Dave Cuthbert wrote
The nick name for MFJ is Mighty Fine Junk.
Yes it is -- or has been. But I'll jump in here to add that while I've in
the past often been underwhelmed by the quality of some MFJ equipment, I
was favorably impressed with my MFJ-259B. And it is quite useful.
I have one of
David Sterner wrote:
UTP has been repeatedly shown to radiate less than STP.
Telcordia GR-1089 exempts STP from intrabuilding surges; the shield is
assumed to carry them. This makes it attractive for US telecom designers
whose equipment uses Ethernet.
Cortland
This message is from the
Dave Cuthbert write
the point at which ferrites are placed will not always have a common
mode impedance of 50 ohms. Here's An example: a large DUT has a 1 meter
long cable that connects
Not always; make that rarely.
Comments about the 150 ohm impedance are on target. That might be
Ghery Pettit wrote:
Chris,
You can indeed make your own, but my bet is that A2LA or NIST NVLAP
inspectors will want to see calibration data, not calculations.
Now, if we just had a published calibration technique...
This is still not rocket science. Using Z = 138*(log OD/ID) --- for an
air
of
cutting off their access to EMC and PSTC matters, or at least rendering it
more expensive, and less frequent.
Cortland Richmond
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc
It's been a long time since I did military-style testing -- TEMPEST in the
1980's -- but it seemed to me then that no great care was taken to control
cable common-mode impedance. The environment INSIDE the chamber was as
unrealistic as one might care to get, and no attempt was made to control or
I was looking over Dries' post again and note that I overlooked something
important: He says that short shield goes on the *wall plate* -- which I
presume to be the shielded room wall. This doesn't change my recommendation
he test with unshielded wires. But it serves as a reminder that support
Dries, it sounds very much as if that short shield is acting as a
(low-value) bypass capacitance. If, as you say, no shield is permitted,
then your customers will audit with no shield, and the product will fail no
matter what tricks you used to get around failure in your own tests.
Then you will
As you can see, the below message arrived unreadable at my compuserve
account. Since I can't read attachments with my off-line reader anyway
without (1) finding a numbered file and (2) opening it with a text reader,
would you be kind enough to re-send the message, this time as ASCII text,
and
Kurt,
Can you be more precise? Is it really 121.000 MHz? There are a number of
low-power, possibly even individually compliant, digital sources which
could be heard around 121 MHz; knowing the exact frequency would allow
pinning it down to one of several clock frequencies, if it is such a
Ghery Pettit wrote:
shall be fitted with ferrite clamps placed on the floor
Well, darn.
Unfortunately, I am not right now in a position where I can procure or look
at standards.
However... I am disappointed that the writers took the route they did. I
thought we were moving away from
Ghery Pettit wrote:
The cables are specified to enter the clamp at the point where the
cable reaches the surface of the
turntable. I don't believe that your suggestion meets this requirement.
Yes, and I thought about that; if ferrites start where the cable enters the
ground plane, that
Tim Pierce wrote:
The new amendment calls out for ferrite tubes on all power cables and
I/O
cables exiting the test site. The problem I'm finding is that the FCC will
not accept this test setup. Does anyone know if the FCC is going to
approve
this test setup?
What of buried cables? It is
Chris Maxwell wrote:
There may be some administrative (follow on testing, factory
auditing...) aspects of EN 60601 which may have costs that are hard to
justify for a mouse tester.
Some years ago a NEMKO engineer recounted (while witnessing a test at my
then employer) having tested a
Lou Aiken wrote
... what practical reasons there are for using PC traces to provide
earth fault circuits.
One practical reason is, to cut costs and simplify construction. Some years
ago a former employer designed and made a computer power supply with the
safety ground on the board, and UL
Derek wrote:
the EUT should have been exposed to simulated shipping and installation
by a user...
FWIW, in the 1980's I worked in an audit lab where we tested samples of
shipped equipment for FCC, vibration, heat, humidity, temperature,
TEMPEST... it was not uncommon for equipment to do
What no one checks -- no one does. Got an OLD PC? One certified under the
old rules, with an ID number? Try THAT.
