Alright, 2 years into this, and now I have a mystery.

Himins

I have 3 strings of 2 panels. with only 2 of the strings attached to the CC, the voltage and amperage reads perfect, but when I connect the 3rd string...the entire voltage drops right in half. Coming from the panels are a total of 12 feet of cable to the CC. At the panel, I get a reading of 68 volts, which adds up perfectly, but when I measure at my first joint only 6' away, the reading drops to 58v. So I soldered up 2 brand new 6' cables, checked continuity, on all cables....even the cables that I removed....and every cable/ connection is cleaned and greased...I am lost here...how can I lose this voltage, and worse yet, when I add this string, why would the entire output drop in half?
Thanks to everyone here...this system has performed perfectly...until just this month, November...the only cloudy month of the year and my batteries are screaming for relief.

NorthGuy

Re: Alright, 2 years into this, and now I have a mystery.

Did you try to measure open circuit voltage and short circuit current on this string when it is not connected?

Himins

Re: Alright, 2 years into this, and now I have a mystery.

Correction, I measured open circuit voltage, and at the panels I got the exact reading I anticipated, 6 feet away, the voltage dropped. I do not know how to measure the voltage short circuit as you say.

NorthGuy

Re: Alright, 2 years into this, and now I have a mystery.

Himins wrote: »

I do not know how.

You just disconnect the string from the controller and then measure.

Voltage is simple. You simply measure voltage between + and -.

Current is more difficult. You need a multmeter capable of measuring currents. Then you also connect one lead to + and the other to -. When connecting/disconnecting, you need to cover panels with something, or you may get sparks and damage leads/connectors.

Or, if you have a DC clamp meter, you simply connect + of the string to -, and then clamp around one cable.

Cariboocoot

Re: Alright, 2 years into this, and now I have a mystery.

You might want to check the polarity on those strings as well.

Measure Voc with the strings connected to nothing but the meter.
Measuring the Isc can be more difficult if it is potentially greater than 10 Amps as that is the typical limit of a DMM. if it isn't, you set the meter to DC Amps and have it be the connection between (+) and (-) under full sun. The current of a string will be equal to the Imp or Isc of one panel (current does not add up in series connections, only in parallel ones).

westbranch

Re: Alright, 2 years into this, and now I have a mystery.

what kind of combiner do you have where the 3 come together? Depending on that it may be easier to test polarity.

Himins

Re: Alright, 2 years into this, and now I have a mystery.

there is no combiner on the system. All panels are hardwired using #6 stranded copper battery cable. I connected just the suspect string to the CC alone, and the string provided the exact same out put as the other 2 strings when they were connected to the cc...one string at a time...all 3 strings performed nearly identical...About the polarity...this has never changed, the system has been stable, if not perfect until this issue came about. I did review the polarity one more time today just to be on the safe side, no issues according to my digital meter. I do NOT have an Amp meter.

The string that is troubling me is pointed slightly different than the other 2 strings, but in today's light, were producing a little more current than the panels pointed due south...

so...here i am...with 3 strings, all producing equal out put when connected independently, when I connect 2 strings...the output doubles...then I add this third string, the entire voltage plummets to one half output cutting the output of 4 panels in half.

BB.

Re: Alright, 2 years into this, and now I have a mystery.

What is the voltage at the connection point (+ and -)? And what is the voltage of the 3rd string you are attaching (before attachment).

If the panels are wired correctly--Adding the string should either add current (and possibly voltage) to the connection point--Or at least not draw it down... Properly wired panels when "dark" look like reverse biased diodes... You will see a little leakage current backwards through the panels (much less than 1 amp), but nothing to tank the common connection point voltage.

I really sounds like the either one or both 3rd string panels are wire "backwards" and you are connecting the wrong polarity panels to your existing two string array.

When you measure the open circuit voltage of the 3rd string, pay attention to if it is +Voc or -Voc... That is the only thing that make sense right now.

Solar panels when connected backwards to a battery bank will look nearly like a dead short and fry the panel with current from the battery bank.

