Its a PITA routing all those CT cables

Don’t know what the rules are, but here in AUS, we need to have the IotaWatt OUTSIDE the power enclosure. Its a PITA routing all those cables, and was wondering
1 - Can I use CAT 5/6 cable (8 wires) to get the outputs from 4 CT clamps
and so all I would need is 3 lengths of CAT5/6 cable to support 12 CT clamps
2 - wire this up to a bank of 12 3.5 mm female sockets, insulated from the enclosure panel, exposing the socket to outside the power enclosure.

Then, all I would need is some male-male leads into the IotaWatt from my bank of sockets

1 Like

Cat 5 extensions should work fine. I know what you mean with the cables. I’ve sourced out a new line of CTs that are smaller than the SCT013 and have longer cords. (1.5M).

I have heard various ideas about AUS installations. The two I heard were (not from official sources, one was a talk from a guy in Melbourne):

  • Installation of CT coils must be done by a licensed electrician
  • Installation of the monitoring device must be outside the meter box

I actually didn’t find any real information about this. All the info was about CT coils used by power companies for installations with really high current usage and there are lots of strict rules here:

Its hard to parse that entire document, but I dont think it really deals with smaller personal power monitoring devices like IoTaWatt and Efergy.

I also looked at Efergy installation manual:

Again it is not clear, but I assume it means that a licensed electrician must install the CT coils, but doesn’t mention anything about the Efergy monitoring device being installed outside the meter box.

Where did you hear about this rule that the IoTaWatt needs to be outside the meter box?

Also, is it feasible to use a single wire for the common CT bias of all CT’s, and get 7x CTs connected to the one CAT5?

Not sure about noise introduced doing this. Don’t imagine it would be all that different to using 2x wires for each as there is no differential used for the twisted pair anyway so probably has similar characteristics to obtain 7x instead of 4x out of the one cable.

Inthe USA, there is some language in the NEC about low voltage wiring originating in the panel and going elsewhere. As I recall, the requirement was that for residential installations, the wires had to be protected so that they were not easily accessible outside the box.

I saw a picture on the OEM site awhile ago of an Australian installation where the IoTaWatt was in a plastic enclosure outside the panel with a short length of conduit connecting to the panel for the CT wires. It was a very clean installation and would appear to conform to the US requirement that I mentioned.

Bottom line though, is to eliminate the guesswork and consult an electrician.

In theory the 7 CT per cat5 cable will work electrically, but I would go with one twisted pair per CT. However you do it, let us know how that works out.

In the US, the requirement I have heard says you need 300V wiring (insulation rating), so CAT5 in an electrical box (with high voltage wiring) would be a no. The rules are there for safety and make sense in that light. I have several blue Smurf tubes going to a large plastic electrical box from each panel. I bought some 300V rated 2 conductor + shield/ground wiring for CTs that will need an extension.

My take on the cat5 usage was that it would be outside the box. I agree, no cat5 in the box.

So it seems that we need 300V rated wiring for ALL CT extensions from inside the electrical box. This would mean that using 3.5 mm headphone plugs is not allowed, unless wrapped in the appropriately rated shrink tubing

I have found 300V rated 30 conductor cable (10mm dia) which would cover 12CTs, getting 1m of it is problematic

I’d highly recommend you consult an electrician to assure compliance with local codes.

Later I’ll post some pictures of UL listed CTs that I am sourcing. They have 1.5M cables with what I believe is code compliant insulation. I also have some UL recognized CTs with better quality cable that may be 300V, I’ll ask the manufacturer about that.

With longer cables, it might be better to simply find a way to organize them neatly rather than make extensions.

With regard to multi-conductor cable, most plenum rated thermostat wire is rated 300V and available typically with up to 6 conductors.


Here are pics of some of the CTs that I will be bundling with IoTaWatt soon.

They all have 1.5M cords with 3.5mm jacks. They are CE and UL recognized. Models shown are 50 and 100A. I have a 200A available for US mains as well.


From where are you sourcing these? They look nice, from a smaller size perspective.

They are made by echun. These are nonstandard in that they have longer cords with 3.5mm jacks and models without TVS diodes standard have them added.

The venerable SCT013-000 is accurate, but I find them to be large and clumsy for a US panel, the cords too short, and the phase shift excessive. These are a slightly more expensive, so we’ll see how they are received.

looking good, no doubt
would be nice to make those 3.5mm jacks as small as possible (in diameter i mean)
The ones I bought and made the extensions from were too bulky around the plug end with the insulation and what not. If the plug end can be as close as possible to 3.5mm would be ideal. Makes for a much easier routing etc out of the power box


Sounds like you would be a proponent of eliminating the jacks and going with screw connectors or level action wire connectors. There are pros and cons each way, but on balance, my sens3 is that for the less technical user, the jacks win out. I had considered 2.5mm jacks, which are smaller, but some CTs already come standard with 3.5.

These jacks appear to be molded to the cords and probably are sourced that way by the CT manufacturer.

Yep, screw connectors would be cool, with just a connection block. Would make the routing of those wires from the CTs much easier, coz the jacks just take up so much space when trying to thread the cable to outside the power board.

1 Like

@overeasy, any chance we could get CT cables sourced from your supplier with much longer runs? My solar inverter is in the garage & ties into the city provided power meter directly. So if I want to monitor my solar production I’ll need to add CT’s to the inverter output probably 250-300 feet away from my electrical panel.
I thought I saw a posting on extending CT cables yourself, but am not finding it now. Regardless, I’d rather have cables that aren’t spliced if possible.

I don’t think that’s practical Andy. Getting custom leads applied in China and then shipping them here would be much more expensive, not even considering that it would probably take a month or more to get it done. If you have a single phase solar inverter, you would only need to have one CT out there, so I’d recommend doing it yourself. I don’t see a downside to using 3.5mm connectors, even if you need to put it in a waterproof box.

I believe the OEM site does get into CT extensions somewhere. As I recall the jist of it was to use quality shielded twisted pair microphone cable. I have used cat5 cable with pretty good results. One consideration with a 300’ run would be the resistance that you add to the loop. 18ga solid copper will add 3.8 ohms. 22ga wil add 9.6 ohms. Not a problem for most CTs, but a small lightweight CT might not be able to generate the VA when you add in the 24 ohm burden of the IoTaWatt. You’re pretty safe with something like an SCT013-000 or an eChun ECS 16100.

Running the wire outside of any conduit carrying high voltage would also be advised, not only because it could introduce noise, but it’s also not allowed by most codes.

Thanks for the reply.
I do remember seeing in another thread where someone mentioned microphone cable or twisted pair telephone cable. I guess I’ll get a better estimation of the length of my run & then try to source the cable for the splice extension.

I am not sure why you would need to put the CT close to the inverter, but rather put the CT where the inverter feeds into the electrical panel ?. Unless your setup is much more complex than mine.

oops, sorry I should have read your full post

1 Like

It was in the OpenEnergyMonitor community: