How to detect Float Mode on Schneider MPPT80

jpederse98

Anyone have any thoughts on how to detect when this or any MPPT charge controller switches from Bulk to Float in a 3 stage charge pattern? My thinking is that in Bulk mode, I know the battery is taking in and I'm getting all the solar output the charge controllers can produce. Once they go into float mode they may only need 1-2k of the 10k available to them to top off the battery from the array. Time to kick in some 48v dump loads and do some water heating or other loads to consume a little more of what's available in this off grid configuration.

Estragon

Between bulk (taking whatever solar current is available), and float (holding at ~53v, taking ~1% of C (bank capacity in amp-hrs) in current to do so), there's an absorb stage.  This stage should last ~2-3hrs, holding voltage at ~58v, with current slowly dropping to about 1-2% of C.  

I sometimes use later absorb time to run loads like water pumping manually.  I know about how much power they'll use, and can keep an eye on the weather while doing it.  Automating in this stage would get complicated.

To run something more automatically, many controllers have low voltage aux ports which can be used to trigger a relay energizing a load such as water heating.  I'm not sure about that specific controller, but I'd guess there's an aux port that can be configured to trigger on transition from absorb to float.

mcgivor

The battery capacity would be programmed into the controller as would be the absorption time, the transition to float would be when the absorption  time expires or current drops to 2% of programed capacity or whichever happens first.

jpederse98

You would think that state changes from Bulk to Absorb to Float would be one of the triggering events for the low voltage relay but NO, not on the Schneider equipment. Basically you can only trigger on voltages, currents, events, faults, etc. so based on the custom logs I have been running every 1 minute since December 2018, it logs 2 minutes of the peak bulk charge voltage before it drops down to float. I'm using the Iron Edison 19kwh 400Ah lithium battery so that peak voltage is 51.8VDC every day for 2 log minutes before it drops down to the 50.4 Float voltage which seems to work well for the little 15 cell battery I have. I'm going to try turning on the relay at 51.8VDC with a 1 minute delay and turn it off when the voltage is headed back down, with a delay, after sundown around 50.xVDC to shut off my "dump" loads (2x2400W 48VDC thermostatically controlled elements in a 100 gal hot water loop doing a water to water heat exchange to a pumped (preferably DC) pair of radiant heat systems with 8 loops each under 5,000 sq feet of basement concrete). 
Any reason I shouldn't burn of some excess DC energy while the sun shines as long as the battery is at a float level?

Estragon

Lithium would be different, with little/no absorb.  

The risks, as I see it, to using a dump load are:
1.  The dump load could use power that in some conditions may not be "excess" (eg stuck relay).
2.  The dump load, if not properly sized and controlled, could itself cause problems (eg stuck thermostat, no pressure relief valve, etc)

Dave Angelini

All of the bulk/absorb/float data is there already. Much of it must be accessed by modbus. Just about everything is there if you can program a PLC. The conext gateway is also used for a canbus bms battery data to xanbus interface for closed loop battery charging. 

Spongology

Not so sure about how to detect float mode on schneider. Looking forward to see what others would say.

jpederse98

Dave Angelini said:

All of the bulk/absorb/float data is there already. Much of it must be accessed by modbus. Just about everything is there if you can program a PLC. 

I have been talking to Rhodes Carter at Iron Edison as to why neither they nor Schneider has developed any tight integration between their Nuvation Energy BMS and the Schneider Conext products. I communicate with the Nuvation BMS on my LAN and keep an eye on the DOD and current in/out of the battery as well as watch the data on the Conext CommBox. A problem with my 15 cell Lithium is that the range from full and floating at 50.4 vdc and as low as I dare go before the battery BMS shows a warning at 48.4 isn't a very large range. 

I have considered adding to the Xanbus/Modbus something that could watch the MPPT modes and help decide how to adjust loads based on available power and battery status. I appreciate you confirming that this could pan out.

