TCP202 with 15mA of ~150KHz noise.


Andy Warner
 

Last week I posted this to the tekprobes list and did not get any responses
(it is a much smaller and quieter list) - apologies in advance for the dups
to any dual subscribers.

I just came into a TCP202 probe, that seems to sense both AC and DC
currents reasonably well but also has about 15mA p-p of noise at around
150KHz: https://photos.app.goo.gl/h3x8JYphgRcznbfX9

When I hit degauss, the noise goes away for a few seconds after the initial
excursion, but returns.

The reason I think most of the probe works is that testing with DC at high
currents shows it working (though it looks like it might need a cal and
hooking it up to the 5mA @ 1KHz loop on my PG506 seems to show a square
wave under all the noise.)

Is this a common/known failure mode ? I'm familiar with "most TCP202
problems lead to the ferrite" situation, and the faces of the ferrite do
look like they would benefit from a full cleaning, but I'm hoping this
isn't a terminally broken device - I cannot seem to find the usual Tek
manuals online for the TCP202, sadly.


I do have a A6302 + AM503 that I have re-built and seem to work well,
wondering how similar/different the A6302 probe is from the TCP202 probe,
clearly the case is identical, and I suspect the magnetics may be also,
which may give me some chance at compare and contrast between the two.

Looking for any experience/advice people have to offer beyond, "maybe the
magnetics are bad".
--
Andy


Mark Litwack
 

Hi Andy,

I also experience high frequency noise on my TCP202 when using a TDS3054. In my case, it was about 20mA p-p and around 50kHz.

Many years ago I reported this to Tektronix, and they confirmed that it was caused by the switching power supply in the TDS3000 series. It seems that the scope's power supply is not filtered well enough. I do not see the noise when using the same TCP202 on an 1103 TekProbe power supply, which is all linear.

Try probing the power supply contacts on one of the TekProbe connectors on the front of your scope. You might find your 150kHz there.

-mark


Andy Warner
 

Thank you, it is indeed a TDS3000 that I am using. Darn, I appreciate the
automatic mode changes and scale annotation (including cursors and
measurements).

Maybe I can add some additional power supply filtering inside the case.
Have to break it open and take a look at my options.

On Thu, Sep 9, 2021, 08:02 Mark Litwack <mlitwack@emeraldelec.com> wrote:

Hi Andy,

I also experience high frequency noise on my TCP202 when using a TDS3054.
In my case, it was about 20mA p-p and around 50kHz.

Many years ago I reported this to Tektronix, and they confirmed that it
was caused by the switching power supply in the TDS3000 series. It seems
that the scope's power supply is not filtered well enough. I do not see
the noise when using the same TCP202 on an 1103 TekProbe power supply,
which is all linear.

Try probing the power supply contacts on one of the TekProbe connectors on
the front of your scope. You might find your 150kHz there.

-mark







Mark Litwack
 

I tried adding some filtering from the main power supply to the main board. Nothing too fancy; just a couple of capacitors across the rails. It reduced the TCP202 noise by maybe a factor of 2.

I had some single-shot measurements I needed to get done (so averaging wasn't going to help), so I bought the 1103 supply.

If you're seeing 150kHz, you might be dealing with noise from POL switchers on the main board in which case you might need more filtering directly at the TekProbe connectors. Alternatively, you could try adding some filtering inside the TCP202 compensation box if you can squeeze it in there. I had to do this with a P5205 differential probe, which is also susceptible to probe power noise.

If you have success, please post back! Thanks,

-mark

On Thu, Sep 9, 2021 at 09:16 PM, Andy Warner wrote:

Thank you, it is indeed a TDS3000 that I am using. Darn, I appreciate the
automatic mode changes and scale annotation (including cursors and
measurements).

Maybe I can add some additional power supply filtering inside the case.
Have to break it open and take a look at my options.

On Thu, Sep 9, 2021, 08:02 Mark Litwack <mlitwack@emeraldelec.com> wrote:

Hi Andy,

I also experience high frequency noise on my TCP202 when using a TDS3054.
In my case, it was about 20mA p-p and around 50kHz.

Many years ago I reported this to Tektronix, and they confirmed that it
was caused by the switching power supply in the TDS3000 series. It seems
that the scope's power supply is not filtered well enough. I do not see
the noise when using the same TCP202 on an 1103 TekProbe power supply,
which is all linear.

Try probing the power supply contacts on one of the TekProbe connectors on
the front of your scope. You might find your 150kHz there.

-mark







Andy Warner
 

My P6248 diff probe doesn't show any noise when hooked up to the same
scope, so I was not expecting this; but just learning that it is a known
problem is half the battle for me.

I do not plan to crack open the case of the scope itself, unless I *really*
have to, so my plan would be to try and fit more decoupling into the TCP202
enclosure that hooks to the scope.

Sadly, I cannot find schematics of the TCP202, so I'll have to reverse
engineer it, but I'm going to assume for now that the +/- 15V rails are the
ones that I should focus on.

I will for sure report back.

