Setting Shunt before the BMS a bad idea? Here is why I think this is GOOD!

Here is another community question: from an accuracy perspective, it makes sense to have the shunt installed first before the BMS, so battery - shunt- BMS - charger/inverter, right?
Is this is bad idea though as most people tend to say the BMS should always come first. Victron says this is correct.
I also show you the setup of my battery box and we will insert the batteries as soon as I know which order I should cable everything.
I'm also keen to hear about the topic ANL vs Class-T fuse 🧨🔋🔥
Please subscribe and join me on my journey of the Off-Grid Garage!
For more information, please check out my website with links to all materials, devices and products: off-grid-garage.com
EVE LF280 LiFePO4 cells 3.2V/280Ah: off-grid-garage.com/batteries/
Solar, Controllers, Inverters and accessories: off-grid-garage.com/solar/
Batteries, chargers and accessories: off-grid-garage.com/batteries-3/
Cables, connectors and electronics: off-grid-garage.com/electronics/
Tools, measurement devices and analysers: off-grid-garage.com/equipment/

[@alsemi-back-up579]

Someone FINALLY shows how th positive wire goes in th shunt to power it. Thanks so much for the Education video also. I will pick # 3 way!

[@tonydickerson999]

Battery negative, shunt, BMS, to exterior cable, Battery positive, switch, fuse to exterior cable

[@mkroach59]

I like #3...especially since it is backed up by Victron.

[@mrbranchatl]

Hey guy, I really like the way you lay out your plan of attack. I'm glad I found your channel. Thanks for sharing.

[@OffGridGarageAustralia]

Thanks Steven!

[@charlesrichards5389]

I'm traveling back in time going back to watch this older video 😃, but anyway, if there is any concern about a blown fuse having a spark-gap... some fuses are designed with spring tension on the elements so they are pulled apart if the fuse blows. And I really like the way you logically came to the shunt-then-BMS conclusion. 🤓

[@rendark419]

Fleißig, fleißig! #3 währe meine Wahl. Nochmals Danke für deinen tollen Videos!

[@curtsweet2581]

I know little about this issue, but number 3, seems the logical and most accurate choice.

[@philbrooke-little7082]

The problem with putting the shunt before the BMS is that, if there is a low voltage shutdown event, the shunt will still be drawing power to transmit and, at low voltage cutoff, the batteries are very close to dangerously flat. A BMS should be able to completely isolate the battery. The voltage drop across the BMS will be minimal and will be to the advantage of the battery when charging as the MPPT side will be very slightly higher. We had this exact discussion on my group the other day.

[@mattr4660]

According to the datasheet, the current draw on the Smart Shunt is

[@philbrooke-little7082]

@@mattr4660 The principle is the same though on positioning. Most monitors are not wireless and have a monitoring head plugged in that uses more current. The voltage drop across the BMS is not an issue because it works in favour of battery protection and so I think it best to keep the BMS as part of the battery and put all other devices after it.

[@dr.dickie1418]

But after voltage cut off, doesn't the BMS also continue to draw power? And likely the loss from the shunt will be less that what the BMS will draw to stay active.

[@philbrooke-little7082]

@@dr.dickie1418 No it doesn't, this is why there is generally a reset procedure after a low voltage cutoff often involving a brief short from B- to P- or C- depending on type.

[@dr.dickie1418]

@@philbrooke-little7082 If the BMS shuts off power, then there is no flow through the shunt. How much power does the shunt really draw with no current through it? After low voltage shut off, there should be an alarm. I would think it would not be a situation that would be left for a long period of time. Even so, as you say, BMS draw should be minimal. Of course, with shunt before BMS, you protect against a BMS going bad and drawing. Each set up has an advantage and disadvantage.

[@paulbarrette2557]

I really like it that you back up your opinion with facts. Everywhere on the internet I see comments and opinions that are 'free' with no explanations such as: 'Do not do X because that is a fire hazard'. Sometimes they are right and sometimes wrong but regardless, the WHY is the most important thing in the answer. Keep up the good work explaining the why. Your video on float charging LifePO4 is another good example. Science always wins!

[@OffGridGarageAustralia]

Thank you Paul!

[@jamess1787]

Your crushing it with the KZbin algo, like 3 videos a week... Its crazy!

[@OffGridGarageAustralia]

Yeah, only because I had holiday. If I would be home all week... wow ;)

[@jsal9678]

Great job. Instructional and informative video. Keep up the great job.

[@OffGridGarageAustralia]

Thanks, will do!

[@JPHER217]

Well your explanation sounds solid to me. 10/10

[@cheapcarstoday]

You've educated me ,I set my RV with bms then shunt but watching your explanation it makes sense the other way around so I'm going to change mine ,after all if I leave it as it is the shunt can't measure the current when the bms is balancing.

[@w3bb0y]

Before watching this I would have bet my left testicle that you would set it up batt>BMS>shunt. I now would go with #3, thanks for the great explanation :)

[@OffGridGarageAustralia]

Sorry for loosing one of your testicles 😯

[@ykchowaa]

Good idea. I will follow the same for my first off grid battery.

[@johnhill3507]

That was a great thought provoking scenario, I thought I would read all the comments to see if if they concurred with conclusion of #3 ....WOW so many comments.. that so great but where do you find the time. Thank you for all of your work and effort making this channel. It is educational and entertaining

[@OffGridGarageAustralia]

Thanks so much John. This is day 3 of reading comments for me and I'm still not through... I will not asking any questions any more 😬

[@chipmeister111]

Thank you for these videos, I will be using #3 scenario for my build. I don't want to hijack your thread but if anyone has any input, I'll be using bus bars on both the negative and positive sides of the battery and was curious where they would fit in sequence. I also have the Victron battery monitor shunt and a BMS for my Lifepo4 cells.

[@musicwiz40]

You've got it figured out with number 3 :)

[@trondeeilertsen5661]

Hi. Will well go for Victron's recommendations and put the shunt first as in alternative #3. I'm going to do that in my setup in my RV

[@OffGridGarageAustralia]

The question still is, do I need this shunt at all?

