Topics

Scope grounding and generators


Chuck Harris
 

The filament transformer complete isolates the primary from
the secondary electrically. So, I put a household plug on
the 120V side, and usually use a Pomona binding post to male
BNC adapter for the low voltage secondary.

Nothing to stop you from simply soldering a piece of RG58 that
is only terminated on one end, or soldering a BNC female to the
transformer, and using a BNC-BNC cable to connect that to the
scope.

-Chuck Harris

Robert Simpson via groups.io wrote:

Chuck,
How do you connect your probe leads to the transformer output. Isn't the output side now independent of a ground reference? At my workbench, I use a 7A22 with two probes in differential setup. But my 465M is what I am using outside near the generator.
I found a 120 to 34V transformer I can use.
Bob






Robert Simpson
 

Chuck,
How do you connect your probe leads to the transformer output. Isn't the output side now independent of a ground reference? At my workbench, I use a 7A22 with two probes in differential setup. But my 465M is what I am using outside near the generator.
I found a 120 to 34V transformer I can use.
Bob


Chuck Harris
 

I usually just leave the 1M scope load to set the
output impedance.

But, since it is easy, try a few different values,
and see what looks the best.

50 ohms would not be a good idea, as it would draw
almost a watt from a 6.3V transformer.

-Chuck

Robert Simpson via groups.io wrote:

Some results:
1. My true RMS meter arrived, so I ran a few basic tests with the meter and the 465M.
After starting the generator and letting it warm up, I connected a 500W shop lamp for a resistivity load. TRMS meter showed about 130 V
Hooked the scope up one lead and took some pictures. Unfortunately although I was in tree shade, general glare resulted in marginal pictures. Using Photoshop I adjusted the brightness and color and got a just readable image
Looks like period is approximately 16ms or about 62.5 HZ.
Peaky barely sine wave

2. I think I have a 120 to 12V transformer. So I might try some tests with motor loads, such as my table saw and hand electric saw.
Question, should the output leads of the transformer be open or have a light load say 10K oms?
Bob






Michael W. Lynch
 

If you are reading 62.5 Hz, then your engine is running slightly fast. Generators typically run at 3600 or 1800 (depends on whether they use a 2 pole or 4 pole rotor). Check your frequency at full rated load, I would guess that it will be closer to correct. Most mechanical governors have a speed adjustment that will allow slight changes to the full load RPMs to correct frequency at full load. Better generators, like HONDA and Yamaha have better governor calibration as well as better regulation.

--
Michael Lynch
Dardanelle, AR


Robert Simpson
 

Some results:
1. My true RMS meter arrived, so I ran a few basic tests with the meter and the 465M.
After starting the generator and letting it warm up, I connected a 500W shop lamp for a resistivity load. TRMS meter showed about 130 V
Hooked the scope up one lead and took some pictures. Unfortunately although I was in tree shade, general glare resulted in marginal pictures. Using Photoshop I adjusted the brightness and color and got a just readable image
Looks like period is approximately 16ms or about 62.5 HZ.
Peaky barely sine wave

2. I think I have a 120 to 12V transformer. So I might try some tests with motor loads, such as my table saw and hand electric saw.
Question, should the output leads of the transformer be open or have a light load say 10K oms?
Bob


Roy Thistle
 

On Thu, Oct 29, 2020 at 07:18 PM, Greg Muir wrote:


