TTWOT - Is it me?

In article , Bill Wright wrote:
and it must be so well
known that yellow sodium lights look red on television that we've just
learned to ignore it.

Blimey! I though they really were red and it was just my eyes that made
them look yellow in real life.

:-) That depends on how you look at it.

Rod.
--
Virtual Access V6.3 free usenet/email software from
http://sourceforge.net/projects/virtual-access/

On Wed, 28 Sep 2011 13:54:16 +0100, PeterC
wrote:

On Tue, 27 Sep 2011 18:00:36 +0100, Johny B Good wrote:

On Tue, 27 Sep 2011 16:30:40 +0100, PeterC

wrote:

====snip====


A bit OTT, but Morrisons sell 30W CFLs with a PF of close to unity.
About
3-quiddish IIRC.

That's interesting. Is the PF figure a quoted one or the result of
measurements taken with a decent digital watt meter?

I've a Maplins plug-in-type meter that gives c. 0.% for the usual CFLs
and
close to unity for the 30W.

That's interesting. Maplin do several such 'watt meters' I'm curious as
to the model you're using. I've got a couple of Maplin digital watt
meters. They're different makes. One describes itself (on the box)as a
"Plug-In Mains Power & Energy Monitor" Maplin code: L61AQ which is a
2000MU-UK (a 240v version of the KAW PM4400). The other meter is Maplin
code: N67FU.

Both these meters were bought at a local flea market from "Maplin Man", a
regular trader, for two or three quid each and contrary to my earlier
experience with cheap ToolMart (and Aldi) meters, they proved surprisingly
accurate. Although the 2000MU-UK's ability to measure _anything_ at all
below the 1 watt level (real or imagined) left something to be desired
(which desire was satisfied by the N67FU meter).

Irritatingly, the more accurate N67FU doesn't have the PF and apparent
power functions, despite the claims on the box label. This means you have
to calculate the apparent power from the product of the voltage and
current readings before you can calculate a PF. Luckily, it's only the
true wattage figure I'm interested in measuring in most cases.

BTW, on those lamps that were giving a 0% PF indication, were you able
get any wattage readings?

--
Regards JB Good

On Wed, 28 Sep 2011 13:56:16 +0100, PeterC
wrote:

On Tue, 27 Sep 2011 20:41:14 +0100, Max Demian wrote:

A bit OTT, but Morrisons sell 30W CFLs with a PF of close to unity.
About
3-quiddish IIRC.

Why does the power factor matter to you?

It doesn't of course (other than hundreds of millions of the things all
with
poor PFs being not too good overall), but as 20W CFLs are supposed to be
corrected I just checked it out of interest.

Unless you've been testing some old CFLs, I think you'll discover that
they too have a PF very close to unity. Now, narrow conduction angles due
to the use of a simple capacitor smoothing cct after the rectifier, is
another (significant) question altogether.

--
Regards JB Good

On Fri, 30 Sep 2011 04:10:33 +0100, Johny B Good wrote:

On Wed, 28 Sep 2011 13:54:16 +0100, PeterC
wrote:

On Tue, 27 Sep 2011 18:00:36 +0100, Johny B Good wrote:

On Tue, 27 Sep 2011 16:30:40 +0100, PeterC

wrote:

====snip====


A bit OTT, but Morrisons sell 30W CFLs with a PF of close to unity.
About
3-quiddish IIRC.

That's interesting. Is the PF figure a quoted one or the result of
measurements taken with a decent digital watt meter?

I've a Maplins plug-in-type meter that gives c. 0.% for the usual CFLs
and
close to unity for the 30W.

That's interesting. Maplin do several such 'watt meters' I'm curious as
to the model you're using. I've got a couple of Maplin digital watt
meters. They're different makes. One describes itself (on the box)as a
"Plug-In Mains Power & Energy Monitor" Maplin code: L61AQ which is a
2000MU-UK (a 240v version of the KAW PM4400). The other meter is Maplin
code: N67FU.

This is a 2000MU - very useful for the energy of, say, the washing machine
(which has a PF of ~0.15 on the faffing part of the spin).