In your case, the solar array has limited Isc current available, and, at this point, probably has not toasted your 3rd string (although, it is possible the string is toasted, or you might have some shorted bypass diodes causing this problem--But you should see poor Voc-3rd-string voltage at that point).

Fingers crossed.
-Bill

Cariboocoot

Re: Alright, 2 years into this, and now I have a mystery.

A couple of things to take note of: Voc is not the same as Vmp and does not necessarily indicate a 'healthy' string.

If string 'A' has a Voc of 'X' and so do strings 'B' and 'C' then when in parallel Voc must be 'X'.
If it isn't, that means one of the strings ('C' in this case) has something wrong with it so that when under load it becomes more of a conductor than a producer. Reverse polarity is one of the things that can cause this, so can a shorted panel or bypass diode(s).

Since you can't measure current (most DMM wil do up to 10 Amps but not beyond) which is the best way of checking the string, check the Voc for change when all strings are connected but not connected to the charge controller (no load).

I am assuming these strings are all made up of identical panels. If not there are other issues that can arise with respect to reverse breakdown Voltage.

NorthGuy

Re: Alright, 2 years into this, and now I have a mystery.

Himins wrote: »

I connected just the suspect string to the CC alone, and the string provided the exact same out put as the other 2 strings when they were connected to the cc...one string at a time...all 3 strings performed nearly identical...

Does your controller tell you input voltage and current? If so, perhaps you can post them here, like:

String 1 only: pv voltage, pv current, battery voltage, battert current
String 2 only: --
String 3 only: --
String 1+2: --
String 1+3: --
String 2_3: --
All three: --

The more information the better.

Also, are your batteries empty enough? If they're charged well, the combined production of 3 strings may be simply too much, so that the controller dials the production down. Could that be the case?

Himins

Re: Alright, 2 years into this, and now I have a mystery.

BB, this all sounds quite possible, but unfortunatly, I'm so far out of my league here, that reading this has only caused me more questions. First, not knowing the terms, or how to use these terms, I am unable to even answer the questions you pose...Then, what would cause a panel to suddenly reverse power. I only have one means to test polarity...if the polarity is now reversed...then it happened by some form of osmosis....Each string on this cloudy day, produced 1.2 amps, then the CC would convert these amps to just about 4.2...So, I'm just lost. The cc likes each string...wouldn't the cc have issues if one of the polarity's were suddenly reversed. Yep...Im lost. I'm not familiar with the terms, diagnosis, testing procedure, or even if I have the right equipment...to test it. I'm confident you folks can put your finger right on the problem...but not so confident I understand what to do with your suggestions.
s

Himins

Re: Alright, 2 years into this, and now I have a mystery.

That test I can do pretty easily Northguy...I will have to test this tomorrow as it is now dark here. The battery is so dead, that I've been using a generator with automotive style batt chargers to bring it up every day...some times 2 tanks per day...but the battery has been drained as far down as 11.2... It very rarely cycles below 12.4...so I'm in full panic mode here.

vtmaps

Re: Alright, 2 years into this, and now I have a mystery.

Himins wrote: »

The battery is so dead, that I've been using a generator with automotive style batt chargers to bring it up every day...some times 2 tanks per day...but the battery has been drained as far down as 11.2... It very rarely cycles below 12.4.

Most of this thread has been about your three strings of panels, and I wish I had something to add on that issue...
but you may have another problem going on that may be even more serious: your batteries.

If your signature is correct you have five 12 volt batteries in parallel. That is a recipe for early death of a battery bank. The problem is that the charging current does not divide equally among the 5 parallel batteries. This means some batteries are getting undercharged and others overcharged.

This causes the batteries to diverge further in their electrochemical characteristics, which causes further imbalance in the charging which causes further divergence in their electrochemical characteristics which causes.... The whole process can accelerate quickly and ruin a battery bank. If you have a DC clamp ammeter you may find that when you stop charging, some batteries are discharging into the other batteries.

What you need to do is separate those batteries and charge them and use them one at a time. You can be using one while charging a different one. Hopefully you will find that some of the batteries have decent life in them.

--vtMaps

westbranch

Re: Alright, 2 years into this, and now I have a mystery.

