Clamp meters: Peak vs. In-rush?
notme
in-rush. The older models have Max (some: Peak). The recent advent in Fluke
clamps is "In-rush".
How do these differ? Isn't in-rush current the short, max current at
motor-turn on? Shouldn't meters with a Max feature capture this accurately?
Compare, for example, my old Fluke 36 (Max):
<http://assets.fluke.com/manuals/36______iseng0000.pdf>
and the 334 (In-Rush):
<http://us.fluke.com/VirtualDemos/330shock.asp>
(click "Explore" then "Selection Guide").
How do Max & In-rush differ? Only in the marketing department?
Or is there a real-world difference?
Thanks,
Dave
notme
"Note: In-rush current measurements done with a 330 Series Clamp Meter will
differ from Min/MAX, Peak, or Peak-Hold measurements which are not triggered
events."
Sounds like other than auto-triggering, the results are the same.
Dave
Salmon Egg
<0001HW.C71FB1C0...@news.eternal-september.org>,
notme <no...@notme.org> wrote:
The essence of one of these meters is a current transformer using a core
that can be opened and then closed. In the closed condition, the
magnetic reluctance of the core is made as small as possible. The
secondary coil is connected to a small resistance. The idea is to have
the current flow waveform in the secondary duplicate the current flow in
the often single turn primary. The voltage on the resistor than
duplicates the current waveform.
How this secondary waveform is used and processed will determine the
features of the meter. Fluke is one of the big names in hand held
oscilloscopes. It is possible to digitize the start of this waveform.
Then various algorithms can be used to characterize the waveform
including the start of it.
Bill
--
As the years go by, dying just before having to fill out a tax return has merit.
PeterD
Like this?
Apply power, inrush current is 10A for .2 seconds.
Current drops to 1.2A for 2 seconds.
Current rises to 12.5A for 1 second.
Current drops to 1.2A until termination.
So the Inrush reading would be 10A, and the Max reading would be
12.5A. I can see some usefulness for this, since a max only meter
would falsly read 12.5A which the user might attribute to the inrush
current, instead of the event at 2 seconds.
Shaun
"notme" <no...@notme.org> wrote in message
news:0001HW.C71FB1C0...@news.eternal-september.org...
In the usual Max measurement, the input (current) is sampled and the maximum
sample is displayed, but the true maximum could have occurred between the
samples and in that case you miss the true maximum or inrush current that
your looking for. In inrush current measurements first off it's a triggered
measurement and measures for a very short period of time and it doesn't
depend on samples, I think it's an analog approach.
Shaun
notme
> sample is displayed, but the true maximum could have occurred between the
> samples and in that case you miss the true maximum or inrush current that
> your looking for. In inrush current measurements first off it's a triggered
> measurement and measures for a very short period of time and it doesn't
> depend on samples, I think it's an analog approach.
>
> Shaun
After a short phone conversation with a tech support person at Fluke, I think
I understand the difference: it's the acquisition speed. (The new clamps also
have triggered event feature, but that's icing on the cake.)
In the clamp meters in Fluke's present product lineup that have the "In-rush"
feature, the acquisition speed is listed as 100 mS. In the older clamp meters
(eg. my model 36) that have the "Max" feature, the acquisition speed is
listed as 250 mS.
In other words, old (model 36) meters sample 4 times a second. New (model
33x) meters sample 10 times a second (overhead aside).
Help me understand the implications of the faster acq. speed. Obviously for a
quick event to be measured, the speed needs to be quick or the event will
pass unnoticed. Having said that, as long as the event overlaps *any* period
of time with the acquisition window, the peak value will be measured. Yes?
It's kind of a random chance of getting the acquisition (for events <
acquisition speed) isn't it? But not impossible.
Thanks,
Dave
Phil Allison
"notme"
> After a short phone conversation with a tech support person at Fluke, I
> think
> I understand the difference: it's the acquisition speed. (The new clamps
> also
> have triggered event feature, but that's icing on the cake.)