[@trondeeilertsen5661]

@@OffGridGarageAustralia Hi. Overall, you don't need the shunt. It's just one device to have more control over consumption. You have a shunt in the Bms, I'd like to think it's going to do the same job. But accuracy compared to Victron, you don't know 😉

[@aussiescotty2950]

Hi Andy, I was reviewing 1 of Will Prowses how to's & he was saying that the Victron Solar Charge Controller does NOT like having the battery disconnected before the solar panels like if a BMS disconnection occurs. He suggests connecting the controller -ve directly to the battery -ve. Just to raise more head scratching issues. Cheers.

[@OffGridGarageAustralia]

The is a statement in the Victron forums and they confirmed the MPPTs can handle it without been destroyed. I haven't read the manual yet, so not sure what that says.

[@chrisfryer3118]

#3 clean designs look good for a reason

[@BenMitro]

Very good! An SSR is usually a mosfet driven switch with isolation between input and outputs. I think all three topologies will work fine. Whichever topology you choose consider what information the charge controller will get if the fuse blows. If you are sensing volts on the shunt on the battery side then the charge controller will think there are volts when there are none.

[@OffGridGarageAustralia]

How do you overcome this though?

[@BenMitro]

@@OffGridGarageAustralia Connect the + connection of the shunt to the load side of the fuse so it sees and reports the controller output voltage. However, you should have a look at what is required by the Victron controller when fuse blows. I imagine the internal fault logic of the controller is based on a truth table that responds to a range of conditions - one being the fuse blown - as to how to connect the power for the remote current sensor.

[@amilww]

You made a very valid point; i.e. BMS uses sensing wires to read the battery voltage. Even though I'm commenting 1 year later, Option 3 is the best.

[@evil17]

Victron shunt first also means u can keep other small essential loads on, like lights,camp fridge for SPAT, etc, going on a seperate circuits of 48v-24v, 12v & 5v, which could be set at a slightly lower setting for trouble shooting, etc , in an otherwise dark situation at night. After all Andy, SPAT is still fine consumed cold in the dark, but warm in the dark is a crime! Victron shunt first is probably ur most accurate measure for ALL loads from battery with backup for when ur BMS shuts down and u dont lose the Victron historical logs. But I know you have well got this all figured by now, another great vid, thanks

[@OffGridGarageAustralia]

All loads need to be connected after the BMS anyway, so with this in mind it does not make a difference in which order we connect. The BMS will have such a small standby current which the shunt won't pick up any way.

[@evil17]

@@OffGridGarageAustralia True Andy, but if the BMS shuts down & shunt is after BMS, dont u lose all ur historical data from the shunt? This is why I thought ur reasoning was good, to have the shunt 1st so u dont lose shunt data in case of a BMS shutdown.

[@Mike_Neukam]

The problem is that the electronics on the shunt will always be powered. You want the shunt to lose power if you turn off the switch or if the BMS shuts off due to low voltage. Imagine if your solar charger controllers failed while you were away on vacation. The BMS could disconnect due to low voltage but the shunt would continue to pull power, potentially damaging the cells. I would recommend the order battery, BMS, switch, shunt. That way the shunt will lose power if the BMS disconnects, or if you turn off the switch. Having a perfectly accurate voltage reading on the shunt shouldn't be critical because the shunt should keep track of state of charge by measuring current flow into and out of the battery.

[@OffGridGarageAustralia]

The shunt has less than 1mA of consumption. no doubt I can run this for months even on 2.5V without discharging the batteries.

[@Sylvan_dB]

The arc can last a lot longer than expected. When the link blows and the end points continue to melt, the air fills with ions that conduct very well. The arc will sustain over a distance of several orders of magnitude greater than would be possible thru clean, dry air. A fuse has three critical ratings. First is the fuse amps. Second is how many amps it can interrupt. Third is the voltage it can interrupt. All three need to be suitable for your application and the ANL fuse is almost certainly not capable of interrupting the current those batteries can provide, and it is highly likely that your car audio fuse voltage rating is less than the maximum voltage on your pack (most automotive fuses are only rated for a max of 32volt but I do not remember what is usual for ANL, and have no idea what your fuse is rated).

[@OffGridGarageAustralia]

Thanks, great comment and explanation. I read that the ANL fuses are safe to use with li-batteries. This one is for 48V (as per seller) but it does not say on the fuse itself. So you would go with a class-t then?

[@SpeakerKevin]

@@OffGridGarageAustralia Victron Energy MIDI and MEGA fuses Rated to 58 Volts. Looks like they also have ANL fuses that are 80V.

[@namechamps]

@@OffGridGarageAustralia I know this is two years old but for anyone else reading it all comes down to short circuit current and the AIC of the fuse. High capacity lithium cells (or multiple cells in parallel) can have huge shortcircuit current. That requires a high AIC. You can calculate it yourself. I = V / R. Fully charged the cells are 3.65V. If internal resistance is 0.25 mOhm then short circuit current (dropping a conductor across primary battery terminals) would be 14,600A. ANL fuses are usually rated for 4k to 6k AIC. Class T for at least 20k AIC. Class-T is really only needed at the primary battery fuse. The branch circuits can use ANL or Mega because the wiring between the battery and any branch circuit (plus switches and busbars, etc) also add resistance which lowers the short circuit current.

[@OffGridGarageAustralia]

@@namechamps Thank you for this additional information.

[@lnxpro]

#3 would be my vote in your scenario.

[@Rigoke123]

Number 3, definitely :)

[@CampPrevost]

#3. Purpose of the shunt is to measure what is going in and out of the battery - nothing more. If you put after the BMS - you could see zero draw and run down your batteries and not even know it. I have heard, however, that the shunt does have a draw and does not read that draw, so same problem. So still should have a monitor alarm if voltage gets below a certain amount. I have a unit measuring my battery voltage and will send a cell signal / email to alert me if batteries reach a certain voltage. Thanks for all the videos!

