Solar controller not playing well with RV converter

System

Had I read the manual for the TriStar 60 more carefully, I would have known that my setup wouldn't work:

Magnetek RV converter DC out wired in parallel with TriStar 60 to a pair of 12 v (parallel)batteries. After a few minutes of the Magnetek plugged into 110vac, I get error LED's on the TriStar indicating overheating (alternating blinking g/y LED's). I hope I have not damaged the TriStar, the RV was plugged in overnight with this (obviously incorrect) wiring arrangement the first time, and I didn't notice the TriStar fault until the next morning. I reset the TriStar after unplugging the RV converter and it seems to operate, but who knows what damage might have been done to it?

I've noticed the status lights on the TriStar seem to not behave the way the book explains, and the EQ operation does not seem to operate correctly. I'll have to place a call to Morningstar monday to get further explaination.

The way I look at it, my options are:
1) disconnect the RV controller and not use it at all. I've never plugged in to shore power before (other than to charge batteries at home, which the Magnetek does not do a good job at anyway).

2) install some kind of 3 position 2 pole switch to select either solar (TriStar) output to the batteries, Magnetek output to the batteries, or everything disconnected from the batteries.

What kind of switch could I use for this purpose?

I put the TriStar in battery EQ mode (just to test the function), thought I cancelled the operation by following the manual directions, and left for a few hours with soler connnected. EQ mode obviously didn't get undone, as I returned a few hours later to find the batteries bubbling vigorously and my gas detector alarm on. Battery liquid level was down, but no plates showing , so I topped up with distilled water.

Now I'm afraid I've damaged not only the TriStar 60 by running it parallel with the RV controller, but also the brand new Trojan batteries by overcooking them. I really should have learned much more about this before risking damaging expensive components.

How can I determine how much damage might have been done to the batteries, or the TriStar?

I feel really stupid about this...

John F
LV, NV

crewzer

Re: Solar controller not playing well with RV converter

John,

It’s generally OK to connect multiple chargers to a common battery bank, although it’s possible that the Magnetek’s relatively low charge voltage may be causing a problem for the TriStar. You may be right about disconnecting the Magnetek when you’re charging from your solar energy system, but I’d double check the connections from the Magnetek to the batteries to verify correct polarity. RV wiring to batteries is often black for (+) and white for (-).

Disconnecting the Magnetek charger will require a DC rated switch. Boat battery switches are a popular solution.

See: http://store.solar-electric.com/basw1300amp.html
And: http://www.bluesea.com/dept.asp?d_id=7492&l1=7459&l2=7492

We’ll see if any of your equipment has been damaged. The TriStar includes a wide range of self protection features (reverse polarity, short circuit, overcurrent, high temperature, overvoltage, etc.), and flooded-cell batteries can take a bit of abuse. For whatever reason, the TriStar’s manual EQ function did not respond to your termination action (push button for five seconds), but, assuming you’ve selected battery type “6” charging program, the EQ function should have automatically terminated after 3 hours. Checking the electrolyte levels and adding some distilled water was a good move.

Here’s a link to the TriStar LED indicator summary: http://www.morningstarcorp.com/products/TriStar/info/TS_LED.pdf

I think there’s some opportunity for double-checking and learning more about your system before contacting Morningstar. You just may want to leave the Magnetek disconnected for a while, and, while you’re learning about your systems’ operation, I’d recommend keeping a close eye on it to make sure it’s behaving correctly.

HTH,
Jim / crewzer

System2

Re: Solar controller not playing well with RV converter

I've confirmed that all wiring is correct (unfortunately). The Magnetek uses a 10 guage red wire for battery pos and a 10 guage white for battery neg.

I'm confused by the TriStar LED's - this morning, with no panels connected and no load, the y/r LED's are lit, which indicates 0-30% state of charge, yet the battery voltage is 12.6, or 90+% SOC. I know that the TriStar docs read that the LED's show SOC is 'when charging' - do the LED's mean nothing when no charging is happening? The TriStar has worked this way from the moment it was installed, so I know there is no malfunction due to the Magnetek or improper wiring.

I've never seen the blinking LED's which should show the type of charge happening with the panels connected, but the LED's will cycle from solid y/r to y to y/g to g over time as charging occurs.

