Tristar TS-45 PWM Battery Sense and Firefly Battery Issue

Cortttt · January 2019

Please note I am a beginner

I have a Tristar TS-45 PWM controller matched up with 250 watts of solar panels and one Firefly battery.

The battery has been superb for the past three years. I've thrown all sorts of stuff at it including accidentally lowering its voltage several times inadvertently to around 4 or 5 volts. Every time it's been able to recover - which is why I paid the extra bucks for it.

Now I'm getting really weird readings.

The Tri-star is giving me a battery sense alarm even though the polarity is correct and the leads are attached. (One of the leads had come unattached and I reattached it.)

The controller is stating that the battery is doing some amazing things: it keeps triggering a high voltage disconnect alarm - as it reports that the battery voltages regularly gets up to 16.1, 16.0 or even 17.0 volts, It's also reporting huge drops in voltage regularly - as low as 3 or 4 volts. This has been going for the last month or so.

Could a faulty battery sense be producing these weird issues? Or is the battery or is it both?

The battery is still usable and some days it performs within spec.

Morningstar had me test the voltage at a number of points and it seemed to be working fine but I have not been able to test when these weird numbers show up.

Both companies seem to feel that the other one is responsible at this point

Float is set at 13.3 I think and bulk at 14.35.

BB. · January 2019

Welcome to the forum Cortttt,

What is the gauge (diameter) of your controller to battery cable, and how long is the wire run? And what is the typical maximum current you see from the controller to battery bank?

Assuming your charge controller to battery connection is roughly correct (heavy/short connections to minimize voltage drop)--Disconnect the remote sense leads (and check that they are in good shape, not corroded at the ends, no insulation breaks).

And go back to the rest of your controller and battery wiring connections. Make sure they are clean and tight. If you pass current through wiring (i.e., sunny day with current flowing), then check the voltage at various points in your system. Ideally, you want the battery bank voltage to be within 0.05 to 0.10 volts of the reading on the charge controller terminals (for a 12 volt battery bank).

The Morningstar controller should work just fine without the sense leads (as I recall, there is no "switch" to select sense leads or not--Typically sense leads connect to the controller through a resistor and/or fuse--To prevent a short circuit in the sense wiring from starting a fire). With the sense leads disconnected, the controller should just read the Vbatt terminal voltages.

Ideally, I would like to see the battery at 12.xx-14.xx volts, and the controller reads 4.x volts--Then we know the battery is good and either the wiring or the controller is bad.

Did you have any indication that the battery bus voltage went below 10.5 volts? Radio or GPS needed memory reset, or you were on the boat when the Morningstar logged a "battery bus failure"?

One suggestion for any "computer" based solar charge controller... Disconnect the solar array. Do a factory reset. Disconnect the controller + lead from the battery bus for ~5 minutes. Reconnect the controller and let it boot. Then reprogram the controller for your voltage setpoints. Then reconnect the solar array. There are always a few reports of a computer based controller "losing its marbles", and a hard reset (to clear the memory and set to factory defaults) and reconfiguring--Sometimes gets everything working again.

Once the system seems to be working again, then reconnect the sense leads.

It "feels" to me like corroded/lose/failing wiring connections between the battery and the charge controller to me. Always disconnect the solar array first from the charge controller before disconnecting the battery bank. And reconnect the array last after the Vbatt connects are made and the controller is booted.

You may have battery issues, but until you prove this with a separate DMM (i.e., read less than 10.5 volts or over 15.0 volts), I would suspect problems elsewhere.

Note, usually a good idea not to discharge a lead acid battery below ~11.5 volts under average loads.

Your float and charging voltage looks OK.

I don't know anything about your Firefly/Carbon Foam batteries to know if they need to be "equalize charged" (typically 15.0 to 16.0 volts--depending) or not... But unless the solar controller is taking the battery above ~14.4 volts (or the generator is taking >14.4 volts), you should be reading ~12.7-13.x volts with the battery charged, or your float/absorb voltage setpoints as the battery is floating or charging (of course, loads and time of day/sun/genset will affect the actual voltages).

Check the battery resting voltage/state of charge (see your firefly battery manual for values) and make sure your battery is (on average) over ~75% state of charge. Avoid going below 50% SoC in normal operation, and never below 20% SoC. Going below 20% SoC can ruin your nice/expensive battery... Also storing a Lead Acid battery (in general) below 75% state of charge encourages sulfation and battery failure. (Foam/Carbon batteries may be a bit different--But a good place to start).

-Bill

stmoloud · January 2019

Maybe turn off auto equalization (if on)? That will be DIP switch # 8. But even so 17 volts is too high, unless the battery chemistry is a factor.
BB's suggestion to do a factory reset is a good start.

