Tek TDS420 help?

D

DaveC

Guest
This scope has the famous "FAIL ++ Acq" error which points to the infamous
failed (and leaking) SMD caps on the acquisition PCB.

I'm moving on to troubleshoot another (unrelated, I think -- and hope) issue.

During the 25 second "on" period, relays click on & off, the monitor goes
through it's boot display (all bright, then dark, then the clock icon), then
the graticule is drawn, which is quickly replaced with the "Tek Running
Autoing" display listing the "FAIL ++ Acq" error.

During the power-up period, the front panel LEDs all turn on for about 25
seconds, then off for 5 seconds, then only the CH 1 LED lights.

The service manual says (in the Front Panel troubleshooting chart):

Do all
of the front panel
LEDs turn on and then
turn off a short
time later?

(Then:)

Does one
front panel LED
turn on and then turn off,
followed by the next LED until
all LEDs have turned
on and then
turned off?

The LEDs do not go through this one-at-a-time sequence. The manual's flow
chart points to a failed CPU PCB if the LEDs don't blink this sequence.

Can someone with a TDS4xx confirm power-on LED behavior?

I'm trying to determine whether, because the LEDs aren't blinking "properly"
and most of the front panel buttons are not working (the buttons bordering
the monitor, plus all the CH input buttons), does this mean a replacement CPU
PCB is in my future, or if it's possibly related to cap failure (ie, other
cap failures beyond the acquisition PCB)?

Has anyone else here have a similar problem (non-functional buttons) in the
past? I'd appreciate hearing your experiences.

Thanks,
Dave
 
DaveC wrote:
This scope has the famous "FAIL ++ Acq" error which points to the infamous
failed (and leaking) SMD caps on the acquisition PCB.

I'm moving on to troubleshoot another (unrelated, I think -- and hope) issue.

During the 25 second "on" period, relays click on & off, the monitor goes
through it's boot display (all bright, then dark, then the clock icon), then
the graticule is drawn, which is quickly replaced with the "Tek Running
Autoing" display listing the "FAIL ++ Acq" error.

During the power-up period, the front panel LEDs all turn on for about 25
seconds, then off for 5 seconds, then only the CH 1 LED lights.

The service manual says (in the Front Panel troubleshooting chart):

Do all
of the front panel
LEDs turn on and then
turn off a short
time later?

(Then:)

Does one
front panel LED
turn on and then turn off,
followed by the next LED until
all LEDs have turned
on and then
turned off?

The LEDs do not go through this one-at-a-time sequence. The manual's flow
chart points to a failed CPU PCB if the LEDs don't blink this sequence.

Can someone with a TDS4xx confirm power-on LED behavior?

I'm trying to determine whether, because the LEDs aren't blinking "properly"
and most of the front panel buttons are not working (the buttons bordering
the monitor, plus all the CH input buttons), does this mean a replacement CPU
PCB is in my future, or if it's possibly related to cap failure (ie, other
cap failures beyond the acquisition PCB)?

Has anyone else here have a similar problem (non-functional buttons) in the
past? I'd appreciate hearing your experiences.

Thanks,
Dave

All my experience is with the 540, but I think you're wasting your time
Click to expand...
trying to troubleshoot ANYTHING until you replace ALL, AND I DO MEAN
EVERY SINGLE ONE on every board, the caps and CLEAN
CLEAN CLEAN and CLEAN again the circuit boards.

Sure you didn't miss any newsgroups with your cross post????
 
All my experience is with the 540, but I think you're wasting your time
trying to troubleshoot ANYTHING
Click to expand...
OK, I suspected that, but non-functional buttons seemed a symptom not related
to caps issue. But I'll take your advice to heart and do *all* the caps.

until you replace ALL, AND I DO MEAN
EVERY SINGLE ONE on every board, the caps and CLEAN
CLEAN CLEAN and CLEAN again the circuit boards.
Click to expand...
Yeah, I plan to do that. Making a complete list right now.

What techniques and chemicals did you use?

Thanks for your reply.
Dave
 
DaveC wrote:
All my experience is with the 540, but I think you're wasting your time
trying to troubleshoot ANYTHING

OK, I suspected that, but non-functional buttons seemed a symptom not related
to caps issue. But I'll take your advice to heart and do *all* the caps.
Click to expand...
If the processor ain't working right, you can't count on anything.
until you replace ALL, AND I DO MEAN
EVERY SINGLE ONE on every board, the caps and CLEAN
CLEAN CLEAN and CLEAN again the circuit boards.

