High Current Diode Question

Hi all,

The other day I had an application at work where a diode was needed for reverse polarity protection. The only problem is, the diode needs to be rated at 15Amps (continuous current). Also, we don't want it in a TO-220 package. We need it to be in line package (like a normal 4004 diode with tails at each end), although I'd imagine it would be considerably bigger.

Has anyone ever come across or used a diode like this? Do they even exist?

Any guidance would be much appreciated. Thanks -Tom

Do a litle simple maths - 15A x 1V (likely 1V drop at 15A) - means 15W of heat to be dissipated.

That's why such diodes tend to be designed to be fitted on good sized heatsinks.

Hi Nigel,

I am posting after a very long time. I see that you have reached 30K posts

Could he use a couple of 4004 / 4007 packages in parallel? Does that cause any design problems?

lord loh. said:

Hi Nigel,

I am posting after a very long time. I see that you have reached 30K posts

Could he use a couple of 4004 / 4007 packages in parallel? Does that cause any design problems?

Click to expand...

You've still got 15W to dissipate - it's got to go somewhere.

I thought the surface area would increase and the 15W is distributed between several diodes causing them to heat up less as each is dissipating 15W/(n diodes). So passive cooling by air convection might suffice (?). My be as long as the diodes are not stacked too close.

@tom81: Why can't you use the TO220? If you can't, the parallel diodes options might not be viable either? Is it size / space issues?

How long do you need reverse protection? Usually when a device is connected backward, it's immediately apparent due to the high current from the power supply and the power is thus removed. So the heat may not build up in the diode(s) enough to damage them.

Edit: You could also use Schottky diodes which dissipate less power for a given current.

I know what I want to say..I just don't know how to explain it in easy terms.

Tom81 said:

Hi all,

The other day I had an application at work where a diode was needed for reverse polarity protection. The only problem is, the diode needs to be rated at 15Amps (continuous current). Also, we don't want it in a TO-220 package. We need it to be in line package (like a normal 4004 diode with tails at each end), although I'd imagine it would be considerably bigger.

Has anyone ever come across or used a diode like this? Do they even exist?

Any guidance would be much appreciated. Thanks -Tom

Click to expand...

Hi,

Why exactly does it have to be in an 'inline' package?
There are bigger diodes out there that come in 'inline' packages rated for 9 amps, but the leads are also thicker than the 1N4004 diodes you are talking about, and of course the body is bigger too. I would think that if you have the room, you have the room, or else you dont have the room for it...either way other packages should work too right?

You may get away without a heat sink if you use a fuse as well. When reverse polarity occurs it blows the fuse so the power in the diode is limited. That's one way to keep the space requirement down.

A common mistake in diode protection circuits comes in the form of a badly designed PC board. The PC board copper traces have to be able to handle the overload without melting. A large overcurrent for even a short time can cause the copper traces to blow right off the board! Just something to keep in mind, and of course a few tests would be a good idea too.

The problem isn't the reverse current, it's the forward current. His application takes 15A continuous forward current.

I too would like to know why the TO-220 package doesn't work for you?

One other thing that I've been debating back and forth in my head is do you put the protection diode on the +V supply, or at the ground? I keep seeing it on the +V side on schematics, but I think the ground side would be better.

My last post may have been a little confusing since I was thinking that the diode was across the input leads (cathode to plus, anode to minus) to conduct any reverse current if the power is incorrectly applied. That way the normal input voltage and circuit efficiency is not affected and the diode doesn't carry current under normal conditions. Of course for this to work, the short circuit current must be limited and the item being protected must be able to withstand the application of the reverse voltage of the conducting diode.

That's often done on the satellite systems I work on to protect the various circuit modules from miswire when the system is being built.

Oh, so like a reverse crowbar type circuit? Never thought of that. I've only ever seen reverse protection done by inlining a single diode with the circuit.

smanches said:

Oh, so like a reverse crowbar type circuit? Never thought of that. I've only ever seen reverse protection done by inlining a single diode with the circuit.