Cortttt · January 2019

Thanks so much.

The wiring is tight, the gauge of the wire is correct and the run is short. This system was put together for me about three years and has worked fine.

The voltage has been checked with solar on and off at the battery and at the controller battery connection and at the solar connection to the battery - that checked out fine.

I read in the manual that the battery sense leads are not needed.

Ideally, I would like to see the battery at 12.xx-14.xx volts, and the controller reads 4.x volts--Then we know the battery is good and either the wiring or the controller is bad.

My guess would be the controller. This morning I got up and the controller read 10.8V 30 seconds later it read 12.4v - this was well before the sun came up.

Did you have any indication that the battery bus voltage went below 10.5 volts? Radio or GPS needed memory reset, or you were on the boat when the Morningstar logged a "battery bus failure"?

What is battery bus failure? If that's an alarm it's not showing up on the controller. I forgot to mention that the system is actually on a van.

Disconnect the solar array. Do a factory reset. Disconnect the controller + lead from the battery bus for ~5 minutes. Reconnect the controller and let it boot. Then reprogram the controller for your voltage setpoints. Then reconnect the solar array. There are always a few reports of a computer based controller "losing its marbles", and a hard reset (to clear the memory and set to factory defaults) and reconfiguring--Sometimes gets everything working again.

Morningstar asked me last night to reset the system by turning off solar and then the connection from the battery to the controller. I did that and the battery sense alarm remains. The polarity is correct - positive is going to red and negative to black....It's a bit of mystery why that alarm is on...I'll see if the high/low voltage problem remains. I'll remove the battery sense leads and see how that goes.

It "feels" to me like corroded/lose/failing wiring connections between the battery and the charge controller to me. Always disconnect the solar array first from the charge controller before disconnecting the battery bank. And reconnect the array last after the Vbatt connects are made and the controller is booted.

I will check this again. I will say the battery sense lead wires are not easy to attach to the controller but they do seem to be tight.

Equalization is not suggested for these batteries.

Thanks for all the suggestions

BB. · January 2019

For lower voltage/high current power supplies, sense leads are actually quite nice (from an engineering point of view--Used them a lot in 5 volt and 3.3 volt power supply systems--Allows you to use much smaller/cheaper current carrying conductors and still have accurate voltage control at the point of use). But if something goes wrong, sense leads can add to the confusion.

Just to confirm wiring... Your negative and positive heavy/short cabling go from the battery bus directly to the charge controller (+/- leads). You are not using (for example) the chassis/sheetmetal of the van to carry the negative (-) current? Vehicle chassis can be a poor option for carrying heavy currents reliably (and chassis to frame grounds are difficult--usually the frame of the chassis is separated from the chassis by rubber sound and vibration isolation mounts--And there needs to be a braided cable to "jumper" the isolation mounts).

It sounds like you do not have a separate voltmeter to check your battery bank voltages?

Anyway, it is always possible that the charge controller has gone bad. Electronic failures after 5 years are not uncommon. And controllers that are 10+ years old are getting towards end of life/replacement time.

At this point, short of having a voltmeter to confirm that battery voltage is 12.xx volts and the charge controller is reading 10.xx (or whatever), you have probably done enough to justifying the replacement of the charge controller.

Good luck,
-Bill

Estragon · January 2019

Is the battery sense lead fused on the positive?

You mentioned one of the sense leads came loose (at the controller?). Any chance it might have shorted even briefly? If this is a possibility, there could be some damage somewhere in the wire which could cause an intermittent problem. As the sense wire would be pretty light (so cheap), I'd be tempted to just replace it.

Assuming heavy and short charging wire from controller to bank, there should be little voltage drop. If the controller works okay without the sense wires attached, and voltage at the battery end of the charge wire is within ~0.1v of the charge controller output terminals with significant charge current flowing, I'd consider not using the sense wires at all.

Being on a van, there may be more potential issues with ground reference, so it may be worth checking equipment grounds as well.

There is also a possibility the battery itself has an intermittent fault. As you say EQ isn't recommended, I'm assuming a sealed (AGM or Gel) type, with can be sensitive to overcharging. If troubleshooting wiring above doesn't find the issue, you might want to try swapping it out temporarily to see if a replacement behaves normally.

Just a few ideas. Problems like this can be frustrating and time consuming to chase down. Good hunting!

mcgivor · January 2019

Check to see if there is a problem with the voltage sense terminal to circuit board, there may be a dry joint, whilst connected gently move the terminal to see if this induces a fault. The solder points are very small so there is a possibility they may have been over torqued at some stage, removal of the wires altogether will eliminate that as a cause, but it really dose sound like a bad connection.

Cortttt · January 2019

Thanks everyone.

The sense wires are very light and I should replace them I agree.