Himins, there is another issue that has come up, the fact that you do not have any circuit protection in 3 the strings coming together. as was mentioned there may have been damage done to the 3 rd string by the other 2, that is the reason we recommend a CB or fuse in each PV + lead when combining, no matter how they are combined. I may have missed it but do you have individual circuit protection?

Cariboocoot

Re: Alright, 2 years into this, and now I have a mystery.

Along those lines if you don't get the batteries charged up fully and properly they are going to die completely.

No matter what else, get them charged.

Himins

Re: Alright, 2 years into this, and now I have a mystery.

VTmaps, I have been unable to change my profile for some reason...I am using 6@2v Telcom batteries, Liberty 900 I believe made by CDS. My target voltage reading at the CC under the very small load used here has been 12.4 in the morning...It never rests above 12.8 without load, and with 6 panels, this has been reasonably simple...November here is historically glum, and this year was no exception..but unfortunately even my back up charger has been unable to bring the battery up to much over 12v this is my fault however because the string may have been bad about 3 weeks ago, when the sky's turned dark...runing on one string for nine days of straight clouds in a row..allowed the battery to drop significantly lower than I knew....The battery does look like it should be equalized, but the overall battery does seem fairly uniform across all units.

Himins

Re: Alright, 2 years into this, and now I have a mystery.

No I do not have individual circuit protection...If the panel/s were damaged however, would there be some repair possibly...I suppose it depends on the type of damage happened,

What baffles me is..that string has excellent out put, when the string is connect alone...so, what kind of damage would have occurred?

Cariboocoot

Re: Alright, 2 years into this, and now I have a mystery.

The "excellent output" you are seeing on its own is Voltage only. You'd be amazed at how damaged a solar panel can be and still produce Voltage. But that is not the same as current nor is it the same as not drawing current when connected in parallel with a good panel.

If a panel is shaded it can supply Voc, but when greater Voltage is put to it it is supposed to go through the diode (bypass the cells). If that diode is a dead short current can flow either way. Thus a 'low Voltage' panel acts as a load rather than a power source. If the cells themselves are shorted you should see a reduced Voc because they are no longer putting out full or possibly any Voltage.

But without running all the nit-picking fiddly tests (which involve disconnecting everything and testing each component and then sequential increases in the circuit complexity as more 'good' components are added) you will be doing little better than guessing, and so will we.

Himins

Re: Alright, 2 years into this, and now I have a mystery.

Thanks Cariboo, you have been instrumental in making this thing work magic so far...the batteries are a little better today, and are being charged with the generator tonight...I removed all but the tinyest load...IE, removed one refrigerator and box fan, leaving only the Inverter, and some low voltage lights on during the evening....The batteries seem to be responding fairly well tonight so far...I will continue the testing tomorrow via the chart suggested earlier...and will post results if something doesn't add up...thanks again every one...till more tests are done...have a great evening.

westbranch

Re: Alright, 2 years into this, and now I have a mystery.

Caution Pilgrim, those appear to be AGM or VRLA batteries.
From what what I know about VRLA batteries, you do NOT want to EQ them.

I searched the C&D Liberty batteries and there is a 1000 series and they are 12 v , and look like a regular battery.
There are 2v MSE series and they are shown here.... Does this look like them? http://www.cdtechno.com/pdf/lit/12_1078_0412.pdf

Float voltage should be ~ 13.6V 2.27V/cell

mike95490

Re: Alright, 2 years into this, and now I have a mystery.

Himins wrote: »

VTmaps, I have been unable to change my profile for some reason...I am using 6@2v Telcom batteries, Liberty 900 I believe made by CDS. ......

Most Telecom batteries are not rated for deep cycle usage. They are sacrificial, high current batteries, to keep the system going till the backup generator gets up to speed. After they have been cycled a couple times, they are changed out for fresh ones. So you may not get all you are hoping for from them.

Himins

Re: Alright, 2 years into this, and now I have a mystery.

