Running IP Cameras Over Existing Cat3 - Will It Work?

We are dealing with a legacy analog system that was installed using 4-pair Cat-3 and video baluns. The customer wants to go IP.

My research suggests that we should be able to use standard switches and cameras as long as both support 10BASE-T. Standard PoE is supposedly supported over Cat-3 as well (PoE+ is not). We have gone through and certified most of the lines (10 out of 16 - the remainder require a lift), and all of them have passed certification.

My next step would be to test a single camera and the intended switch before committing to a price with the customer. Am I being overly cautious here? It really seems that since we passed certification everything should "just work", but I have never used Cat-3 for cameras.


Distance would likely be the problem. Although you can transport IP directly over CAT3, I would think the distance capability would be limited, Also, noise rejection would be nearly nonexistent since that relies on a high number of twists per foot - something CAT3 doesn't provide.

That said, you could use product(s) like the Vigitron Vi2301 or the NVT EC1701 or other EoUTP products. They typically claim to be able to transport Ethernet over just about any wire pairs, including 18/2, shielded cable and even NM power wires.

NVT Vigitron

Nope, Cat3 has the same 100M distance limitation. IP has nothing to do with it, the singalling protocol is *ethernet*, the data encapsulation is more or less irellevant. Ethernet distance is limited to 100M on twisted pair for collision-sensing reasons.

Cat3 actually had pretty good noise rejection, if the cable was run according to best practices at the time. It's designed for a max 16Mhz carrier frequency, which limits it to 10Base-T for Ethernet purposes.

I would suggest using a managed switch that lets you lock the ports to 10Base-T, instead of letting them try to auto-negotiate, which might cause some weird issues. That will also limit your maximum resolution somewhat, as you probably wouldn't want to try going above 3MP or so on a 10Mbps link. It would be worth testing the intended camera on a switch with a port locked at 10Mbps, just to make sure you don't run into any weird issues.

Awesome - I should have thought of that. I just VPNed to a system with similar hardware and locked a switch port to "10 Half" (cisco 300 series switch). The camera didn't even hiccup - things are looking good.

It's worth remembering, as well, that the 100m "limit" is really only a limit in spec: beyond 100m you start to see timing, collision, and crosstalk issues become problematic and there's a pretty quick rolloff to things not working at all, but there's not an enforced "brickwall" limit where things will work up to 10,000cm and then not work at all at 10,001cm. Higher-quality cable in a perfect best-practices installation will get you a fair bit farther; cheap cable with shoddy installation will fail well short of 100m.

The distance limit is 100% timing based, a factor of the propagation speed of the signal down the wire along with the Ethernet spec requirement for how long a device listens for a clear "gap" on the wire before transmitting.

Within the scope of twisted-pair cable, a "better" cable is not going to change the laws of physics about how fast a signal travels down copper (we're talking about all-copper cables/wires here, the standard for any halfway decent twisted pair). In fact "better" cables are sometimes defined by twists per inch, where a "better" cable could have more twists, and therefor slightly lengthen the cable, contributing to the problem.

As you exceed 100M, you increase the possibility that a device at Meter 0 starts to transmit at JUST the same time as a device at Meter 95, thereby causing those two packets to "collide", making both devices need to back-off, wait, and re-transmit. These collisions are expected to a certain degree and are part of the Ethernet spec (CSMA/CD carrier-sense multiple acces / collision detection). You can have a fair number of random collisions and still get your "wire speed" for total expected throughput. But if the number of collisions increases too much, you're spending all your time trying to retransmit the same data, and your effect throughput drops considerably.

Switches reduce your collision domain (basically a segment of network where two devices have the potential to collide with each other) greatly because they create a series of sort of mini-networks, between the switch and each device connected to it. Your device now is mostly only contending with the switch for network acces, and not contending with EVERY other device on the network simultaneously. It's more complex than that, but that's the basic concept. Switches started becoming more affordable and more common about the same time that "better" components and cable started becoming more common, so I think a lot of people attribute functioning 120 meter cable lengths to installation more than switches.

MOST switches and devices WILL work at 120 meters for example, provided the installation wasn't done by a half-blind chimpanzee who tried to invent TIA 568-D on the fly (haha, cabling humor there). However, SOME switches will look at the timing of the link and refuse to establish a link if the propagation times are too long. I've mostly seen this happen with enterprise-class switches, but haven't dealt with over-length segments enough to really fully test this. So, the 100M limit is mostly not a "brick wall", until that time on a new job where it is and you're going crazy trying to figure out why your cameras and cables that worked to 120 meters before suddenly aren't doing so now.

I'm a big fan of and have used Vigitron in quite a few installs ... but the fact of the matter is you're going to get noise and depending on where/how the Cat3 was run, you're most likely not going to be able to do anything about it.

Alon, analog video using baluns on CAT3 = noise. Not so with IP transport over CAT3.

Depending on the qty of IP cameras you have many options

http://www.phybridge.com/

and our favorite solution with cameras galore, some 3,000 away on 50 year old telco pairs in the dirt is

http://www.ethernetdirect.com

speak to Philip

Have a look at

PLANET VC-231 100/100 Mbps Ethernet to VDSL2 Converter

I used once for over 400 feet on CAT 3

works well

Just to counterpoint all this: I HAVE done ethernet over existing Cat3 for workstations in an office situation - it CAN work at 100Mbit, but not reliably (10Mbit was rock-solid except in the case of one cable that turned out to have three splice points in its 250' run, one of them directly beside an HVAC power junction - that one, we got a link light, but nothing else). This was part of a project on multiple sites, switching out token-ring networks for ethernet, and in some cases re-using old Cat3 phone runs. All runs terminated in interchangeable jack frames at the desk end (pop out the T-R or phone jack module, pop in an Ethernet jack module), and BIX blocks at the head end (re-punch jumpers to the appropriate pairs).