I am misunderstanding something here...

John F
LV, NV

crewzer

Re: Solar controller not playing well with RV converter

John,

My interpretation of TriStar manual’s description of the LEDS indicators is that they’re intended to be charge indicators rather than a “voltage meter”, especially since a lit red LED means the battery is discharging.

The Y/R -> Y-> G/Y -> G -> medium blinking green would indicate relative SOC based on charging voltage, with the medium blinking green indicating an absorption (PWM mode) charge voltage of ~14.8 V (ref 77 F). The green LED should have been in “fast blink” mode when your charger was EQ’ing the batteries.

Battery voltages can be all over the place. For example, the absorption voltage may be ~14.8 V, but, once full, disconnected from charging sources and/or discharging loads, and allowed to “rest” for several hours, the battery voltage will be ~12.7 V.

So, a battery under charge always reads higher than its actual SOC, and a battery under load always reads lower than its actual SOC. The difference between the “real” voltage and the indicated voltage is usually directly related to the amount of current going into or out of the battery.

It takes a while to learn batteries in general, and it takes a while longer to learn the nuances of one’s particular system. Accordingly, you might want to consider installing the TriStar’s optional meter.

Here’s a link to its user manual: http://www.morningstarcorp.com/products/TriStar/info/TS_Meter.pdf

HTH,
Jim / crewzer

System2

Re: Solar controller not playing well with RV converter

barshnik wrote:

the TriStar LED's - this morning, with no panels connected and no load, the y/r LED's are lit, which indicates 0-30% state of charge, yet the battery voltage is 12.6, or 90+% SOC. I know that the TriStar docs read that the LED's show SOC is 'when charging' - do the LED's mean nothing when no charging is happening?

---

That's the same way my TS-60 works. I think you'll find the 0-30% is for when it's charging only.
The LED's indicate the existing battery voltage. What that voltage means, depends on if the battery is being charged, discharged, or just resting.
Re the EQ mode, my manual and any other information I can find on it, is rather vague as to how long it stays in EQ. Once the voltage is up to EQ, it stays in EQ until the % of charge is reduced by PWM, to about 30% of what the full charge amps would be, then holds that for several hours type of thing. AND there can't be any large load on the batteries even for a few minutes during that time, or else the time count down starts all over again once the PWM is reduced once again to 30%. I have to control my EQ manually, otherwise it could take days before it kicked out on it's own and the batteries would just boil away. Two reasons: 1) Battery bank too big for the PV's I have, so very hard to reduce to 30% and; 2) My fridge and pump come on from time to time, just enough to reset the countdown. I contacted Morningstar about the problem and they sent me an updated version 10 firmware to install, which I did with great hopes, but nothing changed. So I just live with it. I think it would work OK in my situation, if there was no requirement for the 30% PWM before the countdown could start. That is, just hold the voltage at EQ for the required time, regardless of PWM %.
Wayne

System2

Re: Solar controller not playing well with RV converter

Part of the problem, I believe, is that I've got so much battery for my needs is that I've not seen how the TriStar acts through the various charging stages.  One day of running the water pump, lights, heater at night, only dropped the batteries to over 80% SOC 12.55v) .  The next day, I had the debacle of the unintentional manual equalization which boiled electolyte (but not below any plates in any cell, so I think I'm OK with the batteries) but certainly left the batteries fully charged (and then some.)   The next day and night again left the batteries just below 90% SOC, probably confirming the batteries are unharmed.  Here is my plan, now that our pop-up is back in 'laboratory conditions' (home.) 

Discharge the battery bank over a period of days by leaving lights on in the trailer with known wattage, periodically measuring battery voltage after a 'rest', to get an idea between the battery bank AH rating and actual load drain.  Do this until battery voltage indicates about 50% SOC (12.1 v).  Do some math with the known current draw over known time, try to correlate to battery AH rating.

Then, wait for a sunny day (yes, we are expecting cloudy weather here in LV for a while) and plug in the panels and see what happens with the TriStar, watching LED indicators when possible.  With the batteries at 50% SOC, I should be able to see all charging modes, yes?  All I've seen so far are solid lit LED's (other when I put it in EQ mode), I'm writing that off to the fact that the batteries have been at such a high SOC before plugging in the panels.

