PDA

View Full Version : Oscilloscope with ground pin on wall socket removed?



snuci
May 25th, 2014, 06:06 AM
Hi all,

I recently acquired a Tektronics 465B osilloscope and I'm in the process of learning how to use it. I actually need one to check memory pulses on a bad static ram memory board for a Heathkit H8 that I'm working on so I think this is the right tool. In watching some of the YouTube videos to learn how to use a scope, I did find a video about earth ground short warnings. This scope has the ground pin removed from the plug so I am wondering:

1. Does this affect the usage of the scope or is it mainly for safety?
2. Is this an old way of preventing earth ground shorting or somthing?

I am still trying to figure it out. It has the manual and some probes and parts and was able to check the probes with the calibration loop but I haven't done much further.

Thanks for any help.

offensive_Jerk
May 25th, 2014, 06:31 AM
I would just buy a replacement plug.

Stone
May 25th, 2014, 06:48 AM
It's for safety, not unlike all grounded devices or appliances.

Someone probably removed it because they didn't have a grounded outlet or couldn't take the time/trouble to buy an adapter to shunt the ground pin.

Of course, replacing the plug with a correct one can't hurt. :-)

Dwight Elvey
May 25th, 2014, 07:12 AM
It is for more than safety. If you have a circuit board that has a path
to ground, touching the ground lead to a signal lead can damage a
part on the board. There is enough capacitance in the line filter to
bring the ground lead of a scope to half the voltage of the line.
CMOS parts can be easily damage.
I recall having an almost complete lot of FET transistors wiped out
because someone thought the ground lead on soldering irons was
unneeded and it was cheaper to use a two lead extension cord.
Close to $3000 worth of transistors wiped out in one day.
This is with just the capacitance of the transformer in the soldering
iron base.
There are times when it is desired to float the ground of a scope
but I can't think of a reason one would want to do it on a H8.
Dwight

Tor
May 25th, 2014, 07:43 AM
I can only remember one case where a grounded scope caused a problem.. back in school, somebody set the ground pin of a scope to the chassis of a TV. Back then a TV chassis would be live, so instant short circuit. Boom. Fuses blew. Normally we avoided that by connecting the TV (IIRC) to a separate transformer that we had in the lab for that purpose.

Other than that, yes, as Dwight says. But it doesn't hurt to know exactly what it is you're connecting to.

-Tor

Chuck(G)
May 25th, 2014, 08:10 AM
There's another reason. Look up ground loop (http://en.wikipedia.org/wiki/Ground_loop_%28electricity%29)

g4ugm
May 25th, 2014, 11:21 AM
Its pretty common to run scopes on isolating transformers when de-bugging SMPS but you need to float the SMPS as well. Of course if you float the scope you don't want it grounded....

Some discussion on this here:-

http://electronics.stackexchange.com/questions/37143/isolation-transformer-and-scope-when-troubleshooting-smps

Also try a google search on "running scope isolating transformer" . Lots of hits...

xprt
May 25th, 2014, 11:25 AM
I recall Tektronix had an app note (http://www.tek.com/document/technical-brief/floating-oscilloscope-measurements-and-operator-protection) about floating measurements and strongly advising not to defeat the safety ground:

While the subject of this technical
note is about Floating
Measurements, some definitions
of terms and general precautions
must be understood
before proceeding. Historically,
floating measurements
have been made by knowingly
defeating the built-in safety
ground features of oscilloscopes
or measurement instruments
in various manners.
THIS IS AN UNSAFE AND
DANGEROUS PRACTICE AND
SHOULD NEVER BE DONE!

snuci
May 25th, 2014, 11:42 AM
Thanks for the replies. I just wanted to get some opinions and I'm glad I asked.

I was just wondering if this was some old school "cheap man's way" of isolating the ground on the scope and more importantly, what was the effect of having the ground missing even beyond safety. I did some prior research and asked this question after I saw this video on YouTube: EEVblog #279 - How NOT To Blow Up Your Oscilloscope! (https://www.youtube.com/watch?v=xaELqAo4kkQ) Great video that explains this very well.

I think it's pretty clear now that the plug on the scope needs a ground (and I will replace the plug). As described in that great video and by some of you here, an isolating transformer is a good idea; especially for a beginner like me.

On a side note, if someone here had anything to do with that video, I appreciated that information and it was presented very nicely. Thanks for posting that and thanks for the discussion.

Chuck(G)
May 25th, 2014, 11:50 AM
The fact that Tek would write about the practice illustrates that it was done--and why.

When you're measuring millivolts, there are often potential differences between two of the safety grounds in AC receptacles. Note that I'm not implying that there's a high voltage on any pin being measured. Yes, if you're flush, you can use a differential probe adapter, but I don't think I've ever run across one in the flesh. So, you remove the ground pin on the line plug (or use an isolation transformer) and use the signal ground, tied to both the safety ground and the scope, to avoid any induced signal in the the ground loop.

Even if two AC outlets are only a few feet apart, if a significant current is being drawn by any device downstream, you can get induced currents of a hundred mV between the ground lines of the outlets. One solution is to use the same (duplex) outlet for both the scope and the DUT, but that's not always possible.

Ground loops are very real and in low-level signals, a problem.

Tor
May 25th, 2014, 12:13 PM
Many years ago somebody at work was just about to connect the shielding of an RS-232 cable to ground at both ends because he thought it was a good idea. But somebody stopped him, because it's not. Particularly in this case as it turned out. The cable was very long and came from another building to our computer room, which was on a completely different electrical power system. There was about a 100V difference between the ground levels.. (230V land btw). The RS-232 didn't mind of course because of the optoisolators. As long as the shielding was connected at only one end.

snuci
May 25th, 2014, 12:37 PM
Thanks for the safety note from Tek. Looks like you aren't supposed to use an isolation transformer either. I'll check the probes that came with it but since many of you are still here to write about this topic, I'm guessing it's not super-risky as long as I'm careful wit the ground clip. If I am checking a memory chip, do I just measure between the ground pin and the pin I should be checking? Sounds like making sure both the scope and the H8 are plugged into the same outlet helps too?

g4ugm
May 25th, 2014, 01:03 PM
I would make sure you replace the plug with one with a ground pin before making measurements, and that you check where you connect the ground on the probe. I should have included that as a proviso when pointing you at the other articles.

Chuck(G)
May 25th, 2014, 02:08 PM
I would make sure you replace the plug with one with a ground pin before making measurements, and that you check where you connect the ground on the probe. I should have included that as a proviso when pointing you at the other articles.

You may simply be able to replace the cord assembly if it's one of the older Tek models. There's a place in BC that sells them.

Corey986
May 25th, 2014, 02:51 PM
In case anyone really needs to measure using a floating ground, a battery operated scope is the correct way to go. I have a hand held 60mhz DSO just for that. Too dangerous to bypass the ground on a regular scope.

Cheers,
Corey

Chuck(G)
May 25th, 2014, 02:58 PM
If you have the cash, a differential adapter is also quite useful exempli gratia (http://www.home.agilent.com/en/pd-2317313-pn-N2819A/800-mhz-101-differential-probe-with-autoprobe?nid=-34022.1066109&cc=US&lc=eng)