I don't think I have a separate battery bus. The leads from the one battery go to fuse block which has negative connections on the bottom. There is a negative connection from the battery to the fuse block and a negative connection from the battery to the chassis.

I didn't mention it but those thing may have started out with the 50A fuse on battery terminal blowing about a month ago.

The Firefly is a carbon foam battery: Very expensive battery ($450?) but one which is supposed to be able to handle a lot of abuse - which it has gotten.

From their website - https://www.bruceschwab.com/advanced-energy-storage-systems/firefly-energys-oasis-group-31/

Firefly’s award-winning, patented technology is an innovative material science that removes almost all limitations of current lead-acid battery products. Firefly discovered that much of the lead in the traditional grid structure of conventional batteries can be replaced with a totally new type of grid material, carbon foam. The main improvement of the carbon foam construction is that it resists sulfation and corrosion while dramatically increasing the surface area of the battery plates.

Unparalleled Resistance to Sulfation – Sulfation is what usually kills AGM batteries. The Oasis carbon foam AGM can operate or be stored at a partial state of charge for long periods of time without a loss in capacity. Check out this article by Professional Boatbuilder Magazine to learn more about sulfation.

Depths of Discharge to 80%-100% of rated capacity without any loss of performance

Superior Life Cycle – capable of 3X the number of deep discharge cycles than that of other lead acid batteries

Strong Performance in Extreme Cold and Heat– performance range is -20° C to 50° C

Fast Bulk Charging and topping up is seldom required

Greater Usable Capacity– you can replace your existing bank with a smaller Oasis bank due to its deep discharge capability

Some surprising progress on my problem has been made: two things happened. I discovered some loose wiring was present - from the solar panel to the controller. (It actually came out). That was reattached. I don't know if it had anything to do with anything.

I also redid the system reset suggested but with the longer time frame suggested here - and it seems to have stuck (!). The controller is no longer reading battery sense alarm and everything seems to be in spec for now. I will know more by tomorrow and will update the thread.

I am getting a multimeter that can test the battery under charge.

BB. · January 2019

If you are getting a meter anyway... A very hand model is a DMM (digital multimeter) with a DC current clamp (actually AC/DC current clamp).

Makes it very easy to check current flow anywhere in your system. A couple of links to start looking:

https://www.amazon.com/gp/product/B019CY4FB4 (nice mid-range meter for ~$105)
https://www.amazon.com/gp/product/B07546L9RT ("it is good enough for our needs" meter for $40)

Note there are AC clamp meters too (only measure AC current). They are perfectly good meters, but do not measure DC current (they do measure AC and DC voltage--So don't get confused with AC+DC current clamp).

It is possible that the loose cable on the solar panel side caused some issues/confusion in the meter--Especially if the lead was shorting somewhere in the controller. I would not have expected these symptoms... But possible (confusing the computer/corrupting memory).

If it fixes the controller--Great!

-Bill

Estragon · January 2019

If replacing the sense wire, I'd add a small inline fuse on the positive near the battery to protect that circuit.

Is there an equipment ground connecting the Tristar chassis to the van chassis? Just a thought, but if not, I wonder if the pv+ was intermittently contacting the Tristar chassis, bringing it up to near pv+ potential voltage and confusing the sense circuit?

Cortttt · January 2019

Waahhh!

The alarm is back! After being off all day it came back overnight.

No high voltage disconnect but the logged data shows a low voltage of 3.4 (!)...

I will get new battery sense wires and thanks for the suggestion the multimeter.

The controller ground is going to the fuse block but thanks for bringing up that issue as I didn't know it was a possibility.

Estragon · January 2019

In the meantime, I'd disconnect the sense wires and make sure the controller is charging more or less properly without them.

BB. · January 2019

Run the system without the sense wires until the system is stable (or not...). It should work OK unless you have lots of voltage drop on your controller to battery cables (long cables, not "heavy enough" copper cables, etc.).

Sense wires are usually "twisted" to make it less likely that the leads act like an antenna and pick up stray electrical noise. If you have two sense leads that take different paths from controller to battery bank, it is possible to couple electrical interference into the leads and confuse the controller.

-Bill

Cortttt · January 2019

The latest news is good. With the sense wires gone the battery has been stable for the past couple of days - no highs / no lows - no alarms. After I took the sense wires off I noticed that one of them had been attached to a ring terminal that fit on post attached to the battery. I wonder if that was the problem? The wires were too small to fit onto that ring terminal..

I will probably fit new sense wires tomorrow and report back.

I never thought these little wires could cause so much havoc..

BB. · January 2019

Sense leads are just volt meter leads that do not carry much current. That way you can use "cheaper" cables (smaller AWG) with higher resistance and voltage drop, and the sense leads give you the remote voltage at the battery bank. If the battery bank voltage is too low, it can crank up the output voltage of your charge controller to account for the voltage drop of the heavy current leads.

