Tesla Off-Grid-Charging. Overheat, Overload and Disconnect... IT WORKS!

[@vinniec5286]

Amazing job, Andy! Remember back when you said you were going to do some testing and there would be successes and failures? As for cells 6,7,8, if it were me, I would swap their position with 3 other batteries, but leave the bus bars in place (don't relocate the bus bars). Recharge, rebalance, and try it again. If you get the same result then it is the bus bars at cell location 6,7,8 but if the problem follows the batteries, it is the batteries that are the culprit.

[@GBCobber]

Sry, I should have read your post first. didn't even realized how old this thread is.

[@computerguru]

Whenever I help my electrician friend with wiring, he always applies a coat of antioxidant paste to the connectors. This is especially important when you have dissimilar metals such as copper and aluminum. You have the perfect setup to test the effectiveness of using antioxidant paste between the aluminum battery posts, the copper buss bars, the ring terminals, and the fastening bolts. Your local electrical supply store should be able to supply you what you need. For additional details search for NO-OX-ID made by Sanchem and read their paper on their antioxidant paste. You may find that you can drop the voltage loss to a minimum and also provide long term healthy low loss terminations. Aluminum is notorious for oxidation over a short period of time and the use of antioxidant paste is highly suggested especially in high current circuits. Keep up the nice videos!

[@blackswansystem]

These old episodes are awesome 👏

[@maverickmacgyver]

I love it when a plan comes together!!

[@bruch63]

Es ist genial was Du in diesem und allen folgenden Videos geleistet hast: Es haben wohl viele den Traum ihr Elektroauto mit selbst erzeugtem Strom zu betreiben. Ich bin neidisch. Chapeau!

[@OffGridGarageAustralia]

Danke Dir vielmals. Es ist noch ein weiter Weg. Im Sommer klappt das nun schon gut aber im Winter ist einfach nicht genuegend Sonne da um die Battery zu laden. Ich arbeite daran und kurbele die Produktion kraeftig an...

[@henvan8737]

For those earlier comments when using the MPP solar inverter like my Pip5048GK. I have confirmation from MMP solar this morning I can simply install a jumper between ground and neutral without any damage to the inverter.

[@OffGridGarageAustralia]

Good info for anyone using this inverter.

[@mrzed6597]

I think MPP Solar inverters have an inner relay for that. So when your main power is down relay closes.

[@smdindustries3699]

PIP 5048MK definitely has internal MEN relay. Uses upstream supply when power is on the input terminals. Switches in the internal if upstream supply is disconnected.

[@JeremyAkersInAustin]

Do I get to say I told you so about Smart Shunt? :) The shunt only knows that the battery is supposed to have 280 amp-hours, and only because you programmed it with that knowledge. If you have some weak cells (or cells that get weaker over time), the shunt will claim you have plenty of energy left as the BMS begins to enter shutdown mode. I would raise the under-voltage cutoff and release voltages. Once a cell gets below 3.2 it's very close to fully depleted. Under 3.1 and it's got less than 1% remaining. Going down to 2.5 is putting a lot of unnecessary stress on the cells. I cutoff at 2.9 and release at 3.15 The BMS going through that off/on/off/on loop is the main concern I would have with this set up. Since the BMS can't control the load (inverter) and charger separately it's faced with this dilemma: The cell is under voltage so turn off the relay. But now it senses charge current trying to come in (Which is the only way to *fix* the undervoltage problem) so it has no choice but to turn the relay back on, but that turns the inverter back on, etc. This sort of seesaw could go on for hours (and could potentially end up damaging equipment.) On the polishing question: I've never done that. But I use fortune cells which have massive lugs which I use an impact wrench to put on the nuts. (M13 nuts, quite large) Because the nuts are so large and torqued down so tight I think that avoids any issue with the uneven surface of the terminal. On your cells I notice your screw heads are quite small and don't even cover the hole in the bus bar. So you might just need to get some washers to add some surface area from the top and provide more even pressure.

[@OffGridGarageAustralia]

Much appreciate all your tips and comments, Jeremy. Thank you. I think I can set a delay time for the BMS to when it closes the contactor again to avoid this flip-flop. The Inverter was without load when this happens as the Tesla charger takes 15-20s to ramp up again. I'm not sure why it had this behaviour. If I set the low voltage disconnect to high it may trigger it already in peak situations? Is this something you have come across?

[@JeremyAkersInAustin]

@@OffGridGarageAustralia I haven't had any issues with that, but I will confess that I think some of it could be because the Orion BMS has a really ingenious method of finding the actual cell voltage even while under load. They have some method of finding a cells "open circuit voltage" even while the battery is under load. When you configure voltage cut offs, it compares against this calculated "open circuit" voltage rather than the "under load" voltage which you'd read with a standard voltmeter. This helps remove the need to adjust the voltages lower to avoid tripping them while under load. It now occurs to me that other BMS units may not have this same functionality. The folks at Orion explained it to me once and they basically said that the BMS checks the voltages very rapidly, hundreds of times per second. It then compares the voltage readings against the current flowing in/out of the battery and adjusts the open circuit voltage accordingly. In the case of inverters, since inverters basically pull a rectified sine wave from the battery there are roughly 100 moments per second where the current is actually at or near 0. (Because the inverter pulls more current as the sine wave increases in amplitude and pulls less when the sine approaches 0 in magnitude). Because of this the BMS is able to simply wait for current to get near 0 and use that to find the open circuit voltage. If it can't get that measurement it can calculate it by adjusting for the current flow. Either way: The Orion basically accounts for any voltage drop that would be caused by load so it's less of an issue. However I just went back and checked the history log of my BMS and looked back to the last time I had a good solid discharge (Overnight where solar wouldn't affect the voltage readings) and even the "under load" voltages sagged by only 0.04v when drawing 20 amps from a 200 Ah battery. (Roughly 1100 watt load.) I've honestly never noticed any significant voltage sag on the cells even under much higher loads, not enough that I would necessarily worry about them tripping the under-voltage protection at 2.9v. A voltage drop of 0.1v would be quite unusual I would think.

