Inverters don't like power tools

[@cramesplays]

Testing like this makes me more likely to buy that product. Physical limitations exist, and if your usecase could stress those limitations, knowing what to expect is very important. Shout out to EcoFlow for keeping in touch and providing the blog post.

[@kirkhamandy]

Yeah, everything has limitations and knowing what they are is the most useful part. Hats off to EcoFlow for engaging at the right level to educate us. I'm with you, I'd happily buy from this manufacturer over a competitor now just for this reason. Understanding is key to looking after one's purchases.

[@Flowing23]

You obviously did not watch Matthias video, the video directly before this one. When he originally discovered this issue and brought it to the attention of ecoflow, ecoflow broke off all contact with him for months. When he was not saying what they wanted to hear ecoflow stopped responding to him. No praise at all for ecoflow.

[@ShainAndrews]

@@Flowing23 EE's were telling him to piss up his own rope. Ecoflow should have done the same.

[@cramesplays]

@@Flowing23 I did watch the video before this one. Their response and blog post are redeeming and that behavior should be encouraged.

[@macedindu829]

Yup. EcoFlow has earned respect with me. IMO, they didn't make a mistake sending a unit to Matthias.

[@rklauco]

Turns out this is one of the best advertisements for EcoFlow that they could get. The product seems quite solid to me. Thanks for doing the work. I like the fact you tested the chargers - as I wrote under your previous video, the failure happens most probably only with capacitive droppers, the switch-mode supplies (even the cheapest one) should survive such oscillations.

[@Ultrazaubererger]

Capacitive droppers should also be ok if they have an appropriate zener diode.

[@ratchet1freak]

@@Ultrazaubererger that zener would need to be able to shunt quite a bit of current to survive kHz oscilations, keep in mind the power going through a capacitive droppers is linear with the frequency

[@mrfrenzy.]

@@ratchet1freak It should be rated to survive short spikes of about 1000V which normally occur on the regular electricity grid, however the standards for this probably don't take into account that the spikes continue for quite as long.

[@Ultrazaubererger]

@@ratchet1freak Not sure but I think with the inrush resistor and other stuff taking some of the load and the time of actual high voltage/frequency being just a few cycles it should be able to survive.

[@davidcross890]

Scary 😱 that you think this is solid petformance

[@MeriaDuck]

I think ecoflow cannot wish for a better advocate. For us techies, knowing the limits and protections especially when examined and demonstrated is very useful.

[@kirkhamandy]

This comment wins, I couldn't agree more. Understanding is key.

[@commodoresixfour7478]

Being honest should be a selling point.

[@Splarkszter]

Bit funny that his last 2 videos about this topic damaged the brand (not intentionally but it's what it did). At least he did a correction; and well yes these things happen, the inexperience and basically not knowing well about their reputation make you want to tell people to avoid it. But there is were the ignorance can become a harmful trouble. Because in my viewpoint that brand seems legit, and we don't know how many people that did see those other videos won't ever see this one and avoid the brand. And we need legit brands, we need brands that reach out and do legit stuff. Always compare results when doing your research.

[@Youtube_Stole_My_Handle_Too]

Only true in the eyes of those who understand how difficult inductive loads are but EcoFlow cannot live from those alone. The error with the review is not taking an approach from it being an impossible task yet making something that works pretty well.

[@PatFarrellKTM]

I would love to see you hook up your o'scope to a new Ford F150 electric pickup truck. They advertise it as ideal for contractors, who often carry saws and boards in the bed of the truck. I wonder if the invertor powering its bed outlets have better or worse high load characteristics.

[@filanfyretracker]

The specs I can gather say the F-150 lightning uses AC Synchronous motors, So id bet the power tools plugs are run off that inverter. Probably a transformer in there somewhere too since id imagine the drive motors are running higher voltage than 120VAC. in the case of the truck if my guess on plug source is right at all is the inverter is just so big it soaks up the hit of a tool starting up.