So, I am now ending... with no further comments.[about the discussion so far.]
Here is a crude summary of the discussion so far:
OP (Robert) Is there a safe way to use my 465M to look at my generator output?
Chuck Use a filament transformer
Mike Check your grounds, and use the probe tip (only)
Stevenhorii Check the polarity of the outlets
Mac Check the grounding for stray voltage
Chuck Check the grounding
tenareze32 reports his generator damaged his coffee maker
Jean-Paul claims most generators have a distorted output/ use a Tek 212/ or float the 465
Glen: Check the grounding
Chuck: Don't float the 465
-: reports his generator damaged appliances/ use a constant voltage transformer
Tony: wonders if newer DSOs are isolated/thinks experienced people make grounding mistakes
@Oculus: recommends a Fluke Scopometer
Dave: thinks Tek 222, 222PS, and 224 are appropriate
OP (Robert) relates he has Briggs & Stratton Powermate 5000 (but didn't mention its used)
Carsten: use a differential probe
Greg: says be careful, experienced engineers have got it wrong/ put themselves in harms way/ he has a power analyzer
Roy: suggests a UPS can be used to do a crude check Robert's generator/mentions THD/potential backfeeds/questions the benefit of using a scope to measure THD/recommends an inverter generator
OP(Robert) generator is for temporary use/ wants to understand what the generator is outputing
Jean-Paul: recommends a SMPS (inverter) generator
Harrison: wants to know whether portable generators can damage modern electronics
Chuck:explains difference between contractor and standby generators
Greg:thinks a distortion analyzer is too complex a solution/ test under varying load/ UPS will trip on generators?bigger generator the better
Roy: Small portable Hot Dog stand owners use Honda inverter generators to power business and TV
Ed:cheap transformers saturate easily/ use a good step-down transformer
Roy: recommends a TRMS DMM
Greenboxmaven: get a used military generator
OP(Robert): reports he's army trained/following grounding suggestions/ and not familiar with generators.
Roy: RMS and TRMS meters/OP's generator won't be outputting a sine wave/Robert's old Fluke doesn't have MAX MIN or TRMS
OP(Robert): buys a cheap TRMS
Greg: Posts links about generators and power quality
Chuck: Claims UPS is insufficient
Stephen: Suggests problems could be improper generator field voltage
Paul:Suggests UPS has short run time/developing an electronic governor
Roy: Get a robust UPS
Chuck: scraps thousands of UPS/there are different grades of UPS, according to run-time limitations
Roy: explains what/why use a UPS
Greg: early UPS had square wave output/recommends taken measurements to avoid damage
Robers: Asks about Tek HV probes
Stevenhorii: Suggest optoisolators
Chuck: thinks Tek HV probes are appropriate/465 has sufficient isolation when using them/explains stiff power sources/only low BW needed for low band-limited signals
Roy: claims optoisolators are for isolation
Roy: wondering where there could be DC present on the AC output of a damaged generator/use a DMM first
Greg: relates engineers were mislead by making measurements with an un-shielded DMM in an RF near field/ recommend a low bandwidth analog meter for such cases.
Roy: recommends not using a Tek 465 to measure generator outputs/thinks the discussion so far is appropriate.
Greg: relates that, posts wander off topic/posters are competitive/has made power line measurements with Tek Scope, pigtail, and probes


Greg Muir
 

Roy,

I don’t get into spitting matches here so will be brief.

I use the digest mode of receiving posts due to the wanderings of some subjects getting so far off of track that they eventually end up being in another universe. The measurement and analysis if a simple AC waveform is so easy it is ridiculous to come up with an extremely complex approach. In addition the amount of gyrations discussed reminded me of a term we use to use in engineering called “analysis paralysis” where the project came to a standstill until everyone on the staff had beat it to death analytically with little additional results.

I wholeheartedly understand that there are a wide variety of people on this site who range all the way from beginners to very advanced professionals. And I also realize that the neophytes need nurturing and education on subjects not to mention learning about safe approaches to taking measurements and/or servicing higher level equipment. To that end I try to help. But I also try not to cloud the issue with a multitude of variations or discussions in an effort to try to keep things clear about the subject nor veer off into the politics of the world or other unrelated subjects wholly improper to this group.

I often get the impression that there is more of an air of competition on these sites rather than simply addressing subjects in the most understandable ways to help others. I may be wrong but when topics become hydras and grow like mold on a piece of fruit the original intention seems to become lost for many simple subjects.

I can’t remember how many times I simply connected a scope probe to a pigtail lead with an AC plug on the other end and took a quick look at the waveform. And, yes, a Tek scope was used all of the time with Tek probes. In the event of encountering a portable generator no hocus pocus was performed with having to deal with ground systems in order to make those measurements. And, as I had mentioned in my last post, I have since acquired specific purpose equipment which adds another layer of analysis depth if needed.