A local computer store (of a national chain) is selling computer chassis'
with plastic sides. No complaints, no problems.
It appears no one is CHECKING.
Cortland
This
Blocking a whole ISP is like throwing out mail by postmark. Maybe worse.
Enough said!
Cortland
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/
To cancel your subscription,
I would like to create a check list in order to help designers and PCB
routers of my company to improve their design / EMI and SI.
This is one of the things we get paid to do, assuming anyone is
paying us at all. For this and other reasons given by those who've
already answered your request,
I couldn't resist. After a whole string of I'm out of the office reply
messages on emc-pstc, what shows up in the mail queue?
Where Have They Gone? Who Are They Now?
Click Here: http://( deleted)
Ad for a Search firm.
I LOVE it! Still chuckling, ten minutes later.
Cortland
This message is
I had occasion once to look at the input relay on an NM-37. The set acted
as if relay contacts were corroded -- but the sealed reed relay was fine;
the problem was a cold-solder joint.
Relays sealed in inert gas or vacuum should never fail due to corrosion,
and there's no need to limit their
Neil Helsby wrote:
Has anyone else noticed the tendency of modern switch mode power supply
designers to save manufacturing costs at the expense of harmonic and
conducted emissions measurements?
It's not modern. I ran into it in the 1980's. Saving costs isn't _bad_,
mind; it's simply doing
I believe the term bad, as applied to a fuse which has functioned as
intended, is a report of its (no pun intended) current condition, rather
than its suitability for the purpose. A better word, I think, and still
accessible to laymen, is blown.
On the other side fuses and fuse holders were
Johnson and Jasik in _Antennas_ refer to TV receiving loops in their
chapter on TV antennas, chapter 29. The loop antenna in general is
discussed in Chapter 5. A loop and plane reflector is also discussed.
Amateur experience with the one wavelength loop (search for quad
antennas) suggests
There's a temptation to troubleshoot these problems by applying bandaids
until one works. They're usually the wrong bandaids, and it takes a long
time to learn which one is right, if any. I've found it useful instead to
inject short pulses directly into suspect devices and traces using a pulse
Chris Maxwell wrote
Your customers may not NEED to touch your product; but they COULD touch
it. To me, that is accessible.
And there there are requirements one imposes on his own products, not those
mandated to lawfully market them. How much will it cost me NOT provide
immunity in some area?
Tim Pierce wrote
I have worked with sites that used the hardware cloth (screen) over
concrete
in the past. When the tears would happen, they would patch that area ...
Tim,
If this were a reflector antenna, holes in the reflector would be kept to
0.1 wavelength at the highest frequency. This
Dave Cuthbert (drcuthb...@micron.com) wrote:
If I understand the OATS cal procedure, the RX antenna height is moved
from 1 meter to 4 meters and readings are taken. Now this is the strange
part: The readings are averaged. Is this right? Now think about it- when a
DUT is tested, the RX antenna
My s/c thinks Saddam Hussein is Adams Hissing.
Hmm.
Cortland
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/
To cancel your subscription, send mail to:
,
Cortland Richmond
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/
To cancel your subscription, send mail to:
majord...@ieee.org
with the single line:
unsubscribe emc-pstc
Chris Maxwell wrote:
To me, this whole thread begs the question: Why not put the surge
supressor in front of the UPS?
A lot of us have relied on full-time UPS's to stop transients. But the bit
about a UPS itself producing large enough ones to trigger suppressors, that
is worrisome.
I don't believe there are any power quality requirements imposed on
inverters except those specified by buyers (in the USA, anyway). You can
still buy square-wave inverters, after all, and the modified sine wave
could be called Chock Fulla Harmonic power. Harmful interference is a
limitation, but
Ken Javor wrote:
My way of thinking about this problem and others like it is to reduce it
to the simplest possible problem, and then look at the general case as a
linear extrapolation. In this case I look at the interaction of the direct
and single bounced ray during a site attenuation
Ken Javor wrote:
My thinking is just the opposite. The duration of the pulse should be
long
relative to the time it takes to travel from transmit to receive antennas.