[@rcinfla9017]

Aluminum only takes about 10 seconds to build up a significant oxide layer and it continues to thicken over time. Aluminum oxide is an insulator. Clean with white vineger, then isopropl alcohol then immediately clamp the terminal together. Keep vineger out of the thread holes on battery terminals, only wipe the battery terminals surface. The pressure needs to asserted over the whole area of battery terminal. You may need a thicker washer to ensure distributed pressure and not just under the smaller diameter surface nut. Your fingers have sodium on them so avoid touching battery terminals. Too much contact resistance is a common problem with DIY'ers.

[@photonphil873]

Very, Very cool!!!!!!!!!!!!!!! Well done!

[@beaker2000]

Looks like 6, 7 and 8 may not have been top balanced all the way. Once the pack is "full", you can charge 6, 7 and 8 by themselves to 3.65 volts to top them off before you start discharging again.

[@OffGridGarageAustralia]

I'll try that.

[@georgedemean2228]

I will only assume that a larger screw head or a large flat washer on the battery terminals will apply more even pressure and contact area on the bus bars contact to terminal, therefore preventing energy losses thru the contact area.

[@OffGridGarageAustralia]

Good thinking

[@FKZNL]

@@OffGridGarageAustralia your busbars seems different some square some round. You seem to have different areas covering the terminals. I would aim to cover the complete terminal. The busbars look small for the max amps these cell can handle according the spec. Super build very informative.

[@grahamjones6712]

@@FKZNL the busbars came with the batteries youd think they were suitable for the spec of the batteries

[@grahamjones6712]

Whats the fuse rated at?

[@FKZNL]

@@grahamjones6712 look at the size, looks small to carry 280 amps. I orderd 4 of lishen 272 ah cells with dubbel busbars to carry 280 amps. 280 amps on that distance would require 50sqmm at least in wire. Compare the lugs he crimped to the busbars 1 vs 2mm in thickness 25mm wide vs 20mm. The klauke lugs will carry the amps the busbars wont carry 280 amps. In this installation its fine. The amps the inverter is pulling he can use single busbars. Dont expect the chinees to deliver to max spec of the cells if you have not stated that you want to pull ful amp rating. I think the advised discharge current is 0.5c so 140 apms. That wil be the amps the busbars are calqulated for. The cells are cheap for a reason one of them is they dont handle high C discharge.

[@teardowndan5364]

Any fuse is going to get warm when passing a significant fraction of its rated current since they are engineered to melt on overload after the i2t delay and for that to happen, it needs to get to the fusion temperature of whatever metal it is made from. If that is a problem, use a breaker.

[@partyboarderdos]

moin andy, please find out why some of your batteries drain faster. looks like you cant use the energy stored in the other batteries due to the cell-undervoltage shut off. all busbars should be also the same

[@hommerdalor6301]

Hello. Yes, I polish all my connections, I also use 4 x 2mm thick copper bussbars in parallel, between each cell. Instead off the standart stainless steel screws, I used brass 4mm x 45mm, shoped the heads off to make studs. I wanted to use aluminium, but couldn't find them yet. it would be safer for the cells poles wich can be torqued to 8nm and thats not much, finger tight. car spark plugs need 20nm, when i was apprentice, I could tight the plugs by holding the 3/8 extension. Many youtubers have pull the threads of their poles out. From that the importance of perfectly mating surfaces of contact. If you use 10mm bolts, you can use all your force to tight, and so maybe have a good contact by compression, and deformation of the copper, but if the connector needs to be unbolted and rebolted, now the surfaces are mated in one position, and it will not be easy to find, so, you bolt again with all your force and create new damage to the surfaces ect... I know it seems exaggerate, but use your imagination as a microscope, or imagine the surface bigger, compare touching a table or puch your hand against it. Metals are some king of playdough,a little harder. Cheers Nice test. :-)

[@peterrock2838]

I saw another video posted so I muted the Superbowl to watch this!!! Your videos are much more interesting. I'm thrilled you got the Tesla to charge. That is awesome! The 3kw inverter is doing very good. Now I need to buy a Tesla.

[@Adlay_Eetswah]

You must be a Chiefs fan 😉

[@OffGridGarageAustralia]

Hahaha, wow, Peter. Better than Superbowl... 😂

[@peterrock2838]

@@Adlay_Eetswah Brady fan! I knew the Bucs would win anyway...

[@Adlay_Eetswah]

@@peterrock2838 The Man is a true Champion 👍🏽

[@upnorthandpersonal]

The aluminium terminals have a layer of aluminium oxide on them, and that's an insulator. What I usually do is sand it with a fine sand paper and something like NoAlox to prevent this oxidation. The EVE datasheet also mentions this: "9.3.9 Connection a) Polishing the pole with abrasive paper before use, otherwise it would cause bad contact or failure."

[@jeremy87turbo87]

I have also heard of some medals or terminals having coating on them or oxidation that you need to get off to have better connections same with wire sometimes when you strip the wire back there will be a little bit oxidation or something underneath or coating on the actual copper that sanding it off will help have a better connection

[@clooncellan]

Yes I would use 800 grit grade sand paper on all the bus bar ends including washers and studs' this will improve things a lot....