[@OffGridGarageAustralia]

Thanks for your comment. OK, so I'm not planning of running the battery that low, 2.8 would be the absolute lowest. The shunt draws

[@intheyear2510]

I think 3# and have just installed this way myself, the only thing I can think of as a downside is , in a disconnect situation for low cell voltage the small draw from the victron could carry on using power thus defeating the bms and potentionsly kill the battery if left long enough

[@w3bb0y]

I can't confirm but I'm sure the shunt will have a low voltage shut off? Victron are too smart to allow that to happen :)

[@intheyear2510]

@@w3bb0yperhaps maybe so , but victron say bms after shunt so I followed this instruction for my install and il see how it goes , logically to my mind the bms should before the shunt as it will always defend the cell voltage in event of disconnect

[@JeremyAkersInAustin]

This is a very important point.

[@OffGridGarageAustralia]

The BMS should always come first to disconnect whatever is after it. However the consumption of the shunt is less than 1mA.

[@Mike_Neukam]

@@OffGridGarageAustralia 1mA*(an indefinite amount of time)=potential damage. What do you value more, an accurate voltmeter, or reduced risk of failure? The choice is ultimately yours. Any decent battery monitor should be tracking the state of charge based on current flow, not voltage. I watched a video recently where an RV with lithium was left in storage and due to a power loss, the cells were overly discharged and damaged because of the active cell balancers being powered after the BMS disconnected. Keep in mind that when the BMS disconnects, you don't have much reserve capacity left.

[@alanb76]

Some BMS's don't read the first cell voltage when only the small black wire is connected, they require the B- lead to measure the first cell. So perhaps it varies with the BMS.

[@PowerPaulAu]

Oh now that makes more sense (pun intended)! I didn't realise that BMS had a full set of wires. Brilliant thinking, and I totally agree with the setup of the Shunt first, then the shunt. Nice work! #3

[@vferrandobe]

Victron also makes a 58V 125A MEGA fuse that will fit into your fuse box. I would trust it will do the job if you aren't confident with the one you have.

[@OffGridGarageAustralia]

Arghhh, don't tell me that!!!!

[@vferrandobe]

@@OffGridGarageAustralia Why not? It is a minor change in the system. If at all necessary.

[@jamest.5001]

If you are concerned about the fuse ARCing, put it pid pack, it will be half voltage!! I haven't tested mine at full load, but it pulls somewhere between 225 & 300amps! I'm working on a DC generator to charge the battery bank directly, with the potential to output 560amps, if I have enough engine, I will likely stay around 150-300 amps, allowing it to power the inverter, and charge the batteries, not at full power, normal consumption is about 50amps dc or less, so I could put 100+ amps in to the battery! It's more efficient charging directly from a DC source, without loosing 10%+ every time the voltage changes! Even if the alternator is only 60% effiencent, that's hot effiecency, I'm hoping with a custom regulation, I can get closer to 80%, and with it running colder, although a pma would be more like 90%, (about as good as possible, ) it lacks regulation. I'm hoping a custom regulator, (microprocessor controlled) will improve things, sorry for rambling off topic here! I'd love to have about 40-80 of these cells!! For a 8s system!! 280ah x 10s = AWESOME!! Just enough power to run my house!! Ha-ha, it would be awesome if the price would come down to about $50 each, ha-ha we ain't that lucky!! Maye in a few years I'll convert to lifepo4 from li-nmc, it's a bit spooky having 1200ah of liion in the House!! Great work, it's coming along nicely!!

[@Muppet_Interfector]

Are you serious about the mid pack suggestion? Voltage loops don’t care if the pack is split or not it is the sum of the voltages that you need to deal with.

[@jimmysquires5093]

Go with 3. You want the most accurate measurement to be the Vistron and that is what Vistron says to do. You will find that the BMS will not be accurate to your other meters, but will measure each cell in reference to the other cell just fine.

[@pnoygil]

Switch and fuse is always on the positive side....

[@kheavmady8780]

why? but on my battery, a negative side wire is the hottest

[@aussiescotty2950]

Switch not always - Toyota Coaster Bus -ve battery disconnect, some heavy equipment too. I have seen fusing in between battery cells - worth a think.

[@benjaminishak6192]

Why?

[@pierreconstantin2546]

Fuse are usualy on the positive side because in Canada and US most system are negative ground. A short to ground will blow the fuse if it is on the pos side, and it wont if on negative side.

[@pierreconstantin2546]

@DIY Projects they will still be connected either ways, ((edited)if the load is still present), as the switch disconnect only one side of the power line. There also will be voltage present from voltage sensing lines.

[@launacorp]

a little note to beginners: Your BMS disconnects B- to C- if the balance leads are disconnected. So put your switch into the positive side! You can pull the balance connector and turn of the switch and then you don´t have any connection to the battery from the outside. No + and no - connection and everything is save ;)

[@Sirslayer_X]

Remember more devices you add in series the more resistance and less voltage your load with receive. Therefore more current will reduce your amp hours when load is applied. Testing amp hours is more accurate when load ( your common use) is applied

[@OffGridGarageAustralia]

Yes, I know that, but you need to connect all this somehow, right? There is no way around having quite a lot connections. I hate this too...

[@ronstarner1103]

I am probably going to set up like this but the voltage for the smart shunt I will probably connect to the output of the BMS. This way supposedly if the VMS shuts down the shunt still does not draw from the battery. I will also hook up the secondary input to the shunt to the input of the BMS. That way I can read both sides

[@AMaverickBrother]

You're right option 3 is the cleanest and you'll have a slightly more accurate voltage measurement. The only thing I would wonder is if it would be preferred that the shunt should see its voltage go away similar to that that the MPPT would if your BMS were to disconnect. That would be the only possible corner case that I could think of in which you may have an issue. But your BMS should never disconnect unless you're testing it and under monitored conditions...

[@OffGridGarageAustralia]

Argh, more to think about it...

[@houseofancients]

hindenburg , lovely refrence !

[@waynehobbs5175]

Fyi I suggested option 3 theory in your previous video. Shunt must connect to negative battery alone or it won't be accurate.

