Help me fix this blokes jerry rigged system?

softdown · January 2017

My trucks radiator hose "fell off" on the way to a real estate land closing. Eight miles from home, 17F temps, and with two dogs to get potentially run over by remote country traffic. A bit of a fix really. A good Samaritan gave us a lift back to my home. I owe him. He is also the "King of Cobbling" as you shall see.

Smart guy, knows electricity fairly well, on a really tight budget. He cobbled together a system that works but significant mistakes were certainly made:

Six 230 Ah golf cart batteries on 12 volt system. They were showing 7 volts/battery at 3pm. Outback FM60 charge controller had just kicked from Absorb to Bulk on a sunny day at 3pm. Batteries seem fine though cables are large aluminum cables. He wants to move to copper.

"8000 watt" Power Jack PSW inverter that replaced his "too small" 1500 watt Trace inverter. He is very happy with the inverter and believes their claim that power usage is very "economic".

Panels? This is where it gets interesting. Hopefully my memory is correct on these facts:
Started with four 150 watt 12 volt panels .... three years ago. Added three 100 watt 12 volt panels. Amazingly the voltages are about the same. He is disappointed in the gains realized from adding the three 100 watt panels.

Panels have 12AWG USE-2 wiring of course. This is where his economizing gets really interesting - the wires come together (no combiner box) into an ~50' run to the charge controller by being twisted together. He knows he should have used a large wire for the 50' run but was "unable"....or something. He used two 12AWG wires for the positive reasoning that is where the real power runs. He used one 12AWG wire for the negative.

We quickly agreed that significant gains could be realized by replacing the 12AWG with something much larger. Being indebted, I plan to replace the 50' run with either 6AWG or 4AWG.....after all it is a 12 volt system.

Oh yea...he just bought a new and massive 400 watt 24 volt panel for $100. Give him credit for finding bargains. Says the panel worked fine on the 12 volt system he cobbled together for a long road trip.

Right now I am suggesting that he consider two or three separate strings, on separate charge controllers, to take advantage of the harvest from three different sizes of panels.

It gets better....he is on unemployment. The county is not clearing the roads well enough to get to work with his two wheel drive car.

Hoping to avoid assessing his wisdom. Just wanting to help the chap out.

westbranch · January 2017

SD, can you post a drawing of all he has?

bill von novak · January 2017

A few notes:

1) Consider going to a much higher voltage array. The FM60 will go to 150 volts, so a ~100 volt system would greatly reduce the need for new wiring.

2) Smaller cheaper MPPT controllers like the Victron may help here if he wants to add all sorts of unmatched panels. Again each panel string should be as high voltage as possible.

softdown · January 2017

westbranch said:

SD, can you post a drawing of all he has?

Whoa....sorry but I can't fulfill that request.

The panels are angled at 37 degrees. The latitude is 37 degrees. He did several things right.

That "8000 watt" inverter though. And he seems to love it.

westbranch · January 2017

Bill is talking about the Array voltage to the CC

mcgivor · January 2017

8000 watt" Power Jack PSW inverter that replaced his "too small" 1500 watt Trace inverter. He is very happy with the inverter and believes their claim that power usage is very "economic".

Finally, someone who loves their Yugo...........I mean Power Jack. lol

softdown · January 2017

bill von novak said:

A few notes:

1) Consider going to a much higher voltage array. The FM60 will go to 150 volts, so a ~100 volt system would greatly reduce the need for new wiring.

2) Smaller cheaper MPPT controllers like the Victron may help here if he wants to add all sorts of unmatched panels. Again each panel string should be as high voltage as possible.

Thanks...yes....bringing up the voltage to a high level will surely help. Wish I could remember the present voltage coming into the CC.

He has 7 panels that produce about 20 volts each. 4 are 150 watt. 3 are 100 watt. Then we have the new 400 watt monster that puts out about 40 volts.

We frequently get down to -20F...so 100 volts would still put us below the CC safety ceiling....HyperVOC.

My question is how to "divvy up" the panels so we have an incoming voltage of ~100 volts.
We have seven 20 volt panels plus one 40 volt panels = 180 volts.

EDIT: Do the Victrons rely on built in internal temp sensors or externals or.............?

softdown · January 2017

This may work:
400 watt panel = 40 volts
4 x 150 watt panels into parallel and serial so = 40 volts
3 x 100 watt panels all paralleled so = 20 volts
___________________________________________________________
Total voltage = 100 volts

Why not?

