LED Strip lights (2 Amps) Short Fade on / Fade Off effect

[ThomsCircuit]

I believe you mean square millimetres, not cubic!

yes. and thank you so much.

 

[ThomsCircuit]

Could you identify the pins (C B E) for the darlington NPN?

in post #118 you can see there are 6 spokes. it was the only way (at this time) that i could get enough spokes around the collector. The emitter still has three on the top pour but gains a fourth on the bottom pour.

 

[eTech]

View attachment 135590
in post #118 you can see there are 6 spokes. it was the only way (at this time) that i could get enough spokes around the collector. The emitter still has three on the top pour but gains a fourth on the bottom pour.

Hmmm....is this an NPN as low side switch? I don't remember what the schematic showed.

If so, the highest current will be passing thru the collector (connected to the light) and emitter (connected to GND) of the darlinton transistor. So, I would expect that the emitter would be a thermal pad connected to the ground pour, and the collector would be a thermal pad connected to a isolated pour that includes a light terminal pin thermal pad. The base pin doesn't need to be very wide since it doesn't need to carry much current because the darlington transistor has an enormous amount of gain.

 

[ThomsCircuit]

Hmmm....is this an NPN as low side switch? I don't remember what the schematic showed.

yes it is.

If so, the highest current will be passing thru the collector (connected to the light) and emitter (connected to GND) of the darlinton transistor. So, I would expect that the emitter would be a thermal pad connected to the ground pour, and the collector would be a thermal pad connected to a isolated pour that includes a light terminal pin thermal pad. The base pin doesn't need to be very wide

Yes. That is absolutely what i have done.

 

[eTech]

yes it is.

Yes. That is absolutely what i have done.
View attachment 135592

OK...but don't forget to make the spokes wider at the collector, emitter and light pads, and make the red track wider. After that, should be good to go...

 

[ThomsCircuit]

make the red track wider.

Im going to try to look this up but i believe any existing track, regardless of its width becomes part of the pour. So if you are referring to the thin track from the collector to the pin, i purposely reduced its size so it would not negate the purpose of the spokes. but i did make them wider for the entire board. default 10, now at 16. And I extended the collector pour area to include the LED.

Overall I feel great about this project. Ive learned a great deal about pours, spokes, and a wee bit about current.

Collector pour = yellow
Top GND pour = green
Bottom GND pour = blue
Base track width is 30

 

[eTech]

For future reference...you can create your own custom Thermal pad with wide spokes so you don't have to manually widen the them.

 

[ThomsCircuit]

For future reference...you can create your own custom Thermal pad with wide spokes so you don't have to manually widen the them.

I have found this in the pour settings. As of now I can only make settings per project. While it comes in handy for temporary adjustments on pads for a component things like net track widths are per project. But for now I don't mind.

 

[ThomsCircuit]

If the transistor gets too warm during the fade with your device, either C1 or R1 can be increased to reduce the switching frequency.

At the time i was a bit overwhelmed but i then go back and re-read a few posts to spot things i may have overlooked.
So my current TIP122 reaches 99f under a 1.3 amp load and while it is warm its not anywhere near an issue with itself or in need of a heat-sink im wanting to know if you were referring to the TIP in the above statement or the mosfet.

The fade works wonderful. no flicker. Ive increased the Cap value to 14.7uf (2 caps) for a 1.5 second effect.

 

[rjenkinsgb]

im wanting to know if you were referring to the TIP in the above statement or the mosfet.

That was the MOSFET; its dissipation will be highest during the fades.

The bipolar transistor dissipation is purely down to load current.

 

[ThomsCircuit]

Ive noticed that when this circuit initializes the light connected to the output comes on at 100% then fades off. Is there a way to adjust it so it does not do that? I will have (3)10amp power supplies for the 18 separate light circuits that draw less than 20 amps combined but having them all come on at once worries me a bit. Unless you feel its not a problem.

 

[ThomsCircuit]

N-Channel MOSFET, 80 A, 40 V, 3-Pin TO-220 Infineon IPP039N04LGXKSA1 | RS

uk.rs-online.com

That would dissipate roughly 75mW at 5A, with the circuit running on 12V.

Ive located a Mosfet that very closely matches your recommendation. ​

Can you confirm that this would also make a suitable substitute?​

RDS On = 4mOhms
Gate Source = 4
Gate Charge = 92nC

FDP047AN08AO 75V, 80A​

 

[rjenkinsgb]

Ive noticed that when this circuit initializes the light connected to the output comes on at 100%

That's from reversing the input & output, early on..

Change C3 so its between the control voltage and power, rather than control voltage and ground.

 

[ThomsCircuit]

Change C3 so its between the control voltage and power, rather than control voltage and ground.

You know I brag about you guys.
Thank you.
did i find a good mosfet substitute? #132

 

[rjenkinsgb]

That FET should work; it's a bit higher gate capacitance than the last one you found, so more likely you may need to reduce the PWM frequency, if it gets hot during the fades.

 

[ThomsCircuit]

a bit higher gate capacitance

is this what you are referring to? input / output
EDIT: ive looked at a few of these and I do believe this IS what you are referring to. In any event I know that if it gets too hot during the fade i can increase C1 from 0.1uf to 1.0uf and keep an eye on the flicker.

 

[ThomsCircuit]

That's from reversing the input & output, early on..

Change C3 so its between the control voltage and power, rather than control voltage and ground.

Ive had this on most of the day with this new setting to C3. It initially works perfect but gradually starts to fails. Its connected to a timer so it auto shuts off after 5 minutes so it does not stay on for long. And, i just have the single led connected. If i attempt to turn it on it begins to hesitate then fades up but fading off is rapid. Eventually it will not respond at all. If i change the cap it works again for a while then starts to fail again.
I just reverted back to the original setting and it works perfect.

 

[rjenkinsgb]

You still need to respect the capacitor polarity!

 

[ThomsCircuit]

You still need to respect the capacitor polarity!

Hi.
I just received STP80NF55 MOSFET 8 mOhm 80A 55V N-channel
I added it and changed R7 & 8 from 1K to 470R.
The fade effect is smooth when going from 100% to 0 but the reverse appears to go from 0 to 50% quickly then fade up to 100%. how could I adjust this so its fades on a bit more evenly. What could i try.
BTW there is no heat at all from these mosfets.

 

[rjenkinsgb]

Try increasing C1, to reduce the PWM frequency?