[@JeremyAkersInAustin]

I would do #2 because with #3 if one cell gets to a very low voltage the BMS will disconnect the loads, but the SmartShunt will not be disconnected and thus will continue to draw some power: This could lead to a situation where one cell of the battery drops to 2.5 volts and the BMS disconnects the battery: However since the shunt is connected first it will continue to draw power and could lead to damaging that cell if allowed to continue discharging. I know the shunt doesn't pull much, but it has to power a Bluetooth radio which isn't nothing. I also need to strongly disagree with your statement that the SmartShunt is "more accurate than BMS". *Perhaps* it is more accurate than the particular BMS you're using... (I am not familiar with that one.) However other BMS units I've used and researched have very accurate SOC calculations. For example, in my setup I have both the Victron Shunt as well as an Orion BMS. In my case the Orion BMS is much more accurate than the Shunt: The Shunt *only* measures amp hours: It cannot measure individual cell voltages, and *only* synchronizes with the battery when the battery reaches 100% SOC. It doesn't synchronize at 0% state of charge and so over time it's not 100% accurate, under continuous cycling the SmartShunt SOC will drift considerably from the actual battery SOC. For example: Say you start at 100% and cycle the battery down to 20%. Because the shunt is obviously not 100% accurate it may read 21% instead of 20%. Now say you recharge back to 90%, but the shunt is off a bit so it reads 92%. Now you cycle back to 20% again and the shunt reads 23%... You see how the drift keeps getting larger? I have experienced this first hand with the Victron Shunt. Most BMS units will *also* measure amps going in and out, but they *also* monitor the individual cell voltages. So if you have a 280 AH battery and one cell drops to 3.0 volts while the other cells are still at 3.2 volts, the BMS will sense this and start to adjust the SOC downward to account for the voltage entering the "knee" at the bottom. The BMS will do the same thing as cells rise above 3.3v at the "top" knee. Where-as the smart-shunt doesn't synchronize the SOC until the battery reaches 100% SOC. The other problem with the shunt: As your batteries age the number of amp hours they can store will decrease. So you program the shunt for 280 amp-hours, and it will only read 0% after 280 AH have been consumed. But if one of your cells drops to say 260 AH: What will happen is that after pulling 270 AH the shunt will still read 7%, but that one cell will have reached it's "Low Voltage Disconnect" and the BMS will kill power. However the SmartShunt will continue to draw power, and continue reporting a 7% state of charge to your phone via Bluetooth even though the pack is actually completely drained. The BMS on the other hand should correctly report 0% SOC in this case, and will internally adjust the "state of health" down to 93% so that it knows for future discharge cycles it will base the SOC reading on 260 AH instead of 280. Good BMS units will also monitor the internal resistance of each cell and also factor that into the state of health of the battery to make future SOC calculations more accurate. In other words: The SOC calculation in the BMS becomes more accurate over time as it "learns" the actual capacity and internal resistances of each cell. Where the SmartShunt is only as accurate as the amp hours you program it with. It will not adjust the battery capacity down as the battery ages and the state of health declines. You would need to monitor the state of health yourself and re-program the AH capacity in the SmartShunt from time to time to keep it accurate. One thing to consider: In my setup my BMS can communicate via CAN-bus. I run CAN-bus (Which is very easy to wire) from my BMS to a Victron Color Control GX. The Color Control GX can then read the battery voltage, state of charge, state of health, battery temperature, etc, from the BMS via CAN Bus and can then communicate that info to the MPPT, Inverters, etc. (I use a VE.Direct cable from the CCGX instead of Bluetooth because it's more reliable, but I believe it supports Bluetooth as well) I mentioned that I built my system using both the Victron Shunt and the BMS: But if I had to do it all over again I would simply skip the Victron Shunt and just use the CAN-bus to get the battery voltage and SOC data from the BMS to the Victron MPPT and Inverter products. Any decent BMS will be far more accurate than the shunt because it does more than just monitor amps going in vs amps going out: it also measures the individual cell voltages and cells internal resistances.

[@OffGridGarageAustralia]

Thanks for your looooong comment and sharing your experience with your Victron products. That's very interesting and helpful. The Smart Shunt has

[@Muppet_Interfector]

I don’t believe this was a question about what is better to go with BMS or shunt. Andy asked where would be best to place the bms and shunt relative to each other and the battery. I doubt that Andy’s BMS has CAN bus capability like yours. By the way what BMS are you using? I doubt Andy is going to do without his BMS. No matter what communications capability a BMS has, I expect mine to monitor cell voltages to avoid both an out of balance condition which could eventually lead to either over charging or over discharging an individual cell.

[@JeremyAkersInAustin]

@@Muppet_Interfector "I don’t believe this was a question about what is better to go with BMS or shunt. "

[@JeremyAkersInAustin]

@@OffGridGarageAustralia Sorry for it being so long. I keep thinking of things that I want to mention. I built a very similar project to this about 4 years ago for my RV so I've definitely learned some lessons in the process. Yeah, I saw the SmartShunt has only 1mA of consumption, so it probably doesn't matter much because it would be unlikely to be left in a disconnected state for long enough to hurt anything. But at the same time I thought the same thing about how I hooked up some stuff on my battery and I left a load connected before the BMS that also had a very small current that I thought wouldn't be an issue actually did end up causing me to lose a cell because it over-discharged it. After having that particular battle scar it just makes me nervous to have any kind of "load" (regardless of how small) that isn't controlled by the BMS because you might be off on a 2 week vacation when the disconnect occurs. And if the disconnect also prevents the MPPT from recharging over time that 1mA of current is going to keep pulling that low cell lower and lower. It may not ruin it, but that cell will now be "weaker" and lead to further balance issues in the pack later on. These very large cells are hard to keep balanced as they age. My cells are only 200Ah so I imagine it would be even worse with 280AH. I had to add active balancers to my cells to help keep them balanced, especially since I had to replace a couple cells and so now those new cells are much "stronger" than my other cells which makes balancing even harder. "I'm not sure if the BMS auto calibrates to the degradation of the battery. Maybe your's does that?"

[@gkeith4120]

@@JeremyAkersInAustin I have two 300 amp hour 48 V battery banks with 200 amp DALY. For each. I was considering putting them in series with a 500a shunt positioned before my 30k inverter. However, my inverter has two DC terminals (which I suspect are connected inside the unit each other). At This point it seems to make more sense to eliminate the shunt and rely on my Bluetooth from my each BMS.Any guidance would be appreciated. (Great explanation you are).

