# Questions about LED strip PCB and ground return currents

This question is about the PCB part of a double LED strip 'controller' circuit.

See the circuit at: Questions about LED strip circuit and current.

The arrow shows now the 5V GND pin (JST XH) pin in blue encircled and I tried to put the expected GND flow with blue arrows.

I hope this will be sufficient. Each pin marked in purple has 0.83 A (max) GND from the LED strip.

Are my current flows ok? (except for the one pin JST XH connector that cannot handle the resulting 8 * 0.83 A = 6.64 A.

Below are both sides of the relevant PCB part with the GND planes shown in green (bottom layer) and red (upper layer).

(in the version below the MOSFETs are wrongly connected, but for the question it does not matter).

Calculation for the track widths for the MOSFETs returning path which are 0.5 mm in my PCB:

## 1 Answer

Are my current flows ok?

It depends, the currents will create a small voltage on the ground plane from the resistance of the ground plane. If you don't have any sensitive electronics that are measuring in the sub millivolt range along those current pathways, then it probably won't be a problem. It also won't be a problem for most electronics if the current isn't switching.

If the current is switching or is large (lets say larger than 1A) then you may want to do further analisys on the ground plane to find out how much resistance is from the copper, and how much current is flowing across it and then calculate the voltage rise and see if it will affect the grounds of the components near the currents.

• The current from each of the purple pins can be max 0.83A (but it's GND, not sure if that matters, can it still result in a voltage rise?). There is no switching power (not AC if you are referring to that). I don't have sensitive electronics (except for an ESP32 and 74HC595s and maybe a 74HCT245 on the 'other side' of the (relatively small) board. About those calculations, I don't know how to do them (not having an electronical background). Commented Feb 3, 2022 at 0:22
• Why two gnd layers for only 6A? Why tracks on the inside layers on a 4 layer pcb? The mosfets are wired backwards and your tracks between the mosfet and the connector should be sized to handle the psu max current or fuse otherwise they will be the fuse. Commented Feb 3, 2022 at 6:12
• @Kartman Thanks for the comment. I read that it's best to have as many ground areas as possible, so I used them on both sides. Also since some of the upper/lower GND paths are blocked by signal paths horizontally and vertically there so using both upper/bottom GND areas with vias in between gives always a 'short' way to return for GND. Also, the PCB is 2 layer, not 4 layer. I indeed got already notified about the mosfets wrongly connected (didn't update the PCB for that yet). I did a calculation for the mosfets returning current (maybe not correct) and 0.5 mm should be enough. Is it? Commented Feb 3, 2022 at 9:13
• What is your power supply rated for? That determines the maximum current. If you can make your tracks fatter, then I’d suggest you do that. As for ground areas, you’d be better off laying out the board better rather than doing torturous fills. Understand why more gnd is usually better rather than just doing it willy-nilly. I’m reverse engineering a inverter welder at the moment. They have gnd fill where its needed,no gnd fill and fat tracks for the high current paths (120A). Commented Feb 3, 2022 at 10:22
• Ground fills usually offer the lowest resistance pathway, because you can have the most copper, more copper means less resistance. If you use traces for ground, you run the risk of common mode noise even more than a ground fill in most cases. Commented Feb 3, 2022 at 16:08