Led dimming with 555 IC, radiation problems?

[Teza]

When googeling around about using the 555 timer as a PWM for led's, I found the following comment:

But be aware that you will be radiating like crazy at harmonics of the PWM frequency. Relatively high currents and long leads.

I will end up with controling several meter with LED strips in this project.Will this produce noise that can affect other electronic devices?


Another comment I found was:

...i would multiplex the banks of LEDs. This way you only power one at a time, but they would all "appear" on but you would only power one at a time

Is this something I should look deeper into, would it improve my circiut and be a better circuit design?
This is the schematic I am currently using for my project, tho I use a NA555P instead of a NE555P:

[Seekonk]

With a 10K gate resistor I find it hard to believe you will be radiating anything. The FET might heat up though.

I finally cut a string of 5050 white LED.  Surprised to see there were actually three LED in each package and you could power them individually in the strip.

[Zero999]

I agree, the gate resistor is too high.

The PWM frequency is only around 290Hz, which is low enough to cause stroboscopic effects with moving objects and might be noticeable to some people at low duty cycles.

[b_force]

10kohm is a lot indeed.

If you're using (much) higher frequencies, you could use a snubber network across the mosfet.
Er even in combination with a ferrite in series with the gate.

[Teza]

The schematic is a start point that I found, what is a recomended size for R3?

I'll need to learn the purpose and effect of each of the components, but connected this on the breadboard as I needed some practice and a small breake from the theory 
The circuit works on the bredboard, and is my first real attemt in electronics.

[Zero999]

The schematic is a start point that I found, what is a recomended size for R3?

R3 could be zero and it might work perfectly.

In reality it's a good idea to have a gate resistor to dampen any resonance. About 10R would do.

[CatalinaWOW]

I am not sure what resources you have looked at.  If you haven't already the data sheets and app notes for a 555 timer are a tremendous resource in learning the purpose of each component (particularly on what is inside the chip that is being affected by these components) and recommended values. 

Things you might think about when choosing values for a circuit like this:  What is the power dissipation in the part?  How is it affected by value?  What is the physical size of the part?   Is it affected by value?  Do I want this part to have a very large or very small value relative to some other part of the system so that the interactions between those parts will be minimized.  Is the value I have computed for a component actually something that can be purchased off the shelf?  Does my company have a huge stockpile of a value that needs to be used up somehow?  And so on ad infinitum. 

Choosing a component and value is usually a balance between several factors, and those factors vary from application to application.  The bad news is that it is often hard to get a really good choice.  The good news is that this situation is what keeps engineers employed and with inboxes full of interesting problems.

[Seekonk]

If you are concerned about EMI, I might go with 220 or 330.  If you have a scope, it would be interesting to look at the gate voltage when driven with a 1K resistor.  this creates a RC. You may notice a little step in the rise curve. This is the miller plateau indicating at what voltage the FET starts conducting.

[Teza]

I am still amazed by the knowledge  and willingness to share it on this forum 

The progress of my project is very slow at the moment, as I constanly find new questions to google and check out, but hey, for me this is fun.

The last two days I have created this circuit in Eagle, trying to figure out how to create a pcb, and learned that with my current lack of skills the autorouting is basicly useless.
Love the fact that it is linked directly to farnell/element 14, where I have bought most of the components for this project, so the info is nicely added from farnell and into my schematic and PCB.
Tho found that the footprint of the 50k potmeter is wrong by 2 mm on two legs, so guess checking and double checking is needed.

Finaly created a first edition pcb layout that I liked, and have now spent another day on improving this layout and reduced the size by another ~10x5mm to ~57x32mm (these are made in Illustrator).
Not checked for errors yet, but I am getting close to a layout I am pleased with, will soon fit perfectly into a heatsink that will make out most of the case for this project.





Will then dive into your inputs and learn more about frequens and components when I get back home in a couple of weeks.

Thank you for the inputs so far!

[b_force]

Personally, I wouldn't edge boards myself anymore.
for 10 bucks you will get 10 nicely made PCBs.

Making it yourself is more expensive (and more work)

[T3sl4co1l]

I made this, years ago...

https://www.seventransistorlabs.com/tmoranwms/Elec_LEDs.html

Buh, so long ago, the YouTube video doesn't even embed anymore... they probably downsized the encode too oh well.

Anyway, the drive is pretty bad, but it's enough for the low kHz rate it runs at.  Which is enough for visual purposes.

Despite the weak drive, it could still use some filtering on the output leads.  I notice some interference on the radio when it's running.

Tim

[Zero999]

Personally, I wouldn't edge boards myself anymore.
for 10 bucks you will get 10 nicely made PCBs.

Making it yourself is more expensive (and more work)

Some people find if fun making their own boards.

And as far as cost is concerned, if you need it 'today' then making the cheapest, if not only option.

[Teza]

The PWM frequency is only around 290Hz, which is low enough to cause stroboscopic effects with moving objects and might be noticeable to some people at low duty cycles.

This is very good information for my project.
After looking into this I found that to avoid light flickering /stroboscopic effects with PWM dimming of LED’s, the “IEEE Std 1789™-2015” recommend to use a frequency of 3 kHz.

I now know what frequency I should use in my circuit.