4mm Solar Power cable review and testing

[AussiePhil]

Many of you may already use this cable for connections from your power supplies to fused distribution or even for power injection but may still be interested in the tests.

For others a little up front ramble.
In the world of 5v pixels voltage drop has always been a serious business, the 12v realm has an easier time of it but they are still not immune to the effects with things like flickering happening.
The closer you can have your pixels to the power supply or the thicker the power wire is the better it is, however DC power cable in Oz was stupidly expensive for larger sizes over 1mm sq (18awg) or hard to buy.
With the explosion of solar panels over the years both 4mm sq and 6mm sq DC twin core cable is used extensively in the industry. doesn't mean it's got easier to buy locally or sellers want stupid prices for it.
Not being able to just walk into Bunnings to buy it generally pushes it off peoples list, however it can be found now from various sites, it's still going to cost you close to $2/M or cheaper if you can find deals.
The cable I got is TV1-F 4mm Twin core with red and black marking and UV rated for outdoor use. It is stranded tin coated copper.
So from the 100M drum we sliced off a metre of cable to check voltage drop against the current reading.

Some numbers:
Square mm - 4
Length tested - 1 metre
Manufacture spec for resistance - 10.18m Ohm for the loop (2M equivalent) that's milliOhm
Calculated resistance including spade connectors in circuit - 13.78 mOhm

For a current draw of 5A there was a voltage drop of 70mV / metre
That's 0.7v for 10 metres - 30ft for non metric people

It might not sound great over longer runs but it all helps and is hugely better than most commonly used cables, heck it's even better than the 6mm speaker cable i have that isn't all pure copper.

To interpret the graph, voltage drop per metre is on the left hand side and the current is the bottom axis. These are all real measurements with the same setup.

Edit 3rd Feb: the chart has been updated to include 6mm twin solar cable for comparison


Bonus material:
The cable has a 55A in free air rating and considering it is used normally in 100's of volts systems the voltage drop per metre at 55A would be a concern but may be ok.
at 55A that the measured drop was 0.8v / metre
at 60A the drop was 0.9v / metre

For power supply to a fuse block good practice would be to do dual runs from the PSU.

Now at 60A the cable did heat up, here is a thermal image after it had been under constant load for many minutes.

Note the cable was showing 73c at this point.

Even more impressive though is the MW SE-600, 100A 5v supply suppling the power, it was showing no signs of even a warm spot

Here is the video for the testing

View: https://www.youtube.com/watch?v=KPulNehdblw


Make sure to watch the video all the way as you get to see what happens when you accidently short out the cable after we had tested.

Cheers
Phil

 

[TerryK]

Referring to the chart, there seems to be a change in slope for some of the wire types. Voltage drop being a linear variable (E=IR), unless it's an optical illusion of sorts, does your data indicate a change in slope? If so, the reason for the change in slope? Curiosity only, approximately how many data points did you capture for each wire type?

 

[AussiePhil]

Referring to the chart, there seems to be a change in slope for some of the wire types.

yes it does, likely more obvious with the visually greater vertical axis, the solar cable is bumpy as i only took the readings to two decimal places, the others have a slight upward curve.

Voltage drop being a linear variable (E=IR), unless it's an optical illusion of sorts, does your data indicate a change in slope?

Yes it does,

If so, the reason for the change in slope?

Whilst ohms law holds true what changes is that R increases a little as temperature goes up, temp goes up due to the current flow. Hence the smaller the wire the greater the heating and change in resistance.

Curiosity only, approximately how many data points did you capture for each wire type?

Just because it does take time I just capture every one amp, anything above 30A (not shown here) I will jump 5A at a time.

Cheers
Phil

 

[orchidman33]

With all cable it pays to read the data sheet. The manufactuerer/supplier gives a resistance per meter or kilometer at a given tempurature.
Here is a cable range that we sell where I work

 

[AussiePhil]

With all cable it pays to read the data sheet. The manufactuerer/supplier gives a resistance per meter or kilometer at a given tempurature.
Here is a cable range that we sell where I work

I did read the datasheet to get the manufacturers number of 5.09 per km which is single wire non loop, double that for an actual circuit then add in any connectors loss (insertion loss).
The four spade connectors added around 3.6mOhm based on measurements so 0.9mOhms per spade connector. That may not be a factor over 5m+ in real life but i wanted to show why the measured voltage drops are higher than just using the published Ohms/km number.

where the connector losses will start to add up is the fuse blocks and connectors used in the yard, especially pigtails

Cheers and thanks for the datasheet.
Phil

 

[TerryK]

I expected the cause to be self heating. Must admit though I thought the curves would be more elliptical.

A couple oddities though. 1st, the 'waves' in the 32 x 20. Rounding errors in your data perhaps? And 2nd, there seems to be a definite 'knee' at 10 Amp in the 24 x 20.
And last, while I'm not sure how useful it may be, I wouldn't mind seeing the 18 and 20 AWG sizes 'supposedly' used in the pixel strings. For those that use it, perhaps the Enet twisted pair cable.

 

[AussiePhil]

Hi Terry, thanks for questioning and getting me to explain further.

I expected the cause to be self heating. Must admit though I thought the curves would be more elliptical.

A couple oddities though. 1st, the 'waves' in the 32 x 20. Rounding errors in your data perhaps?

Both rounding errors and possible human error entering numbers in the sheet, planning to re-run that one

And 2nd, there seems to be a definite 'knee' at 10 Amp in the 24 x 20.

Due to scale of the vertical axis in 100mV grad it to may be a rounding error

And last, while I'm not sure how useful it may be, I wouldn't mind seeing the 18 and 20 AWG sizes 'supposedly' used in the pixel strings. For those that use it, perhaps the Enet twisted pair cable.

These are on the list, i know i have "18" wire but can't find it at the moment i used the "20" in the short circuit videos.
it will be useful to do for people that use lengths before they get to the start of the string from controllers. same comment on the Catx cable.
When 18/20 is used in the strings then the losses are different as the current reduces as you move away from the power connection point.
This is a subject for a nearly ready to do video.

Thanks
Phil