Alternative to fuse to protect power supply

[kal.a]

Hello Ladies and Gents,

As I'm an electronics newbie I am protecting my power supply, which is a plug in adapter, with a a fuse and I go through a lo of them as try differ circuitry so I was lookin for an alternative and read about poly switches/thermistors, are they an alternative? Are they rated the same as fuses?

Thanks
Kal

 

[alec_t]

Polyswitches and thermistors can indeed provide protection for some applications. Horses for courses. They are not rated the same as fuses, since they behave differently. Check the datasheets and application notes for typical parameters and ways of using those devices.

 

[Grossel]

You'll always have current limiter circuits. It'll do excactly that, wouldn't they?

 

[kal.a]

:d I don't know th answer to that but I'm working on it. I'm still learning.

 

[spec]

Hy Kal- how goes?

You can make a nice protection circuit with a PMOSFET, an opamp, and a few components. You can have the circuit turn the power line off on an overcurrent or, better still, go in to a constant current mode. You could even make the current adjustable.

If you would like that approach I will knock out a schematic, but first I would need to know the details of your power supply: voltage and current.

spec

 

[kal.a]

Hi Spec,

Thanks a lot. At that time I was experimenting with H-Bridge circuits and was blowing a lot of fuses. I think I was using a wall adapter.
I will take you up on your offer though in a week or so as I am working on another little project.

Cheers
Kal

 

[kal.a]

Where is the reply from "AMWEI Thermistor Sensor" . I received email notification of his or he reply!

 

[audioguru]

Design an H-bridge circuit that does not overload your power supply or design a current limiting circuit for the power supply.

 

[MaxHeadRoom78]

Re-settable circuit breaker.
Max.

 

[Flyback]

spec's idea sounds good, and you could just make it a currnet monitor pcb and a comparator to switch out the cct when current goes over x amps.

 

[AnalogKid]

A polyswitch or polyfuse is a resistor with a very high positive temperature coefficient. Above the rated current for each part, the resistance increases very rapidly. It never goes to infinity, so there always is some current through the part. And because the safety action is thermal, simply removing power for a few seconds does not reset it. If it is not exposed to free air but buried inside a case somewhere, it can be several minutes before it cools off enough to pass operating current again. Also, because it is a thermal device, it does not respond rapidly to an overcurrent situation. In an experimental environment, an active current limiting circuit gives much better protection than a polyfuse.

ak

 

[kal.a]

Thanks guys. I think I'm going to try all of those ideas on at a time for general knowledge.

Max,
I purchased a 500 mA resettable breaker similar to the one in the link below(but 500 mA) and it didn't work. I just wired it in-line with the positive lead from the transformer (wall adapter)and then shorted the two leads momentarily. I didn't want to burn the transformer so it was no more than a second and it didn't trip. Will try it again for sure.

Thanks
Kal

https://www.alliedelec.com/phoenix-contact-0712194/70207677/

 

[audioguru]

The datasheet for the circuit breaker you posted is Slow Blow but they do not say its tolerance (will your 500mA one trip at 400mA or at 600mA?) or how long is the delay.

 

[MikeMl]

kal.a

since you are learning, consider this:

"H-Bridge" implies you are driving a DC motor. Motors are tricky, because the current required to start them is three to ten times higher than what they require to keep running once up to speed. This makes it difficult to size a power supply and fusing. If the motor is rated at 12Vdc@0.2A, then likely it requires a 1 to 2A power supply (and fuse) to start it...

Also, the starting (in-rush) current effects the design of the H-Bridge. The inrush might last a good fraction of a second, so the transistors in the H-Bridge must be sized for the inrush current, not just the running current.

Finally, if you are newbie with H-Bridges, the most likely problem you will encounter is "shoot-through", where both the upper and lower transistor on one side of the Bridge are turned on at the same time (inadvertently, as this should never happen if the H-Bridge is properly designed).

If this happens, there is a huge current spike while both transistors are turned on, which can blow up the transistors, or the power supply. This can happen because the base/gate drive to the upper/lower transistors in the H-Bridge overlap. Post the schematic of your H-Bridge so that we can look to see if this is your problem.

