diy solar

diy solar

System can't keep up after upgrades and reconfiguration

dixonge

New Member
Joined
Jun 18, 2021
Messages
190
For a long time our typical daily use was approximately 10% of our available 460ah. Panels + controller recharged easily.

Recently I upgraded our PWM controller to a Victron Smartsolar 100/30 mppt controller. A couple of days later I rewired my panels from parallel to series/parallel. Most days the voltage coming off the panels is 34-35v and around 20amps to the batteries. Theoretically this configuration is much more efficient. In practice, not so much.

Our usage is minimal. More importantly, our usage did not *change* after this upgrade! No microwaves or A/C. All lighting is LED. We typically run a mini-pc, two modems, one router. TV (brand new low-energy) usage is 1-2 hours/day. We charge devices (phones, watches, laptop, occasionally the tablet). Fridge runs on propane. Don't run the water heater.

The weather has been partly hazy, but we've had full sun almost every day. This morning the batteries are at 70%, -210Ah according to the battery monitor. They are losing the charge/draw battle, slowly.

I am tempted to rewire the solar panels back to parallel. I am also tempted to reroute the solar panels through the PWM controller.

Am I crazy in thinking that my recent 'upgrade' has, in fact, *down*-graded my system's ability to charge?
 
Series is usually always better unless you have partial shading problems on any of the panels.

Panels should have matching current output otherwise you will only get what lowest output panel current output is.
For parallel, the panels mppt voltage must be matching.

PWM controllers just switch panels on and off direct connection to battery so voltage on panels is clamped to battery voltage when charging. An mppt controller with parallel panels will pick best compromise between each panel's true mppt. If not matched in voltage this can be significantly non-optimum for any of the panels.

Based on you saying series connected voltage is 34-35v, your panels may not have high enough voltage in parallel configuration to run an mppt controller. Mppt controller needs 4-6 volts panel overhead voltage from battery voltage to do its mppt search. If not enough overhead voltage the mppt controller may just drop into PWM mode of operation.
 
4x100w panels - here are the specs:

SolarPanelSpecs.jpg

In parallel only the voltage was marginal for mppt - but in series/parallel the Vpm is 35.8v if I am doing my math correctly.

batteries are 4x6v in series/parallel - typical 12v system (FLA)
 
or maybe I should wire the panels in full series? We rarely park where we get any shade...
 
Looks like 36 series cell panels. At sun heated summer temps just barely enough for parallel operation on mppt controller.
or maybe I should wire the panels in full series? We rarely park where we get any shade...
Just watch out for controller max voltage of 100 vdc. It will be Voc at coldest temp x 4 series panels. For 36 cell panel that is about 0.8v x 36 cells x 4 panels = 115 vdc.

You might try to sort the two put in series so their net mppt voltage is closest matched for the other two series panels.
 
Last edited:
Are charge setting on the Victron controller set for flooded lead acid, typically 14.8 volts absorption?

In addition it could be that the battery monitor is giving misleading data.

I would use other criteria, like SG, to establish the batteries are fully charged.

I would look for other reasons in the system for lack of charge retention, assuming it real, and not the monitor giving false data.

Its very unlikely that the PWM controller will out perform the new controller. The Victron controller needs V batt +5 volts to start, and will continue running until the panel volts falls below V batt +1, thus 35 volts is more than enough for a 12v system and usable on a 24 volt battery. Note the start up will be based on the OC volts not the max power volts.

Its possible there is a battery issue, so it would be useful to investigate the battery health and actual capacity.

Mike
 
It will be Voc at coldest temp x 4 series panels. For 36 cell panel that is about 0.8v x 36 cells x 4 panels = 115 vdc.
See specs in previous post. 21.6v Voc per panel = 86.4 max, correct?
 
Are charge setting on the Victron controller set for flooded lead acid, typically 14.8 volts absorption?

Deka gc15 specs indicate 14.6 absorption, 13.9 float - that’s what I have in mppt settings

In addition it could be that the battery monitor is giving misleading data.