Dave Angelini

It is propriety on the xanbus. They have shared with two other battery makers I use. The battery becomes a device in the schneider system without additional shunts, wiring, bat mons, and just stuff I can't sell to my user. I would not want it either as less is always more to me offgrid.

 It is a business decision on Schneiders side and it may be the voltage range design issue you sited also. 

I have a friend I pay to program and build a 48V plc for some of my weird projects. I am hoping to use the gateway for that when it gets finished and be done with modbus.

MarkP

You don't need to know when it is in float so much as you want to know when it is throttling output, as when it is doing that it has "extra" power that could be used for other purposes.  Consider filling a kiddie pool as an analogy.  You would probably unscrew the nozzle from the end of the hose to reduce resistance as much as possible or at least open the nozzle wide.  If there was a pressure gauge at the hose bibb it would show a low pressure as the flow losses are distributed throughout the whole distribution system half of which is upstream of the gauge.  However as you put your thumb over the end of the hose or closed the nozzle the pressure would go up since the flow is decreasing and with it all the flow losses throughout the system.  Same supply pressure ultimately but all the resistance is now occurring at the point of throttling.

The charge controller does the throttling.  The Conext charge controllers have a relay that can be configured to supply up to 200 ma at 12 volts, controlled by high array voltage.  When bulk charging my array voltage is around 90 volts.  During absorption and certainly during float that voltage rises above 100 volts.  Pick the right setpoints and delays and you can shunt current when the shuntin's good, pretty much.  Also, tap the opportunity load power off of the input to the CC (array voltage and current).  Works fine for resistance heating, can't suck from the batteries, only what is coming from the panels.  A 4500 watt 240 VAC water heater draws something like 800 watts at 100VDC.  Get a 250 VDC switch with a 12 VDC solenoid and use that to switch the array power to the water heater.  It is just another circuit breaker in the combiner box only it draws power out instead of feeding power in.

That's what I do.

Dave Angelini

Or the easy and more expensive way is to design a system large enough that you do not need to know all of this.

It is great that you want to do this but I simply can't do this for an average client. They do not want to do this kind of thing and expect high performance without user intervention. The  context gateway will allow one to do what you are doing with wifi load devices and that will make it easier as there will be no external wiring.

Schneider and I tested what you describe in the field test of the mppt-80-600 in 2010. It works very well. This charger has been totally reliable in 100's of my applications.

MarkP

As the price of panels comes down and as I get older and more easily frustrated I see the wisdom of just throwing more panels on the roof.  With enough excess PV capacity you need not split hairs about exactly when to turn on loads.  I might consider running the water heater on 240 VAC for short bursts whenever the array voltage hits 105.  The problem for me right now is that I have only a Conext SW 4048.  The maximum delay to turn off the load is 5 minutes and the 4048 could probably supply 4500 watts for that long but not if I had any other loads running.  Runner up is using 120 VAC instead.  I used to do that.  I really like the idea of using the array current though as it automatically protects the batteries, self modulates consumption as the array voltage sags, and doesn't load up the inverter.

I will go to three Simpliphi Phi 3.8 batteries soon.  The batteries themselves have a limit for current that they will supply so running a 4.5 kW load intermittently is bound to crimp ones style anyway.  It occurs to me as I write that last sentence that I have ignored the issue so far with my FLAs.  Hmmmm, can't say I've noticed any problem though with any of the 3 sets over the last 5 years....

mike95490

MarkP said:

........ I have ignored the issue so far with my FLAs.  Hmmmm, can't say I've noticed any problem though with any of the 3 sets over the last 5 years....

Wow.  That's some pretty harsh battery life.  Unless you are using "marine" batteries, something is wrong, - it looks like your system must be depleting the batteries into the 30% remaining range.  1 set of well treated batteries should last at least 4 years

Dave Angelini

Agree with Mike above and would add that 5 to 10 is what my folks get on L16 FLA's.

The other part of using electricity to heat water is just not sustainable offgrid. It is wrong in concept and over time will lead to electronic doom. If there is a spouse, eventually they will not allow a situation where there is not hot water.
I know people on this forum try to use the excess for heating water and it really just becomes a wiring issue, safety, and adds stress that is not the purpose of offgrid.

