Can I run my inverter off my charge controller?

[John61CT]

I may be wrong, but what I want in LVDs is adjustable setpoints so I can tweak the cutoffs over time,

and put right before the load, so for example the freezer gets isolated long before something more critical (heater) and/or lower current draw (led lights).

Having them associated with a charge source regulator doesn't make as much sense to me, I thought that design harks back to applications like controlling outdoor lighting and such.

 

[tx2sturgis]

I may be wrong, but what I want in LVDs is adjustable setpoints so I can tweak the cutoffs over time,

A couple of these would do it, but they aint cheap:

https://powerwerx.com/low-voltage-disconnect-battery-guard

 

[John61CT]

These guys keep coming out with different products

https://m.ebay.com/sch/i.html?sid=sbs.solutions.sales.usa4

 

[John61CT]

And 12.2V as the highest option is too low for many use cases.

But this *type* of gear is exactly what's needed to protect an expensive bank, when the user hasn't got the skills and/or gear yet to get a good handle on SoC and their usage patterns.

When you see people talking BMS for an LFP bank, this is the most critical part.

It seems even some of those supposedly sophisticated "lithium solar generator" products like Kodiak have no such protection built in.

Because they only care about getting the user through the warranty period, all too happy for a new one to be bought every other year, when with proper care it might last 15 years.

 

[John61CT]

This also looks good to me,

https://www.amazon.com/gp/aw/d/B01EOGTLB4

pretty sure same unit resold for a **lot** more in Australia with the label of a very well respected caravan electrics brand.

 

[John61CT]

Samlex Victron and Votronic are high-end, sold as "battery protect" or battery guard.

Kemo are less expensive, lower ampacity.

Digikey sells Crydom

Not all these are adjustable though.

 

[Trebor English]

Connecting a relay to the load terminals creates an issue because a relay is an inductive load, albeit a small one. 

Relays provide some inductive kickback to the controlling device,  especially when the relay coil is de-energized, although sometimes a suppression diode is used.

I wonder if the controller is rated for a relay at the load output?

Automotive style cube shaped relays have a resistor across the coil.  That makes it so that either end of the coil can be positive.  A diode would be a one way thing.  

The list of no-nos does not include incandescent light bulbs.  Tungsten filaments have a lower resistance when cold.  Like a motor, they have a start up surge.  Avoid fully loading the switch with incandescent bulbs.  

Incandescent 120 volt lamps connected to an inverter will cause a surge on the 12 volt side.

 

[John61CT]

The other use case that I don't see widely used is

Starter batt is just one 12V string out of the the total House bank, IOW no "dedicated starter" batt.

Most simply, two identical large capacity high quality true deep cycling 12V batts.

Most of the time SoC (and thus voltage) is high enough that everything is run off the one big bank,

but as SoC / voltage drops, not only are the bigger / less essential *loads* getting shut down by their LVDs.

*well before* there is any danger of House loads dropping capacity too low for engine cranking,

the Starter batt gets cut off by its own LVD. Similar to the ACR / VSR / isolator /combiner functionality, just a different design POV.

Rest of the bank keeps carrying House loads as needed, maybe right down to 50% SoC, but this does not affect the Starter batt, which is completely isolated.

A little lithium jumpstarter powerpack can be kept charged in the glovebox as a belt-and-suspenders failsafe backup.


This design takes full advantage of Peukert effect for higher total capacity, there is more shallow discharging therefore longer batt life,

maximizing use of space, lower weight less "dead lead", and lower total cost.

 

[Trebor English]

. . .   relay  . . .
This would allow you to be much more aggressive with the Low Voltage Disconnect than most inverters are internally.  I've seen inverter specs thay say they cut off as low as 10.5v.  Rekt!!!

If you set the low voltage disconnect "much more aggressive" then the turn on surge of the inverter or the surge of the load connected to the inverter will drop the battery voltage enough to cause the low voltage disconnect to turn off.

I have ordered one of these.  It is supposed to have a 10 second delay.  That might allow for a more aggressive setting of the disconnect.  

https://m.ebay.com/itm/Low-Voltage-...479318?hash=item2f12c6c4d6:g:ZVEAAOSwwpdW~MoR

 

[tx2sturgis]

The list of no-nos does not include incandescent light bulbs.  Tungsten filaments have a lower resistance when cold.  Like a motor, they have a start up surge. 

Yeah its called in-rush current and its very short, less than a few tenths of a second for most bulbs that we would be using. And of course it's the same reason that older, end-of-life incandescent bulbs tend to burn out right as power is applied. 

This is different from inductive loads that have start up surges lasting a second, or two, or more, and can have a variable impedance the entire time the load is on, and when it is being switched off. Plus, inductive spikes on the DC line are common with this type of load.

