Question on how batteries work in this ACR+Solar situation

[DaveInDenver]

I don't know if we disagree or not, but most consumer-level MPPT controllers still use techniques that will favor higher voltage, particularly when temperatures are elevated and during non-ideal irradiance. The controller efficiency will also decrease as current increases, you know that dwh, conduction and switching losses go up. So your example of 17V/6A -> 13V/8A vs. 34V/3A -> 13V/8A may be more like 17V/6A -> 13V/7.8A vs 34V/3A -> 13V/7.9A with 3% power supply losses.

I do concede it is somewhat inconsequential, since as you point out, it's such a low power system that losses are proportionally low, too.

But also to mention, you also don't want to switch from buck to boost, that will further reduce converter efficiency. You want the panel array (assuming at high temp) to stay well above the charging voltage. The P-V curve favors following the voltage as far right as possible when the panels are used in the summer and at less than 100% irradiance.

The difference in cabling, I dunno, it's obviously not tremendous but I think it's important. Since 10AWG is about 3 milliohms per meter, 50' is about 46 milliohms. For a 170W panel a change from 34V/6A to 17V/12A means you'll lose a couple of watts more in heating the cable. The controller doesn't care, 170W is 170W, so as long as the expected cold temp max voltage is not exceeded (which is 75V for the SunSaver MPPT) there's no reason not to run a higher voltage, letting current be what it is (assuming the panels are all fairly closely rated).

My point is all these little 1% here, 3% there losses do add up system-wide even for small portable systems.

 

[dwh]

I don't know if we disagree or not,

I think we do somewhat. Sunday morning so WTH, guess I'll dig in to it a bit. But I'm using a phone, so don't expect my usual bunch of reference links.

but most consumer-level MPPT controllers still use techniques that will favor higher voltage, particularly when temperatures are elevated and during non-ideal irradiance.

I don't think so. The array is going to have a Vmp (Voltage of maximum power, or Maximum Power Point). Finding the Vmp and operating the array at that voltage works exactly the same no matter if you rig the array 12, 24, 48, 96 volts or whatever, and no matter the operating temp or insolation.

Not sure what techniques you're referring to.

The controller efficiency will also decrease as current increases, you know that dwh, conduction and switching losses go up. So your example of 17V/6A -> 13V/8A vs. 34V/3A -> 13V/8A may be more like 17V/6A -> 13V/7.8A vs 34V/3A -> 13V/7.9A with 3% power supply losses.

Yea, true enough. I can certainly get down and slug it out doing the math to the last decimal place, and when hanging out with the engineering types over on the solar forums I do that.

But when explaining the basic concepts to a new guy, I try to avoid it by just rounding off to keep the distractions to a minimum. Once they have the general idea, then it's time to focus on the detailed math. Just my preferred style is all.

(And technically, to make the 17v/6a round off work, I had to assume 102w instead of 100w. )

But also to mention, you also don't want to switch from buck to boost, that will further reduce converter efficiency.

To the best of my knowledge, the converter in most solar charge controllers is a buck converter, not a boost converter. The array has to be rigged so that the Vmp always remains above battery voltage.

You can use a 24v array and buck it down to charge a 12v or 24v battery, but you can't use a 12v array to charge a 24v battery because the converter doesn't boost.

Thus this talk of switching from buck to boost seems somewhat off target to me.

You want the panel array (assuming at high temp) to stay well above the charging voltage.

Um..."above the charging voltage" certainly. But you have to do that anyway, which is why "12v nominal" PV always has a Vmp of 16v or more.

As for "well above"...doesn't matter. The array Vmp will remain above charging voltage or whoever set it up made a mistake. The MPPT will find the Vmp regardless. And as long as that Vmp is above charge voltage, the converter will buck it down to battery voltage.

Rigging for a higher voltage on the input side won't change any of that.

The CTEK 250s MPPT has a MAX input voltage of 22v. Building the controller to do 12v/24v is a convenience, not a requirement.

The P-V curve favors following the voltage as far right as possible when the panels are used in the summer and at less than 100% irradiance.

The curve is irrelevant. It's useful for illustrating the concept, but the Maximum Power Point Tracker seeks a point, it doesn't care about whatever curve that point happens to slide around on.

No matter what the curve looks like, the MPPT will seek the point where the array puts out maximum power at the moment - which will be whatever the array's Vmp happens to be when the tracker's timer trips and causes it to do a Vmp seek.



Don't see any other disagreements right offhand.

 

[DaveInDenver]

Not sure what techniques you're referring to.

For consumer and automotive I assumed they used the hill climbing techniques (P&O, FOV, INC) or hybrid A/D techniques (BFV, FVI).

To the best of my knowledge, the converter in most solar charge controllers is a buck converter, not a boost converter. The array has to be rigged so that the Vmp always remains above battery voltage.

You can use a 24v array and buck it down to charge a 12v or 24v battery, but you can't use a 12v array to charge a 24v battery because the converter doesn't boost.

Thus this talk of switching from buck to boost seems somewhat off target to me.

You're likely right. My design experience is not primarily in automotive so I'll step aside. I've usually used dual mode controllers for the applications I've been involved.

