Minimising EM noise from PWM switched fairly lights?

On Fri, 25 Oct 2019 13:17:52 +0100, Martin Brown
<'''newspam'''@nezumi.demon.co.uk> wrote:

On 25/10/2019 06:33, Rick C wrote:
On Thursday, October 24, 2019 at 11:02:36 PM UTC-4, Bill Sloman
wrote:
On Friday, October 25, 2019 at 2:30:23 AM UTC+11,
jla...@highlandsniptechnology.com wrote:

At a low PWM rate, say 100 Hz, anything radiated in the AM band
will be ballpark the 10,000th harmonic of the switch frequency.
The energy will be nil.

Not true. AM transmitters are delectable for miles. Even a small
high frequency component from a radiator next door can be big
enough to create problems.

Slow down the switching edges a bit if you like. Add mosfet gate
resistors to soften the edges.

Which increases the switching losses. If you slow down the edges
enough, you will cook your MOSFETs.

Only if they are very weak and feeble. He just meant to avoid having
them snap hard into conduction at their maximum slew rate. The sharper
the edge is the more high frequency content it contains.

We've been talking about maybe chopping at 100 Hz. A few us of gate
slowdown wouldn't wouldn't add much dissipation.

But the whole thing is silly. A string of fairy lights PWMd at 100 Hz
is not going to bother anything.



--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[Rick C]

On Saturday, October 26, 2019 at 8:07:30 AM UTC-4, Tim Watts wrote:

On 25/10/2019 06:25, Rick C wrote:
On Thursday, October 24, 2019 at 1:12:43 PM UTC-4, Tim Watts wrote:
On 24/10/2019 16:37, Rick C wrote:

Why not deal with the RFI by not creating it? I'm not sure what your setup is at the moment but driving with a DC current should be pretty much as simple as PWM. You talk about adding a choke. I'd be willing to bet if you look at the resulting circuit it is a short stone's throw from being a DC/DC converter.

Drive with DC and you only need to deal with residual high frequency noise you can stop with a simple filter.

I'm one of those people who see and are very annoyed by the blinking taillights on so many cars. Cadillac was the first car I noticed it on many years ago and have always been the worst offenders with huge taillights on some of their vehicles. More recently they seem to have changed their design, either by bumping up the frequency or by using DC.

Interesting that you can't effectively google this effect because there is a distinct effect from replacement LED turn signals blinking too fast because they don't draw enough current for the blink circuit to work correctly, called hyper flashing.

Interesting idea - so I'd need a programmable bi directional constant
current drive to do that.

What? Why bidirectional? You can adapt a constant voltage drive to be a current drive very easily.

Because the LEDs are in 2 pairs of sets, alternating down the string,
wired in reverse parallel on a single pair of wires to allow for chaser
patterns and alternating blink effects.


I need alternating drive at 30V peak for this particular type

The current source is just a current source. The bidirectional aspect is what your H drivers are for. I thought you were using PWM to control the brightness which causes the issues I mentioned. As long as the rate of switching the polarity of the drive is fast enough you won't have a problem with flicker. If your current source is controlled, it can ramp up and down the current for each phase and you will get very minor harmonics from the polarity switching.

--

Rick C.

--+ Get 1,000 miles of free Supercharging
--+ Tesla referral code - https://ts.la/richard11209

 

[Rick C]

On Saturday, October 26, 2019 at 7:49:03 AM UTC-4, upsid...@downunder.com wrote:

On Sat, 26 Oct 2019 00:48:28 -0700 (PDT), Rick C
gnuarm.deletethisbit@gmail.com> wrote:

On Saturday, October 26, 2019 at 12:08:50 AM UTC-4, upsid...@downunder.com wrote:

The 8001th harmonic would have 37 mV, spreading through the house
mains wiring, radiating all around. Older equipment without mains
filtering could also suffer directly through their power supply.

The nasty thing about bad SCR dimmers is that while the output
waveform is very dirty, but the same dirty current is drawn through
the dimmer from the feeding mains system, causing strong audio
frequency magnetic fields in the house wiring. In addition, this dirty
current produces voltage drops in both the phase as well as neutral
connector, thus, any equipment connected to the same line will get a
varying voltage. Depending on house wiring conventions used, this may
also insert dirt into protective ground and possibly into signal
grounds, causing all kind of buzzing effects in audio.