Then there is no smearing
It seems to me that you may be overlooking the effect of the reflected wave
on a received pulse's
Neil Helsby wrote
if I am to include a bought in power supply (or any other module) in my
system I need to have a good idea that when I test my system an EMC failure
due to the design and/or construction of the power supply does not cause me
grief.
This is a problem for everyone whose product
Derek N. Walton wrote:
I've often wondered what would happen if the FCC ( for example ) had
under
cover engineers that took products barely passing to test labs.
Specifically,
what the results would be :-)
That is pretty close to what the FCC used to do, remember? Testing indoors
at one
Don Borowski at Schweitzer Engineering Labs wrote:
Loading up the PCI slots puts more capacitance on the PCI bus, slowing
the
logic transistions and thus reducing emissions. Quite likely loading up the
ISA slots will do the same thing, though emissions could go up.
Some years ago, a computer
Many of us here are Amateur Radio operators, and have filed comments of our
own in response to the Notice of Inquiry. By the time I got mine in -- I
waited too long and rushed it, so there are typos and other errors -- over
1600 comments had been filed. Reply Comments are now being taken;
Ken,
A recent article on reverberant chambers mentions a Q of 83,000 or so. I'd
expect a pulse to excite many modes within a chamber as long as either its
length or its transition times are shorter than the time it takes a wave to
travel across the chamber and back. That'll smear the pulse. On
Charles Grasso wrote:
What I was(am)trying to do was match the max voltage
as measured on a scope with the value as measured
on a SA.
Try zero span on the SA. Compare THAT with the scope.
Cortland
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc
Ken Javor wrote:
... but I wonder if the presence of even an rf emission (as contrasted
to 50 Hz) should cause any problem to a land-line phone with a wired
handset. It wouldn't seem likely that the
power available from a couple AA batteries would be sufficient for that.
Now if it were a
CISPR 16-1 and C63.2. I seem to recall seeing a bandwidth mask in CISPR 16
which specified both width and slope of the filter attenuation in three
ranges, from 0 to 3 dB down, 3 to 6 dB down and 6 db to (I think) 40 or 50
dB down.
Cortland
This message is from the IEEE EMC Society Product
with the provisions
of the National Electrical Code, National Fire Code, and use of NRTL listed
electrical devices, too.
Cortland Richmond KA5S
Radio Interference
Where you want it!
When you want it!
(grin)
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc
The Commission requires registration, and high visibility paint and
lighting for Amateur (and other) antennas and towers more than a certain
height above ground level, and coordination with the FAA. Recently, they've
required evaluation for RF safety. Authorities probably have like rules in
Hello everyone,
I am looking for a manufacturer of an active comb filter to notch out
50Hz+harmonics noise from bio-medical signals.
Can anyone help.
Regards
Shaike Raz
I encountered a similar problem almost 20 years ago and thought about using
a digital delay line to generate a powerline
Derek N. Walton wrote:
Two big problems with ALL MB's. Plug in a PS/2 mouse, and the emissions
dramatically increase. Also, this happens with the KB too. It would seem
that the
designers of the boards put a switched mode power supply right next to the
connectors: not a smart move in my
This is an an interesting sort of problem with a number of factors. (A
former manager used to turn gray when I called something interesting.)
One might see a very slow longitudinal DC motion as charges pass over
elevated lines, sometimes enough to fire protectors. DSL equipment uses,
and is open
Vic Gibling wrote
Was the man matching the assistants humour/sarcasms?
OR
Was he ignoring the warnings, it's not going to happen to him?
OR
Was he making a rational risk assessment based on the
information provided by the manufacturer?
I'm in California. I sometimes hear an anti-smoking
John Harrington wrote:
I'm having some fun with an NSA measurement. I get a swing of 8dB
between
30MHz and 45MHz horizontally polarized. At 30 MHz I get too little
attenuation (i.e. I receive too much of the transmitted signal) and at
45MHz
too much attenuation. The rest of the frequency
Derek Walton wrote:
I'm looking for sources of flexible Ferrite sleeving. Can anyone help?
EMC Eupen is still selling it, I think, and CAPCON has already been
recommended to you.
