Forum Discussion

jrnymn7's avatar
jrnymn7
Explorer
Jun 05, 2015

My First PWM Solar Run

Well, I guess you could say I lost my solar charging virginity... finally. :)

Two 140w 12v Poly panels, in parallel, manual tracking, Isc 8.15a (x2), Voc 22.4, 50' (one way) of 6awg cable to Solar30 pwm (set to 14.4Vabs), 4' of 2awg to four 6v GC2 (430Ah).

Bank at 12.53 ocv, 81% soc, -82Ah's.

Conditions were cloud free, except for the odd small, very thin passer by, and about 70-75F ambient, throughout test.

1:10pm:

Isc of 9a each, 19.79 Voc, no idea of panel temps.

I made up adapters to go from panels' mc4's to mini welding connectors, so all V readings begin just after those adapters. So not sure how much loss thru the adapters, themselves. Flipped the switch... calculated about 7% V loss, adapters to cc, at about 18a... 13.47v input, 13.29v output, so 98.5% efficiency thru controller. Cabinet temp was about 80F using a wall thermometer.

2:15pm:

16a (controller), 16.8a (Trimetric), passed thru @ 14.31v adapters, 13.51v input, 13.33v output, and 13.25v at bank...

- adapters to input; 94.4% eff.

- thru controller; 98.6% eff.

- controller to bank; 99.4% eff.

- overall, adapters to bank; 92.5% eff.

2:45pm: 16.4a (controller), 17.2a (Trimetric), Vbatt ?

3:45pm: 16a (controller), Vbatt 13.4v

4:30pm: 16.1a ("), Vbatt 13.53v, 90% soc, -43Ah's.

* 5:05pm: 15.2a ("), Vbatt 13.54v, 60 degree tilt, pointed...

... switched to flat position, laying east to west, amps dropped to 9a @ 13.44v, so a 41% loss compared to tracked.

5:40pm: (tracked) 15.2a @ 13.87v

6:35pm: 13.2a @13.93v, 95% soc, -22Ah's.

7:05pm: (light clouds), 75 degree tilt, pointed, 0a @ 13.35v.

Harvest;

+80Ah's per Solar30, +63Ah's per Trimetric, set to 85% eff.

Observations/Conclusions;

- Need +2v at adapters above ocv to do 16(+)a bulk, and a minimum of +1v at adapters to overcome Vbatt + R at finishing amps.

- Need to do more tracked vs. flat comparisons, throughout next charge. And I found even the smallest of adjustments could make a measurable difference.

- Need to figure out a way to know loss thru adapters?

- Shading only one cell brought the entire array's amps down considerably. Shading an entire row, either way, had the same result as shading a single cell.

- The occasional Isc check showed Isc in excess of 10a, at times, so these panels can really put out, even at mild ambient temps.

13 Replies

  • jrnymn7's avatar
    jrnymn7
    Explorer
    Jun 08, 2015
    grizzzman wrote:
    jrnymn7 wrote:
    Originally, the plan was to go portable, so that's how the adaptors came to be. And they've come in handy doing the test runs, but now that I'm going rooftop, there really is no need for them. I'll just keep them around for further testing, and such.


    Are you planning on tilting them ?


    Yes, tilt and twirl.

    I was originally going to make a frame to hold both panels as one unit, but will likely go with two set-ups of one panel each, each with its own tilt'n'twirl. I have to allow for them not shadowing each other, so I've worked out how far apart they need to be, and I have the room. This way they'll only be sticking up about 30", instead of 60", when tilted high.
  • grizzzman's avatar
    grizzzman
    Explorer
    Jun 07, 2015
    jrnymn7 wrote:
    Originally, the plan was to go portable, so that's how the adaptors came to be. And they've come in handy doing the test runs, but now that I'm going rooftop, there really is no need for them. I'll just keep them around for further testing, and such.


    Are you planning on tilting them ?
  • jrnymn7's avatar
    jrnymn7
    Explorer
    Jun 07, 2015
    Originally, the plan was to go portable, so that's how the adaptors came to be. And they've come in handy doing the test runs, but now that I'm going rooftop, there really is no need for them. I'll just keep them around for further testing, and such.
  • grizzzman's avatar
    grizzzman
    Explorer
    Jun 07, 2015
    brulaz wrote:
    Perhaps measure Voc at the end of cable at NOCT or STD conditions and compare to published values? Crude but will give some idea?
    Ed0
    Would not pierce the sheathing to measure V.

    A $5 tool exists to open MC4 connectors.

    I use MC4 y connectors and have wondered about V drop as well.