So, I am now ending my rant with no further comments.

Regards,

Greg


Roy Thistle
 

On Thu, Oct 29, 2020 at 01:04 PM, Greg Muir wrote:


No oscilloscope or need for any esoteric support hardware required.
Well... I'd rephrase that as "No service type oscilloscope (in the context of TekScopes) should be used; or, no need for any some-what esoteric support hardware required.)
It's been suggested that using an oscilloscope [456], to scope the OP's generator output, was a "reasonable" thing to do.
To me... that proposition was a "wait just a minute while I think about it" moment... and ever since the thread has bifurcated to posts for doing it (scoping the output... albeit with cautions)... and other posts about... what essentially it boils down to me... waiting a minute. With... I'll guest, the OP wondering what's going on.
From what I read... everyone posting is attempting to come up with some way to solve the problem... either pro or con... but, everyone is making an honest attempt.
And, that's what makes TekScopes, and the people on TekScopes great: people doing there best to help other people.


Greg Muir
 

I understand that it has been suggested that this discussion be moved to SMPSs but this post applies to AC quality, not power supplies.

The only problems I ever witnessed regarding anything over basic power line frequency artifacts was once when the local utility company had to move a couple of 3-phase power poles going to a broadcast transmitter site. I responded due to equipment I was responsible for at that site and that the 150 kW backup generator had to come online while the utility power was down. When the transfer to generator had been done that morning I requested the utility crew to give me a call when the job was completed before they were going to reestablish utility power.

The day was moving into late afternoon and I hadn’t receive a call so I responded back to the site to see what was going on. Upon arrival I found several more linemen plus a few gentlemen dressed in casual wear standing around below the bucket trucks. It turned out that when the job was complete the linemen checked the output of the pole pigs and found one with a significant out-of-tolerance voltage level. They replaced the pole transformer and still found that the problem existed. That was when they called the engineering department (guys in casual wear) to come up and look at the problem. But now answer had been found by the time I arrived.

I asked one of the linemen to hand me the DMM he was using to check the voltages. It turned out to be a cheaper yet reasonable brand made-in-China meter. I grabbed my Simpson 260 meter out of the truck, handed it to him and told him to go back up and check voltages again. The result was good – all voltages were correct. A second test with a known-behavior Fluke DMM I gave him also proved to be good.

I then gathered the group around and gave them a quick lesson about high-RF level sites and the effects on certain digital instruments. I asked them if they had any old Simpson 260s around the plant to which they said they had some on the shelf in the warehouse because they didn’t use them anymore. I suggested that they keep one in their trucks when working on sites like the present one.

Obviously above-ground power lines also work as excellent receive antennas. I have witnessed other engineers scratching their heads when working in other high RF power sites finding totally off-center voltage and resistance readings and not understanding why they were happening. Again more education became necessary.

Any concern over significantly high frequency effects from generators has never been seen. I never worry about them if they are even present but the output impedance of a typical generator is so low that any effects are minimal. It’s the distortion that may get you but, again, I have never had any problems with it.

Aside from other dedicated AC test equipment I also have an older BMI 4800 PowerScope that I picked up for $100 with all of the test accessories a few years ago. It had been in possession by the engineering plant down at the NORAD Cheyenne Mountain facility. When power quality is in question I simply hook it to the AC source (up to 3 phases) and print out the AC waveform. Knowing what it can do I can see why they had it. This particular model unit was obsoleted due to a Y2K issue of the internal calendar not being able to increment beyond the year 2000. Otherwise it is a perfectly good unit to observe AC waveforms and disturbances. And the price is right with many being found on that big auction site.

No oscilloscope or need for any esoteric support hardware required.

Greg


Roy Thistle
 

On Thu, Oct 29, 2020 at 07:54 AM, Chuck Harris wrote:


That means, you use the scope's "ADD" mode, and set CH2 to
invert. Both channel attenuators are set to the same value.
You can... but beware that beside the single input maximum voltage rating... there is also the channel to channel (input to input) maximum. (Tektronix gives the input to input isolation a 100:1... but, that refers only to cross-talk? )
We had the input(s) of an expensive scope damaged by using two channels to measure a stiff signal differentially. (That was in an SMPS) (The issue was that a good differential probe was available. The OP probably doesn't have one of those.)
If it were my used generator, I'd use a DMM to see if it is preforming to spec (at least by that commonly used tool), before I'd scope it.