There is a group here in Virginia who have found potentially life threatening electrical ground problems near water. They have measured voltages in the water near docks that can cause a swimmer's muscles to spasm allowing the swimmer to drown.

I remember that story. Apparent they could not get the shore grounding
electrode, the marina metallic parts and water properly bonded
together (equipotential bonding
https://en.wikipedia.org/wiki/Electrical_bonding). Normally it is
required that there shouldn't be "dangerous voltage differences"
between objects that should be grounded. What is considered dangerous,
might be different for swimmer in water and people on dry land.

Yes, when you are immersed in water it takes a much lower voltage to create dangerous currents in your body.

The real issue about it all was not that there was a dock somewhere that had a grounding problem. The issue was that the installations they looked at all were up to code, both local and NEC and the problem still existed.

If everything else fails, install a 240/240 V isolation transformer or
a new pig in the pole close to the marina and connect the secondary
center tap to a grounding electrodes on the ground, in the water as
well as the metallic parts of the marina.

Like I said, the problem was not an installation, but that the codes are not up to dealing with this issue.

These guys have traced the problem to similar currents in the neutral that is then carried on the protective ground. Seems the motor startup transients on the lines can swim upstream much as you are talking about.

I was referring to TN-C-S wiring convention
https://en.wikipedia.org/wiki/Earthing_system
Distribution transfers are wired as TN-C with the start point (or
center tap of 120/240 V) is connected with a single PEN connector to
the houses. For ungrounded loads and ungrounded sockets this is all
that is needed. However for grounded sockets and plugs with
non-polarized sockets, the shared PEN conductor has to be split into
PE and N connections.

The latest point in which this has to be done is in the grounded
socket itself, in which a short jumper connects the PE and N
connections together. After this point the socket and equipment cord
use separate PE and N (TN-S).

Due to the voltage losses in the shared PEN conductor due to the load
current, the N (and hence also PE conductor) has a slightly different
potential. If two grounded equipment are connected to two separate
grounded sockets, the equipment chassis potential may vary by a volt
or two. Now, is an unbalanced signal cable shield is connected to the
corresponding equipment chassis, a strong current may flow, causing
all kinds of hum issues.

The separation of PEN to PE and N can also be done earlier, e.g. in
the main distribution panel and all in-house wiring use separate PE
and N wires and all PE connectors are at the same potential (unless
there is a ground fault) and hence no potential difference between
equipment chassis and no hum issues. This is a TN-C-S configuration.

However, if two buildings with separate TN-C to TN-S splitting points
are connected together, there are again ground potential issues and
hum problems. Balanced signals and grounding cable shields only at one
end helps. The other end of the cable shield can be grounded through a
resistor or a capacitor (to provide RF grounding only).

Now please explain how this relates to the issue at hand?

--

Rick C.

--- Get 1,000 miles of free Supercharging
--- Tesla referral code - https://ts.la/richard11209

 

On Sat, 26 Oct 2019 07:45:08 -0700 (PDT), Rick C
<gnuarm.deletethisbit@gmail.com> wrote:

On Saturday, October 26, 2019 at 7:49:03 AM UTC-4, upsid...@downunder.com wrote:
On Sat, 26 Oct 2019 00:48:28 -0700 (PDT), Rick C
gnuarm.deletethisbit@gmail.com> wrote:

On Saturday, October 26, 2019 at 12:08:50 AM UTC-4, upsid...@downunder.com wrote:

The 8001th harmonic would have 37 mV, spreading through the house
mains wiring, radiating all around. Older equipment without mains
filtering could also suffer directly through their power supply.

The nasty thing about bad SCR dimmers is that while the output
waveform is very dirty, but the same dirty current is drawn through
the dimmer from the feeding mains system, causing strong audio
frequency magnetic fields in the house wiring. In addition, this dirty
current produces voltage drops in both the phase as well as neutral
connector, thus, any equipment connected to the same line will get a
varying voltage. Depending on house wiring conventions used, this may
also insert dirt into protective ground and possibly into signal
grounds, causing all kind of buzzing effects in audio.