The battery shown in your pdf file is similar...these are AGM, 900 AH 2V...From the product sheet given me at the time of acquisition, these were rated for 1200 cycles..I am not sure how deep the cycle...and the information was a tad sketchy. they were purchased and used for 10 years before I acquired these....I paid well under scrap prices for them...they have been working very well indeed so far...but I've had the system matched up fairly well with a pretty small load...1700 ah per day. the CC would go into absorb mode by 10"30 am frequently..and the CC would commonly show 12.4-5 in the am using mostly just lighting, refirg, and 12v water pump during the evening hours. Of course the inverter was also on all these hours. the sheet did confirm your float figures at 13.6.
I've been wondering if these batteries would suffer damage using standard automotive chargers...or worse yet...did they in some way damage this pair of panels.

Himins

Re: Alright, 2 years into this, and now I have a mystery.

North guy,
String 3 was sorely lacking in voltage when connected to the CC under cloudy conditions today. Both string 1 and 2 were identical at 62.8 v output with the inverter off...string 3 produced 62.8 at the back of the panel...but 6 feet away from the panel, only 58 v output and at the cc it read 54...but not enough amps to rise above 0 on the cc output reading...they are dangling now unconnected to the system.
this bizzare voltage drop that occurs between the back of the string, to the cc still has me puzzled... I built and replace all 4 stranded copper #6 wire from the panel to the cc...so I've ruled that out so far.

westbranch

Re: Alright, 2 years into this, and now I have a mystery.

That voltage drop has to be in the wire OR it is because of the lack of amperage.
The Elec. Engineers will have an explanation I am sure...

What are the distances of #6 wire you have? ( panel to CC)

Himins

Re: Alright, 2 years into this, and now I have a mystery.

West branch:
there are 12 feet #6 stranded copper wire for each string.

westbranch

Re: Alright, 2 years into this, and now I have a mystery.

A quick test would be to try one of the other PV sets on the wire you use with the offending set. (or the offending set on know good wires.)
That will tell you if it is the set or the wire.
My bet is it is the PVs that are bad...

ggunn

Re: Alright, 2 years into this, and now I have a mystery.

Himins wrote: »

I have 3 strings of 2 panels. with only 2 of the strings attached to the CC, the voltage and amperage reads perfect, but when I connect the 3rd string...the entire voltage drops right in half. Coming from the panels are a total of 12 feet of cable to the CC. At the panel, I get a reading of 68 volts, which adds up perfectly, but when I measure at my first joint only 6' away, the reading drops to 58v. So I soldered up 2 brand new 6' cables, checked continuity, on all cables....even the cables that I removed....and every cable/ connection is cleaned and greased...I am lost here...how can I lose this voltage, and worse yet, when I add this string, why would the entire output drop in half?
Thanks to everyone here...this system has performed perfectly...until just this month, November...the only cloudy month of the year and my batteries are screaming for relief.

Are you certain that you didn't get the polarity reversed on your newly made up cables?

Himins

Re: Alright, 2 years into this, and now I have a mystery.

ggunn wrote: »

Are you certain that you didn't get the polarity reversed on your newly made up cables?

I am confident the wiring is the same now, as when I first installed the panels..to be sure, I removed one at a time..but this begs the question...what would happen....If I intentionally crossed the panel polarity. What I mean is...if some amature like myself, were to accidently install a string, and connect it to the cc not knowing the polarity was reversed....then threw the switch..would this cause damage...or just a need to correct the polarity...I'm reluctant to just cross the polarity when my dmm reads +voc when measuring the voltage..

BB.

Re: Alright, 2 years into this, and now I have a mystery.

If you have two solar panel strings and hook one up backwards--In theory, nothing bad should happen. Solar panels will just pass the rated current (Imp/Isc).

If you have two or more parallel strings, or are connecting panels directly to a battery bank and connect another string "backwards". It will pretty much destroy the new string being attached.

Solar panels are just massive arrays of diodes. And if you pass too much forward current or hit them with too much "reverse current", like any standard diode, they will be damaged/ruined.

-Bill

bill

Re: Alright, 2 years into this, and now I have a mystery.

Hopefully the designers of the cc took the precaution of placing a diode on one side of the input.