What does "TRUE RMS" mean?

[Abbey Somebody]

"John Larkin" wrote in message


It has a one year accuracy of 0.06% RDG + 0.04% RNG on the 100mV scale. So
for my 5 mV signal it would be 0.06% + 0.0004*100/5 = 0.86% RDG. Not bad,
but not great. There is one on eBay for about $1000 and an 8846a not working
for about $500.

I need to figure out the real accuracy of the others I was considering. But
maybe it would be better to use a 100A 100mV shunt for my low-end
calibration. Then I would have a full 50 mV to work with. Just a hassle to
change the shunt, but I could probably rig them up in series and just use a
high current switch (like a battery cut-off) or a jumper to short out the
100A shunt when I calibrate at higher currents (500-1000A).

Paul

Or just factor it in. Nahhhh...

Make the shorting bar out of Silver stock. Clamp it in tightly.

 

[Phil Allison]

"John Larkin"

Are you measuring AC or DC? Both are a nuisance to measure accurately
if you use a shunt.

These are phenomenal:

** So is the price at over $800.

Very nearly the same for les than 10% of that.

http://au.element14.com/honeywell-s-c/csnf161/current-transducer/dp/7234788


.... Phil

 

[P E Schoen]

"John Larkin" wrote in message

Are you measuring AC or DC? Both are a nuisance to measure
accurately if you use a shunt.

These are phenomenal:

http://www.gmw.com/electric_current/Danfysik/866_867/867.html

We have been using shunts for AC and DC calibration for a long time, and
have had good results. I know that they do have some limitations, but they
work well for the type of equipment we calibrate. One important feature is
that a 2500 amp 50 mV shunt, as we normally use for high current test sets,
will handle surge currents up to 30,000 amps, whereas the LEM transducers do
not have that wide dynamic range. For my own purposes of calibrating
Ortmasters, I can only supply up to 500 amps with an easily handled portable
source, so perhaps an LEM transducer would be useful.

I found an LEM LT1005-T on eBay for $150:
http://www.ebay.com/itm/LEM-LT1005-...986?pt=LH_DefaultDomain_0&hash=item415fb0b342

And the specs:
http://www.lem.com/docs/products/lt 1005-t.pdf

It appears to be about 0.4% accurate, but I'm not sure how that translates
to a 50 amp current. The transfer ratio is 5000:1, so 50 amps will provide
10 mA. The offset is typically about 0.4 mA, so that is 4%. Seems like it
won't really fit the bill.

Perhaps the Danfysik 1000A model might be better with 50 ppm offset,
although at $1500 I think I'll stay with what I have. But thanks for the
suggestion.

Paul

 

[Phil Allison]

"P E Schoen"


I found an LEM LT1005-T on eBay for $150:
http://www.ebay.com/itm/LEM-LT1005-...986?pt=LH_DefaultDomain_0&hash=item415fb0b342

And the specs:
http://www.lem.com/docs/products/lt 1005-t.pdf

It appears to be about 0.4% accurate, but I'm not sure how that translates
to a 50 amp current.


** The LT1005-T is a +/- 2000 amp device.

The transfer ratio is 5000:1, so 50 amps will provide
10 mA. The offset is typically about 0.4 mA, so that is 4%. Seems like it
won't really fit the bill.


** Use hammers as screwdrivers - do you?

The "offset" is due to residual magnetism in the core due to previous use -
long time constant AC coupling will eliminate it completely. As your game
is mainly massive transient overloads - that should not present a big
issue.

LEM and Honeywell current transducers are very cool things.


.... Phil

 

[Phil Allison]

"John Larkin"

Only 3 or 4 orders of magnitude less accurate.

** What absolute crap.

1. Point to the specs that prove that claim beyond doubt.

2. 1% accuracy is easily adequate for the purpose P.E. Schoen has
indicated.