I've looked at both of their products, and ended up not using them. The
problem I have with ferrite impregnated
Ken Javor wrote:
I've seen that design in a D-dot sensor. The elements were a couple of
spheres. The output impedance of such an antenna is quite high, it really
is an electric field probe, and works at frequencies where the elements are
electrically short. These types of devices really sense
There's ESD - and there's ESD. It's not always the zap that gets you.
More than a few years ago, while working at a (now gone) computer
manufacturer, I was urgently summoned to the software compatibility lab.
There I heard the EMI group accused of generating EMP and making evaluation
mother
For this and other fundamental questions, we need to look at the standards
and publications on which the GR's are based, and to which they refer. I
don't have access to standards or GR's at present, but a quick Google(tm)
search shows ASTM B-117 and Mil-Std-810 for salt fog testing, and a wealth
Richard Woods wrote:
An audio amplifier that processes signals over 9 kHz is considered to be
an unintentional radiator and must be verifified.
I don't believe I've heard of the FCC actually requiring analog audio gear
to be verified, though even a cheap record player with 10 KHz audio does
Amund Westin posted
Manufacturer A will put this new system onto the market.
Ir seems to me that you have answered your own question. Whoever puts [a
something] on the market is responsible for its compliance, as marketed.
It's entirely possible that item Z has been shown compliant in a
Jeff Chambers writes:
This removes the 'line of sight' gap into the enclosure. Does this
improve the attenuation? Intuitively it should, but if the leakage occurs
because of the interruption in shielding
conductivity and hence current flow at the seam, it won't.
I've run into this
Robert Macy asks:
Isn't it fraud to sell an item that does not meet spec? As
in, the vendor says the product is compliant when it is
not.
Don't all the customers then have a legal recourse to
return any/all product?
Well -- here's what _I_ think:
Problems arise when a manufacturer has
John Woodgate wrote:
Telling us here won't have any effect. If you are concerned about
violations, tell the regulatory authority.
This is worth some discussion, I think.
Most of the time, EMC engineers are not cops. Among other things, we want
to be sure our OWN employers are regulation
This is sort of off topic -- but it's on topic too. Some years ago, at a
now defunct computer manufacturer, customer complaints of memory
incompatibility lead us to purchase test equipment specifically designed to
evaluate memory cards. What we found was that, although different vendors
cards
Charles Grasso wrote
Can a monitor purchased from an OEM have a company label slapped on it and
sold WITHOUT the company testing it in a system?
Looks like it!
One of my EMI horror stories concerns a monitor I found could make a
computer NEXT to it, not even turned on, exceed the Part 15
A cost analysis can be used to determine the payback time. I know of no
good way to determine the opportunity costs and I have not seen
accountants pay much attention to this.
Some years ago, at a VERY cheap company, I was able to justify the complete
cost of a brand-new spectrum analyzer as
for that
as well.
I'd encourage everyone to make a quick zero-span, linear mode SA modulation
check part of routine calibration checks.
Caveat modulator.
Cortland Richmond
This message is from the IEEE Product Safety Engineering Society
emc-pstc discussion list.Website: http://www.ieee-pses.org
I'd be inclined to look at the design. Bandaids have a way of multiplying
until your product looks like a mummy.
If it's a low-level power problem, make sure the EUT's regulator can
respond to induced ripple. This may be as simple as exchanging a cheap
electrolytic cap for one with better ESR,
telephone products destroyed by all sorts of bazaar events
Where IS that bazaar? (grin)
Cortland
---
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
Visit our web site at:
The important thing is, first average the quantities, then convert to dB.
Ever seen folks doing video averaging on a log-scaled analyzer display?
Sure you have. And it's wrong. How wrong?
Take two samples, 100 dBq and 25 dBq. Sum their amplitudes in dB (100dBq +
25dBq= 125dbq) and divide by
Sounds to me that while the POWER SUPPLY keeps working, it does not
sufficiently attenuate EFT presented to it that your product keeps working.
