Multiplus II is a 3KW inverter, Induction stove is 1500W load, Air Fryer is potentially 1800W. AES has always been disabled.
All AC loads are caried over dedicated 10AWG wires (Boat is only 44 feet long as well so none of the cable lengths are “long”)
12V Battery Bank is 1200Ah of Lithium connected with 4/0 AWG cables with well crimped on ends. Positive lead to the inverter is about 25’. Ground lead is about 10’
Air Fryer suffers a palsy of clicks and false starts for 10-20 seconds before it starts working as expected. You can watch that here.
This does NOT happen on my AIMS inverter that is 9 years older, air fryer just starts up and runs without a problem.
Most of the time if I put another load on the Multiplus II, the air fryer will start just fine. Turn on a induction cooktop to boil some water and it starts just fine. Turn on the water heater (600-900W) and it starts just fine. The problem seems to be if the unit isn’t loaded down, it can’t run this little Air Fryer.
I’ve had some messages back and forth with Pike but so far no solutions that changed anything. Firmware is upt to date as of 2 days ago.
Alarms that are going off are for overload (when I am running the water heater + air Fryer, and DC Ripple (but only occasionally, not with every startup of the air fryer).
AIMS inverter never has a problem unless it is running the water maker at the same time which togheter is more than the inverter is rated for.
DC voltage is stable according to the voltmeter but I don’t have an Oscope so I’ll have to take the word from the inveter that there is DC Ripple.
How do I fix this so my primary inverter (Multiplus) will just work for my boat
I have tried some chokes on the AC end but they made no difference.
How long are the battery cables?
Are your batteries daisy chained or individually bus bar connected?
DC ripple can even be caused by a washer in the wrong place. A terminal not torqued, a battery disconnect that is losing its touch.
While the other inverter was fine, you have thoroughly disturbed the set up.
All 4 batteries go to a battery bus via their own wire which is then connected to the house bus via 4/0 AWG (about 6" long). The shunt sits between the battery bus and the house bus so it can monitor state of charge. Individual battery cables are 2/0 AWG less than 4’ in length. Batteries are 300Ah Lithium 13.2V in the evening 14.6V while charging. Problem is happening when the batteries are more than 90% full.
The other inverter is connected permanently to the house bus with 6’ 4/0 AWG cables, I use it for making water and running the AC and it can stand in for the main inverter if something goes wrong there.
I will try looking for the Ripple using the AC setting of the voltmeter. The DC side does NOT show any significant voltage drop from this very large bank. Less than .3VDC during these events and INCREASING the load on the inverter with the electric water heater makes the problem go away so if it was all about voltage drop then it should happen all the time, not just when this air fryer is turned on by itself.
DC ripple can even be caused by a washer in the wrong place.
If this were true, then the Multiplus should be more resilient, just like the AIMS inverter.
While the other inverter was fine, you have thoroughly disturbed the set up.
This is a new inverter install that has exhibited this problem since day one.
Ok. You do realise we have sight unseen-zero knowlege of installation- shot in the dark diagnosis here? So very general things to looks for based on experience. No idea what has been done or how.
Does the aims inverter even monitor for DC Ripple? Even if it did, the multiplus has a large transformer inside so a complety different topology to the aims. It will behave differently because of physics. Nothing to do with resilience. Or the chicken dance.
We are talking close to 300A at 12v to run a 3kva load. Thats not playing. A cable run over 2m is long
It is only when I drive the current north of that 150A by adding load that the inverter performs as it should.
Your contention that a misplaced washer is the cause of DC ripple is what I was refering to as I showed in quoting your reply.
I don’t know if the AIMS unit monitors for ripple or not, they both have a similar sized transformer, AIMS just delivers the power on the faceplate and the Victron doesn’t. I will try removing the shutoff switch to see if that is where the problem lies. The Cabling looks acceptable length wise as far as I can see based on the charts I’ve found.
And this problem is happening during a time period where heat cannot an issue.
It is indeed a lot of power that’s why the busses are 600A and the inverters are not running over the same wires. The only thing on that 4/0 AWG wire is the victron Multiplus II and the only time it has a problem is when then air fryer is drawing power. I think the Victron is letting noise from the AC side back thru to the DC side, but I’ll know more later today.
[quote=“Ludo, post:5, topic:15993, full:true”]
Your dc wiring is too long and too thing.[/quote]
The long wires are 4/0AWG as the post states
The 2/0AWG wires are from each battery to the battery bus and are only 4’ long. None of the wires are too thin because each lithium battery provides only 200A and 2/0 AWG is good out to 20’. I’m only using 4 feet of 2/0AWG as the posts have said.
Victron specification stops at 30’, for this lenght 2x 2/0 is specified.
I guess if you wire a second set of 4/0 in parallel it might work
No, this would not be a good idea since the fusing around that setup would be impossible to get right. That’s why I’m not running 2/0AWG cable more then 4 feet.
