Diodes in parallel

[madires]

You can parallel schottky diodes, but better not PN diodes, unless they are really close that you can consider then isothermal.

Schottky diodes' internal drift region resistance will increase with temperature due to reduced carrier mobility, thus current will tend to shift to the one that is cooler, balancing the temperature.

Still I see shorted Schottky diodes at the secondary side of SMPSUs with a few diodes in parallel?

[T3sl4co1l]

Although modern power FETs do suffer poor linear SOA due to hotspots as a result. Just saying.

Not as a rule.  SuperJunction types often give DC SOA, despite having higher current density than ever.  I've actually seen a few that don't, not sure if that's a new trend or I just haven't looked at enough of them.

IGBTs have even higher power density still; they largely don't have linear mode ratings at all, but I have seen a few that do, astonishingly enough.

Tim

[Siwastaja]

Still I see shorted Schottky diodes at the secondary side of SMPSUs with a few diodes in parallel?

The issue is you assuming they are dead simply because of paralleling. The corollary is, single diodes can't fail.

Actually, the diodes you see dying die due to overvoltage or overcurrent, basically exceeding the die temperature limit. In other words, the diodes are sized wrong. A single diode can be undersized, as well. If you have a single undersized diode and beef it up by paralleling two, it will do better. But it won't be twice as good.

Of course, as explained by many already, paralleling diodes is tricky because current likely isn't shared perfectly. Derating is needed, and inexperienced or "I don't care" engineer may not have any idea how much to derate. So it's one trap for the young players more. This said, in a power supply design, there are dozens of such tricky areas that require expertise. You can fail at any of them.

BTW I have never seen a failed paralleled double diode in a power supply.

[Harshadb13]

What happens when two diodes are in parallel? if this is a bad idea, why exists shotkky diodes with common cathode? (i. e. to220 package).

Well I believe, having a small resistance in series to the each of the diodes and having this configuration in parallel might help.

Please let me know if what I am speaking is wrong or has some problem

[Gyro]

As has been said, it all depends on how well they are matched. Heck, almost every diode on a IC will be made up out of loads of tiny diodes put in parallel. Same applies to FETs, in the end I can only make a tiny FET on chip, and to get bigger ones I have to put them in parallel, with multiple fingers.

Although modern power FETs do suffer poor linear SOA due to hotspots as a result. Just saying.

I'm no power-semiconductor guy, so I can't be sure, but isn't that more a matter of optimization? Most of those power FETs are targeted at switching applications, so linear-region SOA is not a primary concern in their design.

Yes, I'm not sure either. As T3sl4co1l says, you can get modern devices suitable for linear operation. I guess the devil is in fine the detail. I've rather lost track since power lateral mosfets used in linear operation. They had positive temperature coefficient and so gave pretty much perfect sharing, both across the die and between devices.

[Circlotron]

As has been said, it all depends on how well they are matched. Heck, almost every diode on a IC will be made up out of loads of tiny diodes put in parallel. Same applies to FETs, in the end I can only make a tiny FET on chip, and to get bigger ones I have to put them in parallel, with multiple fingers.

Although modern power FETs do suffer poor linear SOA due to hotspots as a result. Just saying.

I'm no power-semiconductor guy, so I can't be sure, but isn't that more a matter of optimization? Most of those power FETs are targeted at switching applications, so linear-region SOA is not a primary concern in their design.

Yes, I'm not sure either. As T3sl4co1l says, you can get modern devices suitable for linear operation. I guess the devil is in fine the detail. I've rather lost track since power lateral mosfets used in linear operation. They had positive temperature coefficient and so gave pretty much perfect sharing, both across the die and between devices.

When fully turned on the rds has a positive temp coefficient so they will current share somewhat, but if they are only partially turned on the gate threshold voltage has a negative temp coefficient so the hotter mosfet will dig in harder. Worse still, the hotter part of the die will hog the current, as has been mentioned, also because of gate threshold tempco.

[chupocro]

I've just repaired a Telefunken TV which has 17IPS61-2 power supply (common to many TVs) where there are four SB5100 Schottky diodes plus two more UF5402 diodes in parallel. The fault was one shorted Schottky diode.

Here is the schematics:
https://pdfcoffee.com/vestel-17ips61-2-psu-schematic-pdf-free.html

One SB5100 and one UF5402 were not populated by the factory.

The fault is so common they are even selling 17IPS61 repair kits containing those diodes :-)

[MrAl]

What happens when two diodes are in parallel? if this is a bad idea, why exists shotkky diodes with common cathode? (i. e. to220 package).

Hi,

As others have said, some diodes do not do well in parallel, although we do find this arrangement from time to time in commercial equipment.

The problem is with many silicon diodes the characteristic voltage drop goes down with higher temperatures, so if one heats up before the other that one diode could start to take up the whole current load.  That would eventually blow the diode out.  Luckily, there is some other resistance that will increase as the temperature goes up, but it's not enough to compensate for the decrease in characteristic voltage.  This is why some good designs include a small value resistor in series with each diode so that they can share the current better.

The imbalance can be small or large, depending on the type of diodes, their similarities, and how close they are to each other when mounted.  This means it takes a little more effort to use two diodes in parallel successfully.  There are also designs that use more than two in parallel, even four diodes in parallel.  To get this right you have to understand the exact nature of diodes as to how they respond to voltage and current and temperature.  if you are not sure about that, then you should try to use just one diode, or use two and test the hell out of it, but you'd have to test on a piece-by-piece basis because each diode could be a little different.

Diodes that have a common cathode or common anode are not made to be used in parallel, more or less.  They would probably do better than two random single diodes though because the internal diodes would be very well coupled thermally, which means they would have almost the same temperature all the time.  They are not made for that reason though they are made to make it a little easier to mount diodes that have to be connected together, and this application would often be a rectifier with a center tapped transformer which needs only two diodes and one lead from each has to be connected together just like that.