.... Phil

 

[P E Schoen]

"Phil Allison" wrote in message

** So is the price at over $800.

Very nearly the same for les than 10% of that.

http://au.element14.com/honeywell-s-c/csnf161/current-transducer/dp/7234788

That one is only 150A. Here is one of the same series for 600 amps peak:
http://au.element14.com/honeywell-s-c/csns300m/current-sensor-300a/dp/5081798

They do have a 1200A peak LEM for about $60:
http://au.element14.com/lem/htfs-800-p-sp2/current-transducer/dp/9135766
But its accuracy is only 1%.

Another one is a LEM, but it's over $500, and only 1%:
http://au.element14.com/lem/dhr-1000-c5/current-transducer-1000a-0-5v-o/dp/1002707
http://www.farnell.com/datasheets/30908.pdf

This one is a good price, but still not the specs I need:
http://au.element14.com/multicomp/tma400a/current-transducer-open-loop-400a/dp/1898951
http://www.farnell.com/datasheets/1368151.pdf

Thanks for the suggestions.

Paul

 

[Phil Allison]

"P E Schoen"
"Phil Allison"

** So is the price at over $800.

Very nearly the same for les than 10% of that.

http://au.element14.com/honeywell-s-c/csnf161/current-transducer/dp/7234788

They do have a 1200A peak LEM for about $60:
http://au.element14.com/lem/htfs-800-p-sp2/current-transducer/dp/9135766
But its accuracy is only 1%.


** WTF do you need better than 1% for ???

A real engineering need or some brainless " it's in the rules " reason.

Cos you are sounding real brainless right now.



..... Phil

 

[Phil Allison]

"John Larkin is a Fucking MORON

http://www.gmw.com/electric_current/Danfysik/866_867/866_Specs.html

Note that they don't use percentages, they use PPMs.

** I already read that load of ambiguous nonsense.

Point DIRECTLY to the " accuracy " spec.

And it has gotta be better than 0.001%.

Shithead.


.... Phil

 

[P E Schoen]

"Phil Allison" wrote in message

** WTF do you need better than 1% for ???

A real engineering need or some brainless " it's in the rules " reason.

Cos you are sounding real brainless right now.

The Ortmaster is a test instrument with a stated accuracy of 1%. Standard
rules of calibration require the standards used to certify such an
instrument should be at least twice as accurate and ideally three or four
times as accurate. Hence I need an instrument which is about 0.25% accurate
and traceable to NIST. Actually, I will not allow my unit to be shipped
unless it is at least within 0.5%, and I like to set the calibration factors
so that it is better than 0.2%, just to be sure, and because I can.

This is probably overkill, as the reclosers that are being tested are
generally 10% and even 20% devices, but the recloser manufacturer, Cooper,
has specified that they require the instrumentation to be at least 1%
accurate. And they also require at least 2000 samples per second, and timing
within 1 mSec. The Ortmaster, even the original design which was released in
1994, meets this specification, and actually is as accurate as their
laboratory test system which cost them well over $20,000. Not bad for an
instrument that sells for $4000 list, including software, and costs less
than $400 each to make (not counting years of NRE). And it's rugged enough
that one of them still worked (except for a broken DB25 connector) after
being run over by a truck:

http://www.ortmaster.com/ORTM-1_Damaged_01.jpg
http://www.ortmaster.com/ORTM-1_Damaged_02.jpg
http://www.ortmaster.com/ORTM-1_Damaged_03.jpg

Usually the calibration factors are very consistently within 0.25% of each
other, which is understandable since I use 1% resistors. But the low ranges
50 and 100 amp are usually set about 1% higher than the others. This may be
partially due to the gain-bandwidth product of the amplifier at the highest
gain settings (since there is also capacitance in the feedback). But I also
suspect it is the accuracy of the DMM which is really insufficiently
accurate at 5 mV.

Paul

 

[Don Lancaster]

This is the only HP that's currently worth buying:

http://www.amazon.com/HP-Hp-Sauce-9oz/dp/B000EZWKY0

The last decent oscilloscope HP built was the 130C.

--
Many thanks,

Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss: http://www.tinaja.com/whtnu.xml email: [email protected]

Please visit my GURU's LAIR web site at http://www.tinaja.com

 

[JW]

"JW" wrote in message

There are some on eBay for $600-$800. Might be a good investment.