[@aussiescotty2950]

Congrats Andy. All you need now is a generating push bike to make use of all your exuberance. Well done, good job. Cheers.

[@OffGridGarageAustralia]

I'm trying to calm down now!

[@mrzed6597]

I think you broke the balance when you let the BMS burn away power from the cells. Charge all cells to top again and let the discharge empty them. Sanding the terminals is I think in the EVE cells manual.

[@TheDtwilson]

I like homade copper buss bars, because they are softer material and you can use all the surface area, and I like studs with nuts, less ware and tare on the internal aluminum threads, thank you for your videos

[@colinbeckitt9495]

Great job Andy, with only a few minor issues your self built, off-grid solar + battery Tesla charger solution has bought entertainment, admiration and inspiration to hundreds of aspiring off-grid developers. Just as well you didn't waste $1200 on a 3k Victron inverter, from personal experience I can guarantee you would not have been able to output >3000W continuous for even 1min as it would alarm and instantly shut-off. My MultiPlus II 48/3k used to alarm then die within 1min with a >2800W load and it used to get really hot (>45C) in ambient 30C with a constant just 1800W (about its max continuous load). Your cheaper Chinese unit is simply fantastic outputting >3000W for 15mins in 30C ambient at

[@spiritrider3348]

"Very Interesting " - its worth watching multiple times, great work👍👍👍👍

[@lnxpro]

You should do a capacity test on those 3 cells that discharged faster than the other cells. Are they from the same source? That shouldn't happen. I am suspecting those cells are of lower grade or something.

[@OffGridGarageAustralia]

Yep, they came in a pack of 4, cell 5-6-7-8 are from the same delivery.

[@bishwagit]

Yes, I am agree with you. That should not be happen.

[@wadebrewer7212]

@@OffGridGarageAustralia I do t have the time to watch everything, did you top balance batteries before final assembly?

[@kevinmills5293]

Before you condem those cell, remember these cells “fall off the cliff” when fully discharged. The other cells are also pretty much depleted. As I understand it Lifepo4 cell should not be discharged below 20%. Your cells are way lower than that.

[@martinhilburger2971]

I think Basen ship randomly used cells to the consumer for better prices but this destroy the whole package of 48V system 😕

[@FutureSystem738]

Cheers Andy. I hope you can sort out why those 3 cells have fallen off a cliff much earlier than the others - certainly a bit of a worry, especially if they were well top balanced prior. Good luck with it.

[@jahrmichi]

very cool vid, you should get a infrared camera. I bought one from amazon FLIR One and it is amazing. This is about 200EUR and can be connected with usb-c to smartphone. It shows you every thermal weakness especially if you work with high current.

[@OffGridGarageAustralia]

YES, I'm looking at some of them for a while but they were more like $600...

[@robertkosovich1232]

NICE job! Congratulations!

[@OffGridGarageAustralia]

Thank you! Cheers!

[@landonferguson7282]

Great success. Just a bit more fine tuning.

[@howardadams4072]

I am old school. All my battery connections are copper busbar and brass threads and brass washers and nuts. You also need to maximise the contact area and use battery or electrical contact grease. Just look at the electrical resistance of the metals you are using, even solder is a bad conductor compared to copper or brass. Plating on the busbar can have a high resistance. You might consider going completely brass threads into the cells and wider copper busbars to cover the whole cell terminal contact area, held down with brass washer plus spring washer and brass nut. Just a few microohm's at each terminal will add up over all your connections and cause a significant power loss. You need a bigger fuse, as your battery is 280A. The inverter has its own fuses, so the battery fuse is only protecting the battery, so 350A could work.

[@mikehortin9964]

Perhaps replace the fuse with a circuit breaker. The over heating fuses that I have had in the past did not harm the fuses themselves but they will deform the holder, locking the fuse in place and damaging the mount. Do the larger fuses you mentioned better handle heat build up?

[@edc1569]

The BMS will protect from normal overload events, the fuse is there for when it's all gone to crap, the fault current to clear could be ten's of thousands of amps, you won't find a suitable circuit breaker for this application that can deal with that.

[@dash8brj]

You should really set your recovery voltage to 2.9/3.0 volts. Cycling the contact like that in the bms with a heavy load connected (tesla) will eventually ruin the contacts. I just ordered a smart shunt and a victron 12V inverter to play with. Keen to see how long my 180Ah cells can run my desktop PC for. I finished top balancing them the other day - they're still a little out of whack (19mV) but my DALY bms and active balancer should fix that.

[@OffGridGarageAustralia]

It didn't even reach the 2.7V before it kicked in again. Also, the load is zero at the beginning as it takes the Tesla charger around 10s to ramp up again to full load.

[@dash8brj]

@@OffGridGarageAustralia Of course, it was just the inverters fans kicking on as they are on a dumb temperature controller. So it was the SmartSolar feeding current in that was causing the bms to cycle.

[@edc1569]

@@OffGridGarageAustralia the inverter will have a load of capacitance on the input, don't you get a big arc when you connect it to the batteries? That Arc is what the BMS is seeing every time it switches, its not good and it will die quickly, in an automative application where those BYD relays are used there will be a pre-charge circuit to charge the caps before the relay switches in.