[@Dave_Gurman]

The Shunt is a two directional current measuring device. My thought is to use it at the cabinet with battery input on one side and load, inverter and chargers on the orher. Measuring in going and out going current flow. At the battery connect BMS - Master Switch - Fuse.

[@OffGridGarageAustralia]

Another option I didn't thought about. Arghhh

[@randylenart9674]

If you put it in before the BMS you know what it's using to doing good do it your way 👍😎 you payed for it

[@peterrock2838]

That is a very good point to move the Victron. One other thing; why not place the battery shutoff first, between the battery and the Victron shunt so no power goes to the shunt when idle.

[@JoeThePotato147]

The only possible problem I see with switching before the shunt and bms is loosing your programable settings in both, but I could be wrong could be worth testing I might try it on my daly bms and bmv 712 later today.

[@OffGridGarageAustralia]

Thanks Peter. I need to see how much cabling effort this will be. I tried to keep it as simple as possible. But totally agree, it would make sense.

[@twsconsulting]

BMS consist of two parts one for balance and monitoring and the other that is a “switch” to disconnect you’d battery from the rest of the world. Of course pre-made batteries have the shunt at the end because you ere not allow to get inside of it. At the end as long as you connect all the sense wires direct to the cell, the rest is a “glorified” SWITCH and it can be connected at any part or the installation. By the way the only reason why you can’t connect it in the positive side is because internally the circuits use the b- terminal as the negative and connect it to the positive side can result (in some cases) on a fire works show. In my opinion the 3dr option is nicer and you have less cables, which is les “resistance” :)

[@OffGridGarageAustralia]

Great, thank you.

[@energiasolarparatodos128]

Muito bom seus vídeos! , parabéns

[@surplusdoctor]

you are correct about Battery, Shunt, then BMS. try that.

[@howardadams4072]

Yep #3 option, but I would also shorten the lead to the switch from the BMS. Make it a smaller pig tail.

[@OffGridGarageAustralia]

I thought about that but does it matter a lot? Also if I need to change the config again to #2 I've got the cables ready... Maybe later down the track I will shorten these cables and make it look nice.

[@willtaylor8904]

None of the solutions are 100% but then nothing in life is, there always has to be compromise. Downsides as I understand are; #1 MPPT charging and sub-optimal data due to no shunt info, #2 voltage drop at shunt causes sub-optimal MPPT charging and potential shunt data drift due to BMS idle consumption, #3 potential low voltage cut-off override due to shunt idle consumption. So I guess you just have to consider and compromise on the least likely situation. So for me I would go with #3...

[@OffGridGarageAustralia]

Great summary, thank you :)

[@rogergalley2779]

Noting the specification sheet for the EVE280N monthly self discharge rate is quoted as ≤3% I reckon that approximately equates to an internal self discharge rate of up to 12mA. It is therefore more detrimental to leave the battery cells in a disconnected state of low voltage for this reason alone. As this self discharge will continue at a rate of up to 12x that of the 1mA of the SmartShunt, that is the dominant danger. It reinforces Option 3 as the preferred.

[@saidt.8200]

You may want to add an active cell balancer inside the box in addition to the BMS

[@rudikaatz317]

1.Hindenburg, 2.shunt, 3.BMS ist the right version (logisch)sehr gut von dir erklärt! #3

[@OffGridGarageAustralia]

Thanks, Rudi 😉

[@GrizOnTheTrail]

Continuing to learn a lot from your videos. You mentioned the "important loop". Could you explain its importance? Thank you...

[@OffGridGarageAustralia]

Sorry, that was just my sarcasm 🤗 There is no loop necessary for anything in this regards...

[@GrizOnTheTrail]

@@OffGridGarageAustralia After I posted the question, I thought "maybe he was just being a smart @$$"...hahaha. Again, very much enjoy your vids!

[@ronpeters4126]

Why are you switching the negative side? That will leave the fuse hot in the off position when you may need to service it.

[@CollinBaillie]

Great discussion, and I thought so when I first watched this one. #3 seems the best hookup when using a single pack in the system, like this battery 1.0. Battery -> Shunt -> BMS -> Disconnect -> Bus Did you ever cover the change of design philosophy when you went for the 3-pack battery 2.0? If you're using a single shunt but 3 packs in parallel, it obviously MUST be after the Bus! For a multi-pack-in-parallel, #2 would require multiple shunts: Battery -> BMS -> Shunt -> Disconnect -> Bus #3 also would require multiple shunts: Battery -> Shunt -> BMS -> Disconnect -> Bus It would seem option #4 is required: Battery -> BMS -> Disconnect -> Bus -> Shunt So that all packs are paralleled on the Bus and THEN you have the shunt counting coulombs in and out from the system. I figured I would comment here now that v2.0 is done, but you'll get people watching the back log of videos, and they may like to know before deciding their hookup based on this single-pack design.

[@hbmike47]

You mentioned that the charge controller is programmed to use the shunt rather than it's B+/B- leads for voltage sense. I would suggest number #2. The BMS directly to the battery and the shunt afterwards. This way the Charge controller reports correctly if the BMS disconnects the batteries and doesn't try to push current. The voltage drops you are concerned about don't exist when current drops off near full charge so don't impact charging function at all. If you go with option 3 you create a very confusing situation if the BMS fails or disconnects the battery. The charge controller will see battery voltage that is actually disconnected, and will attempt to charge it. Meanwhile what's actually connected to the charge controller leads are the load side so the current will actually be pushed into the load. but the voltage and shunt data still show no current and the (disconnected) battery voltage. That's a very confusing set of meter readings when the BMS is tripped and not what you want to see. I'm also of the mind that you should consider mounting the fuse opposite the switch and connect it mid pack where it can offer some protection against a short mode failure of internal components. Where you have it now, all it's protecting is the wire connecting to the whatever the load is.

[@OffGridGarageAustralia]

Great points. Thank you.

[@surplusdoctor]

your switch goes on the positive terminal with your fuse inline of the positive

[@OffGridGarageAustralia]

That's a bit of cabling effort... I was planning to have the switch in the negative string in this order: battery-shunt-BMS-switch

[@aussiescotty2950]

Any thoughts on cleaning terminals & applying conductive gel between terminal & busbar/ring terminals?