[@eDoc2020]

@@filanfyretracker They wouldn't need an extra transformer to lower the output voltage because the inverter output is PWM. They'd just use a lower duty cycle. However, IIRC the AC outlets can be active when you're driving. If true, this would necessitate a separate inverter.

[@PatFarrellKTM]

@@zlmdragon. Right, that is the ticket

[@MalinDixon]

​@@filanfyretracker The V2L (Vehicle to load) outlets on EVs come from the circuits that are used at other times to charge the vehicle's batteries from AC supplies. The inverters that drive the traction motors are only connected to the traction motors. The AC charging circuits are run backwards to create AC from the battery. The AC charging circuits use active rectification to reduce losses compared to passive rectification, and by controlling the active devices differently, the charging transformer can be used for discharging. The AC charging needs effective power factor correction (PFC) circuitry so that it minimises the current draw and harmonics when charging at 10 kW or so. A wall wart charging an iPhone doesn't need to worry so much, but an EV taking 10 kW with a poor power factor could cause problems. The PFC circuit actively alters the current during each cycle to make it match the supply voltage, giving the good power factor. Running the same circuit in a different way during V2L allows the sinewave of the AC supplied by the vehicle to be produced.

[@AaronBrand]

I’m trying to figure out what happens if I overload my Ranger outlet with a circular saw.

[@anton4video]

Nice with honest people. Have looked at different models and just bought a EcoFlow DELTA Max. The choice became easier after the previous video. Hope EcoFlow appreciates your honesty as much as I do.

[@spelunkerd]

You make a good point about motors that drive power tools. All have an enormous current spike at startup that is a problem even for mains supply, especially if rapidly turning on and off under load. It always surprised me at how well standard breakers accept those huge current spikes without nuisance tripping.

[@mrfrenzy.]

The standard breakers are designed specifically to accept the start current of motors like this. A 20A breaker can take short 100A start currents. There are also special breakers which are more sensitive and others which are less sensitive.

[@InssiAjaton]

My rule of thumb has been for ages that incandescent lamps take an inrush current of some ten-fold their nominal current. But it lasts very short time. On the other hand, squirrel gage motors also take about ten or twenty-fold nominal current for inrush. But the inertia determines the duration. Some fans or blower may take 4 seconds or more. The fuse manufacturers have published detailed graphs of their normal blow and slow blow fuse types. There are super fast fuses for semiconductor protection, typically characterized by a single number of XX ampere squared * second, matching the characterization of many high power semiconductors. Those numbers are typically valid in time span of 0.1 millisecond to 10 milliseconds. As to breakers, the most common (domestic ones) are based on thermal sensing element. For more specialized applications, you use thermal/magnetic combination sensing. You may also come across pneumatic or hydraulic damping (i.e. delay) in some breakers. On the other hand, for high inertia motor loads, you might see star-delta two-step starting or use of wound rotor motors with adjustable rotor resistance. Finally, you may nowadays see variable frequency drives recommended for the difficult start applications. The star-delta start reduces inrush current typically to 1/3 from the direct start value. However, it also reduces the torque to 1/10 of nominal. If you can live with that tradeoff, you can save some fuses. All that is what we had at my childhood farm, and what then later the professor of motor design explained. The rest I have forgotten long since!!

[@TimMcgimpsey]

@@InssiAjaton Led lighting is also pretty terrible in terms of inrush current. Often 50-100x running current for the first few cycles. Causes big issues controlling large led light banks in factories etc. We split the lights up and use cascading contactors to keep inrush in check.

[@chitlitlah]

If it doesn't get the cord warm, it's not going to get the breaker hot. Typical house breakers are thermal only, so it takes a lot of current for a lot of time to trip them. Even at 10 times its rated capacity, it may take a second or two to trip. Industrial breakers like the kind I've designed into control panels also have a magnet solenoid that trips the breaker very quickly when the current exceeds so many times its rated capacity. Going from memory, a class B breaker which is designed for electronics that shouldn't have strong surges trips immediately at 2 to 3 times the rated current, while a class D which is designed more for inductive loads or other circuits with large surges trips at 15 to 20 times the rated current. Those numbers are likely a little off, but you get the point.