Pretty simple to use a tank or on demand and run some collectors with propane as the back-up off grid.

MarkP

I exaggerate for the sake of the story.  The first set of eight golf cart batteries from costco lasted 3 years which I was OK with since I was a noob.  I bought two a few months ahead to run my water system and charged them with a car battery charger, so, mismatched and abused.  The second set lasted only 18 months which was surprising to me but partly due to having swapped from an APC SmartUPS3000 to the SW 4048 halfway through.  There was a week or two where the panels were no longer charging things and I was relying on other charging sources such as the charger built into the UPS.  There was a model of said UPS that could be expanded with external batteries but my UPS was not that model.  It was hard programmed to charge tiny sealed lead acid batteries, which I removed and threw away and hooked up the GCs.  At least twice in that period the UPS shut down due to low battery voltage which being a UPS was a very low voltage meant to save the connected electronics at the expense of the batteries.  The UPS worked great for what it was but there are definitely compromises if you go that route.

I decided to upgrade to L16s and may have bailed on the GCs a little early.  Got a set of Dekas.  They are great and are still going strong after 3 years however I may have yet again snatched defeat from the jaws of victory.  After I got the L16s it became apparent that my parents needed me so I began spending lots of time with them.  Between suddenly having little load, the fact that I foolishly selected voltages from the high end of the published tables, and my distracted state, I think they were experiencing over-charging.  There is swelling around the positive battery posts.

The cost will be a shock but I welcome the relative user friendliness of the LiFePO4s.  At the rate I am going through FLAs the cost difference is not as bad as it might be.  I do however wonder whether I should have simply kept buying GC2s every 2 to 3 years.

MarkP

There is another angle to my situation.  My location gets 15 feet of rain per year, so, lots of clouds.  One day I did work my batteries hard but by the end of they day they were back in float while at the same time we got 1" of rain so it can be intermittent rain and sun, but on the whole there are many days of just plain rain.  On those days it doesn't matter which direction you aim the panels as long as it is up.  I plan on getting plenty more panels and aiming 1/3 East, 1/3 West, and 1/3 South, all tilted at 45 degrees.  During the summer the sun is actually a degree or two north of overhead and the sun will be boring perpendicularly into the East facing panels by 9am.  Same for West panels at 4 or 5 pm.  The South panels will be getting negligible direct sunlight except for around noon when sun will hit them at around 41 degrees.  At that point the East and West panels will all be getting sun at 45 degrees.  In the winter the East and West panels will get poor sun but the South panels will get good sun.  On any day that there are clouds there will be lots less solar energy on the whole but all panels will be contributing equally.  Right now I just suffer during those periods and make electricity with fuel but with extra panels I will do better and burn less fuel making electricity.  Instead I will burn the fuel for hot water.  When sunny I will make hot water with all the excess PV capacity.

My situation is unique in that If I don't put in the extra PV capacity I am doomed to burn fuel.  If I do bulk up on PV then I can do silly things like running a regular electric water heater.

Just keep me away from innocent FLA batteries.

mike95490

MarkP said:

......

The cost will be a shock but I welcome the relative user friendliness of the LiFePO4s.  At the rate I am going through FLAs the cost difference is not as bad as it might be.  I do however wonder whether I should have simply kept buying GC2s every 2 to 3 years.

Properly configured and treated, LiFePO4 works.   But until you get a decent lifetime from a set of cheap flooded batteries, you really risk BBQing a nice set of LiFePO4.   There is NO forgiveness with over or under charge, the battery is toast in a couple minutes or hours, and you may have a highly toxic fume (hydrofluoric acid fumes)

https://www.mdpi.com/2313-0105/5/3/61/htm

There are several failure modes that can result in a thermal event. Some conditions are initiated internally in the battery system, e.g., shorting of the battery cell, either externally or internally, or improper battery management leading to overcharge, overcurrent, or overdischarge, and some are caused