Fuse testing

[Bitsyncmaster]

With the common fuse melting, I've decided to do some testing of how and why the fuses melt the plastic and still keep working.

My first testing was done with my fan fail unit with a thermocouple RTV'ed on the front side to measure the heat. The fuse heating is transferred down into the fan fail unit via the fuse terminals and fuse socket. The heat produced in this test also is some heating components other than fuses. I'm using this test because I have some PPTC fuses ordered for testing.

[Bitsyncmaster]

Got my PPTCs (resettable fuses). Testing looks good so far. Runs my fan fail unit about 25 degrees cooler. But that may because the PPTC is much longer than the regular fuse so the heat is staying inside the PPTC more than the fuse. These PPTC are to long to fit installed in my fan fail unit in the car.

[Bitsyncmaster]

Checking voltage drops which relates to heat. To get the heat in watts, multiply the voltage drop times the current in that circuit.

These drops at 10 amps current cold temps:

Poly fuse 15 amp = 0.063 volts
Poly fuse 20 amp = 0.042 volts
Fuse 15 amp = 0.058 volts
Fuse 20 amp = 0.051 volts

Tested the 40 amp breaker with 20 amps current and it drops 0.070 volts but it varies a little up and down a bit. Must be the mechanical contacts inside changing. This is why I recommend bypassing the fan circuit breaker if you have a fused fan fail. It's just another producer of heat and since it is redundant not needed.

Tested the voltage drop over my fan fail unit with the 15 poly fuse to be 0.093 volts at 10 amps. So it drops 0.030 volts not counting the fuse. That drop is due to the current sensor and the circuit board copper layout. I was just wondering if going to 4 oz. copper would help but it looks like it would not help much to reduce heat.

I'm next going to test the poly fuse for voltage drop after it has been tripped a few times. I'm checking if I could design the poly fuses into my fan fail to replace the fuses and their sockets. It would need to standoff the poly fuses above the fan fail unit to isolate the heat produced by the fuses.

[Exolis]

Interesting, oddly enough the temps seem pretty typical to what I have seen on modern vehicles. Have seen relays heat up to 150F with no issues and within spec. Most likely the plastic is just cheap plastic not capable of that kind of heat. Would love to have an extra one to put it in a temp chamber and see if it meets typical modern temp standards.

[Bitsyncmaster]

Interesting, oddly enough the temps seem pretty typical to what I have seen on modern vehicles. Have seen relays heat up to 150F with no issues and within spec. Most likely the plastic is just cheap plastic not capable of that kind of heat. Would love to have an extra one to put it in a temp chamber and see if it meets typical modern temp standards.

It's not that my fan fail units are failing, I just hate producing heat in the relay compartment. A lot of people are still running OEM fans and it runs only warm with low power fans. My components are mostly rated for 80 deg. C or higher But all the fuses running any amount of current are real heat producers. I'm beginning to think that fuses are only good for short circuit protection and not protecting for marginal overloads.

My testing has been at room temps so running in the enclosed relay compartment will raise those temp readings I recorded.

I'm going to do more fuse testing but am considering incorporating electronic "fuse" circuitry in a new design. But not sure the added cost will make it feasible.

[Exolis]

It's not that my fan fail units are failing, I just hate producing heat in the relay compartment. A lot of people are still running OEM fans and it runs only warm with low power fans. My components are mostly rated for 80 deg. C or higher But all the fuses running any amount of current are real heat producers. I'm beginning to think that fuses are only good for short circuit protection and not protecting for marginal overloads.

My testing has been at room temps so running in the enclosed relay compartment will raise those temp readings I recorded.

I'm going to do more fuse testing but am considering incorporating electronic "fuse" circuitry in a new design. But not sure the added cost will make it feasible.

Yes, I can tell you for a fact fuses are not for marginal over loaded. Depending on who makes the fuses you are using, they will have different characteristics, so you might be able to find something searching for Fuse blow curve. The only goal for a fuse is to protect the wires from catching on fire/being damaged.

I will double check some of the curves we have (granted we mostly use Mini-Fuse low profiles for 5-20 Amps and Fusible Links for 30+), but running 30 Amps on a 20 Amp fuse seems bad. I would almost expect that to blow with 1-10 seconds.

[Bitsyncmaster]

Tested the poly 15 amp fuses with 10 amp fan loads.

Also recorded voltage drop over one poly fuse since I want to see if that drop changes after a few over current trips of the fuse. That is a problem that occurred with poly fuses and want to see if that problem has been corrected.

7:09 76.8 0.063 volts
7:14 87.8 0.068 volts
7:19 96.4 0.070 volts
7:24 100.2 0.071 volts
7:29 102.2 0.071 volts
7:34 103.1 0.072 volts
7:39 103.8 0.072 volts

Again these poly fuses keep my fan fail unit cooler since their taller and that moves the fuse heat further from the fuse pins. Also note these temps were more consistent since I tested it early in the morning before my room AC was tuning on and off.

[Bitsyncmaster]

I set up one 15 amp Poly fuse with a 20 amp load and it took 2 min. and 8 seconds to trip. I then tested the voltage drop after it cooled off. It was 0.063 volts new, it now is 0.086 volts after the first trip.

I did another 20 amp trip time and it was shorter, 1 min. 10 seconds as I would have expected since it's resistance is now higher and it gets hotter quicker. I waited longer for cool down but it still has a 0.083 volt drop. So I think it will probably stay there but will check it again in a few days.

Now I'm going to test a standard size fuse we use in the #7 position (for fuel pump). I want to find what current flowing through that 20 amp fuse will cause it to melt the plastic and maybe start a fire in the rubber fuse holder.

[Bitsyncmaster]

Here is a table of temps of the plastic side of the fuse at increasing currents.

The end result was the fuse started to glow orange the plastic on one side of the fuse melted a hole in the plastic had I not shut off the current, the rubber fuse holder would have caught fire since I saw 500 deg. F on the plastic side that did not melt.

Now this testing was done at room temperature so I'm not sure what effect running in our hot relay compartment would have done to these readings. I wonder what the relay compartment temps run. My guess would be 110 to 120 deg. F in the summer. I would measure mine but it would be cooler than most since I have all solid state relays.

Now the point of this testing was to determine why my friends DeLorean fused fan fail jumpers caught fire. He opened the relay compartment and said the fuses had flames 6 to 8 inches high. That was from the rubber fuse holder sockets. My testing proved it could happen but you would need to be running right near the fuse trip current of 120% of the rating. Anything higher would blow the fuse before quicker.

[Bitsyncmaster]

Test document in PDF

[Bitsyncmaster]

That fuse that I tested and failed was marked BUSS ATO. Don't know what the BUSS part number is.

I wonder if the "LITTLE FUSE" manufacture does the same thing.

[Bitsyncmaster]

I found a 20 amp brand name LITTLEFUSE in my spare fuse bag. Has some numbers marked on the fuse 257 and 0591M so have to reserch those numbers. But what you see is the the element in this fuse is a heavy bar shaped in a "mound" and the BUSSMAN fuse the element is shaped in a "Z".

The LITTLEFUSE popped without glowing and without melting any of the plastic casing. So guess which fuses I'm going to make sure my car has?

Edit.
Looks like my DMCH fuse block came with LITTLEFUSES. I just need to but some non standard amp ratings I now use and did not come in the kit.

[Ron]

Great stuff!!! :thumbup:

[Bitsyncmaster]

The LITTLEFUSE #257 went obsolete a few years back and is replaced by the #287 types.

So the part number for the 20 amp is:
0287020PXCN

Price from Mouser is $0.232 each when you buy 10

Just for "grins" I ran another test on another BUSS 20 amp fuse and got the same red glow and melting plastic after 7 min at 24 amps.

[Bitsyncmaster]

Here is the best price I've found on the internet. If your buying a lot of fuses (replacing them every year) you get them for 7.8 cents each.

http://www.waytekwire.com/products/1367/Fuses/&mfg=111&Volts=32&Style=Blade-Fuse&Type=ATO---ATC-Fuse

[novadmc]

I found a 20 amp brand name LITTLEFUSE in my spare fuse bag. Has some numbers marked on the fuse 257 and 0591M so have to reserch those numbers. But what you see is the the element in this fuse is a heavy bar shaped in a "mound" and the BUSSMAN fuse the element is shaped in a "Z".

The LITTLEFUSE popped without glowing and without melting any of the plastic casing. So guess which fuses I'm going to make sure my car has?

Edit.
Looks like my DMCH fuse block came with LITTLEFUSES. I just need to but some non standard amp ratings I now use and did not come in the kit.

for what it's worth, the 30A fuse in #14 on my car was a LITTLEFUSE (as were most of the ones i recently replaced. All replaced with Bussman ATC fuses) and it was showing signs of melting/bubbling yet hadn't popped. that said, apart from the very noticeable oxidation on the them, i have no knowledge of the age of the the replaced fuses.

[novadmc]

for what it's worth, the 30A fuse in #14 on my car was a LITTLEFUSE (as were most of the ones i recently replaced. All replaced with Bussman ATC fuses) and it was showing signs of melting/bubbling yet hadn't popped. that said, apart from the very noticeable oxidation on the them, i have no knowledge of the age of the the replaced fuses.

i meant fuse #11. god im tired tonight.