Um..."above the charging voltage" certainly. But you have to do that anyway, which is why "12v nominal" PV always has a Vmp of 16v or more.

As for "well above"...doesn't matter. The array Vmp will remain above charging voltage or whoever set it up made a mistake. The MPPT will find the Vmp regardless. And as long as that Vmp is above charge voltage, the converter will buck it down to battery voltage.

Like I say, I don't have the automotive (or recreational) solar design experience you must. I've found that MPP is typically around 70% to 75% of OCV, which for the panels I've used meant the load sometimes required boost in marginal and worse case conditions. I'm primarily a power supply and mixed signal designer, not heavily involved in solar specifically.

The curve is irrelevant. It's useful for illustrating the concept, but the Maximum Power Point Tracker seeks a point, it doesn't care about whatever curve that point happens to slide around on.

Curves are always just a way to visualize, they are useful (at least to me) to close the loop. To validate my analysis against lab data.

 

[Rando]

Remember that you will (almost) never get the full rated power out of your panels, particularly in northern areas. With a flat mounted, dusty panel at 40+ degrees of latitude you will never see more than ~ 80% of the rated output, and much less than that for the majority of the time. I would have no problem using a 300W solar array 15A MPPT charge controller. At least with my Victron controller, on the odd occasion I am getting better than 75% of the rated output, the controller will just adjust the load the panel sees so the current is limited to 15A. It doesn't damage the controller, and it was not worth the extra money, size and weight to purchase a larger controller that is sized for conditions that occur less than 5% of the time.

Not a lot of juice for a $180 piece of equipment, eh? Oh well. 225 should be fine. Not doing anything crazy.

 

[jeegro]

Thanks everyone.

The parts have started to arrive. I will start a new build thread once I figure out where I want to locate each device and create a wiring diagram. It's going to be a very complete system! With all these accessories I think I need something like the Blue Sea SafetyHub to consolidate large amp fuses (inverter, fuse panel x2, etc) vs having a bunch of separate fuses/breakers.

Question on the ML-ACR, in the documentation it says:

"Use circuit protection only if the ACR is not used for emergency cross-connect"

My understanding is that "emergency cross connect" is the feature provided by the remote switch to combine batteries to aid in starting. I plan on using that, so that means I should not fuse the 2 cables between the ACR and batteries, correct? I had planned on using the Blue Sea terminal fuses, but it looks like I should no longer use them.

I assume its because a starter can pull way more than 300 amps to start, so its not realistic to fuse protect, just like the starter to main battery cable is not fused in any truck.

 

[DiploStrat]

Fuse it or lose it. Just fuse to the rating of the wire. The general rule is to fuse anything over about 18". When connecting two batteries, fuse at each battery.

N.B. Your starter doesn't pull as much as you think. dwh can probably give you some guesstimated numbers. (Of course, with a good solar/shore setup and an ACR, the chances of a dead starter battery are very low.)

 

[DaveInDenver]

A typical vehicle starter is rated for around 1.25kW, so an expected typical load current on the order of 130A with a battery voltage depressed to around 10.5V to 11V. Inrush is probably in the 450A to 500A range for 100ms or so. Just had to do some of these measurements. It will depend on the speed of the motor, load, cable size, battery condition. It will vary quite a bit (my guess is about 80A to 150A normally) across different engines.

For battery fuse size always pick for the cable size or smaller. Don't use the load as your first step. For a starter fuse (protecting the starter, not the cable) I would assume some margin to allow for variance over expected temperatures. For example I was measuring my truck and it was warm when I did it. This morning it was -8F and I can guarantee you it was higher than 130A and the battery depressed more than when I did a few weeks ago.

In this case DiploStrat makes a good point, for a short cable you can usually consider it safe. It's about the chance of a cut or short and if it the cable is short enough not to hit the body or other battery terminals then the chance of a massive short is very small.

Something else to mind about Blue Sea's manuals is it's from a marine perspective and there are regulations as to what needs or doesn't need to be protected. Starting circuits on boats don't need to be fused, at least in some cases, but other circuits and batteries may. If the Coast Guard is inspecting your boat then it needs to be right. What you do on your truck may not align totally with that. Your decision to fuse or not can come down to your assessment of how well the cable is protected or the chance of a loosened bolt causing the cable end to hit something. But also think about how bad it would be to get a nuisance trip and either eliminate or size appropriately. For a starter it would be a pain to have a fuse blow but for a winch it might actually be dangerous.

 

[jeegro]

OK. Since the cable will be less than 8" each (probably 6") to connect the ACR to the batteries (with either 2/0 or 4/0 welding cable) then its OK here to skip the fuses as suggested in the instructions. If I were to fuse it, it sounds like I'd need a 200A+ fuse. The cable will also be heatshrinked at each end, then techflex'ed, then heatshrinked again at each end. Might even wrap it all up with silicon self fusing tape or tesa cloth tape as well.

Agreed that I doubt I'll ever need to cross-connect the batteries to start her up, but I figure I might as well wire in that option. I plan on skipping the 3 starter isolation wires as those are designed for marine engines.