Roy Thistle
 

On Thu, Oct 29, 2020 at 07:59 AM, Chuck Harris wrote:


what you want to see...If you want to know the distortion of the "sine wave", you only
need at most DC to a few kilohertz of bandwidth.
If the OP uses isolation (say a transformer, or capacitor)... that will block the D.C. component... but if the OP wanted to investigate for D.C. ...then more modern multi-meters let you measure DC in the presence of A.C. (If anyone tries that with their DMM, make sure you read the manual!)
Why/how there could be DC present on the output of the OP's AC generator?... maybe the AVR?


Roy Thistle
 

On Thu, Oct 29, 2020 at 07:26 AM, stevenhorii wrote:

...wonder whether optical isolation would work...designed for AC input would likely preserve any higher-frequency components of the input signal...noted that opto[-]isolators [optoisolator] are used in medical equipment...
optoisolators...yes, most probably... especially newer ones. They use them in the inputs of some power line analyzers .et al. But, now we are far from the OPs methodology, I believe.
yes, optoisolators are much used in medical equipment, for isolation... much favoured over transformer isolation.
But, is the OP trying to see the transients? (so he/she knows they are there!)... or just trying to block them?... if blocking, and if they are there (somewhat doubtful, I believe cursorily)... then why is the OP scoping... that's usually done to see as much of the true signal as possible. I'd want to know if there are high voltage transients present (my radios thank me in advance... not to mention an expensive DSO!)


Chuck Harris
 

You have to ask yourself what you want to see.

If you want to know the distortion of the "sine wave", you only
need at most DC to a few kilohertz of bandwidth.

If you want to know the RF on the generator's output, then stop
looking at the 60Hz component, and put a capacitor in series with
the generator, and look only at the HF components.

-Chuck Harris

stevenhorii wrote:

I wonder whether optical isolation would work. The photodiode types have
very fast response, but can be nonlinear. Though usually used for logic
signals, there are optocouplers designed for analog signals. Some have a
more complex circuit that compensates for diode nonlinearity. For an AC
input, there are some with two diodes so conduction will occur across one
of the diodes during the AC cycle.

A really fast response optocoupler designed for AC input would likely
preserve any higher-frequency components of the input signal.

An opto isolator is likely a lot less expensive than a scope so destroying
it from overvoltage input would save, at a minimum, having to replace scope
input overvoltage protection elements.

Just a thought. I have noted that opto isolators are used in medical
equipment so the patient is isolated from the high voltages that may be
present in the instruments.


On Thu, Oct 29, 2020 at 02:06 Roy Thistle <roy.thistle@mail.utoronto.ca>
wrote:

On Wed, Oct 28, 2020 at 09:15 PM, Robert Simpson wrote:


I have P6009 probes.
P6009 is rated to 1.5 KV. And for 60 Hz nominal frequency doesn't need to
be derated. (It's flat to around 100 KHz.) The can be compensated to the
465's input capacitance.
So yes and no.
Your generator isn't new, so if there is a problem with the AVR, the
voltage could rise significantly...there could be an over-voltage
problem... but, surely not so much that a P6009 wouldn't protect the 465
inputs.
I'd be more worried about high voltage transients finding their way into
the scope inputs.
I'd be more concerned about that when the generator is started, stopped,
stalls, or is driving inductive loads.
Maybe run a couple of trouble lights, fitted with incandescent bulbs, when
the generator has been running well for a while, and is well warmed up.
(The bulbs are a very crude indication of frequency drift, and voltage
drift, or over-voltage, or under-voltage.)
I'd do a 465/P6009 measurement on the above... if I had to.










Chuck Harris
 

They should be ok, as long as you are using only the probe
tips, and not the grounding clips to connect to the line and
neutral leads of the generator. The probe grounding clips
should connect to the frame ground for the generator, which
should be connected to the same ground as your scope's chassis
ground.

