Electronic Fuse Design Challenge...

Thread Starter

smp4616

Joined Oct 31, 2019
27
I'm working on a project now that involves a microcontroller driving some high DC currents, an example of the problem I'm facing is that one of the outputs is a high side mosfet driven 10 amp output. The load is undefined, but will not be a motor, or anything really tricky like that. It will in many cases though, be an incandescent bulb, the running current of a typical bulb is around 5-6amps, but the cold current is much higher than the 10 amp limit, hard to measure it as it's transient, but based on what it trips, it's somewhere between 12-15amps. The main purpose of the device is to prevent bad stuff from happening due to things like electrical shorts, or bad installations pulling too much current from the device. To throw another wrench in the monkeyworks, it is desirable to be able to pwm (60-100hz or so) the output for variable brightness. Anyone interested in touching this one? My current paths of thought are as follows:

1- inrush current limit controller such as zxct1032, however, when the inrush cap is set high enough to stay on long enough for a bulb to start up, it's on too long to prevent excessive fet heating during a dead short.
2- ina181 type current amplifier, running the output to the MCU ADC, have software monitor current and turn off the output if over current limit, this would allow momentary overcurrent during startup then intelligently monitor current while running. downsides may be processing overhead and possible software bugs causing catastrophic failure

3- current sense amplifier/comparator chip, with discrete components determining the fault current, the flag of which would run to the MCU and trigger a pin change interrupt, which could then determine whether the overcurrent was within the startup time and then intelligently switch off the fet or leave it on until the bulb heated up. downsides would be more discrete parts, upsides might be more reliable software with less overhead (interrupts) ...

it's a cool project, just wondering if anyone else has run into something similar and might have an elegant solution
 

Wolframore

Joined Jan 21, 2019
1,321
It’s not enough information. What’s the voltage, so this is like a light bulb or it is? Heater... etc? Is it a purely resistive load? Have you done the load calculations? Why can’t you measure it? Do you have an oscilliscope? You should be able to calculate it and get somewhat close. Have you considered using an IGBT with a large heat sink... again you will have to calculate this to make sure it doesn’t fail. Switching introduces power dissipation which won’t help but it’s possible. Without more detail it’s impossible to be anything more than vague.
 

Thread Starter

smp4616

Joined Oct 31, 2019
27
It’s not enough information. What’s the voltage, so this is like a light bulb or it is? Heater... etc? Is it a purely resistive load? Have you done the load calculations? Why can’t you measure it? Do you have an oscilliscope? You should be able to calculate it and get somewhat close. Have you considered using an IGBT with a large heat sink... again you will have to calculate this to make sure it doesn’t fail. Switching introduces power dissipation which won’t help but it’s possible. Without more detail it’s impossible to be anything more than vague.
I have a scope, I haven't done a setup to trace the current of the bulb on startup though, I should, it probably wouldn't take too long. The system voltage is 12-15v, automotive power. The size of the module prevents a large heat sink, so operating in switch mode only is required, the FET is a to-220 package, so it will dissipate some power but there will be 8 of them in a small area, so ideally, it can't be allowed to dissipate much. I'll do a setup and get the current curve of the bulb from cold startup today.
 

Tonyr1084

Joined Sep 24, 2015
3,676
Two things: Cold start current??? What's the voltage and what's the cold resistance? That'll give you your startup current.

Second thing: Is this schoolwork / homework?

[edit] Get a higher amperage FET. Say - 50 amps. OR switch to an IGBT. [end edit]

[second edit] Put a similar sized light bulb across the FET. [end edit]
 

Yaakov

Joined Jan 27, 2019
1,604
Hardware controlled soft start allows too much heating in the FET during a dead short.
Soft start allows the filament to heat, reducing the ultimate current. It will never draw in excess of the maximum if done right.

If there is a short, the soft start can detect an overcorrect condition and shut down.
 

Thread Starter

smp4616

Joined Oct 31, 2019
27
the soft start ramps up the gate voltage based on an constant current feeding a known capacitor, after a given T the gate will be past full conduction voltage, during that time, the fet is in linear mode where it's dissipating heat, the higher the current, the more heat. to get a T sufficient to start up from cold, the time where the fet is in linear mode is too high to handle a dead short without generating excessive heat. the overcurrent protection driver IC bases the time between retries on the same capacitor being discharged by another constant current, so the duty cycle is constant if the device is in fault. true, the device, once running in transparent mode after linear ramp up, will quickly trigger to fault if the current exceeds the trip current, but it's the retry cycle where it's dissipating heat. the zxct1032 datasheet gives details on this.
 

Thread Starter

smp4616

Joined Oct 31, 2019
27
Two things: Cold start current??? What's the voltage and what's the cold resistance? That'll give you your startup current.

Second thing: Is this schoolwork / homework?

[edit] Get a higher amperage FET. Say - 50 amps. OR switch to an IGBT. [end edit]

[second edit] Put a similar sized light bulb across the FET. [end edit]
not homework, I do this for fun ;)
 

Yaakov

Joined Jan 27, 2019
1,604
the soft start ramps up the gate voltage based on an constant current feeding a known capacitor, after a given T the gate will be past full conduction voltage, during that time, the fet is in linear mode where it's dissipating heat, the higher the current, the more heat. to get a T sufficient to start up from cold, the time where the fet is in linear mode is too high to handle a dead short without generating excessive heat. the overcurrent protection driver IC bases the time between retries on the same capacitor being discharged by another constant current, so the duty cycle is constant if the device is in fault. true, the device, once running in transparent mode after linear ramp up, will quickly trigger to fault if the current exceeds the trip current, but it's the retry cycle where it's dissipating heat. the zxct1032 datasheet gives details on this.
Using a current sense and the microcontroller, you can manage the soft start yourself. You said you intended to use PWM for dimming, no?
 

Thread Starter

smp4616

Joined Oct 31, 2019
27
yes, that's one of the options I'm considering, using the built in ADC and interrupt capability of the MCU would reduce discrete components, the tradeoff being processing overhead and possible single point failure at the mcu.
 

Thread Starter

smp4616

Joined Oct 31, 2019
27
IMG_20191121_155728318.jpg

I am still working on this p.i.t.a. and took a pic of the trace of the lamp starting up, I figured it's an interesting bit of data for reference to anyone working on incandescent bulbs anyway. This is a 12v 65watt bulb, .001ohm shunt resistor with a 50v/v current amplifier. I'm still trying to sort the best way to handle the protection while I'm working on the software for the controller. If I add an RC to the output of the current amp, I can essentially integrate the voltage so I don't have the huge spike, but it also slows down the detection of a dead short to the point that it will make a mosfet explode. I'm thinking about running the output to a comparator to detect a full short, and shut down via interrupt in the controller, then also using the ADC in the controller to allow for overcurrent based on a running average of the current, to prevent the fet from overheating, and at the lower overcurrent level, the additional time it takes to monitor the ADC should be tolerable as a few amps over the running current isn't going to pop anything...
 

Thread Starter

smp4616

Joined Oct 31, 2019
27
I have been working on an attempt at a simple solution and had some decent results with this, it's basically a 2 stage comparator, with a delay on one stage, it appears to work well for the application, allowing more time for limited overcurrent conditions without immediate shutdown, but hard and fast shutdown when current exceeds a preset limit that would damage the FET. I don't want to use this many discrete components for the finished product, but I realize only 2 of the circuits are meant to run the large incandescent bulbs, the rest of the circuits will be fine with a simple 1 stage comparator with fast shutoff
 

Attachments

Top