Amps vs PCB track width using TO-220 package

[Analog_Fan]

supposedly IRZL44N, IRF4905 can handle handle 50 AMPS and even more?

using a PCB track width calculator.
https://www.omnicalculator.com/other/pcb-trace-width

a standard pcb has 18μm copper thickness.

3 amps requires 5.288 mm width!
but the mosfets are packaged in TO-220.

I can reach the pins with max 1.905 mm track width and have some space between them.
So i could max do 1.4 amps with 1.905 mm track width

How they do that?

Do they thin metal leads of the TO-220 package (IRZL44N, IRF4905) handle something like 50 amps?

 

[ccaudle]

How they do that?

You don't use 17um copper on high power boards, you would use 70um and use traces on multiple layers in parallel. You also route in from the sides so that you can have wider traces

 

[Analog_Fan]

You don't use 17um copper on high power boards, you would use 70um and use traces on multiple layers in parallel. You also route in from the sides so that you can have wider traces

Thnx ... Oh, but don't think Aisler or OSH offer this.

After my post, i figured to use both sides of the boards, but drawing again over the other track over/under the original in eagle 9.x removes automatically the "other" after completing.

I layed the track manually just now via the line tool and it works.

Than i can handle like 2.4 amps (2x 1.905mm, max possible under TO-220 foot), witch is reasonable for my push pull amplifier experiment.

 

[Analog_Fan]

You don't use 17um copper on high power boards, you would use 70um and use traces on multiple layers in parallel. You also route in from the sides so that you can have wider traces

https://community.aisler.net/t/pcb-specifications/101
Base copper thickness 18µm
Min. processed copper thickness 35µm

So it might be even better, not sure.

 

[MidnightArrakis]

I had designed some PCB's for a defense contractor developing a remote power-generator for the U.S. military several years ago and the main requirement for these PCB's is that they -- HAD -- to be able to carry -- 600-Amps -- of current!!! SHEESH!!! Long-story short.....I needed to use 6oz copper -- AND -- these little devices shown below to carry that much current. Of course, my requirement was 12-times more current than what you are looking to do, but the point is.....you may need to use some rather "fat-tracks" using at least 3oz copper and/or also use some devices that are called "PCB Busbars" on your layout. In other words.....you probably more than likely -- ARE NOT -- just going to be able to only use PCB-routing to achieve your end-result and will probably also need to use some external current-carrying devices in parallel with your PCB-routing.

For an aerospace/avionics project also some years ago, I needed to design some PCB's that could carry 30-Amps going to some triacs and that amount of current was kind of at the limit where "standard" PCB-routing techniques could be used that didn't give any heartburn to the PCB-fabrication shop. I would suggest that you spend some time doing a little research on "high-current PCB routing techniques" and you will discover a solution for yourself.

It is 3:30 AM right now as I am writing this and this fuzzy image below is the best I could come up with quickly for you to take a look at. This is one of the avionics PCB's I had designed for a BOEING 737. This PCB controlled a variety of pumps and valves on a private jet to route water around from storage tanks to their usage destination (showers and toilets) and finally to their used holding storage tanks. As you can see, the high-current "routes" are NOT tracks -- per se' -- but, are actually drawn-in "Copper-Pour" areas that eventually were scaled-down as they approached their connectors. And, the connectors by the way, were similar to the BLUE devices that you see further on down here.



https://www.edn.com/3d-bus-bar-an-optimum-solution-for-managing-dc-power-rails-on-pcbs/








>> You can insert these devices into your PCB to either -- replace -- your
PCB-tracks or have them connected -- in parallel -- with your PCB-tracks:


/

 

[ruffrecords]

Just because the device can handle 50 amps does not mean it will happen in the real world. The spec says it will carry a maximum 49 amps continuous but that is with the device turned fully on. You need to design your PCB tracks for your actual application rather than for the spec sheet. Does your application actually require 50 amps of current?

Cheers

Ian

 

[Cqwet Dbdfte]

A device capability of 50A is dependent on device junction being at room temp. Else you will find a current vs. time chart, and the SOA chart.
50A is not much, but describes the resistance of the power transistor.
As an output in an audio amp, expect much lower current, think speaker at 85dB SPL 1W 1m, how much pain a human ear can take. The stochastic nature of music puts the average power at much lower levels than FTC RMS, (who rated audio amp power for the use of driving light bulbs).
The idle current is going to be the largest heat generator. If you were into serious power you would likely not have used TO-220 devices, or used 20 of them.
The PCB trace calculator assumes continuous current, which is great for pumps and light bulbs.
You are not feeding a 50A x 60V load.