[@jameshudspeth7622]

Hey Andy. Exciting times.. but, you've got crappy buss bars. I've seen this problem with my unit, and eventually had to replace with pure copper bars that i custom cut and drilled. You should not have individual cells hitting 2.5V so quickly, as that's a sure sign of some bad connections. I have the same batteries, and really had to work to get good connections with them. I actually "roughed" up the terminals and the copper bars with sandpaper, just to add contact area to the connection. I NEVER had cells collapse as fast as yours did, and it's a great thing you've got a BMS that shuts down at 2.5V, as programmed. I'm sure it saved your batteries from being damaged.. Bottom line, get rid of those cheap Chinese buss bars, and make your own. I used 1/8 inch thickness, and 1.25" wide pure copper.. works perfect, and i have NO individual cells collapse as you did.. really, they should all track together on charging and discharging.. Really enjoy your videos.. thank you.. Jim

[@OffGridGarageAustralia]

Thanks for your comment, Jim. These cells 6, 7 and 8 actually didn't have the 'bad' connections. These were on the other side of the pack. I have to check again.

[@w3bb0y]

@james: be careful as you're using copper busbars and aluminium terminals, this causes galvanic corruption in the aluminium. Ironically the best material may actually be aluminium or steel, the best ones I've heard are copper bars plated in another metal.

[@kevinmills5293]

The bus bars are perfectly fine for 50-60A. The issue is getting good contact with the battery terminals. With only 6mm screws, the surfaces have to be as clean and flat as possible as you cannot torque the screws down very much. I helped install a 4S2P system on a yacht and running the bow 7kW thruster (500A) the BMS clearly showed the bad bus bar connections (the cell voltages were all over the place). After flattening all the bus bars with 800 grit wet-n-dry paper, the losses where drastically reduced and were consistent across all joints.

[@jameshudspeth7622]

@@OffGridGarageAustralia Please do check the connections. You probably actually had "good" connections on those three cells, which is why they drained so quickly. Webb0y's comment is a good one, as with mis-matched materials you can get some corruptive effects given time, humidity, and high power. Remember, it's not just amperage, but power (wattage). 3000 watts is a lot of power, and can move mis matched materials given substantial time. I was using the same technique that Keven discuss's below, with 800 grit, you're actually still roughing up the surface, although to the touch, it seems smooth. The best monitor is still cells that drain quickly- they have the best connection.

[@ChetCheng]

Keep up the good work on the video and trial and error

[@OffGridGarageAustralia]

I'm observing and learning

[@ChetCheng]

@@OffGridGarageAustralia and doing a dam good job keep it up my good sir

[@praveenjain6025]

Very nice information go ahead sir,I assume that silver plated copper bus bar with copper large head bolt and nut will give less MV difference and fuse is warm due to unusual raise amp flowing through the fuse

[@OffGridGarageAustralia]

Yeah, these fuses are not the greatest for these high current applications. A lot of resistance. I should have installed a proper 2-pole DC breaker with 100A. Next time, I guess 😊

[@GBCobber]

My vague understanding of BMS function suggests to me that low capacity cells should located adjacent high capacity cells. I'd assume there would be an optimal arrangement that best balances the balancing. :)

[@jamess1787]

Glad I was wrong about the inverter! Looks like she's hitting her rating!

[@ElectricCarAustralia]

The feeling of charging your EV from your own off grid system is the best. 👍 Well done on the system.

[@ccspack]

What you’ve done Andy, is absolutely amazing for minimal investment. I’m lucky enough to be able to do all my EV charging during the day, from my 28.8KW system, using the Zappi & charging at up to 11KW straight from the Queensland sun. It all happens so quick. But you have your PFIT to protect, while it lasts.

[@edc1569]

28.8kW - wows!

[@itsfpv2561]

As always, love watching your adventure. Couple of questions...Isn't the BMS supposed to balance the cells whilst discharging? By the way what is the impendence of cell 6 and are some of your bus bars differing sizes?

[@mrzed6597]

Only Active balancers do balance all the time. Passive balancer only burns away top cells power while charging.

[@Adlay_Eetswah]

@@mrzed6597 I did notice in one of his previous videos, Andy’s bms balancing cells while he had the battery constructed but not under load or taking any charge, perhaps he needs to tweak his balance settings via the app?

[@mrzed6597]

@@Adlay_Eetswah He set the cell V too low. Should be 3,65V. And because of this the BMS started to balance, burn the load. And so the cells where not equally discharged and gone out of balance. Simple.

[@Adlay_Eetswah]

@@mrzed6597 That’s an interesting argument you put forward.

[@mrzed6597]

@@Adlay_Eetswah I know. He made a perfect top balance ... and then let the BMS broke it. After top balance BMS balance needs to be turned off for at least a full cycle. And only when charging up enabled again. Now I think he charges the pack. I HOPE not to 3,65V, only 3,45-3,50V. So the cells first reaching it will not overcharge. Passive balance with 0,2A is a joke if you charge with 10-20A. Cells in 3,65V and charging are overcharged. That causes degradation.

[@sergioduarte887]

Congratulations again Andy, this is a great entertainment channel, didactic and very useful for lovers of solar and sustainability. Keep it up! I see that now you have some tunning to do but everything will end up spectacularly done, I'm sure. Since the purpose of this channel is also to teach, don't you want to take the opportunity to explain how you can touch the battery terminals, the main-switch, the positive and negative terminals of the inverter ... without having your hair standing on end? And it is not because you have few for sure. :-)

[@OffGridGarageAustralia]

This fine tuning is killing me!

[@rickcooperjr62864]

the inverter isn't overloaded 3049w input would be around 2700w output due to the conversion loss in the inverter of around 80%-90% so the 3040w into the inverter is not overloaded.

[@georgedemean2228]

Very good observation 👍

[@clooncellan]

Personally I would use a min 6kw inverter' to be on the safe side with these higher loads

[@ebenwaterman5858]

Fuses are just resistors. Or resistive heating elements if you will. Don't magnetic circuit breakers exist? Also what are the buss bars actually made out of. Prepping the contact surfaces is a good idea. They make a special paste for hooking aluminum wire up to terminals. Maybe something similar is available for this application.