It is POSSIBLE you might find a power supply that does, but unless you can
control whose PS customers use, it may be smarter in the long run to fix
the
If a test is performed on a slowly cycling state machine, then increasing
the rate of discharge may not help. Knowing how many discharges it takes to
stop something is often as helpful as knowing how much energy it takes, and
if we apply fifty where one would have done, we have gained little by
I think we may be assuming this overload is caused by the EUT. But this is
just as likely to be caused by something else. Medium Wave and Long Wave
broadcasting produces powerful fields at some distance from an antenna.
This has a fix.
If you place a narrowly resonant loop antenna, with feed
Also see Telcordia GR-1089.
Cortland
---
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/
To cancel your subscription, send mail to:
At a former employer, we finally ended up putting resistive load banks in
computer chassis' (as shielding) for testing power supplies. But we found
that power supplies can fail at full load from their own emissions, and
later, installed in working equipment at lower loads, due to passing
through
On 6-Aug-01 John Woodgate j...@jmwa.demon.co.uk wrote
In Europe, the limits are specified in dB(uV/m), but no-one has been
daft enough to propose limits like 53.9790009... dB(uV/m).
Folks HAVE been daft enough; 3 volts per meter is 129.542425 dBuV/m,
right? Result of specifying in two
On 14-Aug-01 Dan Pierce dpie...@openglobe.net wrote:
I have found that if I bypass the earth ground plug I can measure a 80
VAC potential from my chassis to earth ground. I found this out in the lab
when someone touched the chassis and a grounded bench and got zapped. Is
there guidelines
You need to keep fields below the NEBS immunity limits, which are pretty
low, IMHO. However, a 300 mW cellphone is not the same danger as a 5 watt
high-band HT.
Cortland
---
This message is from the IEEE EMC Society Product Safety
Technical Committee
one second
to show up. You want to do the test in such a way that the tester can note
and adjust to this; I once tested something that had a 30 second delay
before a failure showed up. This can't be helped, and, in this particular
case, one second is far too often.
Cortland Richmond
If you want to have your grant of authorization -- or your marking --
challenged by a competitor or discovered noncompliant by the FCC, all you
need to do is play games with the rules. Conservative observance of
reasonable interpretations of the Rules seems prudent to me.
Cortland
If I use a 6 MHz crystal as reference in my 1296 MHz generator chain I
cannot reasonably say that my test must stop at the tenth harmonic of 6
Mhz. I must test to the tenth harmonic of 1296 MHz, and in excess of 10
GHz.
Cortland
---
This message is from
FCC Part 15 goes far beyond digital devices; carrier
current systems, radio and TV receivers, unlicensed
low-power transmitters, and much, much more.
It gives special treatment to digital devices because
they have proven to need special attention. The exemptions
Part 15 contains exist because
Carmen,
FCC requirements don't directly concern 60950.
They are additional to it. Part 68 is now
handled differently than in the past; check
out the FCC Web page. And you should be aware
that meeting FCC Class A emission limits is
not enough. If sold to the public, your product
must meet the
Some monitors from the 1990's WILL meet class B with small H's. I know;
I've done it! At a former job (gad, we use that a lot!) we were able to
get monitors scrolling H's in high resolution to pass Class B. Sometimes it
took modifications, which the manufacturer would copy, but it could be
done.
=There really ARE Class B computers, but you
have to work to make them, and you have to look
to FIND them. I've worked on some of them.
Between CRT's and LCD's, the generators are
different, but the levels are not necessarily
lower for an LCD.
A CRT has a fairly powerful, high frequency
For telecom equipment, one also must bear in mind the required transient
and power-cross stresses. I am just now researching how these differ
between the USA and Europe but one example may suffice: Level 1 common mode
transients of of 1000 volts at 1.2/80 usec are applied to the input under
EN
Lower voltages, because of less corona, tend to have more energy, sooner,
relative to the total. You might pass a 15 kV air discharge test -- but
fail, lower.
