Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

how much can go through a PCB

Status
Not open for further replies.

browningbuck

New Member
So i was going to use a voltage cuttoff like this to power a relay
**broken link removed**

how ever i was not thrilled about the current draw across the relay, so i found that you could use a IRF4905 to make a solid state high side switch no draw :).... so i drew this up
**broken link removed**TV
now i was drawing up my board so i could etch it and it hit me, can the PCB handle the draw im putting through it? the mosfet is rated to 5-10A (total as drawn 10-20A). the battery is 2.3A @ 12V(i know its over rated its an idea right now) now if i were to apply that load to it would the etched board take it with out acting like a fuse and burning up?
 
Use 2 ounce copper helps. Also you can remove the solder mask over the high current traces, then add a heavy layer of solder on the traces.
 
Got any solder wick, solder to the current traces, the IRF4905s I'm using are good for 74 amps and I know for sure I've pushed 50 through them, heat sinked of course.
Kinarfi
 
Use 2 ounce copper helps. Also you can remove the solder mask over the high current traces, then add a heavy layer of solder on the traces.

Hi Ron,

2 onces of copper (70µm thickness) helps in many cases. Heavy layers of solder burry a certain risk.

Solder has 1/10 of the conductivity of copper. If the solder layer is not heavy enough it will melt and flow through the enclosure uncontrolled.

Soldering stranded copper wire of appropriate cross section to the traces is much better (wire should have direct contact with the semiconductor pin) - although this is prohibited according to VDE (and CE) -rules.

For high current applications it's advisable to use copper bars mounted on top of the traces with equi-spaced pins (normally 1/5") produding through the PCB to be soldered into the original copper trace.

They take up the main current flow and use the copper trace just as a carrier.

Regards

Boncuk
 
ahhh another good idea Boncuk, copper trace. now it seems like the next thing undersized would be the leads coming off the Mosfets. i know they are rated for it, but it just seems to me that there has to be a better solution to my problem. i mean so far i need omething that has the ability of a high amp relay, but does not require a substantial draw
 
Just back up the traces to and from the FETs and don't sweat the size of the leads, I did for a while and finally got over it, on the spec sheet for the IRF2805, I read an interesting spec, it basically said the silicone could handle 175 amps but the case could only handle 75 amps. I agree that the leads look to small for very much current, but I'm sure the manufacturers have already covered this. Don't sweat the little stuff, the IRF4905 will handle your 50 amps draw if heat sinked.
Kinarfi
 
ahhh another good idea Boncuk, copper trace. now it seems like the next thing undersized would be the leads coming off the Mosfets. i know they are rated for it, but it just seems to me that there has to be a better solution to my problem. i mean so far i need omething that has the ability of a high amp relay, but does not require a substantial draw

Hi browningbuck,

keeping the copper traces to the best achievable length reduces the trace resistance to a great deal. Of course it's impossible to route a trace of 1/5" thickness to the pad of a TO220 package, which requires reducing the trace width around the component pin.

Here are some numbers based on 1 oz copper weight, ambient temperature 30deg/C, total trace length 40mm (1.57") and 50A current flow (resistance values rounded to the nearest tenth):

trace width 5.08mm (0.2"), R=4.2mΩ, voltage drop=213.6mV, see remark*)
trace width 7.5mm (0.295"), R=2.8mΩ, voltage drop=144.7mV
trace width 10mm (0.394"), R=2.1mΩ, voltage drop 108.5mV

If those values seem too high you might either use PCB-material with 2 oz copper weight or a double sided 1 oz board.

The example shows 3.81 mm trace width for the TO220 package and 5.08mm for the TOP3 package.

Maximum total trace length is 20mm (10mm for a single trace)

* Remark: Although the value seems to be low the power dissipated in the trace is 10.68W.

Boncuk
 

Attachments

  • HC-TRACE.gif
    HC-TRACE.gif
    17.4 KB · Views: 313
Do we know what the load current will be? Somebody mentioned 50A, but I didn't think it was the OP.
 
Do we know what the load current will be? Somebody mentioned 50A, but I didn't think it was the OP.

no i dont know it for sure yet, so i was going to just assume full draw from the battery. they are A123 batteries in series creating ~13V
 
Last edited:
Here's a photo of where the copper traces have been built up with solder, I've used it several time and I fixed one PCB by building it up and bridging where the trace had blown out the surrounding copper on a current handling relay. Soldering down a small gauge wire to the trace increases the current carrying capabilities as well.
Kinarfi
 

Attachments

  • post.PNG
    post.PNG
    2.1 MB · Views: 752
Thanks kinarfi, but i think that i will be using a mix between Shortbus's mosfet (if i can find the pchannel) with boncuk's copper trace...i need to now look into upsizing my resistors, and look up the spec on the V18ZA1 varistors i think i remember them being 100A.

thanks a ton for all the help so far, its been a great help
 
ohhh crap, i am not sure if this will work for me. this system also has to be able to receive charging through the same leads that it supplies from. this is why my initial design had a standard SPST relay....
 
What is that you are doing, if I understand correctly, you have a battery and want to monitor the voltage and drop it out if the voltage gets to low. Is that correct? and are you using a second voltage source for the monitor? If so, I may have some recommendations.
Kinarfi
 
yeah thats the idea,the battery is in a system that uses a stator(second source) to recharge the battery.

Edit: What do you have ????
 
Last edited:
yeah thats the idea,the battery is in a system that uses a stator(second source) to recharge the battery.

Edit: What do you have ????

Years ago, I had to monitor a battery for starting an emergency generator and send a signal back to head quarters via modicon or modicom, a closed set of contacts, and this is how I did it using the power of the battery being monitored to operate the monitor, the battery was on a charger, so the small draw was of no effect. I also used the same set up to monitor the voltage of a UPS system that had 120 volts of battery power, had to dissipate some heat on that one.

This is set to drop out / send a signal at 10 volts.
 

Attachments

  • post.PNG
    post.PNG
    3.7 KB · Views: 256
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top