[@ebenwaterman5858]

@DIY Projects Try silver. Just kidding. ;) Copper is good. :)

[@ebenwaterman5858]

@DIY Projects Hmmm. :) I bought 1/8" copper stock on Amazon. Ni is convenient because it doesn't oxidize quick like Cu. I wish I knew more.

[@wloulittle]

Hi. If you have one of your batteries with a small lug under the larger strap, this is a no no condition. Heat will not dissipate from the battery to the strap but will heat up because there is not enough surface area with the lug. I don't know if you use a torque wrench or not but you may consider it if you do not have one. The one thing that should also be a No-No is to have any kind of Grease between the battery and the straps. I saw that you had a picture of some carbon Grease in your homepage and I would recommend never ever use any kind of Grease on your surfaces as they act like resistors and will heat up your connections and not be as good of a connection as a bare dry surface. Any signs of fingerprints and you have an oil contaminated surface. My 24 volt system uses a bus bar carrying as much as 300 amps through it from my solar array. I have 33 panels at 260 w per panel. 8.6 KW of solar on the roof. My system is by Outback Power Systems running 28.8 volts.

[@OffGridGarageAustralia]

Thanks for your comment Lou!

[@Dirt-Diggler]

I enjoyed that chap ! got as excited as you (well very nearly) lol really enjoy taking this journey alongside you, thank you for making the vids! especialy as i'm builing a shed at the mo and intend to solar power it, just for tool charging, not on the scale you have acheived !

[@pieterbarneveld9518]

Well done 👌

[@OffGridGarageAustralia]

Thank you! Cheers!

[@ArthursHD]

18:43 Was it just a BMS test? Or are cells 6, 7, 8 actually worse in terms of internal resistance or capacity?

[@jeromehelbert2628]

As much of a point that you’ve made in past videos about not charging all the way to 3.65V is limit wear and tear on the battery, I would definitely not let the cells discharge to 2.5V as that is way more damaging to the cells than 3.65V. The discharge curve for LiFePO4 is super steep below 2.9V or so (pretty sure it’s even more steep than above 3.3V) that you really aren’t getting that much extra power. I wouldn’t advise discharging below 2.8 or even 2.9V. Also, about the fuse. You figure that thing has to get hot enough to melt the metal above it’s rated temp, so yeah it’s going to get warm. No way to get around it. Another vote for eliminating the fuse because it’s burning power and the BMS should already be handling that role.

[@OffGridGarageAustralia]

I know, 2.5 is last resort before things get really worse. You mean no fuse at all? Just the BMS taken care of overcurrent?

[@jeromehelbert2628]

@@OffGridGarageAustralia yeah, I guess it depends on the BMS, but I believe most will shutdown the relay/FETs in an overcurrent situation. I don’t know if your model does that or not.

[@jeromehelbert2628]

Yeah, found the product link on AliExpress and it does mention over current protection, but that’s all it says. No idea how effective or reliable it would be

[@JeremyAkersInAustin]

@@jeromehelbert2628 I wouldn't trust the BMS to handle it. His BMS is using a latching relay (Which is why it draws so little power when on) and they can easily fail / get stuck in the closed position. I always include a fuse (along with circuit breakers) because a fuse will not fail in the "on" position the way a latching relay or circuit breaker can.

[@edc1569]

You really don't want that relay trying to open with a 10,000Amp fault current running through it, you still want a fuse, you could probably size it quite high though.

[@mobeale]

Recovery Voltage (20:10): Hi, Can the "Recovery Voltage" be set higher so the batteries can recovery much more, then will the BMS initiate a restart? This would be perfect for charging my vehicle with a small system! Your video's are fantastic. Thx

[@waynehobbs5175]

Well done, pure sine (sign) wave. Great job. Your excitement is contagious. Cheers buddy.

[@ebenwaterman5858]

Looks like you have 3 cells that are significantly lower capacity. The way BMS work is they dump the higher cells through a resistor. That's a lot of dumping for 3 low cells. This has been encountered by so many DIY guys like yourself. I have suggested matching cells by total Capacity and Internal Resistance. In your case (and mine) I would augment low cells with small external cells in parallel with the weak cells to increase their capacity to match the higher ones. Gotta run it for a while to see just how significant the mismatch is. :)

[@OffGridGarageAustralia]

I need to do more testing. I wasn't planning of discharging that much. The situation just came along so I let it go to see what happens. I'm a tester ;)

[@sammaimas155]

Andy, where did you run the green and yellow ground wire from the inverter to?

[@swolath]

what about some conductive paste

[@OffGridGarageAustralia]

Video is following...

[@mkroach59]

you may want to put studs into your battery cells rather than bolts to get a better connection, maybe this would help with the voltage drop issue ? Just an idea. Best wishes.

[@OffGridGarageAustralia]

What would be the difference apart from going into the thread a tiny bit deeper?

[@mkroach59]

@@OffGridGarageAustralia just using the whole depth of thread, then backing off a quarter turn, maybe more contact area and a stronger connection. Maybe ? I really don't know. I know you did a top balance, maybe a bottom balance might help ? I'm confused about this, I've seen some people do top balance and others do bottom balance. What's the difference ? I think a top balance should be good for a solar application and a bottom balance is better for a high discharge rate. Is this correct ? What I'm reading says top balance is fine with a BMS so you should be fine. Bottom balance if no BMS and high C rates.Thanks for posting these videos, very entertaining and educational too !