On Fri, Sep 10, 2021 at 10:34 AM Mark Litwack <mlitwack@emeraldelec.com>
wrote:

I tried adding some filtering from the main power supply to the main
board. Nothing too fancy; just a couple of capacitors across the rails.
It reduced the TCP202 noise by maybe a factor of 2.

I had some single-shot measurements I needed to get done (so averaging
wasn't going to help), so I bought the 1103 supply.

If you're seeing 150kHz, you might be dealing with noise from POL
switchers on the main board in which case you might need more filtering
directly at the TekProbe connectors. Alternatively, you could try adding
some filtering inside the TCP202 compensation box if you can squeeze it in
there. I had to do this with a P5205 differential probe, which is also
susceptible to probe power noise.

If you have success, please post back! Thanks,

-mark

On Thu, Sep 9, 2021 at 09:16 PM, Andy Warner wrote:

Thank you, it is indeed a TDS3000 that I am using. Darn, I appreciate the
automatic mode changes and scale annotation (including cursors and
measurements).

Maybe I can add some additional power supply filtering inside the case.
Have to break it open and take a look at my options.

On Thu, Sep 9, 2021, 08:02 Mark Litwack <mlitwack@emeraldelec.com>
wrote:

Hi Andy,

I also experience high frequency noise on my TCP202 when using a
TDS3054.
In my case, it was about 20mA p-p and around 50kHz.

Many years ago I reported this to Tektronix, and they confirmed that it
was caused by the switching power supply in the TDS3000 series. It
seems
that the scope's power supply is not filtered well enough. I do not
see
the noise when using the same TCP202 on an 1103 TekProbe power supply,
which is all linear.

Try probing the power supply contacts on one of the TekProbe
connectors on
the front of your scope. You might find your 150kHz there.

-mark










--
Andy


Mark Litwack
 

There isn't a published TCP202 schematic I'm aware of, but there is a partial schematic from someone who did some reverse engineering:

https://www.eevblog.com/forum/repair/tektronix-tcp202-current-probe-repair-schematic-and-suggestions-needed/msg3358464/#msg3358464

It uses +/-15V and +/-5V from the TekProbe connector. It looks like the +-/15V supply is probably the one to tackle first.

I do see noise on a P6248. And when plugged in at the same time as the TCP202, I can see the noise is correlated between the two probes. But oddly, it's at 100kHz (exactly double) on the P6248. I hadn't looked at this before.

Best of luck on your hunt,

-mark

On Fri, Sep 10, 2021 at 11:54 AM, Andy Warner wrote:


My P6248 diff probe doesn't show any noise when hooked up to the same
scope, so I was not expecting this; but just learning that it is a known
problem is half the battle for me.

I do not plan to crack open the case of the scope itself, unless I *really*
have to, so my plan would be to try and fit more decoupling into the TCP202
enclosure that hooks to the scope.

Sadly, I cannot find schematics of the TCP202, so I'll have to reverse
engineer it, but I'm going to assume for now that the +/- 15V rails are the
ones that I should focus on.

I will for sure report back.

On Fri, Sep 10, 2021 at 10:34 AM Mark Litwack <mlitwack@emeraldelec.com>
wrote:

I tried adding some filtering from the main power supply to the main
board. Nothing too fancy; just a couple of capacitors across the rails.
It reduced the TCP202 noise by maybe a factor of 2.

I had some single-shot measurements I needed to get done (so averaging
wasn't going to help), so I bought the 1103 supply.

If you're seeing 150kHz, you might be dealing with noise from POL
switchers on the main board in which case you might need more filtering
directly at the TekProbe connectors. Alternatively, you could try adding
some filtering inside the TCP202 compensation box if you can squeeze it in
there. I had to do this with a P5205 differential probe, which is also
susceptible to probe power noise.

If you have success, please post back! Thanks,

-mark

On Thu, Sep 9, 2021 at 09:16 PM, Andy Warner wrote:

Thank you, it is indeed a TDS3000 that I am using. Darn, I appreciate the
automatic mode changes and scale annotation (including cursors and
measurements).

Maybe I can add some additional power supply filtering inside the case.
Have to break it open and take a look at my options.

On Thu, Sep 9, 2021, 08:02 Mark Litwack <mlitwack@emeraldelec.com>
wrote:

Hi Andy,

I also experience high frequency noise on my TCP202 when using a
TDS3054.
In my case, it was about 20mA p-p and around 50kHz.

Many years ago I reported this to Tektronix, and they confirmed that it
was caused by the switching power supply in the TDS3000 series. It
seems
that the scope's power supply is not filtered well enough. I do not
see
the noise when using the same TCP202 on an 1103 TekProbe power supply,
which is all linear.

Try probing the power supply contacts on one of the TekProbe
connectors on
the front of your scope. You might find your 150kHz there.

-mark










--
Andy


Mark Litwack
 

I'll add one additional comment that in looking at the noise with FFT, there's a strong 88kHz component in there as well on both probes, so there's likely multiple noise sources getting into the probe power supply. And the 50kHz is still there on the P6248, but lower in magnitude.