[@eDoc2020]

@@chitlitlah I'm pretty sure standard household breakers have magnetic trip elements. One time I plugged in a shorted light bulb and the Q115 breaker tripped immediately.

[@SilverCymbal]

Such a great video Matthias. Like you I had terrible experiences with Bleutti as a company when they dont get a glowing video, I dropped them as a partner long ago. So glad you did this video!

[@hmspain52]

Very timely video Matthias! The problem a LOT of people face is running AC off batteries that are part of a solar installation. Even with a startup capacitor, the inrush from the compressor startup kills most battery setups.

[@eideticex]

It's overall a problem with electronics. It's relatively easy to go from AC to DC but going DC to AC we at best approximate the AC.

[@putraadriansyah8082]

I think BLDC AC compressor can be a good solution for solar battery use case scenario. But I don't know if such things exist.. all I see on the internet is BLDC refrigerator compressor to run fridges on battery, but I think if we use it for AC it doesn't strong enough.

[@eDoc2020]

@@putraadriansyah8082 There are most definitely inverter-based air conditioning units. It seems to be the norm for mini-split units.

[@PsRohrbaugh]

@@eDoc2020 there are multiple manufacturers of slow-start kits for traditional air conditioners, so you don't have to go replacing the whole unit. I'm not sure if it's a VFD or some other technology, but the one I have (Micro-Air EasyStart) limits peak starting current to 15 amps on an air conditioner that draws 12 amps continuous.

[@johnramirez5032]

@@PsRohrbaugh i asked him if he can hook up what is called a drive to slowly ramp up the motor. Just like the ac soft start kit does. The way you said it was much better. I know someone who sell drives to manage the motors.. i wonder how that would work with the inverter. ? Intresting .

[@AntManBee19]

As an owner of a EcoFlow Delta Max this and the previous video were very informative. It’s nice to see a product put through its paces. Although maybe a little extreme. Nevertheless the company did get back to you and reasonably discussed your concerns. At least I thought so and it made me feel that EcoFlow will stand by their products

[@LawtonDigital]

I think you're performing a great service not only to the public but also to the manufacturers by helping reduce the number of dissatisfied customers. I don't mean that in a snarky way. I'm much happier knowing a product isn't right for me and looking elsewhere than I am buying a product that's a poor fit. Thank you for sharing your time and experience.

[@drewgi7543]

I remember seeing your channel for the first time when you posted the wooden combination lock and your organ, and to think all these years later you're casually doing electronics deep dives on your second channel. Thank you, Matthias.

[@battleshiparmorlord]

Marble machine ftw

[@WmLatin]

I just have to say that your videos continually impress me with your thoroughness and depth of knowledge.

[@JonnyDIY]

I feel your talent is going to waste. This isn't a dig, it's a compliment. Minds like yours are the great innovators of the world. You should do consulting at least, who knows the problems you could end up solving 💕👍

[@patricklyons7683]

They need to make sure in next iteration that eco-flow stipulate or add a physical trip switch(like an RCD breaker). Showing user that the equipment they're connecting is beyond its limit, as a observer your doing a great job at offering up insight.

[@danharold3087]

From the first video the problem is that when you use a motor and electronics at the same time it fries the electronics.

[@Txepetxcc]

Let's clean the air around this answer. First, an RCD protects against live-conductor electric shock due to human body closing a loop between earth and conductor. Since the ecoflow output is inverted from a battery through an isolating transformer , there is no loop, its isolated by design. Therefore RCD useless because there is by design no loop for the leakeage to occur. Secondly, an MCB circuit breaker , depending on the Type Curve you choose can be very well tolerant to inductive / cutting tools inrush current (or startup current). So even that can be useless. So they went with a completelly tunable microcontrolled eFuse / MCB which apparently works pretty well. Wouldn't change a thing.

