Mixed AWG Solar Install

[belizesailor]

Whew...test install of panels and controller all wired up (just strung out on the boat to test). Wire lengths and gauges about the same. Will let it run tomorrow and post numbers before finishing the physical install.

[Cadence]

Originally Posted by belizesailor

Currently upgrading solar to about 25 amps. The run from the equipment arch thru the transom is about 10' and currently wired with 8 AWG and will be a real PITA to rewire with larger gauge. I show 8 AWG over 10' to still be within 3% loss for 25 amps. So, I'm considering leaving the 8 AWG in the arch and installing 4 AWG from just inside the transom to the controller (about 25').

Reasonable idea or not?

I think you have the right idea. Cut the exiting wire back as much as you can and go with it. Every one seems to be an expert making a major project out of it.

[belizesailor]

Pushing up to 17 charging amps by 10AM COT under partly cloudy skys and partial shading today.

Batteries are up from overnight low of about 12.5 to 13.1 V. At this rate I'm gonna have to bleed off some AHs to get a good full on bulk charge test with full sun in the afternoon.

[belizesailor]

Now around mid day with good, but not perfectly perpendicular, sun angle and minimal shading, still light partly cloudy so a bit variable. Charge output has been consistenly over 20 A. Max charge output has been 28.3 A (very close to output limit of 30A).

Partly cloudy conditions make it a challenge to scurry around and get good readings. Really need data loggers at each segment to do this right, but below are approximate readings taken at about 23 A charge output.

15.77 at panel block
15.43 pig tail end (~2.2% loss)
15.40 controller (~1% loss from pig tail, ~2.3% loss total).

Based on the total loss it appears that losses in a mixed AWG install are indeed additive. (Remember? Mixed AWG was actually the subject of this thread...apparently many never noticed that)

Loses would be a bit higher at max panel output, but in reality thats not going to happen often, and would still be at low levels (maybe 4% max?).

Certainly passes the sniff test in my book and easily passes the practical test of meeting (well exceeding) our power usage.

That was interesting....now I have to finish up the damn install. :-(

[Jim Cate]

Great!

Some empirical data that s hows that all the dire warnings were perhaps overstated. The bottom line is that yes, there are some small losses that could be made even smaller by using different cabling, but that as is the performance is satisfactory.

Who woulda guessed?

Good one, mate!

Jim

[Cadence]

Originally Posted by belizesailor

Now around mid day with good, but not perfectly perpendicular, sun angle and minimal shading, still light partly cloudy so a bit variable. Charge output has been consistenly over 20 A. Max charge output has been 28.3 A (very close to output limit of 30A).

Partly cloudy conditions make it a challenge to scurry around and get good readings. Really need data loggers at each segment to do this right, but below are approximate readings taken at about 23 A charge output.

15.77 at panel block
15.43 pig tail end (~2.2% loss)
15.40 controller (~1% loss from pig tail, ~2.3% loss total).

Based on the total loss it appears that losses in a mixed AWG install are indeed additive. (Remember? Mixed AWG was actually the subject of this thread...apparently many never noticed that)

Loses would be a bit higher at max panel output, but in reality thats not going to happen often, and would still be at low levels (maybe 4% max?).

Certainly passes the sniff test in my book and easily passes the practical test of meeting (well exceeding) our power usage.

That was interesting....now I have to finish up the damn install. :-(

[belizesailor]

Originally Posted by belizesailor

...

15.77 at panel block
15.43 pig tail end (~2.2% loss)
15.40 controller (~1% loss from pig tail, ~2.3% loss total).

Based on the total loss it appears that losses in a mixed AWG install are indeed additive. (Remember? Mixed AWG was actually the subject of this thread...apparently many never noticed that)

...

Oops...that should read "not" additive. If so, it would be ~3.2%

[socaldmax]

Originally Posted by belizesailor

Now around mid day with good, but not perfectly perpendicular, sun angle and minimal shading, still light partly cloudy so a bit variable. Charge output has been consistenly over 20 A. Max charge output has been 28.3 A (very close to output limit of 30A).