That means, you use the scope's "ADD" mode, and set CH2 to
invert. Both channel attenuators are set to the same value.

Your scope can survive pretty much anything that your DVM
can handle... as long as you don't exceed the scope's rating
of 500V (DC + Peak AC) derated with frequency: 30V to 50MHz,
and 27V to 100MHz.

Usually, my recommendation is don't exceed the scope's maximums
regardless of the probes "enhancements" to those maximums, but
obviously there have to be exceptions to that very conservative
rule.

At issue is if the probe breaks down internally, it can damage
the scope through over voltage, if the probe is exposed to
voltages that exceed the scope's maximums.

Filament transformers are cheaper than a repair on your scope's
front end, and will reveal everything that matters on your
generator's output signal.

A second thing about instrumentation. When you are measuring
stiff power sources, you need to take stiff protection measures.
When you are measuring weak power sources, weak protection
measures will suffice.

Your generator is a fairly stiff power source. It can dump
way more than 4000W into a dead short circuit.

-Chuck Harris

Robert Simpson via groups.io wrote:

Getting back to a Tektronix question, I have P6009 probes. Wouldn't using those be safe from over voltage outputs (If there are any) of a generator to the 465M oscilloscope?
Bob






stevenhorii
 

I wonder whether optical isolation would work. The photodiode types have
very fast response, but can be nonlinear. Though usually used for logic
signals, there are optocouplers designed for analog signals. Some have a
more complex circuit that compensates for diode nonlinearity. For an AC
input, there are some with two diodes so conduction will occur across one
of the diodes during the AC cycle.

A really fast response optocoupler designed for AC input would likely
preserve any higher-frequency components of the input signal.

An opto isolator is likely a lot less expensive than a scope so destroying
it from overvoltage input would save, at a minimum, having to replace scope
input overvoltage protection elements.

Just a thought. I have noted that opto isolators are used in medical
equipment so the patient is isolated from the high voltages that may be
present in the instruments.


On Thu, Oct 29, 2020 at 02:06 Roy Thistle <roy.thistle@mail.utoronto.ca>
wrote:

On Wed, Oct 28, 2020 at 09:15 PM, Robert Simpson wrote:


I have P6009 probes.
P6009 is rated to 1.5 KV. And for 60 Hz nominal frequency doesn't need to
be derated. (It's flat to around 100 KHz.) The can be compensated to the
465's input capacitance.
So yes and no.
Your generator isn't new, so if there is a problem with the AVR, the
voltage could rise significantly...there could be an over-voltage
problem... but, surely not so much that a P6009 wouldn't protect the 465
inputs.
I'd be more worried about high voltage transients finding their way into
the scope inputs.
I'd be more concerned about that when the generator is started, stopped,
stalls, or is driving inductive loads.
Maybe run a couple of trouble lights, fitted with incandescent bulbs, when
the generator has been running well for a while, and is well warmed up.
(The bulbs are a very crude indication of frequency drift, and voltage
drift, or over-voltage, or under-voltage.)
I'd do a 465/P6009 measurement on the above... if I had to.






Roy Thistle
 

On Wed, Oct 28, 2020 at 09:15 PM, Robert Simpson wrote:


I have P6009 probes.
P6009 is rated to 1.5 KV. And for 60 Hz nominal frequency doesn't need to be derated. (It's flat to around 100 KHz.) The can be compensated to the 465's input capacitance.
So yes and no.
Your generator isn't new, so if there is a problem with the AVR, the voltage could rise significantly...there could be an over-voltage problem... but, surely not so much that a P6009 wouldn't protect the 465 inputs.
I'd be more worried about high voltage transients finding their way into the scope inputs.
I'd be more concerned about that when the generator is started, stopped, stalls, or is driving inductive loads.
Maybe run a couple of trouble lights, fitted with incandescent bulbs, when the generator has been running well for a while, and is well warmed up. (The bulbs are a very crude indication of frequency drift, and voltage drift, or over-voltage, or under-voltage.)
I'd do a 465/P6009 measurement on the above... if I had to.