 

[JohnRoberts]

Sorry another Peavey anecdote.... On high current power amp PCBs Peavey would leave the solder mask off high current traces so they would accumulate extra solder going across the wave solder machine, for lower resistance.

JR

 

[Analog_Fan]

Just because the device can handle 50 amps does not mean it will happen in the real world. The spec says it will carry a maximum 49 amps continuous but that is with the device turned fully on. You need to design your PCB tracks for your actual application rather than for the spec sheet. Does your application actually require 50 amps of current?

Cheers

Ian

Ya, but i would like to design my board so that it can handle some 5 amps.
It's a experiment and i only got cheap unreliable multi meter.
Don't wanna burn stuff if i hook up a "big" speaker.

 

[Analog_Fan]

I had designed some PCB's for a defense contractor developing a remote power-generator for the U.S. military several years ago and the main requirement for these PCB's is that they -- HAD -- to be able to carry -- 600-Amps -- of current!!! SHEESH!!! Long-story short.....I needed to use 6oz copper -- AND -- these little devices shown below to carry that much current. Of course, my requirement was 12-times more current than what you are looking to do, but the point is.....you may need to use some rather "fat-tracks" using at least 3oz copper and/or also use some devices that are called "PCB Busbars" on your layout. In other words.....you probably more than likely -- ARE NOT -- just going to be able to only use PCB-routing to achieve your end-result and will probably also need to use some external current-carrying devices in parallel with your PCB-routing.

For an aerospace/avionics project also some years ago, I needed to design some PCB's that could carry 30-Amps going to some triacs and that amount of current was kind of at the limit where "standard" PCB-routing techniques could be used that didn't give any heartburn to the PCB-fabrication shop. I would suggest that you spend some time doing a little research on "high-current PCB routing techniques" and you will discover a solution for yourself.

It is 3:30 AM right now as I am writing this and this fuzzy image below is the best I could come up with quickly for you to take a look at. This is one of the avionics PCB's I had designed for a BOEING 737. This PCB controlled a variety of pumps and valves on a private jet to route water around from storage tanks to their usage destination (showers and toilets) and finally to their used holding storage tanks. As you can see, the high-current "routes" are NOT tracks -- per se' -- but, are actually drawn-in "Copper-Pour" areas that eventually were scaled-down as they approached their connectors. And, the connectors by the way, were similar to the BLUE devices that you see further on down here.

View attachment 128512

https://www.edn.com/3d-bus-bar-an-optimum-solution-for-managing-dc-power-rails-on-pcbs/




View attachment 128509

View attachment 128510

>> You can insert these devices into your PCB to either -- replace -- your
PCB-tracks or have them connected -- in parallel -- with your PCB-tracks:
View attachment 128513

/

After my comment to ccaudle.
I thought of the idea of 3d printing a piece that holds the pcb aswel some 2.5 mm² solid "installation wire" as bars or busses nd solder the Fets on there or drill a hole and than solder., used in houses. but i don't have a 3d printer.

 

[Analog_Fan]

Sorry another Peavey anecdote.... On high current power amp PCBs Peavey would leave the solder mask off high current traces so they would accumulate extra solder going across the wave solder machine, for lower resistance.

JR

That's a very good idea, i didn't think of but not unknown or unseen.
I switched to version 7 of eagle a and made it 4 layer.
But will do that as well on the bottom side.

It's a 80 mm x 60 mm board.
I used footprints with slotted holes to be able to easy remove parts, but Aisler said it where to many slotted holes.
: )

 

[ruffrecords]

Ya, but i would like to design my board so that it can handle some 5 amps.
It's a experiment and i only got cheap unreliable multi meter.
Don't wanna burn stuff if i hook up a "big" speaker.

Using the calculator link you posted, for 5 amps allowing 20 degrees temperature rise and 2oz copper requires a track width of less than 100 mils. so use 50mil tracks on top and bottom of the PCB.

Cheers

Ian

 

[Newmarket]

For clarification - are you looking to carry max 50A or 5A ?

