4-20mA signal

Slwagner73

Lifetime Supporting Member
Join Date
May 2017
Location
Bellevue, MI
Posts
53
So, gentlemen, I am going to ask if there is any set length that should be followed for a analog signal.
We are installing a system for a plant, and the engineering firm we are working with has not had a very stellar record.
And after discussing with them, (the engineering team), and explaining that the conduit runs, and wire length will be over four hundred feet, would it not make more sense to place a remote IO a lot closer to the end devices, rather than try to push that signal that far? (They are telling the owner I am just being a worry-wart) :oops:

To me, a power supply, some Ethernet cabling, and a remote IO would make more sense than the ongoing issues I foresee with the current design.

So, long story short, is there any set limits for 4-20mA cabling.
For the record, they have also spec. 18-2, (shielded). Which, again, seems small for the distance.
 
4-20mA signals have a big advantage over other signals in that they're inherently less susceptible to voltage drop and other issues prevalent with long cable runs. 400 feet is definitely a long run, but as long as you have a suitable power supply it'll probably work.

But. If there are more than one or two devices in that area, one hundred percent it's worth putting a remote I/O rack down there. What happens when you want to add a couple more devices? Or a whole new cell? If you've got remote I/O down there, you can expand your bus as much as necessary, and the cost is very low. There are also advantages in troubleshooting if the opposite ends of your circuit aren't 400 feet apart.

Of course, you have to consider how you're going to get your I/O rack down there too - ethernet maxes out at 100m (300ft), so you're going to need to look at either fibreoptic or an intermediate switch. The cost of this may make it unattractive to your customer.

If it were my plant, I'd 100% spend the extra money and put an I/O rack in down there. It'd probably save me more money in the long term than it costs in the short term, and it's an objectively better solution in my opinion. But if upfront cost is the primary concern, running 4-20mA signals over 400ft should work OK.
 
It all has to do with loop resistance and 400 feet of copper wire is a breeze for a 4-20mA loop.

If the 4-20mA signals are field devices that are 2 wire loop powered devices, then I can guarantee you that with a conventional, standard 24Vdc power supply that you can run 400 feet of copper wire and have no problem at all.

If the 4-20mA signals are analog outputs from a PLC, then I can guarantee you that they can drive any standard 4-20mA input on a drive, SCR, I/P or valve positioner without any problem at all over 400 feet of copper wire.

What you have to watch out for is when you drive multiple devices in series, like split ranging two I/P's or two valve positioner. That might take a heftier power supply or a 4-20mA repeater.

Buy your remote I/O if you want, but 4-20mA loop issues at 400 feet are not going to present a problem. (Do use shielded twisted pair cable).
 
Last edited:
400’ is nothing especially with 18 gauge. I’ve work with systems that ran 4-20 on 19ga telecommunication cable upwards of two miles using 24v power supplies. Most transmitters will have a voltage/resistance chart that tells you the minimum voltage for a given loop resistance. I’ve come up in the less equipment more copper everything having everything be more centralized for various reasons. Never had an issue with distance.
 
The common sense approach would be to fit remote IO and gain some future proofing, especially as it pretty much pennies these days.
 
In this case, you are being a worry-wart :)
Don't worry, it will work just fine with the shielded cables.
And imho it is a better design then the remote IO in this case.

The common sense approach would be to fit remote IO and gain some future proofing, especially as it pretty much pennies these days.

Were it just an rem IO.. But there is the power supply, CBs, extra cabinet, finding a place for that cabinet, fiber to ethernet on both sides, or yet another extra cabinet with a PS and a switch..

And all that for what? To have more points of failure and a harder to inspect and fault-find? It is not just about the cost, it just complicates things for no reason.
 
4-20ma is fine for 400ft. The big question in my mind though is number or runs, conduit, etc. The electricians are going to charge for every cable and conduit. If this us 2-3 cables in 1 conduit, then homerun it. If this is 30 channels, in 5 conduits then it may make sense to run some cost estimates to compare.
 
Well, I ending up running a 1 1/2" conduit, just so they would have more room to grow.

Based on your opinions, I'll stop worrying about it.

Thanks for the advice.
 
Just pull a few spare cables and you'll probably be covered. No need to over-engineer it. Distance is not a legitimate reason for concern in this application.
 
Were it just an rem IO.. But there is the power supply, CBs, extra cabinet, finding a place for that cabinet, fiber to ethernet on both sides, or yet another extra cabinet with a PS and a switch..

And all that for what? To have more points of failure and a harder to inspect and fault-find? It is not just about the cost, it just complicates things for no reason.

Aside the the fibre to ethernet, which isn't needed, it's all pennies.

Thats the first time I've heard remote io called 'complicated' before.
 
Completely agree with Janner_10. This is particularly true for sites that tend to grow organically over time. I've had projects where the I/O count in a particular part of the plant has doubled between design phase starting and installation being completed. Yeah may be that's a poor reflection on the original planning, but that's life. Remote I/O is an enabler for additional nice to haves.

And given that these additions often happen in the 11th hour during commissioning, being able to put in a 6m sensor lead, a gland and mark up a drawing is much easier than pulling other 100m cable as well as doing all the above.

The tricky thing is determining at what point remote I/O is easier than pulling extra cables. In the OPs case... Maybe not. But in a lot of cases, i think it becomes a missed opportunity.
 
I have had issues with 4-20ma inputs and outputs to VSDs in a couple of sites lately. All over the place. Investigating - have never had the issue before. Installed toroids in one plant and wound the 4-20ma output cables around the toroid - problem solved but I want to find the reason.
 
Aside the the fibre to ethernet, which isn't needed, it's all pennies.

Thats the first time I've heard remote io called 'complicated' before.

I didn't call it complicated. I said it makes things more complicated in this instance..

Remote IO makes sense even on only 20m when planned and with many signals.. But here, we are talking about expanding a system with a few signals (not sure how many, but it seems it might even be only 1 or 2).

And if you think having a remote io with a new cabinet even without fiber for 4-6 measurements is less complicated then just running a few cables, we just have to disagree.

@BobB VSDs are a different thing. They can make problems even on a 2 meter distance if not grounded properly, shielded properly, proper cable to the motor used..
 
Can you keep the loop resistance below ~1k Ohm? If so, it'll work. 18 AWG would hit 1k Ohm at about 15,500 feet length. 400 feet is no problem at all.
 

Similar Topics

Hey folks, I have a situation where I need to connect a 4-20mA transmitter to an AI card that only does 0-20mA. On top of that it only has 8 bit...
Replies
4
Views
2,376
I am trying to debug a pressure transmitter that outputs a 4-20 mA signal. Currently , the value is changing on the HMI screen and PLC depending...
Replies
6
Views
2,292
Just asking for clarification here, but from my research all I need is a 500 ohm resistor across the AI and the common? Heres my schematic. I...
Replies
5
Views
2,546
Any one know of way to split a 4-20 mA signal 4 ways? Do they even make such a device? Thanks
Replies
12
Views
3,609
Hello everyone, I don't have much of an electrical background and so I have these questions: Please check the attachments. I have 4 cases where a...
Replies
15
Views
4,105
Back
Top Bottom