I'm ordering the digital meter tomorrow, so maybe I'll have it by the time I'm ready to charge.  In for a dime, in for a dollar. 

I don't really understand why the TriStar couldn't be smart enough without using the meter to indicate accurate SOC when it knows it is not charging as well as when charging, but oh well.  Maybe with the meter it will be smarter (at least I won't have to drag my VOM with, which will make my wife happy.)  Also, it does seem that the TriStar documentation for LED operation does leave some room for interpretation. All of your comments make sense.

Last night I upgraded the TriStar firmware to the latest on their website (Oct. 2006, v. 10b) with no problems, but I'm having problems using the 'view' software - again, less than great documentation and missing help files - not that I'll be dragging my laptop along camping...

Thanks,

John F
LV, NV

crewzer

Re: Solar controller not playing well with RV converter

John,

With the batteries at 50% SOC, I should be able to see all charging modes, yes?

Should be “yes”. For example, I would expect to see the following:

Bulk mode: Y -> Y/G -> G
Absorb (PWM) mode: Blinking G, medium rate
Float mode: Blinking green, slow rate

I don't really understand why the TriStar couldn't be smart enough without using the meter to indicate accurate SOC when it knows it is not charging as well as when charging, but oh well.

It could be smart enough, but it’s a challenge. For example, the effect of a 5 A discharge rate on a 75 Ah battery is significantly different from the effect on a 250 Ah battery bank. However, The good news is that battery monitor gadgets are available to address this very issue. Take a look at the Bogart Trimetric and the Xantrex Link 10 as examples. The bad news is that they’re in the same price range as many controllers.

Part of the problem, I believe, is that I've got so much battery for my needs...

Sure don’t hear that very often!   

HTH, and good luck!
Jim / crewzer

System2

Re: Solar controller not playing well with RV converter

Crewzer,

as usual, your intreptation of LED operation is just what was explained to me by the nice tech support guy at Morningstar today - no blink means it is in 'bulk' mode, batteries taking in everything the PV's have to offer. Once the battery charge reaches the setpoint (for my Trojans, 14.8 v.) it should go into regulation and show the 1/2 sec. green blink, and after a 'period of time' should go into 'float' which should show the 1 sec. green blink.

I loaded the batteries to drain to 50% SOC, and now have the panels plugged in (partly cloudy today) and I expect it won't reach regulation until sometime tomorrw (assuming we have some sun.) Right now, after 3 hours of charge, battery voltage (without rest time) has gone from 12.2 to 13.4, so I doubt it will reach 14.8 today. So, I won't really know if the TriStar is operating correctly until it reaches regulation. Open circuit PV voltage is 20.2 in full sun (for the 75 w. Siemens and 85 w. Kyocera in parallel.

I ordered the Blue Sea switch (the mini size to save a little space), rated at 350a continuous. when it comes, I'll wire it for 'Solar', 'Shore', and 'Off' (batteries disconnected.)

I also ordered the TriStar meter, along with some other bits (cool flourescent lamp to replace a couple of my incandescents, extra wire, connectors, hydrometer, fuses, etc.) for the trailer. Right now, because of my rewiring, neither battery is fused, so soon both will be.

Those Trojans really hold some liquid - I used 1/2 gallon of distilled water to get both of them filled to 1/8" below the level of the fill well, even though the level in none of the cells dropped to the plate level (this was due to my unintentional EQ episode that boiled away that much of the liquid.) Of course, the batteries might not have been factory filled to that high level when I bought them a couple of weeks ago.

Wondering now if I should have ordered the Kyocera 130 instead of the 85 - maybe I'll add a permantly mounted UniSolar 64 to the A-frame of the hard-sided pop-up and try to always park it aiming south... I'd kind of like to be able to get a full charge in one day. I've figured I've got enough battery to last 3 nights in the winter without recharging (lights, water pump, heater) if somewhat frugal, which is good as long as I'm able to recharge in one full sunny day, thus accounting for cloudy days on an extended trip.

So, not knowing the math, would 220 w. of PV do the trick on my 2 Trojan 225's better than my current (no pun intended) 160 w? Thanks,

John F