Robert Simpson
 

Getting back to a Tektronix question, I have P6009 probes. Wouldn't using those be safe from over voltage outputs (If there are any) of a generator to the 465M oscilloscope?
Bob


Roy Thistle
 

And yet, I do.
If anyone chooses to ignore, then I'll have less to read, less to ponder, and less to write about.
Go ahead make your day, if that's what makes it a better day for you.
My day has certainly been better for participating in TekScopes.


 

May I suggest that the subject of many of these posts should be changed to Switching Mode Power Supplies

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Greg Muir via groups.io
Sent: Wednesday, October 28, 2020 12:21 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] Scope grounding and generators

Chuck,

In days of yore when power supplies were basically linear it was a simple approach to use any type of AC supply available. Switchers sort of changed that scenario to a certain degree.

There are people out there who aren’t aware that most of the early UPS units (and still some from today’s production) have square wave outputs. I have not seen any equipment that disliked that waveform but mostly worry about the high risetimes, overshoot and the stress on power supply components. It is obvious that computers and peripherals seemed to have tolerated that Draconian waveform over the years.

Simulated sine wave UPS units was a more progressive step in the right direction until “pure” sine wave units came along. But I place quotes around “pure” since those, too aren’t so precise with regards to distortion. I guess if one tricked a UPS unit by modifying it into running constantly or procured an online unit and placing either one between the generator and the load they would have a happier arrangement given any distortion produced from the generator itself. But those pure sine wave output UPSs do carry a significant quiescent current draw.

There was one instance several years ago when I was called on to do some RF field testing whereby I connected my spectrum analyzer to a small square wave 12/115V inverter for power. It worked well with no problems but I felt uncomfortable with the 60 Hz noise emanating from the analyzer’s power supply having to deal with the nasty waveform. So I picked up a true sine wave inverter before the next trip out. When on my next outing I connected my new inverter to the spectrum analyzer and things sounded a whole lot better. But much to my surprise the entire RF spectrum of interest was filled with inverter noise!! The problem was eventually rectified by placing the inverter in a shielded enclosure with heavy high frequency filtration on both the 12V input & 115V output leads.

There was one time I was doing engineering for a scientific organization where we were modifying cargotainers to house scientific equipment and living quarters for the scientific staff to drop on the decks of research ships for their use. It was known that the AC power provided by the ship’s supply was rather horrid with regards to noise, transients and other anomalies (something akin to the power found on railroad locomotives). So we procured several large AC-AC converters where the incoming power was rectified to DC (linear-wise) then applied to crystal controlled DC-AC sine wave converters to provide pure power for the instrumentation and other facilities. The system was a real power hog and occupied a lot of space but everything worked well. Besides, the ships have boundless amounts of power to keep things going anyway.

As I mentioned previously I cannot speak of any experiences where the quality of the incoming AC power has jeopardized the operation of my equipment but I am always cautious when coming up against unusual circumstances and will take measurements where necessary to guarantee satisfactory operation. A few simple measurements takes far less time that repairing broken equipment.

Greg







--
Dennis Tillman W7pF
TekScopes Moderator


Chuck Harris
 

And yet, still nobody cares.

More and more I come to realize that ignoring
you completely is a very good policy.

-Chuck Harris

Roy Thistle wrote:

On Wed, Oct 28, 2020 at 10:59 AM, Chuck Harris wrote:


Roy, I am pretty sure that nobody cares that you are a
colonial. I asked some ants
Well in the parlance, the claim that: "There is not one person ("body"), such that that person("body") cares (your vernacular) that that person ("body") is a colonial."... that claim is false. Proof: I care.
If you limit your universe of discourse to the Tekscopes forum, the claim is still false. (Same proof.) If you interpret your claim within the vernacular, the claim is in-consistent... as given.
As for ants... Peng Lei style at most... well... one person may wax metaphorical, even poorly so... yet deny the legitimacy of others to do so.
As for UPS et. al., under-cover, or by the label...yes, they come and go out of the computer center(s)/server-room... and more than a few have been scrapped.