I'm sure we all agree that NO failures is the goal, but to John's point, I
think it would be instructive for some to know what types of failures are
generally found. (e.g. marking, ventilation openings, temperature
limits, dielectric strength, ground bonding, critical components, clause
It's a programmable device, so the code which runs it could serve a safety
critical function and some standards would then refer to UL1998 or perhaps
IEC 61508-1. I don't think 60950 has yet taken that "leap".
Ralph McDiarmid
Compliance Engineering
Residential/Commercial
Solar Business
I have long wondered why solder resist on a PCB cannot be used to reduce
pollution degree and therefore creepage distance. Ever try to scrap it
off? I am sure that all product standards today, assume bare FR4 material
with no coating (no solder resist).
I know that IEC60664 calls out a bunch
A good summary of the state of affairs I think. The last few decades of
independent standards development around the world seems to have produced
a wide variety of non-harmonized requirements. Perhaps things are slowly
improving with wider adoption of IEC standards for safety and for EMC.
A few years ago, I witnessed an electrician shorting 600/347V bus to an
grounded metal duct with his screwdriver. A shower of sparks and then
darkness after the 600V breaker for the entire building tripped. I was
surprised he stayed on the ladder.
Ralph McDiarmid
Compliance Engineering
I must admit I didn't understand any of that, other than references to
MTBF and HALT. It just seems to me that RA is akin to FMEA and FTA (Fault
Tree Analysis) efforts, namely, it's just guess work. There is no way to
come up with a number that can relied on, since there are too many
And to keep things interesting, WG8 of TC77A is trying to define conducted
emission limits between 9kHz and 150kHz. (I think in response to
industry segment responsible for smart wattmeters and other PLC devices
communicating on LV mains)
Ralph McDiarmid
Compliance Engineering
IEC 62109-1 and 62109-2 (at least that's what we are using here for
Europe and Australia)
For USA it would be UL1741, and UL62109-1)
Ralph McDiarmid
Compliance Engineering
Residential/Commercial
Solar Business
Schneider Electric
3700 Gilmore Way
Burnaby
BC
Canada
From:
Grace Lin
Some say 0th fault to mean it is expected to fail, therefore you fault it,
before applying a single-fault. Not the best terminology perhaps.
If bonding impedance test passes, then the circuit is not functional
earthed, it is bonded to earth (to the equipment grounding conductor), In
that
resending (1st one bounced)
From:
Ralph McDiarmid/Canada/Schneider
To:
EMC-PSTC@LISTSERV.IEEE.ORG,
Date:
02/25/2016 08:51 AM
Subject:
Re: [PSES] Insulation testing
It seems that the standards are treating functional ground connections
(those that do not pass a bonding impedance test) as a
However, it is easier to just measure it than argue with the NRTL, which
must follow what the standard says, which all too often include
semiconductors and capacitors. Some of the tables in standards like
UL458 haven't been updated for a long time. (when was the last time you
saw a selenium
We use a small drop of gap-filling cyanoacrylate (like Krazy Glue) to
affix thermocouples. With strain relief (taping) they stay put up to 160C
in our experience. That glue type is nice because it is easy to see bead
contact with the surface.
.
Ralph McDiarmid
Compliance Engineering
I would like to explain to a colleague why the noise floor on a SA does
not look flat as it sweeps across a given frequency range after antenna
factors, cable factors, external gain and external attenuation are
programmed into its display function.
I think it breaks down to these fundamental
If it helps, the Oxford Concise English dictionary defines "ancillary" as:
"a person, activity, or service providing essential support"
Since the AC/DC adapter is essential to the function of the device, it
provides more than merely "essential support"
It is by all means an inseparable part of
And, for another perspective, UNECE Regulation 10, "Uniform provisions
concerning the approval of vehicles with regard to electromagnetic
compatibility", restricts 10m radiated emissions to limits which lie
between CISPR Class A and Class B.
Ralph McDiarmid
Compliance Engineering
" I have been experimenting with this and in several cases using a ferrite
clamp
at the end of cabling seems to "stabilize the spectrum", increasing low
peaks and reducing high peaks. "
Seems to me that it's akin to adding resistance to an L-C network to
provide better damping. So, these
Would a Near Field probe be a better choice?
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
From:
Ken Javor
To:
I though this funny that one of the preferred skills is, "Able to type at
a rate of at least 50 wpm"
I think my mom could do 60 wpm in about 1950 on a manual typewriter. I
think I can do about 40 wpm on a good day on a keyboard; 50 would be take
some training and dedication.
