1
\$\begingroup\$

Noob here. I need to add overvoltage protection to a small, portable solar power system I'm putting together. The solar PV array has a loaded voltage in the 23-26v range and an open circuit voltage that can be as high as 45v. The PV array is connected to a solar charge controller to charge a couple of deep cycle batteries. The "load" output of the charge controller is connected to a voltage regular that has a maximum input voltage of 30v.

The load is a 10A max circuit. I need to limit the voltage to 28-29v to prevent damaging the voltage regulator.

enter image description here

I started my research to solve the problem and have read about crowbar circuits, zener overvoltage protection circuits and TVS diodes, etc. I think that a basic zener+PTC is a good fit for my needs, but I'm absolutely stumped on the values. The problem I'm finding is that my circuit needs to operate at 10A so I need at least a 10A PTC fuse. When I check digikey for 10A PTC fuses they are all VERY slow which from my initial understanding means I will fry my zener before the fuse has time to blow.

I'm stuck. The 10A PTC fuse (or any fuse) will current protect my voltage regulator and downstream parts but the 10A fuse won't blow fast enough to save the Zener and if the zener fails I lose my voltage protection. - Ahh!

How do I overvoltage protect a 10A, 30v DC signal?

Edit: I just realized that I think this is fundamentally flawed: If the batteries are low or disconnected the solar PV can't produce enough current to blow the PTC. I'd have a maybe 1-2A of 50v shunting through the zener and it would fail.

\$\endgroup\$
  • \$\begingroup\$ You could use a series or a shunt regulator or a buck-boost switcher on the output of the charge controller. Do you have a link to the charge controller's manufacturer and part number or data sheet or manual or specs? What's the load, BTW? \$\endgroup\$ – EM Fields Jan 15 '15 at 18:12
  • \$\begingroup\$ Well besides flipping the zener over, if you put the PTC and zener in good thermal contact the ptc may blow a little faster. Other than that I'd think about a different circuit. I'm thinking about monitoring the voltage and turning on a FET (in shunt) when the voltage gets too high. \$\endgroup\$ – George Herold Jan 15 '15 at 18:13
  • \$\begingroup\$ @EM Fields - no part number or data sheet; it's a super cheap Chinese unit. They work, but it's not a high quality device. \$\endgroup\$ – Steve K Jan 15 '15 at 18:39
  • \$\begingroup\$ @George Herold - It sounds like the "simple zener+PTC" isn't a good fit for this application? I'm interested by the FET suggestion, although I don't really understand it ;) I also just realized that if my batteries are dead I won't have the capacity to blow the PTC anyway! \$\endgroup\$ – Steve K Jan 15 '15 at 18:42
  • \$\begingroup\$ Five questions: 1) what is the battery voltage? 2) what is the solar array max short circuit current (Isc)? 3) Does the controller use switch mode power conversion? In other words, is it possible that the controller output current can be greater than the current input from the solar array? 4) What is the capacity (in Amp-hours) of the battery? 5) What is the battery type (lead acid?, and if so, is it flooded or gel or AGM)? \$\endgroup\$ – mkeith Jan 15 '15 at 21:59
3
\$\begingroup\$

PTC fuses can be wonderful in some ways, but they're not very good at providing overvoltage protection for Zener-clamp circuits. Suppose, for example, that one has a Zener that clamps absolutely rigidly at precisely 30.0 volts, the PTC has a tripped power dissipation of 7W (grabbed from the data sheet of a 32V 10A device), and the supply feeds out exactly 30.5 volts. Under that scenario, the resistance of the PTC would increase to 0.035 ohms and remain there indefinitely, passing 14 amps. In such a state, the PTC would continuously dissipate 7 watts perfectly happily, but the 30-amp Zener would need to dissipate 420 watts--not just briefly, but indefinitely. Note that if the supply voltage were to increase, the current through the Zener (and the power it would need to dissipate) would drop considerably, but in many scenarios a low-impedance supply is just as likely to be slightly over the required voltage as to be massively over.

If you want to use a PTC with a shunt, and are only interested in surviving over-voltage conditions rather than being able to operate through them, I would suggest using a circuit that selectively shorts to ground the voltage after the PTC. When such a circuit trips, it will cause a lot of power to be dissipated in the PTC, and relatively little in the protection circuit. Circuit operation will not be possible in such a circumstance, but downstream devices will be usable once everything is shut down and the PTC is allowed to reset.

\$\endgroup\$
  • 1
    \$\begingroup\$ Poly fuses and zeners are commonly used for over voltage protection. The zener does the over-voltage and the poly "blows" to protect the zener from too much heat. They even make devices all in one... look for polyzen fuses. \$\endgroup\$ – George Herold Jan 15 '15 at 21:22
  • 3
    \$\begingroup\$ @GeorgeHerold: An integrated Zener+polyfuse combination could work quite well, if the heat from the Zener would be guaranteed to trip the fuse before the Zener was destroyed. Using separate components for the purpose may be workable when using small fuses at low voltages, but the fact that overvoltage conditions could cause the Zener to dissipate much more power than the fuse would suggest that such designs are only usable when the fuse itself can't dissipate much power, and thus a Zener will be able to dissipate a larger amount of power without costing too much. \$\endgroup\$ – supercat Jan 15 '15 at 21:38
  • \$\begingroup\$ PolyZen makes integrated protection devices. Not sure that they have anything for this level of power but it might be worth a look. \$\endgroup\$ – mkeith Jan 15 '15 at 22:22
  • \$\begingroup\$ @Supercat - "... only interested in surviving over-voltage conditions ..." - yes, this is what I want. The over voltage condition is an exception and if I follow proper procedure, should not happen. However, I'm human and will sometimes not follow procedure and I don't want equipment fried while I'm 4 hours from the nearest highway, deep in the bush. I only somewhat follow your heat/power calculations (I will re-read several times). However your second paragraph sounds interesting and may be what I need. I'm a noobie though and I don't understand what such a circuit would look like. \$\endgroup\$ – Steve K Jan 16 '15 at 17:42
  • \$\begingroup\$ Fun fact - sometimes, you get lucky and your zener fails short circuit, in which case the fuse-zener combo actually works. Not a recommended design technique, though... \$\endgroup\$ – Selvek Oct 26 '18 at 22:19

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.