    I dissagree but understand why you would not pierce the sheathing. If you dont you will not be checking the crimp part of the connections. Repair the with a dab of silicone and electrical tape ( To protect the repair while it drys) Will take care of water issues.
  • jrnymn7's avatar
    jrnymn7
    Explorer
    Jun 07, 2015
    I've eliminated some of the connections, so hopefully v.d. will be decreased significantly. I'll know next time I do a pwm run. I'm hopeful things will work out well, in the end. I really like the combiner box idea, as that will keep things good and dry up there, and eliminate a lot of potential trouble spots. If I do end up with some mc4 connections, I'll definitely grab one of those mc4 tools, thanks!
  • brulaz's avatar
    brulaz
    Explorer
    Jun 07, 2015
    Perhaps measure Voc at the end of cable at NOCT or STD conditions and compare to published values? Crude but will give some idea?

    Would not pierce the sheathing to measure V.

    A $5 tool exists to open MC4 connectors.

    I use MC4 y connectors and have wondered about V drop as well.
  • westend's avatar
    westend
    Explorer
    Jun 05, 2015
    jrnymn7 wrote:
    West,

    So your rooftop connections are all inside the combiner box, and attached directly to the bus bars, with terminal ring ends? So, no mc4 connectors used?

    I can see going that route, as it would still be easy enough to switch things out, from series to parallel, if need be. And I really don't like trying to disconnect those mc4's, anyway. If it wasn't for needing a good finger nail trimming, I probably wouldn't have been able to push those tiny tabs in the other day.

    I used one of those audio CB's, like snowman recommended. I have a 50a on the pos cable to the bank, and it raised voltage drop by 0.08v at 45a, but by nothing measureable at finishing amps. So I'm guessing the 30a on the pos solar cable is likely causing about half that at 20a solar.

    Re; EQ'ing... yes the amps are there, but not the voltage, it seems. BFL and Niner (or Landyacht? I forget) saw a difference of .3-.5v over and above the solar30's setting of 15v, but mine appears to be very close, when confirmed with the dmm. Besides, I prefer to EQ individual 6's with my cc/cv variable adjustable power supply, so it's not really a concern.

    I kept about 6' of MC4 connected to the module's standard box. The MC4 is terminated at the combiner where 4 ga. cable leads down about 6' to the controller. A direct connection to the module would be even better if you don't need or like the MC4 connectors.
    Understood on the EQ. There are a variety of ways to get it done.
  • jrnymn7's avatar
    jrnymn7
    Explorer
    Jun 05, 2015
    grizzzman wrote:
    Kewl Jrnymn7. It appears that it's more important to tilt the further north you are ??
    Perhaps a voltage drop test is in order on your adapters ? Easy enough to do.


    I'm not sure how to test them, as there is nowhere to put the probe north of the adapter. I guess I could just poke thru the 10ga wire's insul?

    Yes, tilting made a significant difference. Even at high noon there was a worthwhile increase in panel power, at a 15 degree tilt. And just turning the panels a couple degrees occasionally made what I would consider a worthwhile difference. And of course, that's all scalable, so the gains on a much larger array could be a game changer.
  • jrnymn7's avatar
    jrnymn7
    Explorer
    Jun 05, 2015
    West,

    So your rooftop connections are all inside the combiner box, and attached directly to the bus bars, with terminal ring ends? So, no mc4 connectors used?

    I can see going that route, as it would still be easy enough to switch things out, from series to parallel, if need be. And I really don't like trying to disconnect those mc4's, anyway. If it wasn't for needing a good finger nail trimming, I probably wouldn't have been able to push those tiny tabs in the other day.

    I used one of those audio CB's, like snowman recommended. I have a 50a on the pos cable to the bank, and it raised voltage drop by 0.08v at 45a, but by nothing measureable at finishing amps. So I'm guessing the 30a on the pos solar cable is likely causing about half that at 20a solar.

    Re; EQ'ing... yes the amps are there, but not the voltage, it seems. BFL and Niner (or Landyacht? I forget) saw a difference of .3-.5v over and above the solar30's setting of 15v, but mine appears to be very close, when confirmed with the dmm. Besides, I prefer to EQ individual 6's with my cc/cv variable adjustable power supply, so it's not really a concern.
  • westend's avatar
    westend
    Explorer
    Jun 05, 2015
    jrnymn7 wrote:
    "Yes, you will need to remove the connector bottle necks, the adaptors"

    West,

    That is yet to be seen. The big losses were thru the cables themselves, and the 30a CB near the cc input. I will be reducing the cabling down from 50' to about 30', once installed, so that should help considerably. And I need some way of connecting to the 6awg, so I don't see any alternative?
    Yeah, shorter will be better. 7% is really a high loss. I went with a DIY combiner box, made from an EMT LB and some buss bars. I ran 4ga. welding cable from that to the controller. My disconnect is a CB manual reset breaker. Truthfully, I didn't measure for voltage drop since I have such short runs and big wire. I had in mind that 235W was not going to be my endpoint. I see a lot of guys that are adding or plan to add in the future. I tried to keep that in mind.

    One solution to getting the higher voltage (I'm assuming this will be for equalization) may be to use a buck/boost device. You probably have the amps to get by with that.

About Technical Issues

Having RV issues? Connect with others who have been in your shoes.24,351 PostsLatest Activity: Jan 22, 2026