Click to expand...

Hi,

Yes, that's right. If the polarity is reversed it blows the fuse and so no more power.

I've also used resistors in series instead of a fuse for lower current applications. If the polarity is
reversed the resistor takes the power (properly rated of course) until the reverse polarity is
disconnected. No fuse to change to get it working again, but the resistor must be sized to take
the full power so this mainly only works with low power circuits. Also, the circuit to be powered
must be able to handle around -1v input without damage.

In extreme high current applications there is also the "switch/buzzer" solution which makes a lot of sense.
A high current switch is wired in series with the input, and a buzzer (with small diode in series) is wired in parallel to the incoming power source so that it buzzes if the input is connected in reverse and a lamp (with a small diode in series) in parallel with the buzzer so that it lights up when the polarity is correct. It's up to the operator to turn the switch to the 'off' position before connecting power. After connecting the power, if the buzzer buzzes and the lamp doesnt light up the polarity MUST be corrected before throwing the switch to the 'on' position. If the buzzer doesnt buzz but the lamp lights up that tells the operator that it is ok to throw the switch to the 'on' position. This is very effective for high current applications but it does depend on the competence of the operator to understand the dangers of not turning off the switch FIRST.

I've only ever seen the single inline diode, usually at the +V supply. I can see the advantages of the crowbar also.

What am I missing here ? Isn't the OP looking for a diode in series with the power supply which will be reverse biased when power is incorrectly applied ? Automotive alternator diodes should do the job.

p-channel fet instead of diode

Hi All,

Thanks for all the responses!

A little more background on the application, as requested:
- It is an Automotive Application
- The Diode is being installed inside a Wiring Harness (not on a pcb etc). That's why we wanted it inline.
- It's on the +12V side in series. Ground (return) wires generally carry less current.
- 15A will be the continuous Forward current. In reality, it's actually 9A, but we've applied a safety factor and rounded up.
- We can't bundle several diodes together (current might find the shortest path and blow that diode. plus it's bulky)
- In an ideal world, we wouldn't be wiring it up backwards. It's just for insurance in case someone does.
- Heat dissipation is probably our main problem. Heat sinks are bulky and aren't really an option. Thicker leads aren't an issue.

I like the idea of an alternator diode. They would handle some big currents. Anyone know of some specs on Alternator diodes? Can they be purchased individually?

Thanks -Tom

Alternator diodes still require a heat sink. The are normally a press fit into holes in the alternator aluminum frame.

ALTERNATOR DIODES ARE mondo current HANDLERS

PERSONALLY i WOULD JUST TURN A 2N3055 TO-3 INTO A DIODE, TIE THE BASE TO THE COLLECTOR.

Diodes 40 Amp 600 Volt Blocking Diode
40A, 600V stud blocking diode.
This diode will work perfectly on many wind generator or solar panel applications. You sh… $5.90
**broken link removed**

Hi all,

Thanks for everyone's input. We found a suitable diode. It's rated at 12A, which should just be enough. For anyone interested, the part number is 1772108 on the farnell website (**broken link removed**).

Cheers -Tom

THESTOCKSGUY said:

ALTERNATOR DIODES ARE mondo current HANDLERS

PERSONALLY i WOULD JUST TURN A 2N3055 TO-3 INTO A DIODE, TIE THE BASE TO THE COLLECTOR.

Diodes 40 Amp 600 Volt Blocking Diode
40A, 600V stud blocking diode.
This diode will work perfectly on many wind generator or solar panel applications. You sh… $5.90
**broken link removed**

Click to expand...

Alternator diodes are indeed capable of carrying large currents. They are on a heatsink in the alternator, as they should be in this application.

A 2N3055 with the base and collector connected together will not work in this application. In the forward direction it still needs to dissipate a lot of power. In the reverse direction the Base-Emitter breakdown is only 7 volts which would FAIL without providing any protection to the load.

Your CAPS LOCK key needs repair.