>
> In the clamp meters in Fluke's present product lineup that have the
> "In-rush"
> feature, the acquisition speed is listed as 100 mS.
** The term actually used is "integration time " - very important .
> In other words, old (model 36) meters sample 4 times a second. New (model
> 33x) meters sample 10 times a second (overhead aside).
** Not at all what Fluke claim.
See page 2 of this pdf.
http://assets.fluke.com/appnotes/1629920_.pdf
The 33x meters are actually sampling the current surge wave a " large number
" of times in the crucial first few cycles of applied AC power, so that the
peak value can be found.
This is quite unlike your typical DMM that *ANALOGUE * samples a DC input
voltage a few times a second - with these, an AC to DC converter ( true rms
or average rectified value ) is needed to measure any AC wave.
.... Phil
notme
Very good description of the In-rush feature. I wish the guy at Fluke would
have recommended it.
Thanks for your observations, Phil. You've been very helpful.
Dave
Phil Allison
"notme"
>
>> http://assets.fluke.com/appnotes/1629920_.pdf
>
> Very good description of the In-rush feature. I wish the guy at Fluke
> would
> have recommended it.
** The link was very hard to find, Fluke's site alluded to its existence but
was not clear on where it was.
Google helped out ....
> Thanks for your observations, Phil. You've been very helpful.
** DMMs baffle the masses, it seems.
Mainly cos the name is so misleading.
.... Phil
Andy
I recently did some tests on the inrush current on a wound rotor
motor. I used two fluke meters, both with inrush capacity. And I
also built my own circuit using CT, a few resistors, a couple op amps,
and a data acquisition card. The data acquisition card was set to
sample at 1000Hz. I ran the tests by starting the data acquisition,
and then starting the motor. The samples were taken for 1 second.
With the data card I was able to get very good graphs of the
asymmetric starting current. However, the max amplitude of the
starting current measured by the data acquisition card was remarkably
different from that measured by the Fluke. The fluke does not
necessarily see the max waveform. The fluke takes a bunch of samples
in the first few cycles, and then spits out the max of what it
measured. I think the flukes are fine, but it should be noted that
they can be off by quite a bit. In my tests, the difference between
the fluke and the data circuit ranged from a few percent to almost
100%.
daestrom
Now you've introduced another nuance to the discussion.
Reading the app note it seems that Fluke designed there 33 meters to
read the symetrical currents for motor starting.
But as you mentioned, large inductive loads often have a DC offset
component to their starting current. This comes as an artifact of
closing the starter when the sine wave is not at zero-crossing
(inevitable in a three-phase motor).
You can see this in oscillograph traces, or hi frequency samping such as
your set up. But it's hard to get repeatability unless you have a
zero-crossing motor starter. Each time the motor/transformer is
energized, it's likely to be at a different point on the sine wave.
(some large motor/transformer protection schemes avoid false-tripping on
this in-rush by using various techniques such has harmonic-restraint, or
simple time delays)
I wonder if Fluke deliberately filter out the DC offset just so they
don't have to explain why the reading changes each time you start the
motor :-)
daestrom
Phil Allison
"daestrom"
>
> Reading the app note it seems that Fluke designed there 33 meters to read
> the symetrical currents for motor starting.
>
> But as you mentioned, large inductive loads often have a DC offset
> component to their starting current. This comes as an artifact of closing
> the starter when the sine wave is not at zero-crossing
** Fraid you have got that all wrong.
The way to *guarantee* very large inrush surges ( with transformers and
transformer based PSUs) is to switch on at the zero crossing of the AC
voltage.
Cos doing this generates the maximum degree of magnetic saturation in the
core.
> I wonder if Fluke deliberately filter out the DC offset just so they don't
> have to explain why the reading changes each time you start the motor :-)
** Fluke make no specific claims about the accuracy of their "inrush surge"
detection circuitry.