[@Txepetxcc]

@@danharold3087 no, only electronics supplied by a ' capacitive dropper type ' rectifier circuit. Which is only used in cheap led bulbs and should have had better protections which the nice ones have

[@patricklyons7683]

@@Txepetxcc l'm obviously a laymen to electronics & no offense meant. But if a tools draw goes past a certain level igniting a breaker trigger would imho be better than breaking the item or internally damaging it. Thank you for the concise information & time taken. I suppose Matthias doing his ultimate tests to see where the limits are & offering suggestions are along the R&D freebie ethos!

[@mrfrenzy.]

@@Txepetxcc a better RCD would prevent the device from blowing up if the user shorts neutral to ground. I really don't think it's acceptable for a device to break on such a common fault.

[@workbenchz]

Impressive that Ecoflow did get back to you and explained the 'noise'; +2 kudo points to them. This also suggests an accessory they can make and market: a 'soft start' plug adapter.

[@Manawyrm]

You can tell your scope that you've connected a current probe (and are measureing Amps) and you can also set the type of multiplier it needs to apply automatically.

[@tissuepaper9962]

I see your point because it is a demonstration, but on the other hand it's kind of more hassle than it's worth if you choose a current sensing resistor that's a multiple of ten. Another thing, if the current trace was correctly scaled it wouldn't be visible most of the time, whereas when we look directly at the voltage across the sensing resistor it's on the same scale as the voltage trace. You could use a separate scale for each trace to make them the same size, but then what was the point of dividing by ten in the first place if you're just going to multiply the scale by ten afterwards?

[@Manawyrm]

@@tissuepaper9962 Oscilloscope axis don't work that way -- each channel has it's own scale. So you could always fit all the content on one screen. The point of dividing anything is usually to lower the effect of the probe on the circuit by reducing the loading on the circuit. "more hassle than it's worth" -- the menu for that is literally open in the video. It would've been on the "Unit [V]" menu item and then another on the "Probe [10X]" item.

[@tissuepaper9962]

@@Manawyrm you're not understanding me, or you didn't read my whole comment. I understand that each trace can have a separate scale, but what would be the point of telling the scope what multiplier to use for the voltage to current conversion, if you're just going to adjust the scale of the converted quantity to be the same size as the original raw quantity? Why even perform the conversion? All that matters is the shape, exact values are mostly irrelevant, so what's the point of wasting your time setting the multiplier, adjusting the scale, and then clearing the multiplier again when you're done? Just move the decimal point yourself, you could do that calculation in your head 100 times before it would have been more time-efficient to set the conversion factor.

[@EcoFlowTech]

Thanks a lot for sharing the Medium blog post about the tech from our Head of User Operations at EcoFlow Tim Dolidze, and don’t worry you haven’t lost us as a sponsor :)

[@WillProwse]

Great follow up video and love the testing. Very cool to see how the over current protection kicks in on the scope, that was super neat. It's amazing these little HF inverters can even handle these inductive loads. I think you would love a nice LF inverter with a toroidal transformer such as victron. And I wasn't trying to be mean but I think the conclusions in the first video were a little unfair haha. If you have inductive loads, these HF inverters are not for you. Even the sol ark has issues like this and it's incredibly expensive and has every UL listing and certification out there. HF inverters can technically power any inductive load if they are sized appropriately. And yes ground neutral bond fried the ac charging circuit. Everything else is still working. They told me not to do that though, but some EV chargers require the bond. Ahh well. Now the pass through issue is what drives me bananas. Ac charging and having two units ac output in series for 240V has issues. It's a tough situation due to the use case that these battery boxes are designed to be used for. For most people and situations, it works great.

[@johnhufnagel]

If you see my comment above... Any chance you could do some testing yourself, with a 'scope and circular saw, and see how some of the big HF inverters fare under this type of load? the EG4 6500 would be the one I'd specifically love to see how it handles.

[@xavierneckor8918]

It doesn't matter if it's HF or LF. If it can't handle the surge load then it needs to shut off instead of putting out crazy oscillations.

[@woo6712]

I suspect that if they were to do that, the box would end up so sensitive that it would shut off seemingly at random during normal use and get a reputation for being unstable. Near the end he mentioned likely needing to design from the ground up for this and I wonder if it just ends up that LF is the cost-effective solution- factoring in it being heavier to boot. Would love to see this tested with an LF inverter even if the battery is separate as a POC.

