Calculating correct fuse amperage

[Circuitous3]

I built the dew heater controller shown below and made the power cord (16 gauge, stranded wire) connection from a 12V lead-acid battery, which connects to the battery with spade connectors and plugs into the front RCA jack. I added a 3-amp fuse on the positive wire between the battery and the plug. I connected a heater element to one of the output jacks on the side of the box which had a 50-ohm resistance and was expected to draw a maximum of 0.4 amp (and supply about 3 watts) when the potentiometer was in the fully open position. In my initial test, I connected the power supply and opened the potentiometer to the channel attached to the heater element. The led did not light up and the element did not get warm. I checked the voltage across the battery with my multimeter and got the expected 12.4V across the battery terminals. When checking the voltage across the terminals at the plug end of the power cord, I got no reading. A little testing showed I had continuity across the negative wire from the battery end to the plug, but no continuity across the positive wire. A little more testing indicated I had no continuity across the 3-amp fuse, which was new. I replaced the fuse with another 3-amp fuse and repeated the process with the same result! Two blown fuses (which should have safely exceeded the expected current level) with no apparent current flowing? I replaced the fuse with a 5-amp one and finally got the expected 12.4V reading across the plug end of the wire. Why should I have blown the 3-amp fuses and how would I determine the correct amperage level (already having taken into account that P=IV)? Now to figure out why I'm not getting current to the output channel.

[Infraviolet]

Quite possibly a fast blowing fuse being broken by an initial surge of current during the startup? If you know you draw <<3 amps (usually a fuse is there to protect against extreme events and dead shorts, so you rate or several times the expected current draw) you could try a slow blowing fuse, which copes with several seconds of overcurrent before cutting off, with a 3 amp rating.

[Circuitous3]

How would you go about estimating what the initial surge on a system might be?

[BILLPOD]

   The perfect use for Min/Max on your DMM

[bdunham7]

I don't see anything that would cause a 'surge' that would blow a fuse.  What is the steady-state current draw of the whole system with the 5A fuse installed?

[tooki]

What kind of heater element? If it’s PTC, then the cold resistance will be much lower (and this current much higher) than once it’s warmed up. The effect can be dramatic, often logarithmic, such that 20 degrees temperature difference cause a tenfold change in resistance.

The rated wattage is, if I’m understanding it correctly, the nominal power at a given temperature. Below that it’ll be much more. One example I looked at is a 12V, 30W@10C unit (so we should expect about 2.5A). But its rated inrush current is 4A, or 48W, over 50% higher. If your heater is anything like that, then it’s not surprising that a 5A fast-blow fuse is needed. With that said, inrush current is precisely why slow-blow fuses exist, and that’s what you should probably be using.

[MathWizard]

Yeah what kind of fuses are they really, maybe they are mis-lableled too.

I was buying regular-slow glass fuses the other day, and I wasn't sure what current rating to get. Since they don't burn out for ages, at like 2x the rated current. I'll have to read something about that.

[Circuitous3]

Thanks, Billpod. I see my DMM has the Min/Max function. I'll have to look into how to use it. 

[Circuitous3]

I don't see anything that would cause a 'surge' that would blow a fuse.  What is the steady-state current draw of the whole system with the 5A fuse installed?

That's a good question. Each heater element should draw approximately 0.2-0.4 amp, depending on the setting of duty cycle (potentiometer setting) for each of the four pulse-width modulator circuits shown in the earlier picture. This was my initial test of the system with one heater element attached, to see if it was working as expected and the heater element warmed up. This heater element consists of 5 feet of 32-gauge nichrome wire wrapped in heat shrink tubing and woven onto a nylon fabric (see picture attached). 32-gauge nichrome has a resistance of 10 ohms per foot, so I made calculations of 50 ohms resistance, with an expected current of 0.24 amp and 2.88 watts from the 12V battery at 100% duty cycle. I had thought a 3-amp fuse would have been safe even in all 4 elements were operating at 0.4 amp each, which was not the case for this initial test. Since the led didn't light and the element didn't warm up, I'm not sure if I was getting current or not. I assume that if no elements were attached there would be no, or minimal, curent drawn. With the battery connected, and the 5-amp fuse, there is no current detected across the battery terminals either with or without the heater element attached (with the 12.4V across the power cord). So it appears that I have an open circuit to the heater element. This is now my main problem to deal with. But it seems there had to be a current draw close to 3 amps initially to have blown those fuses.

[Circuitous3]

Here is the picture of the heater element. 

[Circuitous3]

What kind of heater element? If it’s PTC, then the cold resistance will be much lower (and this current much higher) than once it’s warmed up. The effect can be dramatic, often logarithmic, such that 20 degrees temperature difference cause a tenfold change in resistance.

The rated wattage is, if I’m understanding it correctly, the nominal power at a given temperature. Below that it’ll be much more. One example I looked at is a 12V, 30W@10C unit (so we should expect about 2.5A). But its rated inrush current is 4A, or 48W, over 50% higher. If your heater is anything like that, then it’s not surprising that a 5A fast-blow fuse is needed. With that said, inrush current is precisely why slow-blow fuses exist, and that’s what you should probably be using.