It looks like their Triathlon II models cover a full range of power
ratings.
http://www.georator.com/ProductAdjustableFrequencyConverters.html
___
Ralph McDiarmid | Schneider Electric | Solar Business |
Australia has a similar constraint.
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
From:
Bill Owsley <00f5a03f18eb-dmarc-requ...@ieee.org>
Dear group,
CISPR11 Ed 6 (currently in FDIS) has added emission limits for DC ports,
including PV and Battery.
For products which have DC power ports and which have up to now used the
generic EMC emissions standard, would a change to CISPR11 (EN55011) to now
include DC ports make it the
I assume the relay would have to meet double/reinforced open contact
spacing and dielectric strength in this application. Correct?
Sounds right
We are having a hard time finding a relay that specifies an open contact
dielectric strength for double/reinforced (3000Vac or 4200Vdc).
How about
Ask him to reference a clause in an applicable standard.
Do you know if spacings from N to PE circuit and to dead metal have been
evaluated for 240V ? (in the event it gets cross wired during
installation)
___
I'll only add that the descriptions and examples in the 61000 generics are
pretty clear. Like everything else with EMC, there are shades of grey.
___
Ralph McDiarmid | Schneider Electric | Solar Business |
Agreed
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
From:
John Allen
To:
EMC-PSTC@LISTSERV.IEEE.ORG,
Date:
I expect they would come from a national electrical code. If this is a UL
standard, then the NFPA 70 (NEC)
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance
les list primarily mm sq sizes. “AWG
and kcmil sizes are provided for information only.” So not likely these
are NEC based. But don’t know the answer to the original question.
Perhaps NFPA 70 has table that goes larger but may not help outside of US.
-Dave
From: McDiarmid, Ralph [mail
onding conductor
Take a look at Table 250.122 in the National Electrical Code.
Best regards,
Ron Wellman
From: McDiarmid, Ralph [mailto:ralph.mcdiar...@schneider-electric.com]
Sent: Wednesday, November 25, 2015 3:19 PM
To: EMC-PSTC@LISTSERV.IEEE.ORG
Subject: Re: [PSES] size of protecti
Most residential high-rise buildings distribute 120/208V throughout the
building. Electric ranges and dryers will run from 120/208V or 120/240V
___
Ralph McDiarmid | Schneider Electric | Solar Business |
Isn't that the risk for any measurement using a SA, particularly when
measuring from an AMN? (peak, QP or AVG)
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance
I can see how adding gain ahead of the SA lowers the noise floor, but I
don't see how a preselector can do same. A preselector seems most useful
in tuning only the frequency band of interest, rejecting (attenuating) the
rest.
chneider-electric.com"
<ralph.mcdiar...@schneider-electric.com>, "EMC-PSTC@LISTSERV.IEEE.ORG"
<EMC-PSTC@LISTSERV.IEEE.ORG>,
Date:
10/22/2015 02:32 PM
Subject:
RE: [PSES] 208 split-phase?
I believe NFPA 70 specifies 120/240V for all residential household
voltages.
-Da
Exactly the procedure I posted in our EMC lab years ago.
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
From:
"Kunde, Brian"
Pulse duty ratio must be important too. I've often seen Pk measurements 5
or 6dB higher than QP.
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
Hi Ken, I just scanned through that EDN article.
I see from your graphic that the measurement time was 100ms, so I'm
assuming Peak detection. (not quasi-peak or average) Depending on the
nature of the signal, I assume it might be quite a bit lower when using an
average detector.
When I do
>This lab area has surge protection, so again, we believe the problem is
some kind of low voltage transient, voltage dropout, or >waveform
distortion that we have been unable to detect and simulate.
Was it tested against IEC 61000-4-11 (voltage dips, short interruptions,
variations ) ?
I recall that PFC circuits are generally based on a boost converter and
that circuit arrangement if often challenging to tune for stable
closed-loop response. I think PFC applications try to regulate both input
current and output voltage; one is tightly regulated, the other is not.
Anyone having trouble logging into that site today?
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
-
Clause 5.2.3 in EN50178 (Decisive Voltage) should help clarify. It writes
about "occurs continuously" and "worst operating condition".
I don't think UL508 will be much help, unless it allows use of UL840
(insulation coordination) as an alternate approach to spacings.
If I could download the entire October 2015 issue, it looks like it would
be 100MB. I wonder if they split it into many, smaller chucks for that
reason, because their server has a relatively slow upload data rate? It
took me 30s to download the single-page table of contents.
Thanks David for bringing that detail into focus.