Both these meters were bought at a local flea market from "Maplin Man", a
regular trader, for two or three quid each and contrary to my earlier
experience with cheap ToolMart (and Aldi) meters, they proved surprisingly
accurate. Although the 2000MU-UK's ability to measure _anything_ at all
below the 1 watt level (real or imagined) left something to be desired
(which desire was satisfied by the N67FU meter).

Yes, the TV and FS box each shows 0W on standby; the 2 together show 1W but
this is probably 1.5W I guess.

Irritatingly, the more accurate N67FU doesn't have the PF and apparent
power functions, despite the claims on the box label. This means you have
to calculate the apparent power from the product of the voltage and
current readings before you can calculate a PF. Luckily, it's only the
true wattage figure I'm interested in measuring in most cases.

I'd looked at Aldidl's offerings and not bought them due to lack of these
funcions. Although 1W would be interesting it's not of any use to me.

BTW, on those lamps that were giving a 0% PF indication, were you able
get any wattage readings?

OK, just measured some Philips Genie:
11W reads 9W, PF 0.6
18W reads 16W, same PF
Status lamp:
30W reads 28W, PF 0.95.

The meter is OK on filament lamps.
--
Peter.
The gods will stay away
whilst religions hold sway

On Fri, 30 Sep 2011 04:22:06 +0100, Johny B Good wrote:

On Wed, 28 Sep 2011 13:56:16 +0100, PeterC
wrote:

On Tue, 27 Sep 2011 20:41:14 +0100, Max Demian wrote:

A bit OTT, but Morrisons sell 30W CFLs with a PF of close to unity.
About
3-quiddish IIRC.

Why does the power factor matter to you?

It doesn't of course (other than hundreds of millions of the things all
with
poor PFs being not too good overall), but as 20W CFLs are supposed to be
corrected I just checked it out of interest.

Unless you've been testing some old CFLs, I think you'll discover that
they too have a PF very close to unity. Now, narrow conduction angles due
to the use of a simple capacitor smoothing cct after the rectifier, is
another (significant) question altogether.

These are 2 - 4 yo.
--
Peter.
The gods will stay away
whilst religions hold sway

On Fri, 30 Sep 2011 09:31:53 +0100, PeterC
wrote:

On Fri, 30 Sep 2011 04:10:33 +0100, Johny B Good wrote:

On Wed, 28 Sep 2011 13:54:16 +0100, PeterC

wrote:

On Tue, 27 Sep 2011 18:00:36 +0100, Johny B Good wrote:

On Tue, 27 Sep 2011 16:30:40 +0100, PeterC

wrote:

====snip====


A bit OTT, but Morrisons sell 30W CFLs with a PF of close to unity.
About
3-quiddish IIRC.

That's interesting. Is the PF figure a quoted one or the result of
measurements taken with a decent digital watt meter?

I've a Maplins plug-in-type meter that gives c. 0.% for the usual CFLs
and
close to unity for the 30W.

That's interesting. Maplin do several such 'watt meters' I'm curious
as
to the model you're using. I've got a couple of Maplin digital watt
meters. They're different makes. One describes itself (on the box)as a
"Plug-In Mains Power & Energy Monitor" Maplin code: L61AQ which is a
2000MU-UK (a 240v version of the KAW PM4400). The other meter is Maplin
code: N67FU.

This is a 2000MU - very useful for the energy of, say, the washing
machine
(which has a PF of ~0.15 on the faffing part of the spin).

Other than when measuring sub 10 watt loads, that's an excellent meter.
It's only let down by the fact that they've thrown insufficient resolution
at the ADC handling the current sensor data (and wasted needless
resolution on the voltage sensor data). Since the voltage is the "fixed"
variable in the calculations, a 10 bit ADC will be more than ample to
capture the voltage waveform to the nearest volt.

The current waveform, otoh, being the variable 'Variable' which can range
from as low as 0.4065mA (if you want to capture wattage readings down to
0.1 watt on a 246v supply) to as high as 15A (the original US model's
limit), you need to be able to represent the lowest current waveform value
with enough resolution to approximate a sine wave (probably 4 bit's worth)
plus the additional bits to cover the 3 1/2 orders of magnitude variation
over this range (~16 bits) extra giving a 20 bit resolution requirement
for the current waveform data.