As mentioned above, these are nichrome wire heater elements. Nichrome wire resistance will increase with temperature, but modestly. My understanding is that it has a relatively low temperature coefficient of resistance (alpha=0.00017).  The 30W in your example is quite a bit higher than the ~3W for the heater element. I'm not sure how to calculate the inrush current for the heater element. But I take your point about the slow-blow fuses.

[Circuitous3]

Yeah what kind of fuses are they really, maybe they are mis-lableled too.

They are inexpensive glass fuses in a kit from Amazon, with a range from 0.5 amp to 30 amps.  I'm not sure about mis-labeling, but the 5-amp fuse is working.

[Jwillis]

Fuseology from Little Fuse gives a really good understanding of Fuses and how to determine the appropriate size for application.

https://m.littelfuse.com/~/media/automotive/catalogs/littelfuse_fuseology.pdf

[tooki]

As mentioned above, these are nichrome wire heater elements. Nichrome wire resistance will increase with temperature, but modestly. My understanding is that it has a relatively low temperature coefficient of resistance (alpha=0.00017).  The 30W in your example is quite a bit higher than the ~3W for the heater element. I'm not sure how to calculate the inrush current for the heater element. But I take your point about the slow-blow fuses.

Well what’s the measured resistance of your heater when cold? And what’s the measured voltage of the power supply? (A 12V battery can be a fair bit higher!) Then do some Ohm’s Law juju on those numbers and that should be your inrush current.

Are there any capacitors on the load side of the fuse?

[BeBuLamar]

how much current draws with the 5 amp fuse? Also did you measure the resistance of the heating elements or you simply trust the resistance per length of wire?

[Circuitous3]

Well what’s the measured resistance of your heater when cold? And what’s the measured voltage of the power supply? (A 12V battery can be a fair bit higher!) Then do some Ohm’s Law juju on those numbers and that should be your inrush current.

Are there any capacitors on the load side of the fuse?

The measured resistance of the heater was 54 ohms, close to what was expected.  The measured voltage of the 12V battery was 12.4V.  I=V/R=12.4V/54 ohms=0.23 amp. This is basically what my calculation for the expected current was for the heater element; is the inrush current the same as the steady-state current? As far as capacitors go, I don't think there are any on the PWM circuit board. It has the potentiometer, resistors, some integrated circuits and some other components that I'm not certain of. See attached picture of the underside of the PCB (including the potentiometer which broke off its soldered connections at the bottom edge of the PCB).

[Circuitous3]

how much current draws with the 5 amp fuse? Also did you measure the resistance of the heating elements or you simply trust the resistance per length of wire?

There was no current measured with the 5-amp fuse due to an apparently open circuit to the heater element, although the voltage from the battery through the power cord was measured at 12.4V. The measured resistance of the heater element was 54 ohms.

[BeBuLamar]

I think the inrush current may cause by the PWM board because I am quite sure if you connect the 4 heaters in parrallel directly to the battery the inrush current would still be around 1A. Resitive loads do not have higher inrush current than steady state. The only thing is that as the heaters heat up their resistance increase but not by much and the current would be lower than your calculation.

[Circuitous3]

To follow up, I have solved one problem but have run into another that is confusing me.  The original problem of the open circuit to the heater element turned out to be a loose connection to the negative wire from the power supply into it's socket on the leftmost circuit board shown in the pictures below. After tightening that connection and connecting the 12 V battery to the front jack, the two leds towards the back of the box lit up. The other two remained unlit until I turned up the voltage duty cycle from the potentiometers, which is how they should operate.  The problem is that the two led's controlled by the two middle circuit boards are on even though the potentiometers are set in the off position (that is, there should be no voltage and the leds should be dark). Just as confusing, the voltage output from the jacks controlled by these boards is 12 V, when there should be no voltage. The jacks controlled by the leftmost and righmost PCBs have no voltage when off and show 12 V with the potentiometers in the fully open position, as they should. In the second picture I have turned up the voltage on the rightmost board and you can see the led lit up. The wiring arrangement of the boards is shown in the schematic, which is a screen shot taken from a video presentation by the designer of this dew heater project.

I have checked my wiring at the Phoenix connectors and to the output channel jacks.  I originally thought the two leds could be lighting due to a broken connection between the potentiometer and the circuit boards, but after resoldering the connection (and checking for continuity through the wire to the board) the led remained lit in the off position and 12 V was still measured at the output jack.  I wonder if the problem could be with a component or connection on the printed circuit board itself, but I don't think that's too likely.  On the other hand, when I plug the heater element into the jacks controlled by the problematic circuit boards, they do warm up. So it looks like the dew heater should work even though there is a problem with the operation of the circuits. Any suggestions about what might be causing the faulty operation would be welcome.