I would use other criteria, like SG, to establish the batteries are fully charged.

I would look for other reasons in the system for lack of charge retention, assuming it real, and not the monitor giving false data.

Its very unlikely that the PWM controller will out perform the new controller. The Victron controller needs V batt +5 volts to start, and will continue running until the panel volts falls below V batt +1, thus 35 volts is more than enough for a 12v system and usable on a 24 volt battery. Note the start up will be based on the OC volts not the max power volts.

Its possible there is a battery issue, so it would be useful to investigate the battery health and actual capacity.

Mike
All battery checks have shown all cells healthy. ?‍♂️
Only change was switch of panel wiring and controller
 
See specs in previous post. 21.6v Voc per panel = 86.4 max, correct?
Not when panel is cold. Look up spec for temp coefficient per deg C. Depends on where you live whether cold temp matters or not.

Mono cell are fairly consistant at -2.1 mV/degC per cell or about -0.34% per deg C. Poly cells vary from near zero to -0.34% per deg C.

Freezing can be about 8.5 % higher in voltage which takes 86.4v up to about 93.7 vdc. Probably okay but getting a little close to spec limit on controller. Also need to check the reference temp for the Voc spec. Some panels spec it at 20 degs C, some spec at 40 degs C.

If label reference of 45 degs C is temp of the Voc 21.6v then rise at freezing could be 99.6 vdc.
 
Last edited:
Perhaps a stupid question, but it sounds like you are still using a 12V battery bank. Is the charge controller configured for 12V and are you powering it up in the correct sequence?

I would also double check the rewired panel configuration for mistakes or bad connectors. I just had a thread where two of my panels were dead due to bad MC4 connectors. I had voltage but no current due to high resistance. A bad connector on a series string will kill the whole string.
 
Not when panel is cold. Look up spec for temp coefficient per deg C. Depends on where you live whether cold temp matters or not.
I live in my RV. 48-state limit, but avoid freezing temps.
Mono cell are fairly consistant at -2.1 mV/degC per cell or about -0.34% per deg C. Poly cells vary from near zero to -0.34% per deg C.
these are poly panels
Freezing can be about 8.5 % higher in voltage which takes 86.4v up to about 93.7 vdc. Probably okay but getting a little close to spec limit on controller. Also need to check the reference temp for the Voc spec. Some panels spec it at 20 degs C, some spec at 40 degs C.

If label reference of 45 degs C is temp of the Voc 21.6v then rise at freezing could be 99.6 vdc
Will try to find better specs, but these are unbranded panels from Windynation. Unsure how to better calculate cold temp voc
 
Perhaps a stupid question, but it sounds like you are still using a 12V battery bank. Is the charge controller configured for 12V and are you powering it up in the correct sequence?
this controller auto-senses voltage.
I would also double check the rewired panel configuration for mistakes or bad connectors. I just had a thread where two of my panels were dead due to bad MC4 connectors. I had voltage but no current due to high resistance. A bad connector on a series string will kill the whole string.
I tested all cables at all stages w/ ammeter.
 
Are charge setting on the Victron controller set for flooded lead acid, typically 14.8 volts absorption?

In addition it could be that the battery monitor is giving misleading data.

I would use other criteria, like SG, to establish the batteries are fully charged.

I would look for other reasons in the system for lack of charge retention, assuming it real, and not the monitor giving false data.

Its very unlikely that the PWM controller will out perform the new controller. The Victron controller needs V batt +5 volts to start, and will continue running until the panel volts falls below V batt +1, thus 35 volts is more than enough for a 12v system and usable on a 24 volt battery. Note the start up will be based on the OC volts not the max power volts.

Its possible there is a battery issue, so it would be useful to investigate the battery health and actual capacity.

Mike
I personally would not run absorb of 14.8v but depends on your typical usage. If doing that on a daily basis I would lower it to 14.3 vdc.