I suspect the warning on the Morningstar controller is because of longer start-up surge AND inductive spikes, not the very short start up surges of incandescents,...but, I could be wrong.

Mr Frater has me wondering....maybe a call to Morningstar customer support would clear this up. I have a couple of Morningstar controllers and might like to implement his idea.

 

[John61CT]

I have ordered one of these.  It is supposed to have a 10 second delay.  That might allow for a more aggressive setting of the disconnect.  

https://m.ebay.com/itm/Low-Voltage-...479318?hash=item2f12c6c4d6:g:ZVEAAOSwwpdW~MoR

looks good, yes delay is important with big loads / small banks and great price! Good to see 80A also available.

Just wish they went higher than 12.1V

Please do report back

 

[frater secessus]

If you set the low voltage disconnect "much more aggressive" then the turn on surge of the inverter or the surge of the load connected to the inverter will drop the battery voltage enough to cause the low voltage disconnect to turn off.

I do take your general point, and Mine Doesn't Do That<tm>. 

The LVD for my opportunity circuit is set at 12.7v and my inverter is 300w which is the range OP is talking about.  I drop loads, inverter and otherwise, on that circuit during both Float and Absorption;  the controller adjusts the panels for the increased load in about 1 second.  Vbatth doesn't have time to fall far before the controller adds another 200w or whatever is required to hold the present charging stage.  

Today it was in float and ramped up from something like 2.5A to 16A going into the bank in that second.  It held 16A @ Vfloat  until I removed the load a few minutes later.  Vbatt fell from 13.8 to ~13.5 before the controller caught it. That extra 14A is C/15 load on my bank for one second before recovery.

I also turn on the inverter then the 110 load instead of doing both at once.  Don't know if that makes a difference.

 

[frater secessus]

I'm learning a lot in this thread; I appreciate the discussion.

My takeaway so far is that relays aren't approved to run off LOAD output. Anecdotally, I'll say that so far doing it hasn't caused obvious damage to the five controllers I've done it on. The one with the most time on it was a Genasun with about 8 years of relay duty. Generic 10A HQST PWM for the same amount of time. Two Tracers (A and BN) for 8 months and a shunt controller similar to Trebor's for a couple of months. That suggests that damage from small relays, if any, is cumulative rather than immediate/dramatic.

I admit that experimenting with cheap gear is a lot less risk than experimenting with Morningstar. It will be interesting to hear what they say.h

 

[Weight]

Hi, Tim. This thread went off the rails as so many do in this crowd. Unless you stand and watch, you do not really want to hook anything to a battery positive without a fuse at the terminal. I would cobble some way to have a fuse for the alligator clip. The problem with cigi plugs. They are notoriously prone to bad contact and over heating, voltage drop and most are not designed for more than 10 amps and that only for very short duration. The marine outlets are only a bit better than the original dashboard ones.

 

[Trebor English]

My takeaway so far is that relays aren't approved to run off LOAD output.  

You can connect a diode directly to the load output.  Install it in the direction so that it does not normally conduct.  When the load output goes off with an inductive load the diode will momentarily conduct protecting the load output circuit.  This will dissipate the energy stored in the magnetic field of the inductor.  

A really big relay like a Cole Hersee continuous duty solenoid switch probably needs a big diode like the ones car makers use with air conditioner clutches.  Those Cole Hersee relays take a half amp to operate the coil.  

If there is any other load in addition to the relay the momentary pulse from relay coil will be absorbed by the other load possibly protecting the controller, possibly injuring other load.

 

[tx2sturgis]

My takeaway so far is that relays aren't approved to run off LOAD output.  

{snip}

I admit that experimenting with cheap gear is a lot less risk than experimenting with Morningstar.  It will be interesting to hear what they say.

Yeah one of my Morningstars cost around $150....not terribly expensive but not cheap either....even my cheaper Victron at around $90 is still more than I would like to see damaged, but if you have had good luck with relays then maybe the various manuals can be taken as 'lawyer-speak' when it comes to inductive loads, or at least, the smaller inductive loads.

 

[BigT]

I ended up running it off the 12V terminals on the CC (not the Load terminals).  Seems to be working well. 

Thanks again for all the help.  Funny, but now that it's done it's seem really simple. 

No matter what I tried, there was no room for it on the wall near the CC.  This location will work fine, though, and puts the inverter under the folding table I have mounted to the wall of the van where I'll be using the computer.  

I went ahead and ran the separate ground wire from the inverter to the body of the van, even though the manufacturer said it wasn't really necessary.  
I figure what the heck, why not.