Solid vs stranded wire

[thatguy]

actually calls for 22 gauge

Attachment 817756

And if i recall from their install notes, if shielded wire is used between indoor and outdoor unit, the shielded portion needed to be attached to the ground screw, but ill have to look that info up again to see if there is any reference to its use as the stat wire.

... but only connected to a ground screw on one end of the cable.

When you ground one end of a shield then you get a shield. When you connect both ends of the shield you get an antenna.


Sent from my iPhone using Tapatalk

[ammoniadog]

Called the vendor whom I’m getting the units from.

“We just send 18/4 stranded shielded”

“Manual says unshielded”

“We send shielded with it and no problems”

I’m saving this conversation


Sent from my iPhone using Tapatalk

Did you also tell them that the manual says it needs to be 22 gauge?

18 gauge is better than what LG wants, and you certainly don't want that.

[jbhenergy]

Did you also tell them that the manual says it needs to be 22 gauge?

18 gauge is better than what LG wants, and you certainly don't want that.

lol.

manual also says 18/3 for the wire between indoor and outdoor. 220 on 18 wire. no thanks. yes i know its only 1 amp, but no, im not putting thermostat wire in for 220v

[Juan Madera]

Yup, I buy SO cord with the requisite conductors and works fine. If ya need three conductors, 18/2 w ground. Quite available..

'Stat wire is a no no as it is not rated for line voltage even tho you can buy stranded 'stat wire.

My beef with the chinese mini-cassettes is a totally oddball condensate drain size. Matching up with 1" PVC is a pita.....

[JayMan7]

It's very weird but we have had com problems where installers used solid wire. The com problem is always permanently fixed when we replace the wire with stranded. One theory I had was that the terminal blocks they use are best suited for stranded wire. There's like 10x the contact surface area when you put a stranded wire in there vs a solid wire.

Another theory... a stranded wire is going to be carrying the signal externally but also internally because it will be riding on the "skin" of the internal strands. A solid conductor will carry the signal on its outer "skin". It's true that ethernet is solid but they also use twisted pairs and are not run along high voltage cables. Maybe in a minisplit situation where you have high voltage ac right in the same bundle as the signal wires it's beneficial to have stranded wire carrying the signal deep inside. Maybe that protects it more from the kind of capacitive ghost voltage that ac creates which plagues high impedence sensing circuits.



Sent from my SM-G960U using Tapatalk

[danielthechskid]

It should be noted that stranded wire does not reduce the skin effect unless it is Litz wire where the individual strands are insulated from each other, like by being varnished like magnet wire. If the strands are un-insulated they behave as a single conductor. The problem with Litz is that you have to make sure that all of the strands individually have a good connection at both ends.

[JayMan7]

It should be noted that stranded wire does not reduce the skin effect unless it is Litz wire where the individual strands are insulated from each other, like by being varnished like magnet wire. If the strands are un-insulated they behave as a single conductor. The problem with Litz is that you have to make sure that all of the strands individually have a good connection at both ends.

Yes you're right that was my mistake! It's such a frustrating fact of minisplit installations. By all accounts then, it seems that solid wire should be superior for signals than stranded wire. However, when we had intermittent no com issues, we were told firmly by tech support that we ABSOLUTELY must use stranded wire. Now we didn't have problems with all of the solid wire installs but the ones that we did have problems with were ALWAYS fixed PERMANENTLY by redoing them with stranded wire. Ever since then I have tried and tried to come up with an explanation, but I just can't seem to get anything to stick. It does seem to me though that it may have something to do with the signal wires being so close to the high voltage ac lines. I really wish we could get a definitive answer on this!

Sent from my SM-G960U using Tapatalk

[R600a]

It's very weird but we have had com problems where installers used solid wire. The com problem is always permanently fixed when we replace the wire with stranded. One theory I had was that the terminal blocks they use are best suited for stranded wire. There's like 10x the contact surface area when you put a stranded wire in there vs a solid wire.

Another theory... a stranded wire is going to be carrying the signal externally but also internally because it will be riding on the "skin" of the internal strands. A solid conductor will carry the signal on its outer "skin". It's true that ethernet is solid but they also use twisted pairs and are not run along high voltage cables. Maybe in a minisplit situation where you have high voltage ac right in the same bundle as the signal wires it's beneficial to have stranded wire carrying the signal deep inside. Maybe that protects it more from the kind of capacitive ghost voltage that ac creates which plagues high impedence sensing circuits.



Sent from my SM-G960U using Tapatalk

That is interesting. My boss insists that at Fujitsu school they told him to only use solid wire. I think he probably remembers it wrong.
The current literature says solid or stranded.
I use solid but I remove the screw and loop the wire around it before reinstalling it. It takes an extra few minutes every install but I have never had a com problem.
The only problem we have had with a Fujitsu is a fan error in ice storms with extreme winds. I think the wind and ice are too much for that tiny motor so it doesn't spin the commanded rpm.

Sent from the Okie state usin Tapatalk

[JayMan7]

That is interesting. My boss insists that at Fujitsu school they told him to only use solid wire. I think he probably remembers it wrong.
The current literature says solid or stranded.
I use solid but I remove the screw and loop the wire around it before reinstalling it. It takes an extra few minutes every install but I have never had a com problem.
The only problem we have had with a Fujitsu is a fan error in ice storms with extreme winds. I think the wind and ice are too much for that tiny motor so it doesn't spin the commanded rpm.