Here is the table and the 4/0 AWG is 120mm2 cross section.
1800*12 is 2160W or 150A which is less than the 180A and right at 10m in length (roughly 32.5’)
If I add load to it (600-900W), the clicking stops, so when I exceed the limits in that table it behaves properly, when it is just the air fryer and no other load it fails and at that point I’m 30A under that table.
The cable run is the same gauge for the aims (which you have said is smaller in capacity) as for the victron.
Regardless of the blue sea guide, the installation manual for the victron recommends double the wire for the length run. Since they made the product, I am sure they are in a better position (than a generic guide) to know what the inverter specifically requires to function without problems and maintain a long service life.
The power pack capacitor bank and transformer in the MP make it necessary to use the minimum recommendation at keast -not the load you are expecting to run. The transformer core and pack have to maintain their ‘resevoir of electrical charge’ to work correctly (such as power assist etc), that needs a decent supply.
As an FYI DC ripple will kill the caps in the MP.
Either a different choice of inverter may be something to consider or a wiring upgrade.
Second FYI the cable length in the manual refers to the return trip length.
1800 w / 12 v / .94 inverter efficiency = 159.57 amps. Call it 160 amps.
160 amps @ 12 volts on 4/0 cable over a 50’ round trip distance will have a 3.35% voltage drop. This is above the 3% drop we want to see. My recommendation is that you don’t use wire gauge charts and instead use a wire gauge calculator such as the one found at Bay Marine Supply. Note that this does not take into account the distance from your Battery busbar to the batteries. But it should.
It can be argued that using 12 volts in the above calculations is wrong. It isn’t wrong but it is pessimistic. I prefer to assume the worst, which often results in using a heavier gauge of cable.
As suggested in multiple posts. Double up the 4/0 cable going to the inverter.
With regard to Victron not being “resilient”, I would turn that around and say that Victron is protecting itself better than the AIMS. If you put the AIMS in the same cable position (25’ away) it may or may not throw an error. If it didn’t then I would say that would probably have a shorter life than the Victron because the Multiplus is more picky about the power.
So, I doubled up the 4/0. 2 10´ pieces on the positive side, 2 12´ pieces on the negative side using both terminals on the Multiplus II.
Each battery is connected to the bus bar with a 2/0 cable that is 4’ long for a total of 4 batteries capable of 800ADC continuous current.
After doubling the wires:
First time I turned on the Ninja, no clicking. 2nd, 3rd, 4th times. Clicking, just like the original video, absolutely no change from before. I turn on the water heater (~1000W resistive element) runs over a separate 10/3 cable and then try the Ninja, no problems, no clicking, just starts up. I can restart it several times while the inverter is already under load without issue. There are no warnings about ripple or overload (and the Ripple warning was sporadic at best. I can trigger the overload warning with 2 separate induction cooktops and that is indeed more than 3000 watts combined.
I think the thing that no one is listening to is that this happens when the inverter is NOT loaded but does not happen when it is loaded. This video shows the clicking until I load it up with an 1800W induction cooktop heating a pot of oil in which case, it works as expected.
I’m ordering a different brand of air frier and I’ve opened a ticket with Ninja but I’d swear this is a firmware problem in the Multiplus II.
The Genset doesn’t have this problem but it’s 11KW so no real surprise there. All the connections are tight, I just reattached them today (I pulled the power lose while wiring things up since this modified config lacks a switch or any fusing and those positive and negative lugs are too close together to wire it up hot, even with that little dividing wall between them.
Something seems off and I don’t think it is the wires. Here is the wire sizing from Bay Marine Supply. The wire is from Ancor and the connections were crimped with a hydraulic crimper.
You can’t use 22’ for the length. You must use the round trip length. According to your earlier post, the, “Positive lead to the inverter is about 25’.” Plug 50’ into the calculator.
I don’t have enough electrical background to explain why another load seems to stabilize the system. But that is strange.
So some improvement heading in the right direction.
This is the key to the cable sizing… Round trip.
It is actually a common trick. (To use a heating element).
Yeah not easy to explain. But it does have one which shows how the hearing element load type helps sort out the chopped wave from the air fryer power supply. (If you were to use an O scope on that you may be able to make some electical engineers faint or vomit.)
The way that the victron power assist system works makes its mechanism different to the AIMs they really are like chalk and cheese in how they work.
Ordered a Black and Decker unit and it works marvelously. It’s a bit smaller than I want but it doesn’t piss of the multiplus and it cooks FASTER. French Fries in 6 minutes instead of 15. Not sure what Ninja is doing but I will be avoiding their brand from now on. I sent them the same video and they are saying I have to send the unit to them to test, but unless they have a victron, it’s not going to happen, this oven works fine on wall power because the grid doesn’t care about the PF of one little air fryer.
I still think this is a shared problem but with the Black and Decker is onboard, I have one less thing to worry about. B&D could learn a bit about UI from Ninja, but at least the engineering part of their product is solid.
