(Unorthodox) MPPT Wiring

[shlac]

I used the generally accepted method of wiring my MPPT on my last build (a T1N), connecting a wire from the Bat+ on the MPPT to the physical battery +, and the Bat- wire to the physical battery -.

Now I'm installing an MPPT in an NCV3 that came with a stock aux battery in the engine compartment, and I'm curious if there's any reason not to wire the Bat+ wire from the MPPT to what they call the "Electrical Connector Aux Battery" on p8 of this document: https://sprintervanusa.com/files/2016/07/AuxBattery-Mercedes-retrofit-doc.pdf. This is essentially a positive terminal under the driver's seat, which connects to the battery isolator and then onto the physical battery + on the house battery in the engine compartment.

My concern is that electricity generated by the solar panel could then feed directly into the loads, bypassing the battery altogether. I'm not sure if there's anything necessarily wrong with this, but wanted to check the wisdom of the crowd.

Pushing the unorthodoxy further, I assume it would be a very bad idea to connect the Bat- outlet from the MPPT to a simple ground, and instead should run it all the way to the Bat- terminal on the physical battery.

Thanks as always for your help!

Dave

 

[autostaretx]

There should be no problem with either of your proposed hook-ups, unless you were planning to add a battery monitor.
(2nd thought: even a battery monitor will be OK .. you'd put the shunt on the frame-to-Aux-negative-post wire)

Feeding the "authorized" connection is fine ... but: note that it is not fused.. . i'd add a fuse between the MPPT and that connector.

Having the solar feed the loads is also fine (heck, many MPPT and MPW controllers include a "Load" output that just parallels the battery (through a relay) when the loads are active).

For typical solar currents, a local (to the solar controller) connection to the frame as the negative path is also fine. Just make it "good" (thick metal, scraped clean for good electrical connection).

Check your controller's documentation ... some are "negative ground" (or "common negative") and some are "common positive".
If you run all of the wires that the controller has screws for (2 to panels, 2 to battery, 2 to loads (if yours has that)), you'll be fine.
Where people get into trouble is if they try to "cheat" by tying the panels' negative to the frame instead of running *that* wire as a "private circuit". (this only affects "common positive" controllers, which work their magic by disconnecting (or limiting) the negative connections).

--dick
my cheap setup: https://sprinter-source.com/forum/showthread.php?p=682475#post682475

 

[cc_windsurfer]

that is how mine is wired

 

[elemental]

My concern is that electricity generated by the solar panel could then feed directly into the loads, bypassing the battery altogether. I'm not sure if there's anything necessarily wrong with this, but wanted to check the wisdom of the crowd.

[Ignoring the ground connection question but assuming that a ground path to the aux battery negative terminal is in use] My understanding is that from an electrical standpoint, if your loads are connected through passive electrical components (wires, fuses, connectors) to the positive terminal of the aux battery, and your MPPT output is connected through passive electrical components to the positive terminal of the battery, then your MPPT output is essentially feeding your loads whenever the MPPT voltage exceeds the voltage of your battery. This is true regardless of which thing is attached where in the system (in other words, running the MPPT output directly to the aux battery + terminal but attaching the loads to the Mercedes aux power takeoff connector [that is attached to the positive terminal of the battery] doesn't change the essential interconnectivity of the system). Electrical potential flows through all of the passive components to reach an equilibrium.

As a consequence, whenever your loads exceed the MPPT output created from your PV panels, then there will be a net draw from your battery (assuming no other power input to the system). When your loads are less than the MPPT output, the net result is power flow into your battery, but not at the MPPT output level - just at the MPPT output level minus whatever the loads are drawing. The only way to ensure your MPPT output is ONLY flowing into your aux battery would be to put some kind of automatic or manual disconnect or isolator between your loads and the aux battery. A charge controller with separate terminals for loads and the battery might have more than passive components (internally) so that it could determine battery voltage in isolation from the loads and thus provide better control over the actual battery charging voltage, but the power to run the loads is probably still going to come from the combination of the MPPT controller and the battery.

In the off chance that anyone is interested, I mention using a disconnect because the charge controller that I'm using has a specific full/deep charge cycle built into it, but requires that the aux battery be isolated from the loads in order for it to function correctly. If the loads are not disconnected they muck with the voltages seen by the charge controller, hiding the actual battery potential as opposed to the overall system potential. The purpose of the full/deep charge cycle on this charge controller is to maximize the life and capacity of the aux battery. Performing the field trials to determine whether this maximization is worth the trouble and expense involved is beyond my means; I'm going to this trouble because a) I can easily afford to put the extra components into place, b) I bought fairly expensive deep-cycle AGM batteries and want to treat them well, and c) it's the right thing to do from a "theory" standpoint.

I put in a multi-position switch to either de-power the loads normally powered by the aux battery, or switch them to the vehicle's electrical system. I'll use the latter when the vehicle is underway and the vehicle electrical system is generating enough power to run the loads (refrigerator mainly). I'll disconnect altogether if the vehicle isn't running and I can't risk drawing the starter battery down too low. The full/deep charge cycle isn't needed on a daily basis, but I'm planning on being on the road for weeks to months at a time.

It is quite possible that "in the real world" the increase in aux battery life/cycle time isn't worth the bother, and that the real best "total lifetime system cost" involves buying the cheapest batteries available, skipping any fancy deep/full charge cycles, and just replacing the batteries when their capacity drops so low that they no longer meet my needs. If I couldn't afford to take the fancier theoretical approach, I would probably use this logic. Rationalization - keeping humans sane for thousands of years.

 

[shlac]

Wow - I can't tell you how much I appreciate these detailed responses!

@autostaretx: I currently have the Victron 15a MPPT+ wired to the "authorized connector" under the driver's seat, and the MPPT- wired to the factory ground terminal located in the port/aft/ceiling area, so should be good. I have a 30a fuse between the 100w solar panel and the MPPT, and a 30a fuse between the MPPT and the "authorized connector."

Re the battery monitor, I installed a Victron BMV-712, with the shunt installed between the physical battery negative terminal and the factory ground in the engine compartment, and don't plan to change the wiring after installing the solar panel, leaving the MPPT wiring as described above.

@elemental - yes, that makes perfect sense, and is consistent with what I assumed to be the case based on my (limited, recently-acquired) knowledge of 12v systems.

I can't thank you guys enough - this build wouldn't be possible without the wisdom of this forum!