Horsefly · January 2017

I'm not the expert here, but if you combine all those three in series (implied) then the currents (Imp) need to be close to matching, else you'll get the current of the lowest. Won't hurt anything, but you won't come near getting out of the panels what they will be producing.

Steve

softdown · January 2017

Horsefly said:

I'm not the expert here, but if you combine all those three in series (implied) then the currents (Imp) need to be close to matching, else you'll get the current of the lowest. Won't hurt anything, but you won't come near getting out of the panels what they will be producing.

Steve

Thats what I told him. Then I recommended separate strings for each size panels. Right the first time?

I didn't get the Imp's. The voltages on the 150s and 100s are very, very close. Yet with wattages being either 150 or 100...the match up looked quite dicey to me.

I would imagine that Imp is some type of blend between voltage and wattage...

BB. · January 2017

Pmp = Vmp * Imp

The equation is that simple (Vmp is Voltage maximum power. Imp is current maximum power).

Voc and Isc are voltage open circuit and current short circuit. To match panels, we need the Vmp and Imp (mostly).

We really need the Vmp and Imp of each panel (your numbers sound like the Voc voltage... For example, Voc~20 volts is usually Vmp~17.5 volts).

When you put panels in series, then the Imp of the panels should match within ~10% of each other.

When you put panels in parallel (or strings of panels in parallel), the you want to match the Vmp of the panels (parallel strings) to within ~10%.

If you have "odd" mix and match panels, it is not always easy (and cheap) to make a matched set.

-Bill

softdown · January 2017

BB. said:

Pmp = Vmp * Imp

The equation is that simple (Vmp is Voltage maximum power. Imp is current maximum power).

Voc and Isc are voltage open circuit and current short circuit. To match panels, we need the Vmp and Imp (mostly).

We really need the Vmp and Imp of each panel (your numbers sound like the Voc voltage... For example, Voc~20 volts is usually Vmp~17.5 volts).

When you put panels in series, then the Imp of the panels should match within ~10% of each other.

When you put panels in parallel (or strings of panels in parallel), the you want to match the Vmp of the panels (parallel strings) to within ~10%.

If you have "odd" mix and match panels, it is not always easy (and cheap) to make a matched set.

-Bill

You and your math!

I'm sure you are absolutely correct. Yet I have a suspicion that one may "make do" using simpler figures like Voc/Vmp and rated watts.

We have a lot of snow on the ground that refuses to melt. I have not been able to get my gate open for a few weeks. The snows gets crusty then almost icy and very hard to remove. Makes it harder to work on stuff on the ground. I also hope for a concrete plan before I return.

You saying that if I get the Vmp and Imp figures that we *may* be able to figure a workable solution? Looks pretty tough to me given the disparity of the panels:
4    150 watt at 20 volts
3    100 watt at 20 volts
1    400 watt at 40 volts

If we can do two strings instead of three, I would consider that workable. He doesn't want to do extra strings of course. To be expected from a guy who put an 8000 watt inverter on his system.

He is a real smart guy though. I think his brother welds some. Plus he got the inverter for $169! Some things are hard to resist.

westbranch · January 2017

SD... please keep in mind there is a distinct difference between, 'Well it WORKS' @ 10% and '" making it work to the max..." 99.9%, ... just pulling numbers here. Sort of like you started out from, incredulous that it did much more than turn on... now you are looking at the difference between those 2 positions

BB. · January 2017

It also depends on the charge controller type (MPPT vs PWM) and battery bank voltage.

You can parallel a Vmp~30 volt 200 watt panel with a Vmp~18 volt 100 Watt panel on a PWM controller... But the 200 watt panel will perform more like a 100 Watt panel (I.e., the Vmp~30 volts of the panel will be ~15 volts of the battery voltage--P=V*I -- you set the operating voltage by 1/2, then the power is 1/2 too).

-Bill

softdown · February 2017

It is an Outback FM60...we have that going for us at least. Plus he made pretty good decisions on the batteries. His first set were "Marine Deep Cycle"...they didn't do so well. He thinks he can de-sulfate them. Have we not all tried that?

If I needed to make millions, I might consider "battery rejuvenation".

Help me fix this blokes jerry rigged system?

Comments

Categories