I suspect the 2000MU-UK meter is only using 16bit ADCs for both data
sets, whereas the N67FU has distributed the sum total bit accuracy of 32
bits more intelligently between the voltage and current sensor ADCs. I'm
guessing at a 12 / 20 bit split in favour of the current sensor data (the
N67FU only reads voltage to the nearest volt, despite the 4 digit display
ability to show values of 0.1v increments).

Ignoring the issue of having to manually do the mathematical operations
yourself if you need to know the PF and the Apparent Power figures, the
N67FU model is the one to go for, especially if you want to measure sub 1W
consumption (standby power on modern smpsu wall warts and phone chargers).


Both these meters were bought at a local flea market from "Maplin
Man", a
regular trader, for two or three quid each and contrary to my earlier
experience with cheap ToolMart (and Aldi) meters, they proved
surprisingly
accurate. Although the 2000MU-UK's ability to measure _anything_ at all
below the 1 watt level (real or imagined) left something to be desired
(which desire was satisfied by the N67FU meter).

Yes, the TV and FS box each shows 0W on standby; the 2 together show 1W
but
this is probably 1.5W I guess.

If my test on the power consumption of my Metrawatt analogue watt meter
is anything to go by (the essentially resistive loading from the 50K /100K
/250K multiplier used on the 100 /200 /500 volt settings), I would guess
at an error the other way. The 200v setting showed zero on the 2000MU
whilst the N67FU displayed 0.7W and, using the 100v setting, I saw
readings of 1.7 and 1.2 watts respectively (the calculated wattage figures
being 0.6 and 1.2 watts).


Irritatingly, the more accurate N67FU doesn't have the PF and apparent
power functions, despite the claims on the box label. This means you
have
to calculate the apparent power from the product of the voltage and
current readings before you can calculate a PF. Luckily, it's only the
true wattage figure I'm interested in measuring in most cases.

I'd looked at Aldidl's offerings and not bought them due to lack of these
functions. Although 1W would be interesting it's not of any use to me.

Although the current model of wattmeter now being sold by Aldi might be a
vast improvement over the earlier model (DEM1379), I wouldn't trust it
without being able to compare it against an existing meter of known
accuracy (such as an analogue meter or the N67FU already mentioned).

The 'Apparent Power' and 'Power Factor' functions are really a non
-essential 'convenience feature' since these parameters can so easily be
calculated from the current and voltage readings anyway.


BTW, on those lamps that were giving a 0% PF indication, were you able
get any wattage readings?

OK, just measured some Philips Genie:
11W reads 9W, PF 0.6
18W reads 16W, same PF
Status lamp:
30W reads 28W, PF 0.95.

The meter is OK on filament lamps.

I've just tried a collection of CFLs to compare with your experience.

The first one up is an old Philips SL13 Comfort lamp (old skool inductor
ballast) and the Metrawatt shows 14.7W (less the 0.7W consumed by the
2000MU -UK meter which shows 15.1W (sensitive to mains voltage
variations). The digital meter displays a PF figure of 0.55 (just what
you'd expect on an uncorrected choke ballasted fluorescent fitting), 28.4
VA and a current reading of 0.11A. For some reason the VA figure is
noticeably higher than the 26.4VA calculated from the volts and amps
readings.

Trying one of those cheap and nasty 7W CFLs sold by pound shops under the
brand names of 'Status' or 'Pearl' (you know the type - the ones that give
off a ghastly (ghostly?) bluish light) the real wattage measures 7.3W and
I get the following on the 2000MU-UK meter 7.0W, 14.8VA, PF=0.47 240v and
0.06A. In this case the calculated VA of 14.4 is a closer match to the
displayed VA figure which, in turn, produces a calculated PF value of
0.486.

A brand new unused Tesco 20W stick lamp shows a real power consumption of
20W whilst the digital meter shows 240v, 19.0W, 32.9VA, 0.13A and a PF of
0.49. Curiously the lamp's current is indicated on the base as being 192mA
(Average rms? Average or peak? - it can't be the first otherwise the VA
figure would be 46).

