Xantrex C40 Charge Control Question

[(unknown)]

This weekend, my father and I installed a Xantrex C40 Controller to our solar system in Connecticut. The system in the house is quite small: 3 PV cells, 2 6V "Golf Cart" batteries (in series), used to run a small water pump, a stereo, and a couple of small lights in a cabin. We installed the C40 according to the user manual, and set it up in "Charge Control" mode for a 12V circuit. We left the EQ charge set to manual. We also installed a Battery Temp Sensor and a Remote Readout.

I adjusted the pots inside the C40 to set the Bulk and Float charges to 14.4 and 13.4 V respectively (according to the manual for our battery type).

What's odd is that the remote readout from the C40 shows no current running through the unit. The PV's show about 17 volts in direct sunlight open ciruit, but when connected to the controller they showed only 12.4 volts (the voltage of the batteries). I assume this meant that the C40 was dumping all available current into the batteries, effectively tying them to the same voltage, but we were suprised to find 00.0 amps on the readout. The LED status indicator shows only 1 blink, meaning that the battery voltage is "well below the float voltage". The batteries didn't seem to be charging too fast, if at all.

There are other things in the circuit -- fuses and circuit breakers -- but we didn't measure any appreciable voltage drop across them.

Any thoughts on what the problem could be? Do we just not have enough juice from the PV's? I don't have much model info on them, but they're not terribly large (3 cells, 1' x 3' each).

Thanks.

[Nando]

BEST STEP:

Send this message to Xantrex they will respond with the proper steps, but remove theun-necessary info like stereo etc.

Nando

[iFred]

it sounds like the batts are fully charged. Read the manual that came with it. A minimum voltage of 21vdc need to come from the panels into the charge controller.

 ohhh.. and never set manual mode for gel cells! READ THE BOOK.

[(unknown)]

I read the manual cover to cover several times.

The batteries are not gel cells, but rather liquid electrolyte deep-cycle. We left the EQ charge jumper set to manual precisely because we didn't think we'd need/want it for our batteries. I don't think that's factoring in here, anyway.

If the batts are fully charged, wouldn't the charge controller LED blink 5 times, as said in the manual, rather than just once? Float voltage is set to 13.4, again as per the manual, and the batt voltage read 12.4 volts, which to me says it should be charging and current flowing.

The thought about the minimum 21VDC into the charge controller is interesting. I just went back to the manual and couldn't find this, could you cite a page number or other source?

Thanks for the comments so far.

[halfcrazy]

the light will blink more times in a row up to five then when they are full it will stay on it sounds to me like the c40 has failed in a open state if there is no power passing thru and you have verified all the fuses and breakers are passing power thru have you checked right at the c40 for power on pv side and battery side? if you have voltage on both and the pv side is high compared to battery side it would sound like it is faied in an open state

[windstuffnow]

  If the batteries are fully charged it should be a solid green light and it will drop from 14.4 volts to 13.4 and stay there until there is a reduction in voltage ( i.e. a load pulling the bat voltage down ).   Once it reaches a lower voltage it should increase amps to maintain a trickle charge to the batteries and take care of the load.

  What size are your solar panels?  Check the voltage at the batteries to make sure its reading the same as the read out.   You can also pull the pos line and use an amp meter between the controller and battery to make sure your actually getting some amps to the batteries.   At only 12.4 volts the batteries will take a good batch of amps to get it up to the 14.4.   If your not getting any amps into the battery bypass the controller and connect the panels directly to the battery and check the amp input.   If you get a good reading ( rated amps or there about ) then you may have a problem with the controller.  

Hope that helps...

.

[Ungrounded Lightning Rod]

What's odd is that the remote readout from the C40 shows no current running through the unit. The PV's show about 17 volts in direct sunlight open ciruit, but when connected to the controller they showed only 12.4 volts (the voltage of the batteries). I assume this meant that the C40 was dumping all available current into the batteries, effectively tying them to the same voltage, but we were suprised to find 00.0 amps on the readout. The LED status indicator shows only 1 blink, meaning that the battery voltage is "well below the float voltage". The batteries didn't seem to be charging too fast, if at all.

That sounds to me like:

 - The batteries are low

 - The controller is indicating they are low.

 - The controller has the panels hooked to the batteries to charge them (which is why the panel voltage is dragged down.)

 - The charging current detection or readout in the C40 is defective.

I'd put an external ammeter between the controller output and the battery to measure the actual charging current (if any).

[Ungrounded Lightning Rod]

I'd put an external ammeter between the controller output and the battery to measure the actual charging current (if any).

(As a diagnostic rather than an operating mode.)

[Tom in NH]

I think Ungrounded L.R.'s comments are well grounded. What's your short-circuit current  of the unconnected solar panels? Compare that with the reading you get from an ammeter you insert between the output of the controller and the batteries. Disconnect the load and see if the batteries will charge. Disconnect the temperature probe just to simplify the circuit until you're sure the c40, panels, and batteries are working properly. You didn't say if the controller is being used as a dump load controller, or the shunt mode. Is this a new unit or a used unit? If it's used, you might want to check for bent jumper pins. If they short out, you can fry a trace on the back side of the board (repairable by soldering a jumper wire). --tom

[Flux]

"What's odd is that the remote readout from the C40 shows no current running through the unit. The PV's show about 17 volts in direct sunlight open ciruit, but when connected to the controller they showed only 12.4 volts (the voltage of the batteries"

I suspect you have the answer here. If the volts when connected to the controller are dropping from 17 to 12.4 the controller is doing its job.

