions
about what's working, especially once you get past the "is the CPU
working...."

Harvey


On 1/4/2020 4:29 PM, Panos wrote:
I have read enough service manuals for the oscilloscopes, and till today
I didn't find any of them to mention, that someone can use the same faulty
scope (partially or totally) to repair his self.
But as always we can hope for something like that. After all, don't
they say that hope dies always last? :-)

In my opinion if you have time to play then do it by that way. But if
you want really to repair it, then proceed it with another oscilloscope.





William Schuler
 

Thanks Siggy, I appreciate the pointers. Nice “outer limits“ clip as well!

BTW, the fan runs fine.

On Sat, Jan 4, 2020 at 6:45 PM Siggi <siggi@...> wrote:

On Sat, Jan 4, 2020 at 8:13 PM William Schuler <guitardad1967@...>
wrote:

I am in no hurry, so I will give it a try just for fun. It would be
interesting to compare results with my buddy’s functional scope.

Hmmm, you already see the output of the horizontal amplifier - that's what
your CRT shows.
The only thing is that your horizontal output (
https://youtu.be/8dzxGlcErUo?t=10) is differential. By comparing both
sides, you'd be able to see whether this is due to one side pulling or both
- it's a data point at least. I'd bet that you'll find your problem
elsewhere than in the horizontal amp, though. I'm betting on power
supplies, as that's where ~90% of all trouble happens. Is the fan working
OK?
In any case, while I often^Wsometimes miss the most direct way to figure
out what's up, I find the detours are more than worth their while for the
learnin'.


Now to
figure out exactly where that horizontal amplifier is....
You have the service manual (

http://www.ko4bb.com/manuals/download.php?file=Tektronix/Tektronix_-_2245A/Tektronix_2245A_Oscilloscope_Service_Manual.pdf
)?

Have fun!
Siggi



Harvey White
 

There are some tests you can do with a partially working scope, but you are going to be very limited.

1) assuming horizontal sweep problems, you use an external ramp or the "other" sweep if you have a dual sweep model.

2) if you have a horizontal amplifier problem, using the scope in XY mode, even with the calibrator, can give you some idea of whether or not it's the sweep or the amplifier.

3) if you have a dual channel scope, you can use a working channel to debug the separate part of the vertical amplifier

These generally end up being go/no-go tests and are somewhat crude.

Function generators can make a very crude horizontal sweep.  If they're gated, then a gated ramp might work as a sweep.

Very crude, but better than nothing.

Generally, unless you have a scope specifically designed for self diagnosis, you're better off with another, completely working, scope.

Even then, a scope designed for self diagnosis makes certain assumptions about what's working, especially once you get past the "is the CPU working...."

Harvey

On 1/4/2020 4:29 PM, Panos wrote:
I have read enough service manuals for the oscilloscopes, and till today I didn't find any of them to mention, that someone can use the same faulty scope (partially or totally) to repair his self.
But as always we can hope for something like that. After all, don't they say that hope dies always last? :-)

In my opinion if you have time to play then do it by that way. But if you want really to repair it, then proceed it with another oscilloscope.


Siggi
 

On Sat, Jan 4, 2020 at 8:13 PM William Schuler <guitardad1967@...>
wrote:

I am in no hurry, so I will give it a try just for fun. It would be
interesting to compare results with my buddy’s functional scope.

Hmmm, you already see the output of the horizontal amplifier - that's what
your CRT shows.
The only thing is that your horizontal output (
https://youtu.be/8dzxGlcErUo?t=10) is differential. By comparing both
sides, you'd be able to see whether this is due to one side pulling or both
- it's a data point at least. I'd bet that you'll find your problem
elsewhere than in the horizontal amp, though. I'm betting on power
supplies, as that's where ~90% of all trouble happens. Is the fan working
OK?
In any case, while I often^Wsometimes miss the most direct way to figure
out what's up, I find the detours are more than worth their while for the
learnin'.


Now to
figure out exactly where that horizontal amplifier is....
You have the service manual (
http://www.ko4bb.com/manuals/download.php?file=Tektronix/Tektronix_-_2245A/Tektronix_2245A_Oscilloscope_Service_Manual.pdf
)?

Have fun!
Siggi

William Schuler
 

I am in no hurry, so I will give it a try just for fun. It would be
interesting to compare results with my buddy’s functional scope. Now to
figure out exactly where that horizontal amplifier is....

On Sat, Jan 4, 2020 at 2:29 PM Panos <sadosp@...> wrote:

I have read enough service manuals for the oscilloscopes, and till today I
didn't find any of them to mention, that someone can use the same faulty
scope (partially or totally) to repair his self.
But as always we can hope for something like that. After all, don't they
say that hope dies always last? :-)

In my opinion if you have time to play then do it by that way. But if
you want really to repair it, then proceed it with another oscilloscope.



William Schuler
 

Exactly!

On Sat, Jan 4, 2020 at 9:39 AM Panos <sadosp@...> wrote:

On Sat, Jan 4, 2020 at 02:59 PM, William Schuler wrote:

That makes me wonder – can you use a mostly functional scope to diagnose
itself?
I think that already I see the Star Trek to arriving as to save the earth
after this!!!



Panos
 

I have read enough service manuals for the oscilloscopes, and till today I didn't find any of them to mention, that someone can use the same faulty scope (partially or totally) to repair his self.
But as always we can hope for something like that. After all, don't they say that hope dies always last? :-)

In my opinion if you have time to play then do it by that way. But if you want really to repair it, then proceed it with another oscilloscope.

Siggi
 

On Sat, Jan 4, 2020 at 8:00 AM William Schuler <guitardad1967@...>
wrote:

That makes me wonder – can you use a mostly functional scope to diagnose
itself?

I guess it depends, though a large section of this document (
https://www.sphere.bc.ca/test/tek-parts/troubleshooting-scopes.pdf) is
about "Front-Panel Diagnosis" which largely interrogates the instrument
about its own faults. You can - to a point - use a scope to look at its own
power supply ripple and the like, and *some* of its internal state, like
some portion of the sweep ramp(s). I guess it can be hard to know what to
believe when the instrument itself is faulty.
This scope also has a service mode with some self-diagnosis capability. You
can get the service manual from here (http://w140.com/tekwiki/wiki/2245),
see page 6-10 for the service menu.

Also note that every scope has two built-in "signal sources"; the
calibrator and GND coupling, and most scopes have an additional trigger
signal which is the line. You can also think of AUTO trigger mode and the
holdoff as trigger signals. Then there's X/Y mode, which allows you to
introduce a calibrated signal to the horizontal circuitry. In your case, it
might be interesting to inject DC on the X, and e.g. a sine or triangle
wave on the Y. It might also be interesting to play with line triggering to
see whether the glitching is line-synchronous, which could happen if e.g.
you have excessive line ripple on a supply ... time passes ... these scopes
have a switch-mode power supply, so more likely power supply noise or
ripple will be at the switching frequency.

If it comes to it my friend has a scope for use.


That'll certainly be an easier way to go :). If you poke around in the
power supply, be very, very careful, as everything except the final
low-voltage supplies are line-referenced.

Good luck and much fun,
Siggi