I'm leaning towards dual Odyssey 34m-pc1500 or the more standard 34-PC1500T - reason for the former was to get 3/8" threaded studs for the terminal fuses for the ACR... but since thats no longer needed I can probably do the latter w/ SAE posts.

 

[DaveInDenver]

I didn't fuse between the ACR and two batteries. Highly unlikely to be a problem and not really practical in my configuration anyway. I only used 2AWG cable, which is also what my starter is wired with.

Excuse the sloppiness, I haven't finished making a tray for the aux fuseblock so it's a temporary setup that's going on a few months...

I also don't have a starter fuse but that's only because I couldn't figure out a way to put the MBF on the terminal without it actually be a shorting hazard to the oil filter or hold down bolt. Still thinking on that one. I'd like to fit in a double holder (like I have on the aux battery), one for the starter and one potentially for a winch.

 

[jeegro]

Got it! Thanks. Will skip fuses. Yeah having fuses there adds a lot of bulk - the terminal fuses are the only option that is realistic, otherwise you're stuck with making 4 cables... 8 lugs... ugh.

btw you're not using twisted pair for your shunt

Odyssey's are sealed, yes? I hadn't thought of mounting the ACR right on top. That could be convenient. Manual says not to mount on top of vented batt's

 

[dwh]

Also, both the wire and fuses have a safety margin built into the amperage rating. Neither one instantly melts on temporary overload.

It takes a certain amount of overload for a certain amount of time to blow the fuse. A small overload will take longer to blow the fuse than a dead short overload.

So yea, anything other than short runs should be fused.

Diplo and Dave are right, the starter isn't that big a worry in terms of amp load blowing a fuse. A bigger worry would be winching. IIRC, last time I checked, a Warn 8k winch could draw up to 250a at FLR (full locked rotor).

But of course, your not supposed to stall out your winch.


And the Blue sea 500a ACR is really designed for yachts. You won't be cross connecting trying to emergency start a thousand horsepower diesel.

 

[DaveInDenver]

Got it! Thanks. Will skip fuses. Yeah having fuses there adds a lot of bulk - the terminal fuses are the only option that is realistic, otherwise you're stuck with making 4 cables... 8 lugs... ugh.

If you have the space those MBF types would be slick for sure.

btw you're not using twisted pair for your shunt

I don't have a shunt. You may be seeing the little fuse block on the aux battery?

Odyssey's are sealed, yes? I hadn't thought of mounting the ACR right on top. That could be convenient. Manual says not to mount on top of vented batt's

Yes, they're SLA AGM. I assume they are concerned with corrosion from the venting.

I can't claim this idea as mine. Matt Carter of Off Grid Engineering made the tray. I was thinking this would be a good location but he had already come up with it and I thought it was a good idea. Wham, bam, done.

https://off-grid-engineering.myshopify.com

 

[jeegro]

The brass-ish 2-tower thing on top of the black housing - looks a lot like a shunt! On second glance, it also looks like a fuse. My bad.

Nice call on the ACR mount. This is the tray I bought for my truck: https://columbiaoverland.squarespac...rover-discovery-2-dual-battery-tray-1999-2002

Once it arrives I'll see what I can do to fab a mount. What's the dimension between the bolt holes for your top bracket? I could get lucky and it matches my tray. Or maybe those 4 pre-drilled holes on my tray match the ACR - fingers crossed.

Perhaps a stupid question... what do you all use for ring terminals on the SunSaver MPPT? They look like #8 screws, but the spacing is so tight that I can't imagine fitting a proper 6 or 8 AWG copper lug in there.

 

[DaveInDenver]

Also, both the wire and fuses have a safety margin built into the amperage rating. Neither one instantly melts on temporary overload.

Truth. This is the biggest misunderstanding I think about fuses and circuit breakers and how they relate to wire and cable.

Can do an in-depth explanation if someone would like it but suffice to say you basically pick a fuse for the amount of current and time you want it to hold and not by trying to determine an instantaneous fault condition.

In the case of dwh's 250A locked rotor using those MBRF types I'd probably end up using the 125A (green one) with 2 AWG and EPDM insulation. I'd think most people would look at that and think it's nuts, following NEC's rating that 2 AWG is good for 115A and pick a 250A or 300A fuse. This would be much over sized cables (OK, the NEC's job is to be ultra safe) and much too high of a fuse (not OK).

 

[DaveInDenver]

Perhaps a stupid question... what do you all use for ring terminals on the SunSaver MPPT? They look like #8 screws, but the spacing is so tight that I can't imagine fitting a proper 6 or 8 AWG copper lug in there.

The screws on the SunSaver MPPT 15 are 8-32, the lands are roughly 0.3125" wide.

I used 10AWG wire on mine, but you can get terminals that fit for 8AWG at Del City. The problem is their minimum is qty 50, so it would be about $15. But you'd have plenty to do and redo and redo and redo again.

https://www.delcity.net/catalogdetails?item=254085

If you can wait until this afternoon, I'll check to see if I have any that would work. I have a few bags of terminals at home and might have what you need. I know I have about 42 #8 terminals for 10AWG left.