[@JeremyAkersInAustin]

@@OffGridGarageAustralia Studs may actually help quite a lot. using more of the thread means more contact for the screw (Which means more current carrying capacity for the screw) and it also means you can torque things down tight more safely. By inserting a stud and then using a nut to tighten: As you tighten the nut you know you've already got the stud all the way in, so you can torque down a bit tighter without worrying that the threads of the battery terminal will strip. Where as with a screw you may not have the threads all the way in as you begin to increase torque, and then the only way to get more threads is to apply more torque which can result in stripping threads because you may start applying torque before there's sufficient thread contact to handle that torque.

[@SigmaWorksSW3D]

You have built great system ;) Great goals made true

[@OffGridGarageAustralia]

Thanks!

[@pederb82]

That bms is not usable in your setup at all. It disconnects both the charging and discharging at once. When the load is still present when it detects a charge, and the charge is not large enough to supply the load you get a on/off loop. You must NOT let an inverter or any other gear turn on and off at full load like that. The internals undergo an enormous torture both electrically and mechanically when you do that! Best case is you will get bad solder joints on the main board of the inverter in a short while. Worst case is you blow up caps and fet's(power transistors). So whatever you do - do NOT let it loop like that!! If you are going to keep using that bms, you must turn off the load at once when this happen so you dont get a loop situation.

[@OffGridGarageAustralia]

I know. You're right at the beginning of my videos... there is more... 😉

[@pederb82]

@@OffGridGarageAustralia Yeah. Ive seen that. ☺️ great channel. Keep up your great work. But do think about replacing that bms. I’ve seen you use it in your later videos too. :-D Just a friendly tip. :)

[@OffGridGarageAustralia]

@@pederb82 Still testing it in my production setup. I think it is important to show it off and make others aware of the pros and cons of such devices.

[@stanislav0803]

0.5v it is too large, you need passive balance board to the each cell individually. It help to charge battery to full. Мой английский не очень))

[@unliyou]

It is normal for the BMS to turn on again when charging voltage is present but no discharging is allowed until the 2.7V setting is reached.

[@OffGridGarageAustralia]

That was my question too. That's the first time I tested it. Might be that special BMS only?

[@unliyou]

@@OffGridGarageAustralia , my DALY BMS 16S does that, the SRNE 5kw inverter also have the same option to set the voltage. I believed it is done that way that battery cell should not stay long at that level of charge that is should be charged right away to a certain level and discharging should only be allowed at some higher level only that the chemistry have stabilized to a better level after being drained.

[@mrzed6597]

@@OffGridGarageAustralia It is a Common port BMS. It has to turn on to let you charge the battery. For a DC coupled system like this a Separate port BMS would be better. That can allow charge without allowing discharge.

[@netpok]

@@OffGridGarageAustralia The problem is that it's a relay based common port BMS and the relay can only be on or off. So either needs to be a separate port with two relays or a mosfet based, because mosfets have polarity, so it can decide which way is allowed.

[@kuhrd]

Even at 12A charging to the Tesla you were only drawing about 2880 watts from the inverter. The inverter was drawing 3040 watts from the battery so the inverter is about 94-95% efficient. The inverter should be able to put out 3000watts continuous from the AC output and you will see 3150 watts or so being pulled from the battery. The beeping almost sounded like it was coming from your meter possibly an auto power-off feature. The high voltage drop across some of your battery busbars could be many things from the torque of the bolt to the contact area on the terminals. Your busbars could also have some variations that affect the quality of the electrical contact. I would verify the torque of your bolts but if they are not really heating up at max load then you are likely fine and should just check them over every so often to look for hot spots as time goes on.

[@OffGridGarageAustralia]

Yeah, correct, I looked at the shunt, not the inverter output.

[@jmaus2k]

You should have a BMS with separate charging and discharging relays. Then only inverter disconnects with low voltage and solar disconnects at high voltage. You could also move the solar to the battery terminal instead of bms lead. Then set recovery voltage much higher so inverter wont cycle.

[@mrzed6597]

Yep. Separate port BMS is better for a DC coupled system. Can charge without discharge.

[@colinhercus7765]

I wonder if using liquid metal on contacts could reduce the resistance. This is stuff used on CPU coolers.

[@OffGridGarageAustralia]

Uhm, I'm not sure what this does under high currents.

[@diydan6045]

Those must be grade B batteries, getting so unbalanced, correct?Maybe try a quality fuse and holder to see if it makes a difference.

[@francisdoll1253]

ANL fuse is not for continious high loads,it's for peak currents in audio car systems,use mega fuses instead,it's better for high continious loads

[@OffGridGarageAustralia]

Yeah, I don't have high loads on this battery. 60A max is the highest...

[@alexandergunda8916]

Great Test! And also the frogs at the end, perfect! I'm concerned about those 3 bad behaving batteries. as others already suggested, a capacity testing of those would be a good idea. Also i think that an suplementary active balancing would help a little bit - not much because of the big loads, but at least during charging.

[@rotspitze1]

Herzlichen Glückwunsch aus Bayern ! Congratulation, now you have a real Solar Tesla.

[@DIYwithBatteries]

Batteries inverters shunts again inverters to bucks and the to batteries. I thought why don't just connect the both batteries in parallel. So we don't need any inverters and Voltage drops. 🤔

[@daveduran-up6kn]

Can the Tesla charge directly from any DC power? A DC-DC stepup buck converter could push up your battery voltage😊

[@OffGridGarageAustralia]

Yes, it can charge from DC power.

[@dickthompson8018]

Congratulations!!! Well done, so cool! I was smiling with you .

[@nilobualoy117]

Do u have a link of shunt and battery monitoring youre using on it?