Sent from the Okie state usin Tapatalk

The other variable here too, is the type of com used by the manufacturers. It could vary wildly. The designs I have seen seem totally stupid...but what do I know? It seems to me that they should have hired some electrical engineer from the automotive world to design the com system. Those guys have been doing that for decades and the way they do it just seems to make more sense.

Sent from my SM-G960U using Tapatalk

[R600a]

The other variable here too, is the type of com used by the manufacturers. It could vary wildly. The designs I have seen seem totally stupid...but what do I know? It seems to me that they should have hired some electrical engineer from the automotive world to design the com system. Those guys have been doing that for decades and the way they do it just seems to make more sense.

Sent from my SM-G960U using Tapatalk

Yeah and it would probably hold up better when I mount one on a buss.

Sent from the Okie state usin Tapatalk

[mks]

It's very weird but we have had com problems where installers used solid wire. The com problem is always permanently fixed when we replace the wire with stranded. One theory I had was that the terminal blocks they use are best suited for stranded wire.

Yeah, that's almost certainly related to the terminations. Solid wire can vibrate out of a slightly loose terminal, especially if it's a clamp type designed for stranded. Could also be contact area as you said, especially if there's some oxidation on the solid copper. I think ferrules + stranded is the best combo. Clean terminations and the flex of stranded.

[JayMan7]

Yeah, that's almost certainly related to the terminations. Solid wire can vibrate out of a slightly loose terminal, especially if it's a clamp type designed for stranded. Could also be contact area as you said, especially if there's some oxidation on the solid copper. I think ferrules + stranded is the best combo. Clean terminations and the flex of stranded.

I still really think that could be it! But what was weird is that I worked on one where I took the solid wire and sanded it and looped it around the terminals and then got the terminals wicked tight. It worked for like maybe several weeks and them threw another no com code. Now maybe it's corrosion or something but the terminals were all still tight and outlets which carry high amperage and have similar terminals will work for decades without having connection issues. So it's still got me wondering what exactly is going on.

Sent from my SM-G960U using Tapatalk

[RLJN]

Just a long shot on what might be different on solid and stranded wiring to mini inside heads.

The stranded wire is usually in a cord with a twist of the conductors.....might cancel out the " cross talk" of the data conductor and the power legs.

Solid wire would most likely be in raceway and the conductors end often end up parallel....side by side.
Possible inductive cross over??

I have always used stranded within a cord and no communication problems. (30 plus inside units)

[CHAINIK]

I'm no expert on this, but when it comes to communication wiring, what I've gathered from some controls training is, the reason that smaller gauge wire is spec'ed has less to do with its resistance characteristics but more to do with its capacitance. Larger diameter wires = more capacitive effect between them, interfering with the actual shape and characteristics of the data signal. I would bet solid wire would be even worse in this regard.

When I took some Carrier factory-provided Bacnet training for installers, they harped on this point about using the proper wire over and over. They give you some very specific characteristics regarding the comm wire you should be using, and tell you of all the grizzly results if you do not heed them.

[DeanKS]

Electrically it will not matter. Mechanically ...

Inside a blower compartment, single strand will not move and stranded may be moving and fatigue/break. I secure wires to eliminate loose rats nests and so not use wire nuts. Have seen wire nuts loose in blower box, exposed twisted connections.

Stranded wire under conventional screw terminals can easily not bed down on all stands and there may be loose strands, too much dependency on skill.

When stripping stranded wires, strands can be damaged and lost. Room for error.

When I use stranded wire, I twist and tin the stripped wires to have a solid wire at the terminal. And bundle with wire ties to eliminate movement. Yes time consuming.

[Poodle Head Mikey]

Can you detail these two out for me please?

PHM
--------

. . . the reason for using non-shielded wire on any wire carrying communications, which since this is a mini split it is likely it sends data on this wire,

"is that a lot of people don't know how to properly terminate the shield wire"

"which can cause more issues than no shielding."

[pecmsg]

Can you detail these two out for me please?

PHM
--------

I’ll go out on a limb. I’ve been taught shielded communication wire gets grounded or terminated on 1 end only, not both!

[thatguy]

Electrically it will not matter. Mechanically ...

Inside a blower compartment, single strand will not move and stranded may be moving and fatigue/break. I secure wires to eliminate loose rats nests and so not use wire nuts. Have seen wire nuts loose in blower box, exposed twisted connections.

Stranded wire under conventional screw terminals can easily not bed down on all stands and there may be loose strands, too much dependency on skill.

When stripping stranded wires, strands can be damaged and lost. Room for error.

When I use stranded wire, I twist and tin the stripped wires to have a solid wire at the terminal. And bundle with wire ties to eliminate movement. Yes time consuming.

You should not tin stranded wire when using them in lugs. The surface of the wire after being tinned is bumpy and has poor contact when fastened. It’s better to use a properly crimped tip or just twist it.

[Adlerberts-Protege]

I just took the training for the new RP17 side discharge Rheem Heat pump, basically a mini.

Install instructions state that all connections indoor and outdoor must be crimped fork connectors. Didnt specify stranded or solid but you must not loop the wire, ground or not. Hot, hot, comm, and ground must all be crimped fork connectors.

[Poodle Head Mikey]

Posted by BALloyd:

. . . the reason for using non-shielded wire (for communication) . . . is that a lot of people don't know how to properly terminate the shield wire . . . which can cause more issues than no shielding.

. . . shielded communication wire gets grounded or terminated on 1 end only, not both!

========

OK. But what does "properly terminated" mean exactly?
What is proper?
What is improper?
Both for the conductors - And for the shielding.

Also: what specific malady occurs when the shielded wire / shielding is not terminated properly?