[Bitsyncmaster]

The fuse is a resistor that produces heat. When that heat reaches the melting point of the metal element in the fuse than it should open the circuit. Lower amp rated fuses have more resistance to produce the heat required to melt the element with less current. It’s just a smaller cross sectional area of the element.

So a fuse running near it rated current will be running the hottest temps. So if your #11 fuse is melting I would check the current in that circuit and decide if the fuse should be increased in it’s rating to run cooler. That is assuming your fuse connections are clean and tight. Elvis did a study of that #11 fuse melting about 8 years ago and decided it should be increased from a 20 amp to a 25 or 30 amp. The Europe owners had to install higher power high beam lights which pushed the current to high in the 20 amp fuse.

Just ran a voltage drop test over a 20 amp LITTLEFUSE. The voltage was measured at the base of the fuse after running at each current for 2 min.

10 amps. 0.0373 volts = 0.373 watts of heat
15 amps. 0.0625 volts = 0.9375 watts of heat
20 amps. 0.0990 volts = 1.980 watts of heat

Same test with a 20 BUSS.

10 amps. 0.0419 volts = 0.410 watts of heat
15 amps. 0.0689 volts = 1.0335 watts of heat
20 amps. 0.1127 volts = 2.254 watts of heat

[Bitsyncmaster]

You would think I'm testing the fan circuits for a manned space craft mission.:)

I made up a short circuit connector to plug into the two pin fan connector near the fans up front. I attached my clamp on current meter onto that short circuit jumper and fed the current meter output into my oscilloscope.

I install the following fuses into my fan fail unit, started the engine and switched on the AC. Note these are LITTLEFUSE fuses but the BUSS brand looks like they have the same type of protection element so BUSS brands may also be good for the ATM type fuses.

1o amp fuse, 55 amps peak current held for 12 ms.
15 amp fuse, 55 amps peak current held for 32 ms.
20 amp fuse, 55 amps peak current held for 41 ms.

I did this testing to decide if I can make an electronic fuse that works that fast or faster. Also wanted to test if the fuses work fast enough to protect the current sensors in my fan fail unit which they did.

[Bitsyncmaster]

Well I made a mistake when I said the currents peaked at 55 amps. It's really 110 amps which explains why the fuses popped so quickly. My clamp meter puts out 0 to 2 volts which relates to the selected scale of 20 amp or 200 amp ranges. Doing the modification on the scale on my scope I made the mistake.

Anyway, I wanted to test it with a 30 amp fuse.

30 amp, peak current of 110 amps, popped the fuse after 124 ms. And my current sensor in the fan fail is still working fine. It's rated for 20 amps continuous and one 100 ms transit at 100 amps.

[Bitsyncmaster]

Had my friend over the house to install a set of low power fans (Hervey's fans I had sitting on the self). This was the guy that started all this fuse testing because his fan fail jumper fuse caught fire one day.

I first tested the current draw with his OEM fans and one was 13.7 amps and the other was 13.5 amps. I looked at his burnt up jumpers and the one fuse that did not burn was a BUSS brand 20 fuse. I check all his other fuses in the fuse block they also are BUSS brand fuses. So why it caught fire is beyond me. I can only guess one fan motor somehow started drawing more current then I measured. Or the fuse holder itself had a marginal wire connection.

Replacing the fans took about 3 to 3.5 hours. The new fan currents were 7.4 amps and 7.6 amps. My order of about 1000 LITTLEFUSEs did not come in yet. So I will ship him a full set when they do.

[Bitsyncmaster]

It's been a few weeks since I posted here. I've found that the fuse holders add quite a bit of heating problem with my fan fail or the fused jumpers most use. Using the same fuse holder and 20 amp fuse that most users have installed, I did voltage drop testing which equates to heat generated.

The LITTLEFUSE 20 amp resistance is specified at 0.0038 ohms which is what the element has measured from the test points on the top of the fuse at room temp. So I ran 15 amps into the fuse holder with that 20 amp fuse and measured a voltage drop starting at 0.1499 volts on the ends of the fuse holder wires. That results with a resistance of 0.0100 ohms which is so much more than the fuse resistance of 0.0038 ohms.

It gets worse after running for 20 minuets. Voltage drop at fuse holder wire ends is 0.1790 volts and fuse drop on test points is 0.0657 volts.

Yes these sound like negligible numbers but fan fail jumper produces 5.37 watts of heat if your fans are drawing 15 amps each. Don't grab the fuse or you will burn your fingers.

So I went and redesigned the fan fail unit with electronic fuses. I'm still planning on selling the older units since if your running low power fans the heating of the fuses and holders is not as great. The solid state fuse circuitry adds about $15 more in parts cost.