[@WillProwse]

@@xavierneckor8918 if an inverter is properly sized, it won't. The ecoflow is not designed to run the load in the video, plain and simple. A LF inverter can manage inductive loads easier because it has a massive transformer, which increases its surge capacity considerably.

[@WillProwse]

@@woo6712 They could do that, but it would increase the weight considerably. Hysolis actually sells an LF inverter power system with NMC chemistry. Worth checking out if power tools or well pumps are to be run.

[@rickhuang4493]

There are two types of inverter design commonly in use. The one used by EcoFlow first convert the battery voltage to ~170VDC and run it though an inverter to generate 60Hz using PWM signal. The output filter inductor is usually small to ensure the overall system weight is small. The downside of the small inductor is it can saturate easily, which can damage the inverter when that happen. The other common design use a big output transformer and switch the PWM signal at battery voltage. The downsize is the output transformer is big and heavy. The advantage is the high inductance which should let it handle over current easier. It would be interesting to see how one of these type of inverter handle spinning up a big motor.

[@beyondfossil]

Good explanation of High Frequency (HF) vs Low Frequency (LF) inverters. I too would love to see how an equivalent power LF inverter handles these tests.

[@viyusavery248]

Other downside to LF inverter is the crazy high idle draw expect for vicron's hybrid LF inverter

[@LitchKB]

Kudos for following this up. I beleive most phone chargers are flyback topologies - both for mains isolation, and because they (depending on the standard) need to deliver over 30W. A capacitive power supply is only useful for sub-1W type of power (5V at 50mA, for example) because it's just not that efficient (it's only benefit is cost, really). Also, I'm glad EcoFlow got back to you and took this as an opportunity for everyone to learn something - I definately find that attitude appealing in a company, and it builds trust.

[@zxxirt]

The non-bias agnostic videos you make about this product has been encouraging me to get an eco flow. Never had advertisements work on me but the very realistic tests you have done has been making impact lol

[@WJCTechyman]

Nice to see Eco-Flow is still working with you, so that's good.

[@jbfedison]

Very nice demonstration and this battery box does seem like a robust product. Kudos to you also for your candor, what a refreshing change compared to many others on KZbin.

[@1pcfred]

I give that inverter a lot of credit for handling the loads as well as it does.

[@jllaine]

Compared to "mains power" was missing from the previous video, thanks for adding it to this one.

[@TheMobilefidelity]

Oh, those rude youtubers. Don't listen to them. Fantastic video BTW. great walk-through all the possible mitigation measures to starting induction motors.

[@rayj4362]

Good to see that the overcurrent protection on the Ecoflow works. It does prove that it is pretty well built.

[@shimmerite_ua]

Thanks for the video! Invertors and charge stations now became a highly sought after necessities in Ukraine because of blackouts, this information will help me choose suitable solution for my needs

[@christopherjordan8472]

This video does prove that my Ecoflow power station is very well built and the overcurrent protection system really does work.

[@HobbyHalloween]

To Eco flow's credit and yours, I like the open discussion on the design and operating behavior. What I see is an opportunity to solve this problem, a big win for you, them, and everyone! *Here is my $1M+ idea:* I think the focus should be on the tools that don't have that soft-start feature... like your skill saw. Make a product that would be an external retrofit box that you'd plug the tool into that is then plugged into the Eco Flow. That way you don't burden the cost of this for the main box for those that don't need this... and besides, this only applies to certain power tools. The retrofit box could employ some strategy that it does nothing until it senses the power tool being turned, that way it wouldn't be drawing all that standby power like that 25uF filter cap you added. I believe it is well within the capabilities of engineers to come out with a design... I know it's not easy, but that's what will make a solution like this valuable! It's kind of funny, but the more you're talking about this product and seeing the technical details, the more I'm liking it!

[@JessicaFEREM]

yeah this seems like a good solution. have a "powertool" addon module.