Edition 2 writes about unwanted voltage drops from coupling/decoupling
networks, while Table 6 allows for a shorter front time for higher line
currents, presumably to reduces impedance required by the decoupling
network ( a smaller
And there are going to be more stringent requirements in the 2017 NEC,
related to rapid shutdown.
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
Like very low-drift analog circuits, high impedance circuits require
attention to detail normally dismissed as negligible in 'everyday' circuit
design.
___
Ralph McDiarmid | Schneider Electric | Solar Business
Could you mark it 220/230/240V ?
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
From:
Scott Xe scott...@gmail.com
To:
If it helps, the EMC Handbook Vol. 4 (White) suggests that absorbing
material (6dB loss) in an SAC reduces the error caused by all reflection
(reinforced/cancellation) to about 4dB peak to peak.
Single bound path with no absorption seems to result in only about 3dB
(peak to peak) error,
See the less well-known CISPR 17 standard.
Or, any decent text book on a.c. circuits. The reason I still have some
of my aging, but still useful, college text books.
___
Ralph McDiarmid | Schneider Electric |
We are looking for an EMC consultant or an EMC lab with a good knowledge
of international requirements for certifying host products to which an
approved WiFi transceiver module has been added. We know the basics for
the US, Canada, and the EU but need help for a long list of countries
outside
IEC 60335-1
13 Leakage current and electric strength at operating temperature
Protective Impedance and radio interference filters are to be disconnected
before carrying out the tests.
I suppose they are merely checking for an obvious no-no, like using
accessible chassis metalwork as a
Maybe I'm missing something, and I'm certainly no statistician, but on
page 11, it states,
A manufacturer, knowing the typical standard deviation of RFI
measurements on his products, can – using the 80%/80% rule - estimate the
required margin to the limit already with a single prototype.
I'm drifting ever so slightly off topic now but . . .
legislation certainly keeps NRTLs in business. I've long admired the EU
model, where manufactures declare compliance and are responsible for it.
Do we really need 3rd party certification in USA, Canada, Australia, etc?
I think the new
Can anyone recommend a calibration lab for current tranducers rated to
3,000A (ac and dc) to ISO17025 with a certificate?
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory
Now that the cat's out of the bag, the grow ops will quickly become EMC
experts.
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
From:
Ken Wyatt
The overhead wires for the electric trams (buses) here cause strong RFI on
fringe stations on the AM band, but not strong enough to interfere with
local broadcasts.
___
Ralph McDiarmid | Schneider Electric |
, August 08, 2014 3:47 PM
To: EMC-PSTC@LISTSERV.IEEE.ORG
Subject: Re: [PSES] question with regard of inverter
IEC 62109-1 considers “PV circuits in general” to be OVCII.
Regards,
Peter Tarver
From: McDiarmid, Ralph [mailto:ralph.mcdiar...@schneider-electric.com]
Sent: Friday, August 08, 2014 13
My understanding as well, namely, OV categorization in power distribution
is based on indirect (near-by) lightning strikes.
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory
Certainly appears that they did. As I understand it, the DofC doesn't
even need to accompany the product, only that the product bear the CE
mark.
___
Ralph McDiarmid | Schneider Electric | Solar Business |
Do you mean the e-mark or the E-mark?
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
Phone: +1-604-422-2622 | Address: 3700 Gilmore Way, Burnaby,
And IEEE1547 doesn't required a disconnection for unintentional islanding,
merely that the DG ceases to energize the local electrical power system.
(stop a negative power flow from the DG)
Some utilities in the USA and in Canada are suggesting that 500ms is a
better response time, to
Linear techniques these days are costly, heavy or just unfashionable.
Or mysterious to some of the recent graduate engineers today?
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Dear EMC-PSTC forum members,
Given a need to measure conducted emissions from 150kHz down to 9kHz,
would it be better to:
1. characterise 50ohm LISN for insertion loss down to 9kHz
2. use a current clamp rated for that frequency and use transfer impedance
to determine dB(uV)
I can see that
. . . or wire in free air, or buried, depending on what table you look
at.
As as gut feel, I would say that 3 inches of 26AWG would have a very low
temperature rise at 500mA, but as John stated,
it's best to run a test.
Californians ?
___
Ralph McDiarmid | Schneider Electric | Solar Business | CANADA |
Regulatory Compliance Engineering
From:
Ed Price edpr...@cox.net
To:
EMC-PSTC@LISTSERV.IEEE.ORG,
Date:
07/11/2013
62 matches
Mail list logo