But I would not doubt is does the job required, as far as motors and
circuit breakers are concerned.
.... Phil
Shaun
"Phil Allison" <phi...@tpg.com.au> wrote in message
news:7m3ik3F...@mid.individual.net...
Switch on power at or close to the voltage maximum on the AC sinewave
creates the biggest transient ( surge current) in equipment with a magnetic
core. Think about it.
Shaun
Phil Allison
"Shaun"
> "Phil Allison"
>> "daestrom"
>>>
>>> Reading the app note it seems that Fluke designed there 33 meters to
>>> read the symetrical currents for motor starting.
>>>
>>> But as you mentioned, large inductive loads often have a DC offset
>>> component to their starting current. This comes as an artifact of
>>> closing the starter when the sine wave is not at zero-crossing
>>
>> ** Fraid you have got that all wrong.
>>
>> The way to *guarantee* very large inrush surges ( with transformers and
>> transformer based PSUs) is to switch on at the zero crossing of the AC
>> voltage.
>>
>> Cos doing this generates the maximum degree of magnetic saturation in the
>> core.
>
> creates the biggest transient ( surge current) in equipment with a
> magnetic core.
** Fraid that is a very silly and persistent myth.
You have obviously NEVER checked it out.
> Think about it.
** That is very stupid and very rude.
Read this then fuck off.
http://en.wikipedia.org/wiki/Inrush_current
.... Phil
Phil Allison
"Shaun"
> In inrush current measurements first off it's a triggered measurement and
> measures for a very short period of time and it doesn't depend on samples,
> I think it's an analog approach.
** I see no sign of thought here whatever.
.... Phil
Proteus IIV
THERE IS NO IN-RUSH APPLICABLE WITH CLAMP ON METERS
WHAT ARE YOU A TROLL FISHING FOR COX TOO ?
I AM PROTEUS
Andy
Shaun is correct in his assessment of inrush current being at a
maximum if the breaker is closed at a zero crossing of voltage.
Because the magnetization current is inductive, the current lags by 90
degrees. So, if the breaker is closed at V=0, and current is lagging
by 90 degrees, then this timing corresponds to a max value of current.
Phil Allison
"Andy is a Fucking Tenth Wit "
> Shaun is correct in his assessment of inrush current being at a
> maximum if the breaker is closed at a zero crossing of voltage.
** Fraid he said the direct opposite - pal.
" Switch on power at or close to the voltage maximum on the AC sinewave
creates the biggest transient .. "
Stinking FUCKWITS like YOU have no reason to live.
Top yourself, ASAP
Rat bait would be ideal.
And appropriate.
... Phil
Shaun
I looked up inrush current for transformers and I stand corrected, it's
maximum will occur at the voltage zero crossing point. I thought it was the
same as an L R circuit in which case if switch closes as the peak of the ac
waveform it causes the maximum transient.
Shaun
BTW: Phil you are very RUDE!!
Phil Allison
"Shaun"
>
> I looked up inrush current for transformers and I stand corrected, it's
> maximum will occur at the voltage zero crossing point. I thought it was
> the same as an L R circuit in which case if switch closes as the peak of
> the ac waveform it causes the maximum transient.
** Now you owe me a big apology.
> BTW: Phil you are very RUDE!!
** Drop dead you tedious wanker.
..... Phil
Archimedes' Lever
The meter READS the inrush signature, you idiot. Not create it.
Proteus IIV
wrote:
>
> - Show quoted text -
WELL IF IT ISN'T SCUMMANDER FUCKTARD AGAIN
YOU DO NOT EVEN FOLLOW THREADS ANYMORE DO YOU ?
GET A LIFE YOU FAGGOTY LITTLE TROLL
I AM PROTEUS
Paul G.
wrote:
>
>"Shaun"
>>
>> I looked up inrush current for transformers and I stand corrected, it's
>> maximum will occur at the voltage zero crossing point. I thought it was
>> the same as an L R circuit in which case if switch closes as the peak of
>> the ac waveform it causes the maximum transient.