You need an independent ammeter to see what is really happening.

Firstly use it to check the short circuit current of the panels and that will likely show a connection problem or trouble with the panels. 17v oc is rather low anyway and you may be measuring leakage from a high resistance.

If the s/c current is ok then connect the panels directly to the battery with the ammeter in circuit, I suspect you will see the same current when you do the same thing via the controller. Don't blame the controller at this stage, I think it is likely to be working fine.

Flux

[iFred]

All I can tell you is that it seems to be correct and is doing exactly what mine has been doing since I purchased it. I have it wired as charge control mode. I have 2-1120ah 12vdc lead acid batts in parallel; this is hooked to a 1 kw solar array and produces 22vdc going into the controller. on power up it blinks 5 times. after full charge and delivering 15 amps to the batts in 4-6 hours the batts are fully charged, and goes solid green for a short time and trickle charges around 2-4amps, at night I get a single blink. normal, it's waiting for a charge.

There are two potentiometers on the board for low and high voltage disconnect, if these aren't set properly then the batts do not charge the batt according to bat specs, here's the kicker, each batt is different and because of that you need to find out exactly what the correct settings should be for those batts, you may be undercharging or equalizing them without knowing it, this can also offset the controller. The newer controllers are beginning to remember charging states and cycles. pushing the equalize button should only be done on a sunny day and for an approximate duration of 4 hours and no more for most standard lead acid batts and is done once every 3 to 4 weeks. (hence the automatic mode and the reason for remembering states), the temp sensor can be off by ~.50mv , meaning it's almost at the point of useless, what is supposed to do and what it does is two different things, it is supposed to measure the temperature of the batts and determine an appropriate offset voltage accordingly. (which can also offset you blinks)

The manuals are wrong when it comes to the batts! hate to say it, don't depend on them, bulk is around ~13.8-14.8 depending on the batt, do internet check on your batt type, all types are listed with all the info. float charge is the battery disconected and not connected to any loads for a period no less then 12 hours, read the voltage, this is your float voltage. charge voltage depends on the battery again and can vary from 13.8 to 14.8. equalize voltage is above 15v.

Most problems I have found (More then one solar install) has been atributed to the wrong data on the batterys and wrong settings on the controller, sulfated batterys or batterys that have been sitting around. battery stats are critical.

What is the manufacture of the battery? the type and any ID numbers, then I might be able to help you.

Hope this helps.

[(unknown)]

Wow, thanks for all these great comments.

We did try an ammeter in series (replacing the fuse between the PV array and the controller). In direct sunlight it measured about 180mA, which seems low as it should have been bulk charging. Also, if the charge current read on the remote readout was correct, shouldn't it have at least read something like "00.1"? It was good at least to see that current was flowing SOMEWHERE.

I didn't get the chance to test the SC current from the PV array. That's a good idea and one to try next.

The unit is new, and this was our first installation. I wouldn't blame the unit yet, either. It could be an issue with the battery settings as well. I hope that Ungrounded LR is right (that everything's workign but the current readout).

At the moment, I'm a 3 hour drive from the house where it's set up, so I might not get to test any of this until a week or two from now.

I'll get the manufacturer data for the PV's and the batteries from my dad and post them. Thanks for offering to help, iFred.

Thank you all very much for your comments and suggestsions. More soon, I hope.

[SamoaPower]

I have to disagree with your definition of float voltage (I don't know what float 'charge' means). The float part of the charge cycle algoithm is to allow the charge to be maintained with minimal water loss while compensating for self-discharge. What you defined is the no-load resting voltage and will normally be between 12.6 and 12.8 volts for 100% charge with the surface charge removed.

The float voltage will typically be between 13.1 to 13.8 depending on battery type and temperature. I run my T-105 golf cart batteries at a float of 13.2 at 80F.

[iFred]

ummm, Modbus specifications list it as...

http://www.morningstarcorp.com 11/27 8 September 2003

Regulation voltage @ 25ºC.

The battery will charge at 100% charge current until battery voltage reaches this setpoint. The controller will begin to taper input current so that this setpoint is maintained, but not exceeded

EV_float [57346][0xE001](V). Float voltage @ 25ºC After some period of time in regulation when the battery is fully charged, the battery will drop down to this lower setpoint to reduce gassing.

Et_float [57347][0xE002] time before entering float Defines the length of time in regulation before dropping down to the float stage

Equalize voltage @ 25ºC The voltage setpoint to which the battery will be equalized. Periodic equalization equalizes cell voltages, bubbles the electrolyte, and helps prevent sulfation of the battery.