TBH, I can't see how an electronically ballasted lamp can show a PF much
below unity since, as is the case with all SMPSUs, the mains voltage is
directly rectified and the 100/120Hz ripple on the output smoothed by a
simple capacitor filter. Whilst this results in narrow angles of
conduction on each voltage peak, the current still remains in phase over
that conduction period and no stored power ever flows back into the mains.

The so called (passive) PF correction inductor used in more sophisticated
smpsu designs is simply a means to take the sting out of that current
spike and extend the conduction period so as to reduce the gross harmonic
distortion in the current waveform which, if left unchecked, would present
a serious problem to the mains supply network.

Looking at the mains waveform on a 'scope already reveals the effect of
narrow conduction angles from the millions of smpsu powered gadgets now
plugged into the UK National Grid. This is quite evident by the very
obvious flat topping of what is supposed to be a sine wave of reasonable
purity. This issue is now serious enough to make it obligatory that PC ATX
PSUs and the like incorporate 'Power Factor' correction circuitry to stop
the problem getting any worse.

I suspect electronically ballasted CFLs below the 20W rating may be
exempt from this requirement for pretty much the same reason that
conventionally ballasted fluorescent lamps below the 20W limit were exempt
from the need to fit a PF correction capacitor[1]. It's quite possible
that the 30W lamp you tested may not be exempt and incorporates active PF
correction circuitry as an integral part of the electronic ballast. Even
so, it surprises me that these meters detect the (in phase) current spikes
as some species of sub-unity power factor.

[1] The reason being that such low power lamps, even in their millions,
would only form a very small part of the total loading on the National
Grid and so not be a problem.

--
Regards JB Good

Woody wrote:

Has anyone noticed that more and more light bulbs - especially the
18/20/23W PLCs - are now ES?

I think it's to do with internationalisation. ES have long been more
popular in America and Europe, with bayonets here in the UK.

To me it seems like the manufacturers want one product that will work
anywhere, so they go with the majority.

Have you noticed how taps (bathroom, not electrical) have changed in
the past decade or so? Lots of them now have 8mm or 10mm entries,
which is fine if you use them with mains pressure water (like in most
of Europe). Unfortunately, here in the UK with our traditional
fondness for header tanks, they flow too slowly.

--
SteveT

On Sat, 01 Oct 2011 10:19:11 +0100, Steve Thackery
wrote:

Woody wrote:

Has anyone noticed that more and more light bulbs - especially the
18/20/23W PLCs - are now ES?

I think it's to do with internationalisation. ES have long been more
popular in America and Europe, with bayonets here in the UK.

To me it seems like the manufacturers want one product that will work
anywhere, so they go with the majority.

Have you noticed how taps (bathroom, not electrical) have changed in
the past decade or so? Lots of them now have 8mm or 10mm entries,
which is fine if you use them with mains pressure water (like in most
of Europe). Unfortunately, here in the UK with our traditional
fondness for header tanks, they flow too slowly.

I've noticed ;-) The worse case, IMHO, is with toilets where they
take 10 minutes to refill.
--
(\__/) M.
(='.'=) Due to the amount of spam posted via googlegroups and
(")_(") their inaction to the problem. I am blocking some articles
posted from there. If you wish your postings to be seen by
everyone you will need use a different method of posting.

On Tue, 4 Oct 2011 08:52:03 UTC, Mark
wrote:

snip
I've noticed ;-) The worse case, IMHO, is with toilets where they
take 10 minutes to refill.

Which is usually a mains pressure nozzle in the ball valve rather than
a low pressure one. Most ball valves these days come with both - but
one has to fit the correct one.
--
Regards
Dave Saville

On Thu, 6 Oct 2011 06:27:51 +0000 (UTC), "Dave Saville"
wrote:

On Tue, 4 Oct 2011 08:52:03 UTC, Mark
wrote:

snip
I've noticed ;-) The worse case, IMHO, is with toilets where they
take 10 minutes to refill.

Which is usually a mains pressure nozzle in the ball valve rather than
a low pressure one. Most ball valves these days come with both - but
one has to fit the correct one.

If the fitter bothers to do this. Mostly they don't.
--
(\__/) M.
(='.'=) Due to the amount of spam posted via googlegroups and
(")_(") their inaction to the problem. I am blocking some articles
posted from there. If you wish your postings to be seen by
everyone you will need use a different method of posting.