Generally, I like remote sense lines for low(er) voltage systems. However, they can cause issues if not installed or maintained correctly.

If your voltage drop is not much on the main cables (less than 0.05 0.10 volts at max/rated current), then don't even worry about the sense leads.

Normally, there are a pair of resistors from the controller Vbatt terminals to the Sense Meter input... That way, if a sense lead breaks, you should not have a wild/uncontrolled output (meter reads Vbatt terminal voltages through resistor).

I do not know how MorningStar designed their systems, so I can only guess why are are seeing these issues.

Most solar charge controllers and installers/end users do not use sense leads (from what I have seen here). Not having sense leads is only an issue if you seeing >~0.2 or 0.4 volts drop (or more on a 12 volt battery system) between Vbatt on controller and Vbatt-bus connections.

-Bill

Cortttt · February 2019

Thanks.

The news is actually not so good I think. I'm not sure about the voltage ranges I should be seeing in this battery but over the past couple of days the highest voltage has been 12.8 and lowest around 11.8. Today was sunny all day - I'm in Arizona - and the highest it got was 12.5. I have 250 watts of solar and am running a fridge and powering up a laptop and cell phone on it.

mcgivor · February 2019

Can you provide details about the battery capacity and the refrigerator used. The 250W of PV strikes me as being extremely low to support a refrigerator, which makes me wonder if the bank is overdrawn and supply can't meet the demand.

Cortttt · February 2019

It's an Engel 84Qt refrigerator.

https://www.amazon.com/Engel-Portable-Voltage-Fridge-Freezer/dp/B003TM2K0Y/ref=sr_1_7?ie=UTF8&qid=1549069520&sr=8-7&keywords=engel+fridge+freezer

From Amazon

When power and space allow, consider an Engel MT-series fridge/freezer. Every Engel fridge/freezer is capable of keeping contents fresh - or frozen - wherever you roam. Freeze food to last you on long trips, or set the temperature control to keep fresh food and drinks chilled. All Engel fridge/freezers feature the Engel Swing Motor Compressor. Highly efficient, it typically draws around 1 to 2 Amps per hour (check the specs for each individual model), about 40% less than a traditional compressor

The Engel website says: Power Consumption: Variable from 0.7 - 3.6 Amps (12V DC)

It's been working for 3 years with no problem. It's one 12 volt Firefly battery.

BB. · February 2019

At least the numbers you are seeing on the charge controller seem to be accurate. As Mcgivor says, you are probably way under solar panel'ed for your loads.

Using a Kill-a-Watt meter or some other WattHour/AmpHour meter to measure your loads would be very helpful... But some first pass estimates.

Full size energy star fridge--Around 1,000 WH per day
Regular full size laptop--Around 300 WH per day (for example 30 Watts * 10 hours = 300 WH)
Smart Phone running 24 hours per day and internet hot spot ~ 20 WH per day
AC inverter running 24 hours per day (15 watts * 24 hours

~360 WH per day
=============================================================
Total WH per day ~ 1,680 WH per day (pure guesswork).

Say near Phoenix AZ, fixed array, tilted for winter:
http://www.solarelectricityhandbook.com/solar-irradiance.html

Phoenix AZ
Average Solar Insolation figures

Measured in kWh/m2/day onto a solar panel set at a 42° angle:
(Optimal winter settings)

Jan	Feb	Mar	Apr	May	Jun
5.28	5.74	6.34	6.33	5.87	5.58
Jul	Aug	Sep	Oct	Nov	Dec
5.26	5.22	5.78	5.93	5.57	5.14

At best, the minimum array you would need is:

1,680 WH per day load * 1/0.77 panel+controller derating * 1/5.28 hours of sun (Jan) = 413 Watt solar array minimum

And, nominally, I would be suggesting a larger array (to allow for cloudy weather, not using a genset, etc.).

As McGivor also asks, what is the battery bank details (volts/AH/type of battery), and brand/model of AC inverter (if you are using one).

-Bill

Cortttt · February 2019

I looked at the logged data - very low amp hours (5-10) logged every day! I checked the solar array - one of the connections was off. It's amazing the torture I put these systems through...Hopefully tomorrow will be better.

BB. · February 2019

First rule of debugging--Make sure "it" is plugged in.

-Bill

Tristar TS-45 PWM Battery Sense and Firefly Battery Issue

Comments

Phoenix AZ
Average Solar Insolation figures

Categories

Tristar TS-45 PWM Battery Sense and Firefly Battery Issue

Comments

Phoenix AZAverage Solar Insolation figures

Categories

Phoenix AZ
Average Solar Insolation figures