Yeah, I plan to do that. Making a complete list right now.

What techniques and chemicals did you use?
Simple green, scrub,scrub, scrub with toothbrush.
Click to expand...
Rinse, blow dry with high pressure air, repeat with isopropyl alcohol.
The 99% drugstore alcohol works well. Less water to dry out.
Check for crud between EVERY IC lead.
Gotta get it all out. Even the stuff under the
SMT chips. If any of the tented vias have their tent missing,
there's crud in the hole. Get it out or you'll have open vias.
Of course, poking around can CAUSE open vias. Use judgment.
Don't wash the front panel encoders, cover dip switches, the usual.

Dry, Dry, Dry the boards. I stick 'em in the oven at LOW temperature
for a while. If you can't hold the board comfortably, it's too hot.

I've had boards I was sure were squeaky clean. Had to wash 'em three
times before it passed self-test. Lots of high-impedance stuff
doesn't like conductive stuff on the board.

Open traces and vias are not unusual. Schematics unavailable. It's
a bitch to fix if you can't see an internal trace.

Stop by your favorite psychic and buy some luck...you'll need it.

Don't forget the calibration problem. Unless you have the procedure,
the software, the gpib controller, and the specific calibration
equipment, you may still end up with a scope that doesn't work.
Not sure about the 400 series, but the 540 cal procedure demands
that you use specific equipment. If the reference frequency
you use isn't the one required, it bitches and stops.
I had to modify a 067-0532 to move the reference frequency
so the scope would let me continue the calibration. I used a
power supply and DVM for the DC reference. Worked, but took
HOURS to do the procedure.

FWIW, this is definitely a hobby project. If you can get someone
to pay you for your time, you're better off taking the job and
using the money to buy a new scope.
Thanks for your reply.
Dave
Click to expand...
 
DaveC wrote:
All my experience is with the 540, but I think you're wasting your time
trying to troubleshoot ANYTHING

OK, I suspected that, but non-functional buttons seemed a symptom not related
to caps issue. But I'll take your advice to heart and do *all* the caps.

until you replace ALL, AND I DO MEAN
EVERY SINGLE ONE on every board, the caps and CLEAN
CLEAN CLEAN and CLEAN again the circuit boards.

Yeah, I plan to do that. Making a complete list right now.

What techniques and chemicals did you use?

Thanks for your reply.
Dave

On the subject of technique.
Click to expand...
You have significant risk of lifting pads when you remove the old caps.
The solder will be corroded. You need to "tin" each joint to expose
clean solder for heat transfer.
To remove caps, use two soldering irons. One on each end but on opposite
sides.
Gentle pressure provides a twisting motion that rotates
the cap off the pads. This is MUCH less stressful to the pads than
trying to pull upward on the cap.

I'd suggest you NOT try to use a heat gun unless you clean
the boards before removing the caps.
 
On the subject of technique.
Click to expand...
Mike, Thanks very much for your advice and experience. Not having done more
than a few SMD caps the techniques will be valuable.

Thanks again,
Dave
 
A little more info on my non-functioning button problem:

I disassembled the front of the scope -- bezel, buttons, front panel
assembly.

The elastomer button strip (the "rubber' buttons) have a conductive backing
which presses down onto the menu flex circuit (flex pcb).

The back of the button looks good (very light wear indications). The flex
circuit traces are not metal, which surprises me. It's made from pretty
fragile carbon-like traces.

When I measure the continuity of the closed contacts of a button that works,
I see approximately 200 ohms. Closed contacts of a button that doesn't work,
I see 400 ohms (sometimes more).

What resistance should I see at the contacts when the button is pressed?

Can I clean these contacts with alcohol? Or...?

Can the traces on the flex circuit be renewed (ie, silver-bearing paint)?

Thanks.
 
The method I use for cleaning those conductive pads on button switches
(keyboards, IR remote controls, etc) is very lightly scuffing them with a
pen-type tool that has strands of glass fibers in it.

It can be difficult to determine the "normal" resistance of these types of
contacts, because typically, the resistance decreases as more pressure is
applied to the button.

