7
\$\begingroup\$

I want to use LM78xx for various applications. What I am concerned about is its failure modes.

enter image description here

1. Failure to Short

Not shorting Vcc to ground!! I mean shorting V_input to V_output.

Assume that you have a delicate and expensive piece of electronic equipment that works with 5V. You are using LM7805 to reduce voltage from 12~13VDC to 5VDC. For some reasons your LM7805 fails to short and your delicate electronic device sees the 12VDC instead of 5VDC and fries. I have seen this all around the web. How to protect against this?

2. Failure to Open

It is not as critical as the failure to short. It can happen as a result of over temperature (temporary) or permanent failure. What are the other things that can trigger this?

3. Reverse Bias & Reverse Discharge

enter image description here

I have seen that too big an output cap can cause reverse current going back to input. It is recommended to add reverse discharge diode protection. What else can cause this kind of reverse bias voltage?

From Various datasheets:

With the LM7805, the output capacitor should not exceed 1mf, as larger values could damage the 7805 due to backfeeding of current when power is switched off.

When a surge voltage exceeding maximum rating is applied to the input terminal or when a voltage in excess of the input terminal voltage is applied to the output terminal, the circuit may be destroyed.

4. Flyback

Does the diode on the above picture also protect against flyback current of an inductor? Or is it better to add a separate diode across the inductance load?

5. What else?

What are the other possible causes of LM78xx failure?

\$\endgroup\$
  • 1
    \$\begingroup\$ Are you designing some kind of product or it is for hobby? In my experience an LM317 is more robust and its application circuits are only slightly more complex (a couple of resistors) than those of the '78xx. Moreover it is much more versatile. Where I live there is no much difference in price also (at least for low quantities). Sometimes I found it is even a bit cheaper. Have you considered this option? \$\endgroup\$ – Lorenzo Donati supports Monica Jul 8 '15 at 12:24
  • \$\begingroup\$ Thank you. It is for a piece of lab equipment. Actually I don't even mind using a switch mode power supply to step down from 13V to 9V and 5V but I wasn't able to find a good DIY project of schematic. I will definitely look into LM317. \$\endgroup\$ – arudino.tyro Jul 8 '15 at 12:28
7
\$\begingroup\$
  1. Failure to short- crowbar + fused input (a thyristor and trigger circuit) or at least a TVS. The TVS might be able to limit the voltage to something like 8V which your delicate equipment might have a fighting chance of surviving (newer chips with 5.5V abs max supply may not live). An open GND connection will cause the output voltage to rise so make sure the soldering is solid.

  2. Failure to open- bad solder connections, destroyed chip. Don't depend on the thermal protection as a matter of course.

  3. D1 is unnecessary for a 7805 or 7806, only for higher voltage regulators (and only where the input voltage can be actively discharged). It provides another possible path for failure to short, so I suggest leaving it out unless there is some way the output can (say) get connected to a 12V battery and you don't have a TVS or crowbar on the output.

  4. The power supply does not care about load inductance- the inductance will only tend to make the current continue to flow in the same direction. Any flyback voltage appears across the switch so you need a flyback diode to be placed across the switch or the load.

  5. You might want to put a reverse-biased diode across the output in case two supplies are connected in series and one is driven negative. If you use a unipolar TVS that can serve both functions with a single part.

TL;DR: Put a polyswitch fuse in series with the input (before the capacitor) and add a TVS to the output, and lose D1 (if 7805/6). For extra points mount the polyswitch close to the regulator so it sees the heat.

\$\endgroup\$
  • \$\begingroup\$ Thank you, Could you elaborate on number 3? How exactly this D1 is going to make a problem? I want to use it in an application that steps down 13Vdc to 9Vdc. \$\endgroup\$ – arudino.tyro Jul 8 '15 at 15:14
  • 1
    \$\begingroup\$ If D1 fails short your output will be at 13V. Since your regulator is >7V you may need to use the diode. It's unnecessary for 7805/7806. \$\endgroup\$ – Spehro Pefhany Jul 8 '15 at 15:39
6
\$\begingroup\$

If you have an application where the failure of the linear regulator causes a higher output voltage then either use a fuse and zener diode (5V6 maybe) or design a small crowbar protection circuit. Crowbars are virtually mandatory on switch mode power supplies to prevent a greater output voltage than what would be allowed under the European Low Voltage Directive (LVD).

Failure to open is lost on me - maybe you can explain that.

Reverse biasing you have covered fairly well except for the scenario of a battery charger - usually extra precautions need to be made to prevent damage when primary charging power is removed.

The flyback situation totally depends on the target circuit you have in mind.

\$\endgroup\$
  • \$\begingroup\$ Thank you, Failure to open like a open switch. Could you elaborate more on the battery charging precaution? I didn't get it. \$\endgroup\$ – arudino.tyro Jul 8 '15 at 12:42
  • 1
    \$\begingroup\$ "Failure to open like a open switch". Nope i still don't know what you mean. If you are charging a battery, it remains producing a voltage when the power to the regulator is removed - this causes a bigger problem with reverse-discharge than just a capacitor. Extra precautions might be needed because the battery is almost a constant source of reversing voltage. \$\endgroup\$ – Andy aka Jul 8 '15 at 12:57
  • 1
    \$\begingroup\$ Here's a forum on TI that talks about it: e2e.ti.com/support/power_management/linear_regulators/f/321/p/… \$\endgroup\$ – Andy aka Jul 8 '15 at 13:03
1
\$\begingroup\$

If you are open to using other regulators, you might want to consider more robust automotive linear voltage regulators, such as Micrel MIC2940A or MIC2941A.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.