-mark

On Fri, Sep 10, 2021 at 01:25 PM, Mark Litwack wrote:


There isn't a published TCP202 schematic I'm aware of, but there is a partial
schematic from someone who did some reverse engineering:


https://www.eevblog.com/forum/repair/tektronix-tcp202-current-probe-repair-schematic-and-suggestions-needed/msg3358464/#msg3358464

It uses +/-15V and +/-5V from the TekProbe connector. It looks like the
+-/15V supply is probably the one to tackle first.

I do see noise on a P6248. And when plugged in at the same time as the
TCP202, I can see the noise is correlated between the two probes. But oddly,
it's at 100kHz (exactly double) on the P6248. I hadn't looked at this before.

Best of luck on your hunt,

-mark


On Fri, Sep 10, 2021 at 11:54 AM, Andy Warner wrote:


My P6248 diff probe doesn't show any noise when hooked up to the same
scope, so I was not expecting this; but just learning that it is a known
problem is half the battle for me.

I do not plan to crack open the case of the scope itself, unless I *really*
have to, so my plan would be to try and fit more decoupling into the TCP202
enclosure that hooks to the scope.

Sadly, I cannot find schematics of the TCP202, so I'll have to reverse
engineer it, but I'm going to assume for now that the +/- 15V rails are the
ones that I should focus on.

I will for sure report back.

On Fri, Sep 10, 2021 at 10:34 AM Mark Litwack <mlitwack@emeraldelec.com>
wrote:

I tried adding some filtering from the main power supply to the main
board. Nothing too fancy; just a couple of capacitors across the rails.
It reduced the TCP202 noise by maybe a factor of 2.

I had some single-shot measurements I needed to get done (so averaging
wasn't going to help), so I bought the 1103 supply.

If you're seeing 150kHz, you might be dealing with noise from POL
switchers on the main board in which case you might need more filtering
directly at the TekProbe connectors. Alternatively, you could try adding
some filtering inside the TCP202 compensation box if you can squeeze it in
there. I had to do this with a P5205 differential probe, which is also
susceptible to probe power noise.

If you have success, please post back! Thanks,

-mark

On Thu, Sep 9, 2021 at 09:16 PM, Andy Warner wrote:

Thank you, it is indeed a TDS3000 that I am using. Darn, I appreciate
the
automatic mode changes and scale annotation (including cursors and
measurements).

Maybe I can add some additional power supply filtering inside the case.
Have to break it open and take a look at my options.

On Thu, Sep 9, 2021, 08:02 Mark Litwack <mlitwack@emeraldelec.com>
wrote:

Hi Andy,

I also experience high frequency noise on my TCP202 when using a
TDS3054.
In my case, it was about 20mA p-p and around 50kHz.

Many years ago I reported this to Tektronix, and they confirmed that
it
was caused by the switching power supply in the TDS3000 series. It
seems
that the scope's power supply is not filtered well enough. I do not
see
the noise when using the same TCP202 on an 1103 TekProbe power supply,
which is all linear.

Try probing the power supply contacts on one of the TekProbe
connectors on
the front of your scope. You might find your 150kHz there.

-mark










--
Andy


 

On Fri, Sep 10, 2021 at 07:40 PM, Mark Litwack wrote:


I'll add one additional comment that in looking at the noise with FFT, there's
a strong 88kHz component in there as well on both probes, so there's likely
multiple noise sources getting into the probe power supply.
I checked my TDS3000B/TCP202 combination. In normal ‘scope mode, all I see is an 88 kHz waveform at about 2 mA pp.

Raymond


Andy Warner
 

Thank you for the datapoint, I wish I were so lucky.

On Fri, Sep 10, 2021 at 16:08 Raymond Domp Frank <hewpatek@gmail.com> wrote:

On Fri, Sep 10, 2021 at 07:40 PM, Mark Litwack wrote:


I'll add one additional comment that in looking at the noise with FFT,
there's
a strong 88kHz component in there as well on both probes, so there's
likely
multiple noise sources getting into the probe power supply.
I checked my TDS3000B/TCP202 combination. In normal ‘scope mode, all I see
is an 88 kHz waveform at about 2 mA pp.

Raymond





--
Andy


Andy Warner
 

I tried adding some ferrites with decent impedance in the ~100KHz range to
the +/- 15V rails:
https://photos.app.goo.gl/RaxHFTVdvF6T7LAJA

No significant change, next step repeat for the +/- 5V rails, and if that
doesn’t help, move on to augmenting the decoupling…

On Fri, Sep 10, 2021 at 16:17 Andy Warner <andyw@pobox.com> wrote:

Thank you for the datapoint, I wish I were so lucky.

On Fri, Sep 10, 2021 at 16:08 Raymond Domp Frank <hewpatek@gmail.com>
wrote:

On Fri, Sep 10, 2021 at 07:40 PM, Mark Litwack wrote:


I'll add one additional comment that in looking at the noise with FFT,
there's
a strong 88kHz component in there as well on both probes, so there's
likely
multiple noise sources getting into the probe power supply.
I checked my TDS3000B/TCP202 combination. In normal ‘scope mode, all I
see
is an 88 kHz waveform at about 2 mA pp.

Raymond





--
Andy





--
Andy