The button pads that are shiney, somewhat glossy, are the ones that need
scuffing, IME.
I lightly scuff the conducive pads just enough to remove the shine,
sometimes just a few passes across the pads.

I can never be certain how long the scuffing treatment will last. Most times
the problem is corrected for at least a couple of months to about a year,
maybe longer.

The tool I use was found in a camera department of a retail store years ago,
as a battery contact cleaning tool. I've read numerous comments that the
same type of tool is available at autobody shop supply places, as a scuffing
tool for small chips in in painted surfaces on car bodies.

I would also like to know if any conductive paint-type products work
effectively, and are somewhat permanent.

--
Cheers,
WB
..............

"DaveC" <invalid@invalid.net> wrote in message
news:0001HW.C78BD8D601BE4002B08A39AF@news.eternal-september.org...
A little more info on my non-functioning button problem:

I disassembled the front of the scope -- bezel, buttons, front panel
assembly.

The elastomer button strip (the "rubber' buttons) have a conductive
backing
which presses down onto the menu flex circuit (flex pcb).

The back of the button looks good (very light wear indications). The flex
circuit traces are not metal, which surprises me. It's made from pretty
fragile carbon-like traces.

When I measure the continuity of the closed contacts of a button that
works,
I see approximately 200 ohms. Closed contacts of a button that doesn't
work,
I see 400 ohms (sometimes more).

What resistance should I see at the contacts when the button is pressed?

Can I clean these contacts with alcohol? Or...?

Can the traces on the flex circuit be renewed (ie, silver-bearing paint)?

Thanks.
Click to expand...
 
On Sun, 31 Jan 2010 12:49:09 -0800 DaveC <invalid@invalid.net> wrote in
Message id: <0001HW.C78B2CC50195EC48B0B2D9AF@news.eternal-september.org>:

All my experience is with the 540, but I think you're wasting your time
trying to troubleshoot ANYTHING

OK, I suspected that, but non-functional buttons seemed a symptom not related
to caps issue. But I'll take your advice to heart and do *all* the caps.

until you replace ALL, AND I DO MEAN
EVERY SINGLE ONE on every board, the caps and CLEAN
CLEAN CLEAN and CLEAN again the circuit boards.

Yeah, I plan to do that. Making a complete list right now.

What techniques and chemicals did you use?
Click to expand...
I use Simple Green with a good non-metallic brush, then follow up with 90%
isopropyl alcohol. For the rubber membrane contacts VERY gently clean with
alcohol and pat dry.

On scope ACQ boards that have been badly contaminated I run 'em through
the dishwasher rinse cycle first with no soap. I position them vertically
which allows the sprayer to gut under the chips themselves, as well as
allowing the contaminant to run off. Then I put it in front of a fan for
an hour or so, after which I do the cleaning mentioned above.
 
On Sun, 31 Jan 2010 14:14:59 -0800 mike <spamme0@go.com> wrote in Message
id: <hk4vf4$jdr$1@news.eternal-september.org>:

DaveC wrote:
All my experience is with the 540, but I think you're wasting your time
trying to troubleshoot ANYTHING

OK, I suspected that, but non-functional buttons seemed a symptom not related
to caps issue. But I'll take your advice to heart and do *all* the caps.

until you replace ALL, AND I DO MEAN
EVERY SINGLE ONE on every board, the caps and CLEAN
CLEAN CLEAN and CLEAN again the circuit boards.

Yeah, I plan to do that. Making a complete list right now.

What techniques and chemicals did you use?

Thanks for your reply.
Dave

On the subject of technique.
You have significant risk of lifting pads when you remove the old caps.
The solder will be corroded. You need to "tin" each joint to expose
clean solder for heat transfer.
To remove caps, use two soldering irons. One on each end but on opposite
sides.
Gentle pressure provides a twisting motion that rotates
the cap off the pads. This is MUCH less stressful to the pads than
trying to pull upward on the cap.
Click to expand...
I've had 100% success just using a pair of wire cutters. I cut the caps
off where they dimple in at the bottom, then use a probe to remove the
remaining metal and plastic bit. After that, a soldering iron to remove
the remaining leads. When I cut the caps, I keep the blades parallel to
the leads to avoid any more stress than necessary.
 
DaveC <invalid@invalid.net> wrote:

This scope has the famous "FAIL ++ Acq" error which points to the infamous
failed (and leaking) SMD caps on the acquisition PCB.

