Over Current Protection For Solid State Relay (SSR) - Fuse Is Not Fast Enough

Thread Starter

Mahonroy

Joined Oct 21, 2014
406
Hey guys, I am having some solid state relays fail during short circuit at the output, and I am wondering what a better way might be to protect them?

This is the relay I am using:
TLP3546A(TP1FCT-ND
3.5A continuous current draw
10.5A current draw for a maximum of 0.1 second pulse

And this is the fast blow fuse I am using:
0698Q3150-01
3.15A Fast Blow

They are typically used at 12 or 24 volts, AC or DC.

At modest and large current overloads, the fuse trips no problem, and nothing bad happens to the SSR.

During a short circuit, the fuse also trips, but the SSR sustains permanent damage and is now permanently in the ON position. Some times the SSR survives this, but most of the time it does not. It seems the fuse is not fast enough for short circuit protection. Here is the schematic:
ssr_protection.jpgIs there a better way of protecting against short circuits?It seems like a "Ultra-Fast-Blow" fuse would work, or a semiconductor fuse, since they react so quickly that it would save the relay.... however, I can't really find any reasonable ones for a 2.5 to 3.25 amp protection.

I also considered incorporating a series resistor on the output to try and limit current... but even then it seems I would have to use a massively sized resistor, and it would emit a lot of heat.

Are there any other options you can think of? I was almost wondering if there might be a way to accomplish this with some sort of a current draw sense circuit that could react faster than the fuse, and shut off the output before damage occured. Not really sure though.

Thanks and any help or advice is greatly appreciated!
 

crutschow

Joined Mar 14, 2008
34,280
Yes, I was told when I first studied electronics that the transistor is (almost) always faster than the fuse. :eek:

What is the maximum supply voltage and is it AC or DC?
 

Thread Starter

Mahonroy

Joined Oct 21, 2014
406
Yes, I was told when I first studied electronics that the transistor is (almost) always faster than the fuse. :eek:

What is the maximum supply voltage and is it AC or DC?
I have it displayed as less than 48 volts, AC or DC. The relay itself is rated for up to 100 volts, AC or DC.
99% of the time its being used with either 12 or 24 volts, both AC and DC.
 
How about putting the TVS at the load side? Add a snubber?

At work we had some AC SSR modules, but the heaters were in vacuum and shorts were possible.

I did use over-rated modules like 25A and 25A I2t fuses, but I did size a 3AG fuse for the load value or about 3A usually.
Most of the time, the cheap fuse blew and not the $25.00 USD one.

The SSR's never blew. They were SCR;s back to back.

I made sure the SSR units had current limit as well. These units were 120 V and were operating into a a Variac.

What's the nature of your load(s)?
 

Thread Starter

Mahonroy

Joined Oct 21, 2014
406
How about putting the TVS at the load side? Add a snubber?

At work we had some AC SSR modules, but the heaters were in vacuum and shorts were possible.

I did use over-rated modules like 25A and 25A I2t fuses, but I did size a 3AG fuse for the load value or about 3A usually.
Most of the time, the cheap fuse blew and not the $25.00 USD one.

The SSR's never blew. They were SCR;s back to back.

I made sure the SSR units had current limit as well. These units were 120 V and were operating into a a Variac.

What's the nature of your load(s)?
The issues are not with over voltage. The problem is over current... specifically when an accidental short circuit happens.

All kinds of different things are connected to the relays (motors, relays, contactors, heating elements, both resistive and inductive loads). The fuse works as intended if someone connects up too large of a motor, puts too much load on the motor, connects up too many contactors in parallel, etc. The problem is during accidental short circuits.
 

AlbertHall

Joined Jun 4, 2014
12,343
This is probably not any help to you but I have seen a fuse which operates quickly enough to protect transistors but this was in a very much higher current circuit. The fuse is a piece of normal fuse wire submerged in water. At low currents the water carries the heat away from the wire. When the current increases sufficiently the water cannot carry the heat away quickly enough and the water immediately around the wire boils and now the wire is in a gas bubble and heats up very quickly and blows. Cunning huh?
 

crutschow

Joined Mar 14, 2008
34,280
DC current limit circuits can be fairly simple, but become quite complex for AC current.