[craig110]

Even though the C40's current amperage display has a ".X" digit, it doesn't use that digit.  (I wish the manual said this, as it really confused me when I first started using the C40.  I've concluded that it doesn't use the .X digit as I've never seen a non-zero in that location, even when the sun is emerging from clouds and the displayed amperage is going up.)  Therefore, if your panel is putting .18A into the C40, the C40 will show 0.0A.  Since I've never tested my panels with an ammeter, I don't know if the C40 rounds to the nearest whole amp (i.e. .6A in = 1.0A displayed) or if it truncates (i.e. .99A in = 0.0A displayed).  Either way, though, .18A will show as 0.0A.

Something to consider when pondering why the batteries are showing low voltage when the C40 is hooked up is the power taken by the controller.  The base C40 takes 15mA (page 70 in the manual) and I can't easily find what the optional display panel takes, but it is probably about the same.  Adding those together, your C40 is using nearly 20% of your maximum solar power output (!) and, most likely, is connected 24 hours a day.  You might have the case where the C40 is using more power per day than the cells are producing, and this deficit would be made up by the C40 taking power from the batteries.  The net result would be to eventually have depleted batteries.  One way of testing this is to borrow a panel from someone (or, perhaps, hook the output of a couple of amp 18-20 volt DC adapter to the C40) to boost the input side to well above what the C40 will consume.  Then, if the batteries still aren't getting charged, you know there is a true configuration problem somewhere and not just a case of having too tiny of input.

Craig

 

[iFred]

Craig, on my c-60, it does charge in fractional units of less then 1 vdc.

These charge controllers take very little power. If the voltage/current is higher (unbalanced system) then the controller can handle then the mosfets trun off (overcurrent protection), they really don't disapate much power, the displays are lcd and 15ma drain would be nothing to even a 30ah golf battery. just my oppinion.

[(unknown)]

Some additional info about our setup:

By calling the battery supplier, the best I can tell is that the batteries are 225 AH hour 6 volt golf cart batteries made by Crown for Atlantic, model # GC, but the web site for Crown batteries does not show it exactly.

The PV array is three 1' x 3' amorphic panels made by Chronar Tri Solar.

[craig110]

Oh yes, the voltage of the batteries shows to the tenth of a volt.  I was talking specifically about the current (i.e. present) amperage.  Have you ever seen yours with a non-zero tenth of a current amp?  (It is odd that three out of the four numbers displayed use the tenth digit, but the present amperage figure doesn't.)  And agreed, the C-40 doesn't normally take up much power, but on a percentage basis it takes up a lot of power when the charging source is only giving it 180mA.  The author of this string, Litchfield, only has a total of 9 square inches of solar cells connected to the C-40 and batteries and is wondering why the batteries aren't getting charged.

[iFred]

I don't have the display you have connected to the unit. I use voltmeters and shunts to determine amperage. as far as the amps are concerned through my shunt I see a linear upward or downward slope on the meters, these are digital voltmeters that are set to mv scale across the shunt and read just about anything, I have seen it go to the lowest point of around  .70mv or .700ma and 1vdc. At this point it drops to zero.  

[richhagen]

I have to double check, as I have a C40 in load control mode here at the house, in that mode its resolution is only about a half an amp.  I have one in charge control mode at another location, so I can't check its resolution right now.  I'm not sure what type of panels you have, but if you have 1 by 3 foot amourphous panels, you may want to look at that, start by testing the short circuit current.  if it is low, check the connections and the panels individually.  I generally do not like the amourphous panels that I have had, they are 1 foot by 3 foot, they reach charging voltage sooner than my mono and poly crystaline panels, but they don't really output much once they get there.  They also have a bad habit of having the backing with the silicone and aluminum (I think) layers start peeling off after a while.  I only have one of these types of glass panels mounted right now.  I have some damaged ones I took down just sitting.  I had off brand and ICP panels of this type.  I have some Unisolar amourphous panels as well, I haven't checked the current from them in a while, but aside from a little yellowing of the plastic they seem to be holding up OK physically.  Rich

[craig110]

Seeing the nice linear changes with the shunt makes sense.  Perhaps someday I'll measure mine the same way and determine whether the display panel is truncating or rounding the amps.

[(unknown)]

I have a C60 charge controller operated in the charge control mode and the display is installed in the unit.

I have had a problem on occasion when the controller "would not Charge" this sometimes occured after using a small delco aternator to provide backup power.

The pulsating DC input from the alt. (remember there is no filtering here) seems to be the problem. Disconnecting both the DC input and output to the battery bank for

several minutes and then reconnecting has always solved the ptoblem for me.

In my system I have the float and bulk voltages set to 13.8 and 14.4 ( i use sealed AGM batteries) I would think that your float voltage is set too low, try 13.8,

It does seem strange that your AMPS display shows zero.

The WATTS display should show something if juice is going into your batteries.

When the controller is limiting the amount of current going in to the batt. bank you should be able to hear a "whine or singing" noise from the contoller as it rapidly

turns on & off.

The only other thing I can think of to ask is where are you measuring the solar

panel output? at the panels or at the controller.

Good luck

Wizard   fframbob@hotmail.com