I'm moving on to troubleshoot another (unrelated, I think -- and hope) issue.
Click to expand...
Tektronix has an excellent user forum covering these sort of repairs.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------
 
On Sun, 31 Jan 2010 12:49:09 -0800 DaveC <invalid@invalid.net> wrote in
Message id: <0001HW.C78B2CC50195EC48B0B2D9AF@news.eternal-september.org>:

All my experience is with the 540, but I think you're wasting your time
trying to troubleshoot ANYTHING

OK, I suspected that, but non-functional buttons seemed a symptom not
related
to caps issue. But I'll take your advice to heart and do *all* the caps.

until you replace ALL, AND I DO MEAN
EVERY SINGLE ONE on every board, the caps and CLEAN
CLEAN CLEAN and CLEAN again the circuit boards.

Yeah, I plan to do that. Making a complete list right now.

What techniques and chemicals did you use?

I use Simple Green with a good non-metallic brush, then follow up with 90%
isopropyl alcohol. For the rubber membrane contacts VERY gently clean with
alcohol and pat dry.
Click to expand...
Thanks JW for your reply.

Serious cleaning ahead!

Dave
 
The method I use for cleaning those conductive pads on button switches
(keyboards, IR remote controls, etc) is very lightly scuffing them with a
pen-type tool that has strands of glass fibers in it.
[...]
WB
Click to expand...
Thanks for that input Bill. I'll see if scrubbing lightly helps.

The flex pcb switch contacts are more what I suspect than the button. And
those are seriously fragile. It looks like a tiny bead of carbon-filled
rubber cement. I dented a trace (actually chipped off a *tiny* piece) with a
DMM test probe. Yikes!

Thanks again,
Dave
 
Tektronix has an excellent user forum covering these sort of repairs.
Click to expand...
Nothing I could find regarding the button switches.

Can you point me to a Tek forum topic about button contact repairs?

Thanks,
Dave
 
DaveC <invalid@invalid.net> wrote:

Tektronix has an excellent user forum covering these sort of repairs.

Nothing I could find regarding the button switches.

Can you point me to a Tek forum topic about button contact repairs?
Click to expand...
I'd register and post a question there.

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------
 
Wild_Bill wrote:
The method I use for cleaning those conductive pads on button switches
(keyboards, IR remote controls, etc) is very lightly scuffing them with
a pen-type tool that has strands of glass fibers in it.

It can be difficult to determine the "normal" resistance of these types
of contacts, because typically, the resistance decreases as more
pressure is applied to the button.

The button pads that are shiney, somewhat glossy, are the ones that need
scuffing, IME.
I lightly scuff the conducive pads just enough to remove the shine,
sometimes just a few passes across the pads.

I can never be certain how long the scuffing treatment will last. Most
times the problem is corrected for at least a couple of months to about
a year, maybe longer.

The tool I use was found in a camera department of a retail store years
ago, as a battery contact cleaning tool. I've read numerous comments
that the same type of tool is available at autobody shop supply places,
as a scuffing tool for small chips in in painted surfaces on car bodies.

I would also like to know if any conductive paint-type products work
effectively, and are somewhat permanent.

--
Cheers,
WB
.............

"DaveC" <invalid@invalid.net> wrote in message
news:0001HW.C78BD8D601BE4002B08A39AF@news.eternal-september.org...
A little more info on my non-functioning button problem:

I disassembled the front of the scope -- bezel, buttons, front panel
assembly.

The elastomer button strip (the "rubber' buttons) have a conductive
backing
which presses down onto the menu flex circuit (flex pcb).

The back of the button looks good (very light wear indications). The flex
circuit traces are not metal, which surprises me. It's made from pretty
fragile carbon-like traces.

When I measure the continuity of the closed contacts of a button that
works,
I see approximately 200 ohms. Closed contacts of a button that doesn't
work,
I see 400 ohms (sometimes more).

What resistance should I see at the contacts when the button is pressed?

Can I clean these contacts with alcohol? Or...?

Can the traces on the flex circuit be renewed (ie, silver-bearing paint)?

Thanks.
Click to expand...


Don't scuff them! You don't want to remove material, the problem is skin
oils or oils from the rubber that leach out and reduce the conductivity.
All you have to do is clean it thoroughly and it will work like new and
be fine for years.
 