I would suggest a small resistor, such as a 0.1Ω, 2W, in series with a fast-acting fuse.
That should likely limit the current sufficiently for the fuse to blow before the SSR.
 

Thread Starter

Mahonroy

Joined Oct 21, 2014
406
DC current limit circuits can be fairly simple, but become quite complex for AC current.

I would suggest a small resistor, such as a 0.1Ω, 2W, in series with a fast-acting fuse.
That should likely limit the current sufficiently for the fuse to blow before the SSR.
Thanks a lot for the suggestion!

Do you know if there is a good way to simulate this, or a way to run through some numbers to verify if it might work or not? I would normally just connect up a 0.1 ohm resistor in series with the device and try it out... problem is in the past when I was doing tests, a lot of the time I could not damage the SSR by shorting it out (the fuse would protect it).

With that 0.1 ohm resistor, it looks like it would dissipate 878 mW of power @ 24VDC and 3 amps during normal max operation, so I agree that a 2 watt resistor would be good here.

And it looks like during a short circuit (assuming a short of 0.08 ohms, hope that's a good assumption?), a max of 133 amps would flow, which is less than the rated breaking capacity of the fuse which is 150 amps, so does that mean the fuse should definitely stop current before the SSR breaks?

Also during that short circuit, it looks like the 0.1 ohm resistor would dissipate 1.78 killowatts. Do you think the resistor can handle that for the short amount of time it takes for the fuse to stop the current?

Last, for 24VAC RMS, which is 34VAC peak to peak, it looks like I would need a 0.3 ohm resistor to stay below the rated breaking capacity of 100 amps (for AC)? But 0.3 ohms under normal 3 amp operation would dissipate about 2.8 watts continuously. So in this case, would I want to use a 0.3 ohm, 5 watt resistor to cover all bases?

Thanks again for the help!
 
Last edited:

MaxHeadRoom

Joined Jul 18, 2013
28,617
The issues are not with over voltage. The problem is over current... specifically when an accidental short circuit happens.
Some fuse manufacturers have fast blo and also Rectifier (semi-conductor) fuses, these are super fast sweep through types, I believe Littlefuse sells them.
We used to use them on the old Triac 3 phase drives that otherwise when a power outage occurred would often take one high $$$ device with it.
Max.
 

Thread Starter

Mahonroy

Joined Oct 21, 2014
406
Some fuse manufacturers have fast blo and also Rectifier (semi-conductor) fuses, these are super fast sweep through types, I believe Littlefuse sells them.
We used to use them on the old Triac 3 phase drives that otherwise when a power outage occurred would often take one high $$$ device with it.
Max.
Hi Max, thanks for the response. I had already mentioned that I am using a fast blow fuse, and that I could not find any reasonable semi-conductor fuses in the 2.5 to 3.5 amp range. They are all huge and very expensive, and typically a lot higher amperage.
 

crutschow

Joined Mar 14, 2008
34,280
And it looks like during a short circuit (assuming a short of 0.08 ohms, hope that's a good assumption?), a max of 133 amps would flow, which is less than the rated breaking capacity of the fuse which is 150 amps, so does that mean the fuse should definitely stop current before the SSR breaks?
There is likely also some other line/wire resistance in the circuit, so I think you are okay, for example the 24V supply impedance.
You said you've already tested it with no added resistance and the fuse blows.
I don't think you need to go to a higher resistance.
Also during that short circuit, it looks like the 0.1 ohm resistor would dissipate 1.78 killowatts.
I think you are okay for the short time that will occur until the fuse blows.
You could use a wirewound fuse, which as a lot of overload tolerance.
 