 

[Tony Stewart]

The size of protection volume gives you some indication of time lag.
Fastest: Active Current limiter most accurate but not suitable for motors or Cap bridge rectifiers.
Fast: Fastblo fuse
Medium: Std Fuse
slower: Polyfuse , Magnetic breaker

The criteria is how long does it take for the device to burn out and choose anything faster.
A small Polyfuse will heat up to 85'C and trip (become 100-1000 x more R) which requires it to be free of convection air blockage must faster than a large inductive device unless it saturates too early. (Remenance), greater care in selecting trip point. Magnetic breakers are more lossless, and reliable for safe turn of but not resetable.

The problem with H bridge motor drivers is the stall or start current can be 5-8x max rated current and if Bridge is only capable of 2x Max Current and 10x for brief start, stops , a polyfuse may not be adequate and depends on inertia and surge time, thus smarter active limiters with filter to adjust threshold depending on duration of overcurrent is best protection driver AND load.

Murphy's Law says, unless you think carefully, your bridge driver will be undersized worst case and you will fail, not the driver.

 

[kal.a]

Thanks gentlemen.

The H-Bridge was designed here with lots of help from experts. I was blowing fuses (and MOSFETS and transistors) while working with MOSFETS and transistors as I was learning how they worked and how to use them in a circuit and it got so much that I thought there must be a cheaper way to protect my power supply. ( I had already burned one which prompted me to fuse the next one). The H-Bridge works fine but the power supply will have to be replaced as I sized incorrectly and now the V-Reg heats up when I run the motor with load or on and off too frequently. I knew that going in but I focused on learning H-Bridge circuitry.
So my idea was to protect the power supply that I'm using for my experimentation which powers a variety circuitry.
But now with all the suggestions I'm interested in trying different ideas on separate power supplies powering different projects.

Cheers
Kal

 

[spec]

In the real world overloading power supplies is frequent during development- that is why lab power supplies have current control.

spec

 

[Tony Stewart]

Your **broken link removed**, sorry to say , lacks necessary expertise in power control and understanding of ESR for every part (incl. diodes and RdsOn of FETs. (rookie design). Schematic is missing dots on diodes to motor and link on one gate All diodes must have ESR same or less than RdsOn of FETs which must MUCH lower than DC resistance of motor. (not just motor current rating but >>20x motor rating and well heat-sunk.

If you change direction at full speed with mechanical load, devices will burn out !!!

PWM is necessary to regulate energy demand from almost short circuit condition when starting or stopping motor. It is 4x worse when changing direction at full speed due to twice the voltage. (Vcc + |Vbackemf |). where both are equal.

If your design requires such abrupt acceleration and braking power, then your drivers all rectifier diodes and transformer DCR must be < 5% of motor coil resistance. pref, <1%, to minimize losses.

Better is to use PWM with 2pi*f*L raising motor impedance from DCR to 10xDCR using measurements of motor for DCR and L(f).

Say, if you had a motor rated at full load of 4A @12V or 1/3 Ohm equivant load, Rm, the motor coil DCR will be 10-20% of this Rm and 5% of that DCR for the active devices, means the Diodes and drivers must be rated for ESR of 0.5-1% of motor Rm and preferrably 0.1% to 0.2% of Rm, or 3.3 to 6.6 milliohm, which raises cost of active devices. (bad design).
NTE2395 is 28m !! For snubber diodes to be effective ESR of 10mOhm power rating must be 100W !! ESR~=1/W_rated

MOSFETs may not necessarily blow from heat but poor snubber voltage limiting and overvoltage on Vgs.

This is too expensive... bad design.

..PWM is lowest cost speed control which also limits current from rate of change in speed control.
Also using simple unregulated transformer yields saturation on overcurrent and more problems. PWM speed control must be slew rate limited in case of contact bounce.

Think what Murphy's Law really means. Not enough thought in design, process and material quality issues.

Polyfuse may work for a locked rotor but not good enough for 5x -8x surge current or double that if change direction at full speed. Get a bunch anyways. They function at 85'C approx.

 

[Tony Stewart]

If you regulate the current with PWM rather than voltage regulation, you can achieve better efficiency and torque is proportional to current with a V limiter to protect the Vgs limits safely and limit average Vout to motor to limit RPM and motor power. Then unregulated voltage say no more than 50 % of DC motor rating Vrms is possible with careful choice of design.

Does that make sense?

Then the supply looks more like a SMPS speed control than an unregulated pulser driving a linear reg, driving a DPDT bridge with massive surge currents overloading everything in its path.