If you're patient, you can get them for less. Probably in the $400 to $450
range. I was lucky on the one I got for my home lab - the guy didn't know
how to test it and was unsure if it was working. Got it for $260 and it
worked fine.

That also looks quite good, and are available on eBay for as little as $200
to $400 (but some are $1000). If the $200 unit is OK (and it is described as
working), that may be the best deal. At least HP/Agilent usually have
extensive schematics and cal/repair documentation.

Yes, schematics are available for the 34401A, unlike the Keithley 2000. I
wish I could get my hands on the 2000 schematics!

Thanks!

Paul

You're welcome.

 

[Phil Allison]

"P E Schoen"
"Phil Allison"

** WTF do you need better than 1% for ???

A real engineering need or some brainless " it's in the rules " reason.

Cos you are sounding real brainless right now.

The Ortmaster is a test instrument with a stated accuracy of 1%.


** How boring.

You are not answering my question - or maybe you are .....


Standard
rules of calibration require the standards used to certify such an
instrument should be at least twice as accurate and ideally three or four
times as accurate. Hence I need an instrument which is about 0.25% accurate
and traceable to NIST.


** Right - so it IS an " it's in the rules " reason then.


Actually, I will not allow my unit to be shipped
unless it is at least within 0.5%, and I like to set the calibration factors
so that it is better than 0.2%, just to be sure, and because I can.

** How boring.


This is probably overkill, as the reclosers that are being tested are
generally 10% and even 20% devices, but the recloser manufacturer, Cooper,
has specified that they require the instrumentation to be at least 1%
accurate.

** How fucking boring.


And they also require at least 2000 samples per second, and timing
within 1 mSec. The Ortmaster, even the original design which was released in
1994, meets this specification, and actually is as accurate as their
laboratory test system which cost them well over $20,000.

** Yaawwwnnnnnn ........


http://www.ortmaster.com/ORTM-1_Damaged_02.jpg


** How fascinating - a bent metal box.........

You are sounding real brainless, right now !


Usually the calibration factors are very consistently within 0.25% of each
other, which is understandable since I use 1% resistors. But the low ranges
50 and 100 amp are usually set about 1% higher than the others. This may be
partially due to the gain-bandwidth product of the amplifier at the highest
gain settings (since there is also capacitance in the feedback). But I also
suspect it is the accuracy of the DMM which is really insufficiently
accurate at 5 mV.

**That would be true.

DMM accuracy figures are generally based on full scale readings, plus or
minus a couple of digits.

Linearity is usually very good, but you will need about 1000 counts on the
display to guarantee 1%.




.... Phil

 

[P E Schoen]

"Don Lancaster" wrote in message

The last decent oscilloscope HP built was the 130C.

http://www.hpmemory.org/wa_pages/wall_a_page_12.htm

I still have my HP 130A, with fast blue phosphor for camera. My father gave
it to me in 1969 after his lab was flooded by a hurricane and it was sent
back by the factory as FUBAR, or perhaps the estimate was unreasonable. I
found that two octal plugs were reversed and when I fixed that, it worked,
but the display was dim. I replaced a HV capacitor and generally cleaned it
up and it worked fine!

It died again in 1983 and I never did fix it, but I kept it. I really liked
the layout of the controls and the overall appearance. Much nicer than
Tektronix.

My first scope was a Triumph, which had an optional "wobbulator"! It was
IIRC only AC coupled and very low bandwidth. I think I gave/sold it to a
friend who was a Ham.

http://antiqueradio.org/trium01.htm

Ah, yes, the good old days!

My father showed me an early digital meter which used a mechanical means to
adjust the reference voltage through decade dividers and had flipping metal
or plastic numbers, like old clocks. It made a lot of noise especially when
it was hunting for a signal between 0.999 and 1.000 volts. I don't remember
how many digits or how accurate, but it was pretty cool!

Paul

 

[Spehro Pefhany]

Their printers are good. I finally retired my 6Ps after many years,
and cut over to CP1525 color lasers. They installed easily and work
great. They *should* come with a free flashlight so you can read the
LCD.