[@jamesthomas4080]

I had a similar thought, but as you might guess they already exist. Search for "power tool soft starter", you can buy them as a control module to hardwire in or as a plug in box. That said, they (the plug in style) seem expensive enough that there may be room for ecoflow to sell one as an accessory at a lower price point.

[@JessicaFEREM]

@@jamesthomas4080 this is great info. maybe matias can get one of them and test it out with the ecoflow!

[@blenderpics]

I had an issue with the inrush current of certain electrical devices in my home, due to old fuses. Once i started a device that required a high amount of inrush current, the fuse blew. But there was a real simple and cheap solution: there are devices called inrush current limiters. You plug them in between the mains socket, and the device you want to power. The device switches a resistor between the mains and your device, until the current lowers to a reasonable amount. It then connect mains power directly to the device. It solved my problem and wasn't expensive at all. The thing cost me 20 or so bucks and was in stock at my local hardware store. This should solve your issue with your power banks quite neatly.

[@1pcfred]

You just stuff pennies in the fuse box. Problem solved!

[@OmmerSyssel]

@@1pcfred Fried my entire installation following that advice, pure luck the house didn't catch fire 🙈

[@1pcfred]

@@OmmerSyssel it'd only work with old pennies. New pennies aren't copper anymore so they'd blow like fuses do. Well, maybe they wouldn't blow as good as a fuse but it wouldn't hold like an old copper penny either.

[@Kamodomon]

It is soooo important to understand the limitations of these things so you know how far you can go and not exceed that. Thank you for this.

[@FrankGraffagnino]

I appreciate the response from EcoFlow... this makes me more willing to buy their hardware!

[@budmartin3344]

Great video. ECOFlow should hire you to test their products before releasing products to the market.

[@TheodenEdnewDoesDnD]

I bet Ecoflow actually really appreciates that you are doing all this testing for them. Their R&D department probably has a different perspective and could use your thoughts.

[@MozTS]

R&D already knows all this shit, its marketing and execs who just dont care

[@ckm-mkc]

Their Medium reply was from one of the operations directors and was very, very detailed, way more than I've ever seen from any manufacturer....

[@sheph7]

I’ll guarantee the design team already knows this and a lot more. I’d bet they can design anything that doesn’t defy the laws of physics but other real-world constraints dictate the chosen path. The critique by Matthias reminds me of audiophiles criticizing power amplifier designs.

[@XantheFIN]

True, their value raised HUGE time in my eyes when they send email about they get interested to make product better after showing the fault (though after KZbin video just).

[@tissuepaper9962]

@@sheph7 this box advertises itself as an off-grid power source, and sells itself on the premise that you won't need all those extension cords anymore. It's perfectly reasonable to expect somebody to stick a power tool in there so that they can work on something without a nearby power source, and it's a very valid critique of the device that it will kill certain electronics if they're used at the same time as a big inductive load. I would have definitely been dumb enough to charge my phone with an old charger at the same time as using a power tool, maybe that's just me or maybe the hazards aren't obvious to everybody. Matthias's use case is still within the realm of what a reasonable person would try, and he has been very clear throughout these videos that this is a very good product with one notable flaw that only appears in the worst case. His coverage isn't comparable to audiophile nitpicking at all, when was the last time an audiophile actually presented oscilloscope traces as evidence for their criticisms?

[@stellamcwick8455]

I think the problem here is that these consumer grade battery back-up supplies aren’t designed for certain applications. Lighting, charging, radio devices, laptops and networking hardware, etc. fine. Large inductive loads? Probably not. There is a point where you could engineer a solution to this problem but at what cost and is it worth it when the current design covers 90% of the intended user base. What I like about this series is that Mathias was willing to put in the extra effort to properly evaluate a specific usage case that could be problematic to relatively few users and do so in such a way that was honest but fair. Note to EcoFlow. Mathias pointed out a flaw in your design but that flaw is an extreme case that also exists in other manufacturers products. If i had the need for such a device I would still STRONGLY consider your product. If for any other reason than Mathias did the teardown on it. I just wouldn’t use it to run inductive loads without soft start capability.