Float is compromise between what is good for positive plates to avoid too much plate grid corrosion (< 13.4 v) and what keeps negative plates fully charged (about 13.8vdc) to avoid sulfation. Above about 13.5v the bubbling increases which consumes more electrolyte water and bathes positive plates in oxygen which accelerates plate support grid corrosion which increases battery internal series resistance.

Floating at 13.25v with once a week full charge to 14.3v is a good profile for batteries mostly on float.

A true deep cycle lead-acid is lead-antimony alloy grid which is stronger and have thicker lead plates, but they have greater self discharge rate. Auto and mixed use marine batteries are lead-calcium alloy grid which have lower self discharge rate of 10-15% C per month. Most all lead-calcium battery have thinner plates and don't take to deep discharge or excessive overcharging. Obviously, a sealed 'maintanance free' cannot take the heavy outgassing of high absorb or equalize overcharging. They will just relieve the pressure from their vent port and lose unreplaceable water from electrolyte.

AGM's are pure lead plates supported by glass mat compression. They are 'electrolyte starved' meaning they cannot afford to lose water from available electrolyte due to overcharging. They usually die from electrolyte water depletion or sulfation due to lack of recharging.
 
Last edited:
I personally would not run absorb of 14.8v but depends on your typical usage. If doing that on a daily basis I would lower it to 14.3 vdc.
These are four 6v golf cart batteries. Sam’s sells under Duracell EGC15 label, mfg. is Deka GC15. They are discharged/charged daily. Would you still use 14.3v in this scenario?
 
These are four 6v golf cart batteries. Sam’s sells under Duracell EGC15 label, mfg. is Deka GC15. They are discharged/charged daily. Would you still use 14.3v in this scenario?
Yes I would. If you only get limited time to charge or concerned about generator gas consumption would I go above this.

If battery is suitable for equalization controlled overcharging you can do it every three to six months to keep electrolyte specific gravity equalized in all cells. 14.8v to 15.2v for about an hour or two. Don't do it too often as it is stressful on battery and eats up water.

My experience with golf cart batteries is they do not like high long period float voltage and die from high cell impedance due to positive grid corrosion.

Left on float for long period, a lead acid battery builds up a thick lead-oxide layer on positive plates. This increases their resistance. It is good to periodically burn this build up off with some, 5% C, discharge periodically.
 
Last edited:
If you only get limited time to charge or concerned about generator gas consumption would I go above this.
Time is limited only on cloudy days. There are no charging options outside of the solar panels.
 
Time is limited only on cloudy days. There are no charging options outside of the solar panels.
Those battery’s like more voltage then that , I would up your voltage to 14.6 and try absorb for 2hours .
I’m charge my 48 v system at 59 volts and 59 .8 in the winter plus temp compensation .
My battery’s 16 GC Duracell start to drop SG if they don’t get there voltage ? .
I’m in 3 seasons on these battery’s and they are going strong .
I shoot the battery’s with a heat gun and they don’t go over 78o .
I do drop 2 strings so I charge with 2700 watts so the battery’s get to full charge at around 300 so there is less heat .
You have to watch the SG as you charge to figure out when you reach full .
14.3 will leave you sitting in the dark ?
 
One other thing if you are not loosening water your voltage is to low .
I lose about 1/4” of water in in the cells in 2 months this time of year .
I eq 2 3 times a year . all cells stayup around 1.280 but one , in the winter i have one cell that stays low 1.267 +-
 
Those battery’s like more voltage then that , I would up your voltage to 14.6 and try absorb for 2hours .
I've had it on 14.6v for months. It has been many days since the controller dropped out of bulk/absorption State.
 
The spec sheet for my GC2s says to charge them in the absorption phase at 7.65V. That's 15.3V for the bank. The bank floats at 13.4V. I've been using this bank in my RV with those settings for 6 1/2 years with good results.
 

Attachments

  • GC2 charge parameters.jpg
    GC2 charge parameters.jpg
    221.3 KB · Views: 10
Back
Top