Supplying 4-5 amps from stripboard - is it safe?

[mattybigback]

Hi there

I'm building a 12 LED strip dimmer/driver at the moment, and I've hit a bit of a problem. My circuit works fine, but I'm going to need to scale it up to power 5 5m strips. Each strip should theoretically draw 1.5 amps, plus about 700mA for the driver itself.

I don't have the facility to make my own circuit boards, but I have a lot of stripboard here. I can't seem to find any info about how much current stripboard can handle, so I'm a bit concerned about building the circuit up.

The circuit works by taking the PWM output of an Arduino and driving a STP16NF06L MOSFET as a high side switch using optocouplers to pull it high and low. It's a design I saw on YouTube that I've adapted to my needs. I've attached a schematic of how it is all hooked up (some of the parts have changed but it still works the same way).

What would be the best way to build this up using stripboard without the whole thing melting?

[jeremy]

Use this program: http://www.saturnpcb.com/pcb_toolkit.htm and put in the dimension of the trace. It will tell you what the maximum current rating is.

For stripboard I would probably assume 1oz copper unless you have a reason not to.

[Simon]

you will need to try and break down the current draw onto more than one trace if possible. I have loaded stripboad traces and even made two parallel ones bridge together.

The important thing is to bridge the wholes and that is your trace width bottleneck.

[timb]

Applying solder to the traces increases current handling as well. The thicker you layer the solder on, the more current it'll carry. (You see this in a lot of test equipment such as power supplies, loads and meters; especially cheaper Chinese ones.)


Sent from my Smartphone

[tggzzz]

You might need to consider the EMI that repetitivvely switching 5A might cause to nearby circuits, particularly if the edge rates are fast.

[G7PSK]

I have run Vero strip board at 30 amps by soldering a piece of bare copper wire onto the strips that are going to carry the high current.

[bktemp]

Hi there

I'm building a 12 LED strip dimmer/driver at the moment, and I've hit a bit of a problem. My circuit works fine, but I'm going to need to scale it up to power 5 5m strips. Each strip should theoretically draw 1.5 amps, plus about 700mA for the driver itself.

You have a much bigger problem: You are driving the optocoupler from 12V without any series resistors! The two optocoupler leds and the two external leds are in series. But their forward voltage is much lower than 12V. This explains why the circuit draws 700mA.
The next thing is, your bootstrap circuit does not work with only leds as a load, because of their forward voltage. 12V led strips have a forward voltage of about 8V at low currents. The bootstrap cap will therefore charge to about 4V and this is not enough for driving the mosfet. You need to add at least a resistor in parallel to the led strips.

[tszaboo]

I think once I run 15-20A on two of those traces to test a component. It was defiantly more than 10A. Why dont just try it before making your circuit. You have a lab supply? Please provide feedback!
Depends on the configuration also. If you only run it through like 2 cm and the rest is just a heatsink, it can handle more.

[Simon]

There is not much trace width around the holes, you can have PCB traces run up to 125C in some cases (FR4) not sure if the paper based stripboards can. Just tin the tracks and your fine.

[mattybigback]

You have a much bigger problem: You are driving the optocoupler from 12V without any series resistors! The two optocoupler leds and the two external leds are in series. But their forward voltage is much lower than 12V. This explains why the circuit draws 700mA.
The next thing is, your bootstrap circuit does not work with only leds as a load, because of their forward voltage. 12V led strips have a forward voltage of about 8V at low currents. The bootstrap cap will therefore charge to about 4V and this is not enough for driving the mosfet. You need to add at least a resistor in parallel to the led strips.

That is a mistake in the schematic, the optocoupler is being driven by a separate 5v supply. I found out the hard way what happens when you hook it up to the 12V rail! Also, the MOSFET is working fine, even at very low duty cycles

[mattybigback]

You might need to consider the EMI that repetitivvely switching 5A might cause to nearby circuits, particularly if the edge rates are fast.

Hm, that's something I didn't comnsider. I take it I'll need to shield the circuit somehow? Would a low pass filter on the ouput of the MOSFET help?

[tggzzz]

You might need to consider the EMI that repetitivvely switching 5A might cause to nearby circuits, particularly if the edge rates are fast.

Hm, that's something I didn't comnsider. I take it I'll need to shield the circuit somehow? Would a low pass filter on the ouput of the MOSFET help?

Yes, but the details are all-important; EMI can be surprisingly intractable to track down and eliminate.

You'll have to look at many application notes for the FET and the load, and textbooks on what causes/ameliorates EMI.