[@OffGridGarageAustralia]

Nope. I just connected the cells with the busbars. I think we're overthinking some processes here.

[@alanb76]

I'd be skeptical about the actual ratings of the fuse if there is no manufacturer's part number to reference specifications. You could always test one into a hard short and see what damage there is. Some fuses explode and make a mess because they really aren't rated for the voltage and don't quench the arc quickly. I don't think your batteries would mind a fun test. :)

[@stevebailey1682]

Maybe someone else has pointed this out, but your shunt is a very low ohm resistor (in the milli-ohm range). To measure current, what is being measured is the voltage drop across this resistor. If one side of the shunt is at ground (connected to negative battery terminal), the voltage drop is simply the voltage on the other side of the shunt (call it V1). Call the battery side of the shunt V2. V1 - V2 = voltage drop. Since V2 is at 0 volts, Voltage drop is V1 - 0 = V1. That is why a shunt is right next to battery...so only one measurement needs to be made. Your smart shunt might measure the voltage on both sides...therefore it could go anywhere...but I would check. I know my cheap $20 shunt needs to be next to the battery because it only measures voltage on one side.

[@5UPRAH]

Yep - #3 as you have already done :)

[@OffGridGarageAustralia]

You need to catch up 😂

[@ChetCheng]

Put in anderson connectors so you have quick detach for the battery box also do what victron tells you

[@fillg]

👍 I was going to recommend anderson connectors as well. I will definitely be using them on my system as soon as I gather up the money to build it

[@ChetCheng]

@@fillg I made mine out of old 18506 batteries that I got from battery hookup made 2.4kw batter for less that 200 bucks

[@OffGridGarageAustralia]

Can I connect a 35mm/2AWG wire to an Anderson connector?

[@fillg]

@@OffGridGarageAustralia Absolutely, if you get the ones that are made for 2AWG

[@dobrzpe]

i honestly see this as a moot point - especially if you plan on having more than 1 battery pack in your bank, you would want the shunt outside the battery pack... OR you would be putting a shunt on EACH pack separately? no, i don't think so. so for 1 pack, it's a good thought experiment and get's ya thinking, but if multiple packs are involved in your bank, BMS should always part of the battery pack, THEN shunt. cheers!

[@richardsirois6975]

Just read most of the comments. Amazing to see the majority now voting for #3. Where they the silent majority in the previous vid?!?!

[@OffGridGarageAustralia]

I was expecting #2 as well...

[@mrl22222]

there's actually a reason they use sand filled fuses in these high amperage low impedance applications. If there is a dead short, the fuse will "melt" ...but one of the way it does this is by vaporizing the metal in a plasma ball. this vaporized metal is then re-deposited on the inside of the fuse, recreating the circuit, only this time with a lot more metal available. This is particularly problematic in very high amperage DC applications, which is why you see the max ratings for DC less than half of the AC. The sand disrupts the plasma and prevents the plating recreating the circuit..

[@OffGridGarageAustralia]

Great, didn't know that... Thanks for sharing.

[@zigzagzac]

Hi there, btw you are able to calibrate the shunt for BMS losses in the shunts settings so I would still put it after the BMS as a rule. If the BMS only takes voltage measurements and doesn't have low and high SOC disconnect then it won't make much difference. The only thing you would slightly compromise is over current protection but that's so minor its not a worry. At the end of the day the SOC reading you get from the shunt is only an estimate so I would stick to convention, unless there are other benefits?

[@Banszi1]

#3. Just to be safe I would cover the exposed metal parts of the shunt. Does the Victron Shunt have a idle draw (through the voltage sense lead)? If so the BMS LVD would not trigger in this setup.

[@OffGridGarageAustralia]

Yep, current for the shunt is

[@Muppet_Interfector]

For a battery used and disconnected then left like that until next use like one used for camping I would place the switch first. This is to prevent the battery from discharging through the bms and shunt. You might have to get a few more low current poles on your switch to interrupt the current draw of these devices. For an energy storage system where the batteries are always going to be charged by the PV system the minor current draw of the bms and shunt will not matter. In your case I would place the shunt before the bms. The shunt is going to be the most accurate measurement instrument in that setup, I would want it to provide the most accurate information to all the other components that are relying on the information provided by the shunt.

[@OffGridGarageAustralia]

Thanks Gabriel.

[@archangel9568]

Old way of doing it - a Smart BMS does everything and there is no need for a shunt

[@Muppet_Interfector]

But he has a shunt and he needs it to interface to his other Victron devices and he can monitor ir on his mobile phone. Yes Smart BMS can do some of that too.

[@TheGalifrey]

I will be doing the same as #3. Other youtubers think otherwise and that is their choice.

[@jmaus2k]

1-3 will all work. But I still would suggest #4 using the shunt to just measure charging loop or inverter usage. You won't know how much you are charging or how much is going to the inverter if you just have the smartshunt on the battery loop. Since you already get data from the BMS, put it on a different loop.

[@offgridwanabe]

True

[@petervanbeekum]

I agree, why use the shunt to get more or less the same data as the BMS? The MPPT charge controller will work perfectly without and in previous videos it was made clear that the top charge is going to be well below 3.65v so why does the alleged accuracy of the shunt matter? If you absolutely want to us a shunt put it in another loop like the inverter so you can monitor consumption accurately (the BMS only shows the sum of wattage in & out)

[@OffGridGarageAustralia]

That's what someone said a few videos back. The shunt and BMS will double up and give the same information.

[@mjlentjes]

#3 + Schalter und Sicherung am positiven Kabel

[@itsfpv2561]

#3 , inline with Victron's recommendations, though don't think it makes that much difference. Question: Should you shorten the cable lengths were possible to reduce loss?

[@hommerdalor6301]

Hello. Where it is possible, you do that anyway, but too short cable can be difficult to connect, and if you don't get the crimp connector fitted perfectly at the first time, you can cut a bit from a too long cable.