I used to think that cleaning the conducive pads would solve key or button
problems, but the provblems returned quickly, several days to a week.
I tried using some of he milder solvens like 90% isopropyl alcohol and
denatured alcohol on swab tips, and may have even tried lacquer thinner, as
that would generally be my next choice. The problems kept returning.

For items like the rubbery button material in IR remote controls, the stuff
does get dirty and ooze oils from the plastic, but that was easy enough to
clean off.
Then I noticed all the faulty buttons conductive pads had a glossy
appearance identical to the pattern they're pressed against, so I gently
scuffed them just enuough to remove the shiney pattern, and they worked
perfectly for much longer periods.

They get shiney again after hundreds of presses, and when the faulty ones
were examined, they looked exactly the same as they did before they were
scuffed.
Buttons that are rarely used on keyboards, for example, keep working while
the most frequently used key conductive pads develop the glossy pattern.
After a very light scuffing, they work fine.

As I mentioned before, the very light scuffing procedure may involve just a
few passes across the conducive pad area, nothing any more severe than a few
brush strokes.

I'd still like to know of any recommendations for an effective conductive
paint-type product, for some of my older gear that I'd like to keep using
for many more years.

The best key switching design I've seen was on a vintage (maybe mid-1960s)
deskop calculator with Nixie tubes. The keyboard had a reed swich under each
key, and the hard keys had a small magnet in them.. very nice assembly, and
I expect that it was very reliable.

--
Cheers,
WB
..............


"James Sweet" <jamesrsweet@gmail.com> wrote in message
news:hk8jis$3r7$2@news.eternal-september.org...
Don't scuff them! You don't want to remove material, the problem is skin
oils or oils from the rubber that leach out and reduce the conductivity.
All you have to do is clean it thoroughly and it will work like new and be
fine for years.
Click to expand...
 
Wild_Bill wrote:
I used to think that cleaning the conducive pads would solve key or
button problems, but the provblems returned quickly, several days to a
week.
I tried using some of he milder solvens like 90% isopropyl alcohol and
denatured alcohol on swab tips, and may have even tried lacquer thinner,
as that would generally be my next choice. The problems kept returning.

For items like the rubbery button material in IR remote controls, the
stuff does get dirty and ooze oils from the plastic, but that was easy
enough to clean off.
Then I noticed all the faulty buttons conductive pads had a glossy
appearance identical to the pattern they're pressed against, so I gently
scuffed them just enuough to remove the shiney pattern, and they worked
perfectly for much longer periods.

They get shiney again after hundreds of presses, and when the faulty
ones were examined, they looked exactly the same as they did before they
were scuffed.
Buttons that are rarely used on keyboards, for example, keep working
while the most frequently used key conductive pads develop the glossy
pattern.
After a very light scuffing, they work fine.

As I mentioned before, the very light scuffing procedure may involve
just a few passes across the conducive pad area, nothing any more severe
than a few brush strokes.

I'd still like to know of any recommendations for an effective
conductive paint-type product, for some of my older gear that I'd like
to keep using for many more years.

The best key switching design I've seen was on a vintage (maybe
mid-1960s) deskop calculator with Nixie tubes. The keyboard had a reed
swich under each key, and the hard keys had a small magnet in them..
very nice assembly, and I expect that it was very reliable.

--
Cheers,
WB
.............
Click to expand...

I suppose the surface could wear over time, I've just never seen it
myself. Certainly try just cleaning before doing anything more drastic
though, liquid dish soap and warm water works as well as anything I've
tried.
 
On Sun, 31 Jan 2010 13:33:24 -0800, mike <spamme0@go.com> wrote:


Don't forget the calibration problem. Unless you have the procedure,
the software, the gpib controller, and the specific calibration
equipment, you may still end up with a scope that doesn't work.
Click to expand...
Assuming your scope does work, it might not be a bad idea to backup
the contents of the Dallas NVRAM module. It has an internal lithium
battery that is only rated to last about 10 years. Since many of
these scopes are approaching 20 years old, they could fail at any
time.

I was able to read the NVRAM from my TDS544 using an EPROM programmer.
Hopefully, when the battery does fail, I will be able to replace the
battery, and reprogram it from my backup. When I read it, I tied the
WP pin to the VCC pin so that there was no chance of corrupting it.

Andy Cuffe

acuffe@gmail.com
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top