For shared office use, the Lexmarks (T654DN B&W and C925 tabloid
color) are performing admirably. We've got a narrow (24") carriage HP
Designjet as well. For SOHO use, I've been using mostly Brother and
Samsung, with good results, and my workhorse HP2015DN is still
chugging along despite severe problems with the first one. Perhaps the
thought that there is a new Brother equivalent sitting on the shelf,
drivers installed, and waiting to replace in an instant it is keeping
it going.

Most of our computers here are ProLiant server boxes with hot-plug
RAID on the front panel and redundant power supplies, fans, and BIOS.
They have been wonderfully reliable.

John

Good to hear. I've had bad luck reliability-wise with their SOHO level
computers and the small printers are essentially a free box to hold
some very expensive proprietary "chipped" cartridges. And I've had
major troubles with reliability and grudging support on their
mid-range printers. They're up to two strikes and a couple foul
balls..

I would definitely still consider their workstations. Some even come
standard configured with a professional-level CAD-certified video
cards, and I think they've got a pretty good reputation.

 

[tm]

I chose my Fluke 45 multimeter as the best choice for true RMS reading at
the time I bought it about 20 years ago. It has an accuracy of 0.2% + 100
counts from 50 Hz to 10 kHz, but that means that on the 100 mV range,
specified for 15,000 to 99,999 counts, it may be an additional 100/15000 or
0.67%. It seems to be actually much better than that, although it has an
offset of about 0.240 mV with the input shorted. I have been using this to
calibrate my Ortmasters, which are rated at 1% from full scale ranges of 50A
to 10,000A, using a 1000A 100 mV shunt. So when I calibrate the 50A range, I
am reading about 5.000 mV, which could be as much as 100 counts off, or 2%.
The Ortmaster typically shows an offset of 0.10 A on that range, which is
only 10 uV. So I may need to get a better standard!

The Fluke 8520A has a 1 year accuracy of 0.15% reading plus 0.05% FS, which
in this case is 1.99999 V, so the true error is as much as 1 mV. So its
accuracy at 5 mV is only really 20%! No winner here.

The Fluke 289 has a true-RMS accuracy of 0.3% + 25 counts, 45-65 Hz, on the
50.000 mV range, which gives me an effective accuracy on a 5 mV reading of
0.8%. An improvement, but I'd really like 0.5% or better.

The Fluke 8808A is similar to the Fluke 45, but it does not seem to be true
RMS.

The Fluke 87 has a true-RMS accuracy of 0.7% +2 digits and 0.1 mV
resolution.

The HP 3478 has a true-RMS accuracy of 0.46% + 163 counts for the 300.000 mV
range.

I really have not found a meter that is significantly better than the Fluke
45. Any suggestions?

__________________________

How about putting a good low noise instrumentation differential amplifier on
the
measurement points of the shunt? It should be easy to calibrate it for 20 dB
(or 40)
of gain and using some good low TC precision resistors should make it very
stable.
Then you will move the measurement out of the low order least sig digits of
the meter.

??



tm

 

[P E Schoen]

"tm" wrote in message

How about putting a good low noise instrumentation differential
amplifier on the measurement points of the shunt? It should be
easy to calibrate it for 20 dB (or 40) of gain and using some good
low TC precision resistors should make it very stable.
Then you will move the measurement out of the low order
least sig digits of the meter.

Yes, that would work, but I'd have to build it and add a power supply and
put it in an enclosure, and then it would be another piece of equipment to
calibrate so that everything is traceable to NIST. And calibration companies
may be hesitant to certify customer-built equipment.

In fact, I have essentially that in the Ortmaster devices I calibrate. I
have an AD620 (or INA129) at the front end with some resistors, capacitors,
and TVS for protection. Then I have an 8-stage amplifier using an LT1112
dual precision op-amp and a DG408 and a resistor string in a 1-2-5
configuration for gains of 1 to 200.

Much easier to use a 100A 100mV shunt which shifts the readings by a factor
of 10, and it's a very durable, stable, and easily calibrated device that
needs no batteries or external power and is immune to most forms of
degradation and failure.

Paul