Short circuit protection for 4-20mA outputs?

[scoobyrollz]

I have 4-20mA outputs that need to be short-circuit protected and a resistance of 100 - 500 ohms. What are some suggestion for the protection circuitry and to comply with the required resistance?

 

[ScottyUK]

Short-circuit protected? The source delivers a controlled current into, ideally, a short circuit. The burden resistors make things harder for a current source.

Presumably the 100R - 500R requirement is the minimum burden that the current source can drive?

Can you clarify exactly what you want the circuit to do?



----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[peebee]

Assuming ScottyUK is right, and he usually is, I'd think you could put a 100-home resistor in series (unless the total series resistance of load plus the extra 100 ohms was too high a total load).

 

[Skogsgurra]

But.

It IS inherently short-circuit proof! It needs no short-circuit protection!

Scotty told you.

 

[felixc]

If you need to protect a 20mA circuit against anything it is about spikes and overvoltages.

 

[buzzp]

Usually, the driver of such a circuit is only capable of sourcing so much current before the voltage supply collapses. However, this is for SOME resistance in the circuit. If their is no resistance in the circuit (and there always is with 4-20mA circuits) then you have a short circuit. This would take out the driver circuit of your 4-20mA circuit unless it had some current limiting capability built in AND all the ones I have seen do have this protection.

I never had anyone ask this question but it seems feasible that some badly designed products may attempt to supply current above 20mA. If it makes you feel better, put a 1A fuse in the circuit.

 

[Skogsgurra]

Objections, buzzp.

If it is a 0-20 or 4-20 mA device - and that is what we are talking about - it can be short-circuited until Hell freezes over. And then still some time.

And a 1 A fuse is equivalent to a piece of heavy gauge wire in this case.

Made my point?

 

[ScottyUK]

Hi BuzzP,

Don't forget that a current source varies its outout voltage to accomodate the load, within the available range of voltage. This usually places a top limit of about 500R or so on commercial devices, which run out of voltage to drive the demanded current.

The bottom end of the range can go down to zero resistance because the device is not trying to regulate output voltage, only output current. A good current source should be able to maintain control of output current down to a dead short across its terminals. A bad current source... may be a different story!



----------------------------------

If we learn from our mistakes,
I'm getting a great education!

 

[buzzp]

That was my point, a bad design could supply more than 20mA. Chances of it supplying enough current for a fire are minimal but probably still possible in some designs. However, given the wide variety of manufacturers and products using current loops, I would have to verify the current sourcing capability of suspected designs or companies.

My point is it is possible to supply more than enough current, depending on design. Chances are slim but still possible. A one amp or five amp fuse for that matter is not going to effect the current loop. If the load resistance is that marginal that a few ohms causes issues then your operating too marginal already.

If you want to prove your design, short the outputs together and see what happens. I would bet that the current will not go much above 20mA but it is an easy test to verify the design.

Skogs, Just because it sources 20mA does not mean that this is the maximum. There are bad designs in all industries and given the size of this industry, I would not say 'There is no chance of this happening' in any design. If the device is made by a reputable company then there probably is no chance of this happening. Never say Never.

ScottyUK, you understand my point.

 

[DRWeig]

Guyz,

If it's overcurrent protection in question, there's a simple little circuit to limit the current to whatever. We found that a lot of European 4-20 mA stuff would fail in a mode that drove 40 mA or better, thus burning out sensing input circuits.

I'll have to dig it out, but it's a TI voltage regulator chip wired in series to regulate voltage across a small resistor that's also in series...

I'll post it here if I can find it this afternoon.

Good on ya,

Old Dave

 

[buzzp]

DrWeig, I dont think the failure of the circuit is in question, only the failure of the circuit to handle low resistance shorts on its output. Maybe I am wrong.

I am sure most designs are fine with shorted outputs but it is an easy test if there is cause for concern. Afterall, there are many ways to design a current source.

 

[mgopalan]

You want either a ZENER Barrier OR a Galvanic Isolaator (sometimes called a repeater power supply) and possibly a Surge Arrestor.

Buy these and you are protected.

The Galvanic Isolator is

Part # MTL5041
from

http://www.MTL-INST.com

Basically you could have a dead short in you primary (hararzous) side and this puppy will clamp you off and not let you blow things up. Rated to Class 1 Div 1. It needs 24V though....And you need to be careful how you mount and use it if you want to retain the certification.

Some of the above discussion seems to talk about current limiting. A resistor is a current limiter and a resistor with a Zener diode is the cheapest protection for current loops. The same company sells Zener Barriers. You can also buy Zener Barriers on OMega.com and a million other places.

The 24V you provide to the Galvanic isolator is used to "mirror" the loop current and also provide enough compliance (about 32V) in the circuit to allow you to easily use a 500 Ohm sense resistor (which is what I think you are saying). 500 Ohms at 20 mA is only 10V and that still leaves you with more than 20V for the actual sensor etc.

As for Spikes and other protection. If you plan to have the loop in an environment where you will have potential lightning strikes etc, look at Surge Arrestors. Again
MTL makes a surge protector for 4-20mA current loops.

Part # SD32X

I love these two things and use them all the time. I would talk to other companies and see what they recommend. Maybe something cheaper than these perhaps.

MG

 

[ScottyUK]

Hmm,

I'm not sure you really read the original question. A current source doesn't care if it is shorted. If it could talk, it would probably thank you for putting a short across the output because that makes things easier. If we discount failure, a current source is inherently current limiting because the current is the controlled variable. It has a range of 4mA to 20mA. You might conceivably want voltage and or power limiting, which a zener barrier would achieve.

What stops the output of the galvanic isolator from going high under failure conditions? Ok, I know MTL's gear is well designed and well built, and is pretty tolerant of abuse. We use a lot of their stuff with I.S. equipment for hazardous areas. But the principle of protecting one current source by replacing it with .... another current source? What have you achieved, unless your original current source was a poor design?



----------------------------------

If we learn from our mistakes,
I'm getting a great education!