>
A bit off topic....
At first thought, it seems counter-intuitive that zero-crossing is
the worst case for inrush current, so I used LTSPICEIV to simulate the
situation. As usual Phil is correct! here's the .asc file for the
simulation, note that you must set the inductor current to zero for
the initial condition.
Version 4
SHEET 1 880 680
WIRE 192 80 48 80
WIRE 336 80 272 80
WIRE 48 112 48 80
WIRE 336 144 336 80
WIRE 48 224 48 192
WIRE 336 224 48 224
WIRE 48 256 48 224
FLAG 48 256 0
SYMBOL voltage 48 96 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value SINE(0 160 60 0 0 0 10)
SYMBOL res 320 128 R0
SYMATTR InstName R1
SYMATTR Value 5
SYMBOL ind 176 96 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 5 56 VBottom 0
SYMATTR InstName L1
SYMATTR Value 1
TEXT 80 272 Left 0 !.tran 0 0.2 0.00001 0.00001
TEXT 88 304 Left 0 !.IC i(L1)=0
In the sim above, is a 1H inductor in series with 5 ohms, with
110vac (160v peak) applied. The voltage is set to start at zero
crossing.
Afterward, set the phase of the voltage to 90degrees, and you'll see
that the current is less. In the above sim, the AC current has +/- 400
mA peaks (800mA p-p) when you start at sinewave of 90 degrees.
Starting at sinewave zero degrees, the current STARTS at 0.0 mA,then
goes to about +800mA, then gradually loses the DC component.
When the AC supply is switched off, you also get interesting
results! When switched off at zero-crossing, there is still
considerable DC current flowing in the inductor. It will cause a lot
of trouble if not taken into account. When switched off at the +/-
peak of the sine wave, the inductor current will be zero. In other
words, inductive loads are happier being switched at the +/- peaks,
which is not what you might expect, if you are used to capacitive
input power supplies.
>
> ** Now you owe me a big apology.
>
>
>> BTW: Phil you are very RUDE!!
>
>
> ** Drop dead you tedious wanker.
>
>
>..... Phil
>
Phil has pretty good advice, but he sure gets irritated if you
don't know what you're doing. You'll get absolutely nowhere biting him
back, unless you can decently support what seems to be your mistake.
There's a lot to learn from his "critiques", but some of the back &
forth rants are pretty silly. It's always fun to watch a "Phil"
thread!
Paul G.
ehsjr
> On Sat, 14 Nov 2009 08:07:37 +1100, "Phil Allison" <phi...@tpg.com.au>
> wrote:
>
>
>>"Shaun"
>>
>>>I looked up inrush current for transformers and I stand corrected, it's
>>>maximum will occur at the voltage zero crossing point. I thought it was
>>>the same as an L R circuit in which case if switch closes as the peak of
>>>the ac waveform it causes the maximum transient.
>>
> A bit off topic....
> At first thought, it seems counter-intuitive that zero-crossing is
> the worst case for inrush current,
To make it more intuitive, consider that starting at 0 crossing
provides more time charging the inductance, so higher I.
Ed
Phil Allison
"Paul G."
"Phil Allison"
>"Shaun"
>>>
>>> I looked up inrush current for transformers and I stand corrected, it's
>>> maximum will occur at the voltage zero crossing point. I thought it was
>>> the same as an L R circuit in which case if switch closes as the peak of
>>> the ac waveform it causes the maximum transient.
>>
> A bit off topic....
> At first thought, it seems counter-intuitive that zero-crossing is
> the worst case for inrush current, so I used LTSPICEIV to simulate the
> situation. As usual Phil is correct! here's the .asc file for the
> simulation, note that you must set the inductor current to zero for
> the initial condition.
>
>
> In the sim above, is a 1H inductor in series with 5 ohms, with
> 110vac (160v peak) applied. The voltage is set to start at zero
> crossing.
