Max current spikes in diodes

  • #1
828
1

Main Question or Discussion Point

If I consider a simple example of charging a capacitive load with a low impedance source, and charge the capacitor through a diode, I will have a very high current spike through the diode while the capacitor charges to the supplied voltage.

When I look for diodes and see their peak pulse current, most of them are quite small compared to the "theoretical" infinite current you would have into a capacitor to charge it.


Say the ESR of a capacitor is 100 mOhms, and you charge it with a 20V source through the diode, the current spike in the diode will be about (20-.7)/.1 = 199 amps, and I have not found a normal diode that has this spec.

Yet I have seen, for example, a car battery with a diode after its positive terminal, to protect a reverse voltage connection, and then feed into a capacitive load (ceramic capacitors in front of a SMPS) and these diodes are fine.

Why do these diodes have small peak pulse currents, but can handle surge currents?
 

Answers and Replies

  • #2
247
1
In theory most power supplies will have an extremely high in-rush current to the filter capacitor. Rectifier diodes generally have a much higher Ifsm or "Surge Rating" than Iav or "Average Current Rating". For instance: http://www.diodes.com/datasheets/ds28010.pdf has an Iav of 10A, but an Ifsm of 600A. But we need to be clear here, this does NOT mean the diode can handle a 600A repetitive current, it simply means that the device can handle a brief spike of 600A (and this figure is a bit misleading, the actual maximum current is limited more by the junction temperature Tj).

A lot depends on what the diode is designed to do. In your example of a car battery with a protective diode, the diode does not need to be "fast" or have a "low forward voltage drop", it needs to have a high surge current and a sufficiently high Average Current to supply the SMPS. In the case of the output rectifier for an SMPS, other characteristics like reverse recovery time and low forward drop may be far more critical to efficient operation.

The key is selecting the diode that best fits the needs of your circuit.

Fish
 
  • #3
828
1
Thank you that is very helpful. I haven't seen a surge rating, but I wasn't looking for it either. I'll check some datasheets again.
 
  • #4
205
0
In the situation you describe, with ceramic capacitors of probably quite low values, the inrush current will flow for only a brief period. For instance, with R = 0.1 ohm and C = 10 micro farads we get a time constant RC = 1 microsecond. This is much shorter than the 8.3ms often quoted in a diode Ifsm rating.

The situation is of course different where very large capacitances are concerned (big electrolytics or even "supercapacitors"). In such cases, the inrush lasts far longer and so has to be treated more seriously. A deliberate series element or even an active "soft-start" arrangement may be necessary to keep the current to a tolerable level.
 

Related Threads on Max current spikes in diodes

  • Last Post
Replies
2
Views
2K
  • Last Post
Replies
1
Views
8K
Replies
3
Views
4K
Replies
19
Views
20K
  • Last Post
Replies
2
Views
4K
  • Last Post
Replies
2
Views
737
  • Last Post
Replies
10
Views
6K
Replies
7
Views
1K
  • Last Post
Replies
6
Views
2K
  • Last Post
2
Replies
28
Views
9K
Top