Thread Starter

Mahonroy

Joined Oct 21, 2014
406
There is likely also some other line/wire resistance in the circuit, so I think you are okay, for example the 24V supply impedance.
You said you've already tested it with no added resistance and the fuse blows.
I don't think you need to go to a higher resistance.
I think you are okay for the short time that will occur until the fuse blows.
You could use a wirewound fuse, which as a lot of overload tolerance.
Thanks again for the response!
What about the scenario for 24VAC (34v peak to peak). Do you think I would still be good with the 0.1 ohm 2 watt resistor?
 

crutschow

Joined Mar 14, 2008
34,280
Thanks again for the response!
What about the scenario for 24VAC (34v peak to peak). Do you think I would still be good with the 0.1 ohm 2 watt resistor?
Yes.

But I would test the circuit while monitoring the time it takes for the fuse to blow with a digital oscilloscope, to get a better feel for what happens during a short.
 

MisterBill2

Joined Jan 23, 2018
18,167
I have questions and I have what could be the solution, if the relay is only used for DC. The solution would be a regulated power supply that will go from voltage regulated to current regulated, or current limited, almost instantly. That is not hard with a DC supply, with an AC supply it gets more complex. The plan would be for the current limit to be set at some point above what the load would ever require but below the safe limit for the relay, and then the voltage set at the required voltage for whatever was connected as the controlled load. If the response of the current regulation is fast enough, when there is a short circuit the voltage will drop as the current limiter takes over and the result will be that the current never exceeds the desired maximum level. The theory is quite simple, the actual circuit is a bit complex, but such power supplies are quite common and available commercially from quite a few different companies. Success depends on the speed of the current regulator and the length of overcurrent that the relay can tolerate.
 

TeeKay6

Joined Apr 20, 2019
573
Thanks a lot for the suggestion!

Do you know if there is a good way to simulate this, or a way to run through some numbers to verify if it might work or not? I would normally just connect up a 0.1 ohm resistor in series with the device and try it out... problem is in the past when I was doing tests, a lot of the time I could not damage the SSR by shorting it out (the fuse would protect it).

With that 0.1 ohm resistor, it looks like it would dissipate 878 mW of power @ 24VDC and 3 amps during normal max operation, so I agree that a 2 watt resistor would be good here.

And it looks like during a short circuit (assuming a short of 0.08 ohms, hope that's a good assumption?), a max of 133 amps would flow, which is less than the rated breaking capacity of the fuse which is 150 amps, so does that mean the fuse should definitely stop current before the SSR breaks?

Also during that short circuit, it looks like the 0.1 ohm resistor would dissipate 1.78 killowatts. Do you think the resistor can handle that for the short amount of time it takes for the fuse to stop the current?

Last, for 24VAC RMS, which is 34VAC peak to peak, it looks like I would need a 0.3 ohm resistor to stay below the rated breaking capacity of 100 amps (for AC)? But 0.3 ohms under normal 3 amp operation would dissipate about 2.8 watts continuously. So in this case, would I want to use a 0.3 ohm, 5 watt resistor to cover all bases?

Thanks again for the help!
When evaluating overload, consider that these all have resistance:
SSR
Wires
Fuse
Added R
Power source
 

MisterBill2

Joined Jan 23, 2018
18,167
I re-read the initial post and realize that we have no information about the application or the nature of the short circuit. In post #5 we find that all different kinds of loads get connected. So now I am wondering about the rest of the details, such as the possibility that the short circuit also connects to some other circuit, as might happen when a cord or conduit is accidentally crushed, or in a test setup, if a test lead accidentally contacts a ground or some other voltage. It might possibly be that adding a second fuse on the other side of the SSR would solve the problem, if the damaging current is taking a different path from what you think it is. At least that is a possible solution, given that we do not know all of the details. My guess is that the circuit is actually more complex than what we are shown.
Simulations seldom give the correct solution if the model is not actually the circuit being considered. Sneak paths are almost always different from what is known.
 
Top