[@OffGridGarageAustralia]

I'll show this in the next video (or the one after). I took the BMS apart to show what;s inside (😋) and I could shorten the cables if necessary. But do 10-15cm make a huge difference? Not sure if it's worth the hassle, crimping and effort.

[@Sanwizard1]

Wouldnt putting the shunt after the BMS also enable to find any issues with the BMS using the shunt data? If the shunt is before BMS, then the BMS itself is not monitored at all. This is very interesting stuff. I am building a similar setup. I think the switch is overkill though. You dont need an on/off switch in the battery itself. The switch should be on the positive lead going to the inverter. It makes it more flexible to add another battery for the inverter. I will be placing two 16S 280ah strings, with two BMS's, and 2 of the cheaper AiLi shunts in a larger metal Truck toolbox. I like the idea of placing those before the BMS. The AiLi monitors will monitor each 16S battery in the box, so two holes cut for two monitors. The 2 sets of battery cables then go to external fused and switched terminal blocks. My Victron shunt then sits on the negative of BOTH batteries in the steel box, so I get to monitor the batteries directly with the AiLi shunts, and the batteries and BMS's with the Victron shunt. So actually 5 monitor points all together. Two Heltec BMS's with bluetooth, two AiLi shunts with physical monitors of each battery on the battery box, and then the Victron to monitor everything. I am glad I saw this upload, as I was going to put the AiLi shunts after the BMS's in the box, but tou explained it very well, and it makes sense.

[@AndyM...]

Hey Andy, Would it be advisable to put the switch on the battery side before it goes to the BMS/Shunt, so that you can isolate the complete system in case of failure/maintenance etc. as it stands once connected there's no way to isolate the BMS besides physically disconnecting the battery. I'm also assuming you're going to connect the charge controller to the neg/pos leads going to the battery box at the actual inverter ? Looking good ! And like with Windows there's usually a few different ways to achieve the same result and they all work ! Andy :)

[@OffGridGarageAustralia]

Yeah, that's probably the best way to do it, but the cabling gets messy. I have to sit over the box and think about it for a while how to get this done properly and in a clean way...

[@jasonm914]

Have you considered using a Victron Smart Battery Sense to monitor the battery voltage? I believe these can be setup to transmit voltage and temperature information via Bluetooth to the Victron Smart Shunt and MPPT Solar controller. Then you would be able to put the Shunt on the load side of the BMS.

[@OffGridGarageAustralia]

No, I haven't considered that. I looked into it but decided the shunt is the better product. Don't tell me I have to look again 😑

[@jasonm914]

Look for a KZbin channel called Freely Roaming, he used one in his battery build and explained what it does and why.

[@tomcherry9093]

Alternative #3

[@aussiescotty2950]

Hi Andy, I'm thinking BMS 1st, because shunt will continue to communicate with the charge controller when a lo voltage shutoff has occured therefore charge controllers will get incorrect information & deplete the battery further as well if the shunt is before the BMS. Should the shunt be oriented in a particular plane ie fins facing vertically not horizontally, what does victron say. How accurate do you want this system to be, knowing you you will probly monitor it on an hourly basis (haha).

[@OffGridGarageAustralia]

Good point. I will have a live feed of all data on my phone 24h a day, hahaha. The shunt has got less than 1mA of usage, so I would say that negligible. The orientation does not matter for this device.

[@waynewallace3316]

Hi, I built my Daly BMS with similar cells as you have in the video. I have the older Victron smart shunt installed with the system. I cannot get my system to work with the Shunt installed and then the BMS after that. When I take the battery and BMS out and put it on charge the BMS and bluetooth comes online. If Input it back in the camping trailer the BMS is not working. Am I missing something here?

[@jws3925]

Been watching intently this series of videos. I have a question and will apologize in advance if you clearly mentioned this. I looked on your website and went back thru a couple videos to find the answer but could not nail down which BMS you settled on. Could you just clarify what one you settled on and why or direct to the video where you did so?

[@OffGridGarageAustralia]

Thanks for your feedback. I haven't linked the BMS yet, the video is still to come. Very soon...

[@Muppet_Interfector]

QUCC from what I saw on this video. What model?

[@Jsak666]

#3 all the way

[@andreiapostolache3487]

3# opción shunt first and switch on positive firt before fuse

[@philipbotha6718]

With this setup it makes the most sense to have Battery, Shunt and then BMS. If the shunt also had had a negative reference lead like the BMS, then it wouldn't really matter.

[@lennieadi]

Hi Andy. I'm intrigued with this project. Question. I'm completely non technical but would love a diy set some day in the future. If one had a all in one inverter, can you still have the smart shunt and victron mppt CC? Would the inverters mppt conflict with standalone mppt and smartshunt?

[@OffGridGarageAustralia]

Thanks Lennie. The AIO unit has a solar charge controller, inverter and all the cabling already included. It just needs the solar panels and the battery to work. It's very easy to setup. The unit needs the correct settings for the battery and you're done. I probably buy one in the future to show how it works.

[@lennieadi]

Off-Grid Garage Hi Andy. Thanks for reply. I get the concept of how it works although programming parameters for lithium would be a challenge. My question was basically with an. AIO set up I would I still need a smart shunt and addition mppt if I had a large array and large battery bank?

[@Gnagnie]

Ok, the Shunt has an Error LED ? then I would place it outside of the battery-box, to see the signals. If the Victron Solarloader comunicates via blue tooth the metal-cabinet may cause problems? May be the cabinetdoor(outside) will be a good mounting place for the Solarloader ;o) --- This type of fuse is used in elektrical fork lifters by more than 48V ! What will happen? a fuse 100A will transfer 1000A for what time till the metal is vaporisised ? Blowed out by his own steam ! Warning: Don't observe such sparks without suitable eye protection!!!

[@OffGridGarageAustralia]

All Bluetooth... it shows all stats on the phone/website. Comms between MPPT and shunt should be fine, they are close together

[@Gnagnie]

@@OffGridGarageAustralia That's nice if BT works, but in case of an error I prefer the status leds ! The MPPT is monted inside a faraday cage or will you let the cabinetdoor open?