 

[Analog_Fan]

For clarification - are you looking to carry max 50A or 5A ?

50 Amps would blow the main fuse of my home on a 1 to 1 transformer.
: )
Locked behind seals.

I have a "cheap" 10 inch sub woofer that i use in my cars, just wanna know what it takes to make it move, as well other speakers i have around.

I made the board 4 layers and it should handle 4.8 amps as is currently in Eagle.

But i did wonder how a TO-220 package is rated for 50+ amps with these rather thin metal strips.

 

[Analog_Fan]

Using the calculator link you posted, for 5 amps allowing 20 degrees temperature rise and 2oz copper requires a track width of less than 100 mils. so use 50mil tracks on top and bottom of the PCB.

Cheers

Ian

I got 75 Mills on 4 layers, aswel used "bstop" layer to lay some solder if needed.

I used used 50ºC in that calculator, assuming to board would heat up that much.
I have another earlier mention task to explore, so it will be some kind of universal power supply pcb.

But thank you for your response.

 

[john12ax7]

But i did wonder how a TO-220 package is rated for 50+ amps with these rather thin metal strips.

The reality is it can't, the pins and the bonds inside are an issue Most likely someone just calculated it, 50W with 0.022 Rds gives you 47A.

 

[Cqwet Dbdfte]

The reality is it can't, the pins and the bonds inside are an issue Most likely someone just calculated it, 50W with 0.022 Rds gives you 47A.

It could very well do 50A, the test was likely done with ~100us pulses

 

[MidnightArrakis]

50 Amps would blow the main fuse of my home on a 1 to 1 transformer.

[50 Amps would blow the main fuse of my home on a 1 to 1 transformer] -- What part of this Planet Earth are you on? In my part of this spinning blue-marble whizzing around the Sun at approximately 17,000 MPH, the typical home main circuit breaker is rated at 200-Amps. Of course, your home could be very, very, very small and has a wood-stove for heat during the winter and open-windows with a prevailing wind for cooling during the summer. But, still.....

>> 200-Amps is typical for homes around 2,000 square feet, with smaller homes utilizing 150- or 100-Amp varieties. In the case of an emergency, you can turn-off power to your entire home by flipping this switch.

Common household electronics use 15- and 20-Amp breakers, while larger appliances require higher amperage breakers: typically 30-Amps for water heaters and clothes dryers; 40-50 Amp for stoves; 70-Amp breakers for an HVAC unit, etc.

/

 

[Analog_Fan]

[50 Amps would blow the main fuse of my home on a 1 to 1 transformer] -- What part of this Planet Earth are you on? In my part of this spinning blue-marble whizzing around the Sun at approximately 17,000 MPH, the typical home main circuit breaker is rated at 200-Amps. Of course, your home could be very, very, very small and has a wood-stove for heat during the winter and open-windows with a prevailing wind for cooling during the summer. But, still.....

>> 200-Amps is typical for homes around 2,000 square feet, with smaller homes utilizing 150- or 100-Amp varieties. In the case of an emergency, you can turn-off power to your entire home by flipping this switch.

Common household electronics use 15- and 20-Amp breakers, while larger appliances require higher amperage breakers: typically 30-Amps for water heaters and clothes dryers; 40-50 Amp for stoves; 70-Amp breakers for an HVAC unit, etc.

/

it's more than 106.000 km/H, (orbit radius · 2PI) / annual hours.
However in the winter time goes slower, we are somewhere between 3/4 million km closer to the sun.

I can't open the black box locked with seals or stamps, to see what's there.
I have 3 circuits capped at 16 Amps + 2 differentials, so i assumed the main fuse is like 40 Amp, witch is common value, the next "standard" one is 63 Amps, this one is probable used.

I do have several empty tubes and empty boxes where these tubes lead too in the walls in ll the rooms and stuff to expand the system

 

[Analog_Fan]

The reality is it can't, the pins and the bonds inside are an issue Most likely someone just calculated it, 50W with 0.022 Rds gives you 47A.

Their datasheet ramble even about -74 Amps.

Continuous Drain Current, V GS @ -10V -74 Amps.
Avalanche Current -38 Amps.
for the Infineon (Siemens) IRF4905.

Someone on YouTube flashed with a fiber laser to decap a TO-220, the bonding is bit more substantial.
Not the typical bonding wire, but rather the metal it self in direct contact.