Not something I'd suggest doing on a regular basis today, but at the time (1999), Cat5 was stupidly expensive and used only when there wasn't existing cable, or when the existing cable didn't work (as in the above example... but there were only two or three instances like that, where cable had to be replaced, that I can think of).

here I 'm using technology which can work on cat 1,2 or 3 and even untwisted pairs ... where VDSL can't workSome Power Line communication manufacturers (German, devolo) have released some specific solution over coax or telephone/audio wires which can stand crosstalk errors (where Ethernet can't ..) and reach much biger distances than 90 meters (400 meters with PLC onTP)

Anyhow, regardless the technology (Ethernet, xdsl, PLC, Eoc..) or the wire, I test the segment with Iperf (freeware) and 2 pc to make sure I can get the bandwidth. With PLC you can get a 50 Mbits real shared throughput and 180 Mbit/s on Kx6 coax !

On Ethernet , it's not shared bandwidth , but it should reach the same levels as Cat5E or 6A : 94 mbit/s real throughput (on a 100 mbit theorical bandwidth ) . If you get a 20 mbits it's wont ' be normal at all.. This test is better than a fluke and much cheaper...

Carl - Supposedly 10BASE-T is rated to 100 meters even on Cat3. The furthest run in our situation is 190 feet.

Matt - Thanks for the on point response. I believe that all of these are straight runs with no splices. The installation is not that old and was done during constrution (and the coinciding inspections). The workmanship appears good. These all lead me to be hopeful that it is not run next to mains power anywhere. Did you use PoE in your situation? How did that work out? When you say 100Mbit was unreliable, what do you mean? Did it gracefully fall back from 100Mbit to 10Mbit, or should I somehow force 10Mbit (a switch port setting somewhere I am guessing)?

Everybody Else - I know proprietary specialty hardware can solve the problem, but I would like to avoid the expense if possible. If you have got specific experience with Cat-3 ethernet & PoE please share.

Actually, thinking back... seems to me there WASN'T a problem with 100Mbit. As I recall now (15 years later, so some of it may be a bit fuzzy), one of the drivers behind the project was to up the available network speeds, since T-R maxed out at 16Mbit (although, it was still pretty fast, being a non-collision protocol).

If memory serves, most of the sites were using ATM connections to the corporate WAN, into a router, then to a *hub* (switches were also ridiculously expensive then, compared to hubs), with separate routers for each group of hubs up to 24 ports (for collision mitigation).

I think PoE was just a glimmer in the IEEE's eye at the time as well; I can't recall even having heard of PoE back then. The only devices that got ethernet were PCs, and printers that had interchangeable network cards (pull out the T-R card, slot in ethernet). There were a ton of parallel-port-only printers running on Lexmark MarkNet T-R printerface boxes, that it seems to me stayed on T-R until they died, at which point they were replaced with Ethernet-capable printers. (Almost all printers, BTW, were either commercial-grade LaserJets (II/III/IV), or Lexmark Optras).

You're thinking of 100-BaseT4, which was a 100Mbs variant that used all 4 pairs, you won't find that anywhere today, especially not in an IP camera.

Cat3 was good for 10Mbps Ethernet, or 16Mbps Token Ring.

PoE was first discussed around 1998, IIRC, a friend of mine at Nortel was working on part of the spec.

Definitely wasn't baseT4, as only two pairs were used for T-R, and when we switched those jacks to ethernet, we only moved the two jumpers on the BIX block; we never added any.

IIRC, there were 3 or 4 100Mbps competing options around that time. One of them I think was from HP. So it might have been some other quasi-Ethernet variant.

Or, it's possible that you ran actual 100-BaseT on Cat3. While the cable wasn't neccessarily rated for it, if you didn't have a lot of interference sources, it would work. It's one of those things though that I wouldn't say "since it worked here, it should be good everywhere". More like "well, we saved some cost on THIS one".

Back when I started in the cabling industry, we ran "100Base-T" on Cat 3 in a couple instances. I put it in quotes because sure, the port on the hub (yeah, hub) lit up green instead of orange, but I doubt we were getting anything close to what we'd have gotten on better cable.

One of the first projects I did when I started was a complete Cat 3 cabling system running 10Base-T, though, and you can absolutely get that reliably, and be following cabling standards, even.

It's possible, too, that the workstations on Cat3 were only running at 10Mbit... trying to remember if they had both 10 and 100Mbit hubs, or dual-speed hubs, or exactly how that was done. I wasn't involved with the system design, I did some installation and some coordination of other installers on-site, so while I did pay attention to what the system guys were doing (for the sake of learning!), I don't remember a lot of the details of their setup.

It was a wide variety of sites all over BC, from small two-person offices to large sites with three or four dozen staff, and some were "more important" than others and thus more likely to get wiring upgrades along with the system upgrades.

Then disable all switch ports from AUTO-NEGOTIATE and manually set them to 10MB. Note not all devices like cameras, phones, LAN cards support 10MB. 10/100 10/100/1000 is common but we've seen plenty that are 100/1000 and no 10. We learned the hard way.

If you are using non-managed switches, the devices may support 10 only

An update.

We have now 30+ cameras working on legacy Cat3. It all just worked. No hiccups.