[@archangel9568]

A good solar battery controller you can dispose with the high price vitron gadgets

[@waynehobbs5175]

Number 3

[@caspleijsier8983]

Hi, nr3 please, the reason is you want to measure the correct voltages of the battery and the cells. Every other way of connecting gives a less accurate measurement. Good luck installing!

[@897aa33]

Correct voltage measured by balance cables not the b-. B- used for load connect disconnect scenarios.

[@guohuaye12]

I might worry about the fuse arcing if I was using lead acid batteries in an area that was not well vented, because of the fumes. JMHO Yes, I like the Shunt before the BMS, I'm using a Separate Port BMS because of my Solar Panels Charger. Yea, OK #3.

[@paulbaker3144]

What seperate port BMS are you using? I haven’t found any.

[@PentaxLife]

I think #3

[@ricardomarcelino8388]

your explanation #3 makes all the sense, but I have no knowledge to fully state or understand the downsides of this choice. It is in fact true that a battery has a shunt after the BMS, because shunts are not included in the battery, so, there is no option than doing that way, is there? By the way, I noticed you refer the blue sea main switch is for 40V max! You are running a continuous (more or less) load true that switch of over 48V (16x3,65V=58,4V maximum charge) which is above 20% maximum rated current. Isn't that a safety issue??? Loving the set up and looks of it. All the best and keep it up!!!

[@OffGridGarageAustralia]

Thanks. 48V is the nominal voltage for the switch to disconnect under load. I'm not planning to use the switch during operation. It is more for maintenance purposes.

[@thomasgriffin4714]

#3, that is what Victron recommends. BTW, the voltage rating of your switch is less than your battery voltage. Even if you only charge the battery to 3.4 v/cell, the total voltage is 54.4v and the switch is rated for 48v.

[@JeremyAkersInAustin]

The switch is labeled for it's "nominal" voltage. All "48v" batteries are well over 48v when fully charged. A "48v" lead acid pack (Which is what the switch was originally intended for) would exceed 60v during equalization charging.

[@dobrzpe]

@@JeremyAkersInAustin i believe this has been debunked - as in Blue Sea Systems rated the switch for the (rare) 36V system. it's listed as a 48V "max"... this has been, and probably will, be debated for a much longer time. with that being said, i'm contemplating putting one of these switches on my 48v LiFePo4 bank myself...

[@nigelcharles511]

They ought to quote the switch as capable of 60v as if 48v was the limit the limit would be exceeded with every 48v system when it is on charge.

[@philwhitehead8706]

No low voltage disconnect for the shunt also. If put in after the bms the bms will turn off the shunt if the voltage gets too low

[@OffGridGarageAustralia]

Good point!

[@marccohen5579]

Sunt 2nd as Victron advises. How can the BMS monitor the cells if you add in the 50 MG Ohm of the shunt? They do it with their batteries.

[@OffGridGarageAustralia]

Victron actual recommends to have the shunt directly at the battery. The BMS does have balance leads through which it senses the voltage.

[@jean-baptistemackenson5630]

What is the brand/model of the charge controller you are using?

[@OffGridGarageAustralia]

The Victron SmartSolar controller can be found on my website: off-grid-garage.com/solar-charge-controller/

[@rocktech7144]

That is a series battery only about 900 amps at short not 10,000. #4 From battery to shutoff to bms to shunt to terminal. This isolates battery from potential bms failure as well.

[@OffGridGarageAustralia]

LiFePO4 cells have a super low resistance and can easily deliver 10-15kA when shorted. This is not lead acid tech.

[@Swenser]

I'm afraid to use lithium because of the unlikely event they go up in sparks. Lead acid don't fire up and not likely to burn down a log cabin. I'd recommend them since costs are down for those.

[@OffGridGarageAustralia]

I guess you're driving an car with internal combustion engine and around 60l of highly combustible liquid inside? I would be much more afraid of that!!!

[@gggg-xv7nb]

The shunt is so beefy it might as well be power cable. My BMS is only 120A and thats enough for charging current and my smaller 700W inverter, but the big 3000W inverter cannot go through the BMS. So i connected the shunt first, then the 3000W inverter, then BMS, then the smaller loads and all chargers. 2 weeks so far. No problems

[@OffGridGarageAustralia]

But if the BMS turns off your inverter will keep working?

[@neliosamch3195]

#3 will be the most accurate reading because it will measure the consumption of the bms, perfect for a small system. #2 is best for a large system because the consumption of the bms is very low. The problem with #3 is that the chunt gets very hot and has to be as far as possible from the batteries and the plastic box.

[@aussiescotty2950]

By my calcs 100A x 0.05 volts would be 5 watts at maximum output. How hot is too hot?

[@neliosamch3195]

@@aussiescotty2950 I don't understand your calculation but my 48v system over 4kw gets very hot to touch.

[@aussiescotty2950]

@@neliosamch3195 Interesting. Your 48v syst 4kW = I=P/V = 4000/48 = 83 A. The shunt (Andy's) has a volt drop of 50mV @ max current. Woops my bad. P=IV yes. P = 500 x .05 = 25W. If a draw of Andy's max 100A, shunt would drop 10mV so P=IV= 100 x .01 = 1 watt is what the shunt power it would consume. Please correct me if I'm wrong.

[@neliosamch3195]

@@aussiescotty2950 okay, 12v system the amperage is higher than the 48v system for the same load. I have the same shunt 500A 50mv from victron. The time I noticed the hot temperature in the shunt was in the summer with 90⁰ ambiente temp while charging 2 ev and an a/c on which was less than 100A. Cable is 1/0 awg. Other shunts overheated has been attributed to loose connection and defective terminals. Mine has neither. The only thing abnormal is no voltage drop between shunt terminals. So my assumption is based on experience and not technical data.

[@hommerdalor6301]

Maybe you don't see defect on the connectors, but how perfect are the contact surface between them and shunt posts? Are they with a miror finish, or as they were, just bolted on, and go? Imagine only some points are touching, it's as if you have thin wires, the higher the amps the more temp, little fusion points are created, the terminal could become loose, and further arcing damage the surface. If you thight the connector with damage surface, the show goes on again, the connector will get loose, endless loop. Always trouble and energy loss.