[@OffGridGarageAustralia]

All links are in the description. Shunt: off-grid-garage.com/battery-accessories/ BMS: off-grid-garage.com/battery-management-systems-bms/ Battery monitor: off-grid-garage.com/measurement-tools-and-analisers/

[@elpihendi6793]

Why it would not possible charge tesla with that inverter pls ? I didnt understand

[@OffGridGarageAustralia]

It has to do with the safety system of the Tesla EVSE. It needs a proper grounding/M.E.N. link to work.

[@vspadrver]

55mV x 60A = 3.3W ... I guess the fuse is doing fine. Great job

[@neliosamch3195]

You need a 10A Lipo Lifepo4 LTO Battery Active Equalizer to keep all batteries cells within 5mv. One of my battery terminal was getting hot and hot wire resistance because it wasn't tight enough.

[@OffGridGarageAustralia]

no no no.

[@grahamjones6712]

Your supposed to shout "BANG!" as you plug it in.we once had a sparky work on some electrics in our house and he kept shouting "BANG" as a joke as he stuck his screwdriver in something even tho the electrics were off at the mains.

[@PeterMilanovski]

Whoa! Look at the various states of charge in the batteries!!! For the sake of one battery reaching it's cutoff voltage, the rest become useless! Now you might understand why I chose to go with a more complicated solar charging setup! As I have mentioned earlier in a previous post, I charge my batteries from a separate charge controller per 12V block! So basically that means I am using 4 charge Controllers in the series connected 12V batteries to get the 48V required output! But of course my batteries are single 12V blocks with their own built-in BMS. If I was using your batteries, I would be using a charge controller per cell that would charge each cell to it's rated voltage individually from the other cells! Automatically keeping each cell in balance! The charge controller will stop charging the cell if it's full without affecting the other cells that are a bit slower. Your BMS on the other hand, when one cell reaches full capacity before the others will begin sending energy that the cell should be getting to the big resistor to burn off that energy so that current from the 48V can continue to flow through the whole battery bank in order to charge the rest! From what I saw inside your BMS, there's only one resistor and that to me means that it can only discharge one cell at a time! So what happens when only one cell needs to be charged while the rest are done? Does it stop charging and begin systematically discharging the rest before reconnecting the solar input to finish off the complete charge? Who knows? I don't have this particular problem with my setup! This is the biggest problem with charging batteries in series, some BMS systems are able to discharge multiple cell's simultaneously to keep them in balance... This is where you should see the rise in temperature of that 10W resistor as you charge the cell's, it might be setup to top balance which means that the resistor will only be used when the first cell reaches full capacity... You might not see a temperature rise on the case of the BMS but I sure that if the PCB board was exposed, you will see it working when the first cell reaches full capacity.... I could be wrong but I feel that you might need another 48V string to add in parallel for some extra capacity to cover those overcast days and to increase current sharing which would increase the lifespan of the batteries at the same time! I personally would want as less current draw from each cell as financially possible! It's all about redundancy and lifespan in the end! Hope this helps you will your quest for energy independence!

[@gd2329j]

Cells 4, 5-6-7-8 are not balanced ( or these are lower capacity cells ) . The spec sheet should give a tolerance for cell capacity . It maybe +/- 5 % ( This is basically how the 90% DOD is derived ) . This bank is well out if it's just not properly balanced …. So the question should be at what point does the BMS start balancing cells . What method is it using ( At the end of charge & a input voltage is needed ) . How fast can it balance a pack that's 5 % out .

[@OffGridGarageAustralia]

We will see with the next test...

[@johnnycagiva737]

Ich habe die Pole poliert und hatte keine thermischen Probleme. Beste Grüße aus Berlin

[@OffGridGarageAustralia]

Wie hast du das gemacht? Schleifpapier?

[@johnnycagiva737]

@@OffGridGarageAustralia Ich habe mir 1000er Schleifpapier mit einem Tropfen Öl genommen und dazu ein Endmaß.

[@HiltonT69]

I like that the passive BMS is able to show up what are possibly bad cells, however even if those cells are slightly higher resistance than the rest, an active BMS would give you a higher output from the battery as a whole. This makes me lean towards an active BMS.

[@OffGridGarageAustralia]

But on the other hand you will never identify these bad cells if the get constantly balanced from the other cells.

[@eildertferwerda2163]

What is the plan, charge the battery during the day and then charge the car from the battery overnight.

[@mikeleopold7028]

how much cycles you have on this pack, andy? ich muss mal deutsch weiter schreiben, da mein englisch nicht ausreicht. ich belas mich mal zu zellchemie lifepo4... die cellen brauchen so 5 cyclen um an anode (oder kathode) einige lithiumlemminge abzulegen, daher ist die kapazität der einzelnen zellen erst nach 5 zyclen genau zu messen (wird etwas weniger, aber wohl stabiler in der differenz der zellen zueinander).....vielleicht ein gedanke.

[@mjlentjes]

Ich denke, da die möglichweise schlechteren Zellen 5-6-7-8 ja zum verfrühten deaktivieren des BMS führen, ist die sprichwörtliche Kette nur so stark wie ihr schwächstes Glied. Sprich die Kapazität der besseren Zellen kommt nicht (nie?) zum Tragen. Hmm.

[@OffGridGarageAustralia]

Muss ich testen. Das war der erste Test...

[@martinhansen9517]

Hi Andy Really great job you have done so far. You might have 3 faulty cell's. If you want to watch you battery from every where you could build in a old smartphone or tablet, hook it up to your Wi-Fi and install vnc.

[@henvan8737]

Looks like you have a number of faulty cells, as the wont track during heavy discharge.