We ALWAYS want margin, and others ALWAYS want none. My experience in a
previous area of the industry is, this lasts until a rash of
Since the contact method requires penetrating thin, cosmetic coatings, it
is a bad idea on membrane keypads; bore that sharp point into the contacts
and it's ruined even before you hit the electronics. Not that they'd fare
all that well with direct discharge! In any case, there is often some
A lump of coal will heat up in a microwave oven. No water needed. It is the
bulk resistivity of an object which allows circulating currents to generate
heat, and while water can help (when we wet something, its chemical bonds
are a source of electrons) it is not the only thing that is heated in a
An STA is basically a permit for operation not normally allowed a license
holder. However, RF immunity testing covers so many frequencies, at such
power, that unless the nearest town is over the horizon, and you have
jungle canopy overhead to absorb RF and keep from jamming satellites, I
think it
in
that event call you to task, claiming you
knew, or should have known, that this would
happen when you made a device recognizably
fake. Another reason to add the mark!
Cortland Richmond
== Original Message Follows
(headers and trailers snipped)
Kevin
HP used to have some information on each probe type's actual impedance in
the instructions packaged with them; it was educational, to say the least.
A 10 meg probe turned into a rather greater load at 200 MHz. Then, too, the
'scope's -3 dB point was much modified by the probe one hung on it.
Ed, and list members,
I have for some time been using a digital camera to include data in reports
and e-mail form. It turns out to be as convenient, sometimes more, as
dumping data directly from a digitizing instrument into a computer. I can
use any kind of display -- analog or digital readout,
Larry,
I had no idea there even WAS such a standard.
Interesting, the things it's possible to learn
here; an assemblage of experts, indeed!
Thanks!
Cortland
== Original Message Follows
Date: 24-Jan-01 15:19:07 MsgID: 1078-329 ToID: 72146,373
Debbie,
Doug Powell explained it as venting, where slots or holes are added to a
plane in order to let vapor out. However, it is my belief that this is also
done to keep copper balanced during the etching process. I may well be
mistaken, but the EMI and signal integrity concerns we have with the
Logically, if it can make emissions worse, it can make them better, too.
Suppose your original board exhibited some resonance, and thieving changed
it. That would certainly lower emissions due to a resonant board. Anything
that resonates will radiate, and this can even be a whole board in its
Yes indeed. My preference is for thieving to be done with dots or
islands small with respect to the shortest wavelength of concern. This is
because thieving can be constructed so as to resonate and aggravate an EMI
problem.
Some years ago, at an employer far away (grin), we had obtained
Ken,
When you ask how members feel, you open a Pandora's box!
We must still meet some kind of installed bottom line; our equipment must
not generate fields above some limit. (We can argue what that should be
some other time.)
However, when _designing_ an EMC solution, we can estimate field
I'd say either a comb generator, or a sweep generator but use them to
excite a test object of the same general size as the equipment you wish to
test. The smaller your chamber, the more it will be affected by the size of
an EUT sitting in it. If you can be pretty sure what you will test, add its
Years ago, I carried some equipment to Marble Falls, Texas, to test at
Professional Testing, Inc's OATS there. I remember one of the sites having
a taut cloth weather shield stretched over a number of bent PVC formers. I
don't know how long those PVC ribs lasted in the Texas sun, but I DO
Yes, it's important to recognize the limitations of a test method before
relying on it. In this case, the A-B method was used at power frequency and
below, to observe discharge time for UL testing. The MAIN advantage was not
balance so much as that the chassis was isolated from the potential
Can you please explain why? The receiving antenna just responds to the
field strength at its position; it doesn't 'know' anything about the
source - it cold be an EUT at 10 m or a distant TV transmitter or even a
cosmic source.
This isn't the issue. The receiving antenna, as you say, can't
Sam,
I think you did it right with one AF and one gain. There's a problem with
that method. You need more information needed to make the _results_
right.
Given a certain power at the antenna terminal, and a known gain and
efficiency, you can calculate the free-space field strength at some
One can also use a pair of probes known to be well balanced and take the
difference between the A and B channels, or
use a differential input accessory or plugin. There is then no possibility of
high-voltage on the instrument chassis.
Cortland
(my own thoughts, and nor those of my employer)
John Shinn wrote:
If we refer back to the series, and refer to the n-th term, we
would all be on the same page (and harmonic).
And in harmony!
Cortland
(disclaimer: my views, not Alcatel's!)
---
This message is from the IEEE EMC Society Product
1 - 100 of 513 matches
Mail list logo