<deleted story about marina grounding>


>Now please explain how this relates to the issue at hand?

See above.

 

[Rick C]

On Saturday, October 26, 2019 at 11:19:41 AM UTC-4, upsid...@downunder.com wrote:

On Sat, 26 Oct 2019 07:45:08 -0700 (PDT), Rick C
gnuarm.deletethisbit@gmail.com> wrote:

On Saturday, October 26, 2019 at 7:49:03 AM UTC-4, upsid...@downunder.com wrote:
On Sat, 26 Oct 2019 00:48:28 -0700 (PDT), Rick C
gnuarm.deletethisbit@gmail.com> wrote:

On Saturday, October 26, 2019 at 12:08:50 AM UTC-4, upsid...@downunder.com wrote:

The 8001th harmonic would have 37 mV, spreading through the house
mains wiring, radiating all around. Older equipment without mains
filtering could also suffer directly through their power supply.

The nasty thing about bad SCR dimmers is that while the output
waveform is very dirty, but the same dirty current is drawn through
the dimmer from the feeding mains system, causing strong audio
frequency magnetic fields in the house wiring. In addition, this dirty
current produces voltage drops in both the phase as well as neutral
connector, thus, any equipment connected to the same line will get a
varying voltage. Depending on house wiring conventions used, this may
also insert dirt into protective ground and possibly into signal
grounds, causing all kind of buzzing effects in audio.

deleted story about marina grounding


Now please explain how this relates to the issue at hand?

See above.

Sorry, I thought you were replying to me.

--

Rick C.

-+- Get 1,000 miles of free Supercharging
-+- Tesla referral code - https://ts.la/richard11209

 

[Klaus Kragelund]

Instead of guessing, you can:

Measure it with a log periodic antenna, example:

https://www.edn.com/electronics-blogs/the-emc-blog/4403451/PC-board-log-periodic-antennas?utm_source=AspenCore&utm_medium=EDN

Or calculate, as others mentioned

Or simulate, plugging in the waveform and using the attack/decay properties defined in the final test in the standard

Cheers

Klaus

 

[Klaus Kragelund]

33 USD for a log periodic antenna:

http://www.wa5vjb.com/products1.html

 

[Tim Watts]

On 26/10/2019 15:49, Rick C wrote:

The current source is just a current source. The bidirectional aspect is what your H drivers are for. I thought you were using PWM to control the brightness which causes the issues I mentioned. As long as the rate of switching the polarity of the drive is fast enough you won't have a problem with flicker. If your current source is controlled, it can ramp up and down the current for each phase and you will get very minor harmonics from the polarity switching.

Ah - I see...

The current source would have to switch in time to the PWM though to
drive the alternate sets at different brightnesses.

 

[Rick C]

On Sunday, October 27, 2019 at 7:27:34 AM UTC-4, Tim Watts wrote:

On 26/10/2019 15:49, Rick C wrote:

The current source is just a current source. The bidirectional aspect is what your H drivers are for. I thought you were using PWM to control the brightness which causes the issues I mentioned. As long as the rate of switching the polarity of the drive is fast enough you won't have a problem with flicker. If your current source is controlled, it can ramp up and down the current for each phase and you will get very minor harmonics from the polarity switching.


Ah - I see...

The current source would have to switch in time to the PWM though to
drive the alternate sets at different brightnesses.

The current source has to ramp down and up to give the resulting signal low harmonics.

Current output
/--\ /--\
/ \ /----\ / \ /----\
-/ \-/ \-/ \-/ \-
Phase 1 2 1 2
________ ________
_________/ \________/ \_
H bridge control

If you want to get fancy you can slew at adjustable rate to round the corners further reducing the harmonics. The PSU will be whatever you want, linear or switching, but there you don't have to worry about RFI if it is properly designed. The current can be sensed by a small resistor in the ground leg and otherwise the entire supply is the same as a voltage regulator.

For a fixed rise time many PSU chips have a slew rate control via a cap. To directly control the output current a variable reference voltage can be used, say from an MCU DAC output... the same MCU that can control the H bridge.

--

Rick C.

-++ Get 1,000 miles of free Supercharging
-++ Tesla referral code - https://ts.la/richard11209