[@offgridwanabe]

I am impressed with that inverter frog approves lol. those 3 batteries are not performing as good so I would charge and balance the bank and then change the position of those 3 low batteries put them at the end of the chain then discharge again to see if it is the set up or the batteries. Cheers

[@OffGridGarageAustralia]

Yep, I also get my capacity tester soon, so can test these three cells individually.

[@michaelb.1192]

You may need a additional balancer to balance the batteries with a higher current than your BMS can do.

[@howardadams4072]

Oh! and your BMS connections. I would remove from under the busbar terminal as they are not helping I would use a bigger copper busbar and drill and tap the busbar in the mid point but off centre and use a brass screw or thread for the connection to maintain low resistance. Also I notice the elongated holes in your existing busbars, these are limiting your contact area.

[@normantomlins1742]

I would be concerned about the clicking that's happening when the BMS is switching on and off. You need a way to allow the solar to charge the batteries up to a safe voltage level before enabling the inverter again. Otherwise, you might get some arcing on the relay and it could stick closed due to the cycling. Great job on the whole system!

[@edc1569]

Let me guess, when you were top balancing the cells and shorted things out, that was nearest 6, 7 & 8? I think you need to get them all parallel and balance charge them again, give it at least a week, maybe longer with those alligator clips.

[@DazGeary]

Great result, I am as happy as you. :-) Would be great if all the busbars had the same resistance. Copper would be too expensive eh. Unless you buy a length and DIY. definately try the polishing or even copper washers between. whatever works.

[@millzee60]

Did you measure the voltage drop across the Bluesea isolation switch? I'd use a dielectric paste on the bus bar connection as well as making sure the bars are flat. Something like Duralac.

[@Adlay_Eetswah]

So Andy did you come up with a solution to the rouge voltage from the previous video?

[@OffGridGarageAustralia]

Well, I ordered a different inverter. I'm not happy with these 70VAC on the DC side and don't know what trouble it may cause later on...

[@BischesseHunting]

Wow, thank you for bringing us All to a New Level. How can we support you?

[@moglivishal6656]

You should replace these bus bar to thick copper one

[@jasondevine6014]

Everytime that relay kicks in and out with high currents it is burning the contacts. Great inverter. I did scratch the terminals. You can get Smart fuses basically MOSFET based with a trigger circuit. Might be better. I just use ANL fuse.i still think you need to top balance some more with a 3.4v or higher setting in BMS.

[@OffGridGarageAustralia]

It's only 40A for a 300A relay, so...

[@jasondevine6014]

@@OffGridGarageAustralia true. I am thinking you are on 24v for some reason. I am finding with my system there is a tricky balance between cell balancing voltages (3.4), charger setpoints(3.45) and not letting the BMS high volt cutoff(3.6) bang the relays. A little in-balance takes a large amount of time to get right before the voltages are even across cells. I think these cheaper cells tend to mismatch easily. This is top balancing of course.

[@rickcooperjr62864]

your cells are not balancing properly you got a major issue you need to address with a active cell balancer or something because those few cells dropping so low and others staying so high the BMS cannot handle it and this will kill the cells and cause all kinds of issues.

[@neliosamch3195]

A 10A Lipo Lifepo4 LTO Battery Active Equalizer will keep the cells within 5mv.

[@OffGridGarageAustralia]

If the cells are not good, a cell balancer will not help at all.

[@neliosamch3195]

@@OffGridGarageAustralia lifepo4 batteries have some peculiar characteristics as acting crazy above 3.4v or below 3.2v. Give each cell a name and switch their place. If always is the same cell them is bad but usually I have observed all of them do the same crazy behavior and the only remedy I have found is taking charge from the highest cell and giving it to the lowes charged cell by using capacitors on the Lipo Lifepo4 LTO Battery Active Equalizer. Regular bms just doesn't work that way.

[@rickcooperjr62864]

@@OffGridGarageAustralia you already tested the cells and got full capacity from all of them and you top balanced them either your BMS isn't balancing properly or you got other issues a active cell balancer will help this issue theyre not expensive and wont be a issue for the BMS.

[@romorris]

My meter beeps if i leave it on and drives me nuts sometimes..you sure it was the inverter beeping ?

[@OffGridGarageAustralia]

Mine beeps only once before it turns off. It's a long beeeeeeeeep.

[@andresgodinho]

You really shouldn't use the inverter with that earth connection... I explained this in a comment on your last vídeo! And the cells are way off on terms of matching capacities... Great work doing all that!!

[@OffGridGarageAustralia]

I know, the link is dodgy and only for test purposes, proof of concept.

[@mjp0815]

Your frogs are green and love to sit in the sun all day... but they don't do photosynthesis, maybe you are giving them ideas...

[@DSmartLife]

Looks like those 3 cells are not in a good shape, esp cell 6. What do you think?

[@OffGridGarageAustralia]

Need more testing. I'm recharging atm so will see then if they also hit the top first, then they are bad, I guess.

[@DSmartLife]

@@OffGridGarageAustralia Yeah, please let us know. Thanks.

[@saurabhverma3332]

One question- why didn't you use DC circuit breaker of 100 rating instead of fuse which was at 55 degree. And you can decrease resistance of battery connector strip by adding the bolt washer ring ( that has hole in middle) to increase surface area of contacts.

[@OffGridGarageAustralia]

🤷‍♂️ I just used a fuse...

[@computeification]

🐸 👍

[@mondotv4216]

Hey Andy - have you changed the ground yet for your off grid setup so it’s not connected to the grid ground? Shame about the crook cells.

[@JPHER217]

Haha at 12:00 thats how we roll! :)