** Your simulation is of an inductor - and NOT a AC supply transformer
primary as the question requires.
The differences are many and great and the switch on transient behaviour
very different - mostly because a transformer's laminated iron core will
saturate hard when a frequency just a little lower than it is designed for
is applied.
I doubt that LTSPICE IV can even do such a simulation.
Its why I said to TRY it !!
> Phil has pretty good advice, but he sure gets irritated if you
> don't know what you're doing. You'll get absolutely nowhere biting him
> back, unless you can decently support what seems to be your mistake.
** Yep - there is nowhere to stat demolishing a fallacy if no sensible case
is posted that supports it.
Posting inane drivel like " .think about it " or " I am surprised you don't
know this " is just pouring fuel on the fire.
..... Phil
Phil Allison
"ehsjr"
>
> To make it more intuitive, consider that starting at 0 crossing
> provides more time charging the inductance, so higher I.
>
** But we are discussing the behaviour of a AC supply TRANSFORMER - where
core saturation IS the cause of inrush current surges.
The simplest way to think about it is to first note that the core of nearly
any AC supply tranny operates on the edge of saturation ALL THE TIME when
mains power is applied.
( Proof of this is that if one raises the applied voltage by about 20% or
lowers the AC frequency by the same - the magnetising current drawn by the
tranny will dramatically increase. )
The note that switching the supply on at a zero crossing means the *average
value* of the wave will not become zero until a whole cycle has passed -
while switching on at a peak means only half a cycle need pass for the same
result.
The second case produces a very much smaller inrush surge as a result.
Interestingly, if the AC voltage applied to a supply tranny is half or less
its rated operating voltage - inrush current surges barely exist.
..... Phil
Paul G.
wrote:
>
LTSPICE has 2 methods for simulating saturation, if you run
LTSPICE, and look for saturation in the help menu it gives you the
details. One method is "based on a model first proposed in by John
Chan et la. in IEEE Transactions On Computer-Aided Design, Vol. 10.
No. 4, April 1991 but extended with the methods in United States
Patent 7,502,723". It uses Hc (Coercive force), Br (Remnant flux
density), Bs (Saturation flux density), Lm (Magnetic Length), Lg
(Length of gap), A (Cross sectional area), N (Number of turns). That
will be tricky for an inductor that's aready built.
The other method uses a "flux" statement:
L1 N001 0 Flux=1m*tanh(5*x)
I1 0 N001 PWL(0 0 1 1)
this didn't make a lot of sense to me..... Fortunately a search
came up with:
http://www.plcdrives.com/forum/f34/re-simulating-non-linear-magnetics-ltspice-29584/
which has a file that has many examples inside it to play with. It
explains how to set up different saturation scenarios. A quicky
simulation showed an enormous inrush current.
Another link which seems useful is:
http://ltwiki.org/index.php5?title=Main_Page (a WIKI for LTSpice)
more:
http://www.electronicskb.com/Uwe/Forum.aspx/design/45226/Inductor-saturation-in-LTspice
and another:
http://www.electronicskb.com/Uwe/Forum.aspx/cad/538/Simulating-non-linear-magnetics
whew! this is a lot of reading, and getting into the guts of
LTSpice. Many years ago, I used PSPice (the full release, that worked
under DOS, cost me well over $1000), it would do inductor saturation
as well.
Apparently the saturation models don't allow you to do mutual
coupling, so you need to make the transformer equivalent circuit in
which you can place the saturating inductance. There are details in
the above links.
Of course, it will be quite difficult to set up the parameters to do
a reasonable simulation, you need to know a lot about the device you
are simulating.
Paul G.
Phil Allison
"Paul G."
>
> Of course, it will be quite difficult to set up the parameters to do
> a reasonable simulation, you need to know a lot about the device you
> are simulating.
>
** Takes far less time to simply measure what happens with a real
transformer - the results are always 100% trustworthy too.
.... Phil