Partly cloudy conditions make it a challenge to scurry around and get good readings. Really need data loggers at each segment to do this right, but below are approximate readings taken at about 23 A charge output.

15.77 at panel block
15.43 pig tail end (~2.2% loss)
15.40 controller (~1% loss from pig tail, ~2.3% loss total).

Based on the total loss it appears that losses in a mixed AWG install are indeed additive. (Remember? Mixed AWG was actually the subject of this thread...apparently many never noticed that)

Loses would be a bit higher at max panel output, but in reality thats not going to happen often, and would still be at low levels (maybe 4% max?).

Certainly passes the sniff test in my book and easily passes the practical test of meeting (well exceeding) our power usage.

That was interesting....now I have to finish up the damn install. :-(

Hmmm...

those voltages seem awful low. Not from cable losses, just low compared to the 17.6v that the panels were posted to produce at max. power. Considering how low that input voltage is compared to my set charging voltage (15.5v) and losses inherent in any MPPT controller, I'd prefer to rewire the system in a parallel/series configuration to bring the input V up to 30.8v for the controller and make cable losses even less of an issue.

Even with a claimed Voc of 21.xV, if you double it to 40.xV you still aren't close to max. input V of 50V.

[belizesailor]

Originally Posted by socaldmax

Hmmm...

those voltages seem awful low. Not from cable losses, just low compared to the 17.6v that the panels were posted to produce at max. power. Considering how low that input voltage is compared to my set charging voltage (15.5v) and losses inherent in any MPPT controller, I'd prefer to rewire the system in a parallel/series configuration to bring the input V up to 30.8v for the controller and make cable losses even less of an issue.

Even with a claimed Voc of 21.xV, if you double it to 40.xV you still aren't close to max. input V of 50V.

A good, but not perfect solar day, hot panels, less than perfect alignment...I assume those factors explain the 15.77 at the panel.

For now the configuration works for me. Plenty of juice and my venerable old gels cells only charge at a max of 14.2 anyway. I've gotten an impressive 10 years out of them, still very much serviceable, but I expect in a few more seasons they will be ready for retirement. I will likely go with another battery technology then and need higher charging volts...so probably swap out controllers then and go series rather than a hybird wiring scheme to accomodate the existing controller.

For now, I'm tired of working on the f'ing boat, ready to get the hell off the dock, and not about to wait for another controller to be shipped in. It works...time to go sailing!

[colemj]

If you are heading East, looks like a pretty decent window coming up next week.

Mark

[belizesailor]

Originally Posted by colemj

If you are heading East, looks like a pretty decent window coming up next week.

Mark

Hope to be East bound within w 2 weeks. Taking a break in the City for a few days, finish up, then head out.

[svlamorocha]

Originally Posted by belizesailor

Oops...that should read "not" additive. If so, it would be ~3.2%

Kirchhoff's voltage law (KVL)[edit]

Voltage drop is additive, that is embedded in the definition of voltage/potential.. "Kirchhoff's second law", "Kirchhoff's loop rule" or "Kirchhoff's second rule".

[HoppySailor]

P


Sent from my iPhone using Cruisers Sailing Forum

[barnakiel]

Note:

> with 20 Volt at source,
> introduce 5 Volt drop from wire (= 25%),
> is 15 Volt at junction point,
> introduce 1.5 Volt on terminals (10%),
> is 13.5 Volt at next junction point,

Now:

PLS note 20 - (5+1.5) = 13.5 (loss 6.5 Volt or 32.5%) (!!! NOT 35%)

BUT 20 x (25+10 Sic! WRONG)% = 13 (Sic! WRONG)

Or something like that.

Point: Voltage drops are additive. BUT one must use proper math rules when calculating cumulative PERCENTAGE changes.

Just talking.

b.

[Cadence]

I wish some would stop trying to impress everyone with their knowledge, on all of the treads, related to electricity. It is not black magic. Be helpful but not smug. If someone asks for help they are not looking for formulas. Maybe asking what V drop they are getting would be more helpful. Just a thought.