[@Beakerbite]

A high inductive load would work better with a gas generator than a battery backup. That's how builders have been doing it for years.

[@stellamcwick8455]

@@Beakerbite agreed.

[@joebrodie]

You are absolutely correct that power tools is an extreme case. Quick charging 18V/20V batteries for power tools would be more reasonable expectation. Also, a power outage and plugging in an appliance to run (refrigerator, freezer) might be a good scenario. For me, I would put this in a shed to run DC lights, have an occasional 120V system and charge batteries for tools.

[@chitlitlah]

I'm trying to figure out what kind of yahoo would decide that a power outage is the perfect time to go to the shed and cut wood.

[@robertbackhaus8911]

Power tools are an extreme use case, but it is also a use case they use to market the devices. And the power tools do work!

[@ericwazhung]

Awesome test spinning up the bandsaw by hand. Very informative results, all-round.

[@stormriderkaos]

I was about to ask about soft start motors when you addressed them. Good to know, perhaps especially for people using these backup systems in a winter storm power outage with forced air heating driven by an induction motor.

[@jeanious2009]

Love watching your videos, I feel a bit smarter after watching them. Thanks!!

[@thomask4836]

Thank You Matt! I think it would behoove EcoFlow to link your videos to their advertisements to help consumers discover the great vs not so great applications for their power source. That way when they are purchased, there are no surprises!

[@matthiasrandomstuff2221]

well this sort of info does turn off some potential customers - just read some of the comments.

[@Brian-ly9xm]

Thanks for this informative video. I was not aware that these Ecoflow power stations have an overcurrent protection system and it is reassuring to know that it has.

[@earnestmarshall4694]

Pretty good tests you did with that Ecoflow and it shows that if I accidentally overload it nothing bad will happen and the protection will override the system.

[@CaptainSeamus]

This all goes back to inrush current values versus surge loading. The NEC allows up to 6x full load current for a "short period" - so if that saw pulls 10A normally, it could draw up to 60A for a few cycles as the electro - magnetic interactions happen in the inductive load (the motor) as it comes up to speed. Unfortunately, most inverters only have a 2x surge load capacity. So, it can't keep up, and the sine wave breaks down into the noisy patterns you see. The real issue is, no one explains inrush current on inductive loads well, and inverter manufacturers use a different idea for "surge" (that actually does work most of the time) from electric motor manufacturers. I actually am pretty impressed how well those battery/inverter units hold up under load. Kudos to them, and thanks for a great real world example of the how / why / what of what the ragged edge of inverter power looks like.

[@rhoderzau]

I’d love to see how other inverters deal with this. Perhaps a Victron Multiplus II, Jackery and a Bluetti to cover some of the big names.

[@nokidding152]

They would all handle it the same. You need a lot of filtering or a massive inverter to absorb those surges.

[@bradley3549]

A great follow up. Connecting neutral to ground sounds like a pretty rare situation unless you A. have equipment fault or B. intend to feed a sub-panel from one of these devices. I wonder if they warn against this in the documentation anywhere as the latter isn't an uncommon use case for backup power.

[@samTollefson]

That is an excellent point! I hook my generator up to the main panel and switch off the circuits it can't handle so I don't have extension cords running all over the place. This would be a problem in that scenario. Thanks for bringing it up!

[@AndrewMoizer]

Agree this is an important point. Backfeeding the house supply is a very useful application. Even a 3 pole disconnect that also disconnected the neutral coming in to the panel as well as the hots would still leave the panel neutral bonded to ground.

[@samTollefson]

@@AndrewMoizer Agreed, however, most folks would only use one of these battery packs to power their fridge or small freezer through a power outage or to keep the oil furnace going in the winter up North. Where I am in Florida, I have to deal with occasional power outages for days or weeks at a time so I have a generator system to power the house essentials and a stash of 50 gallons of fuel to get us through. I do like those power boxes though!

[@mrfrenzy.]

What if you cut the extensin cord with your hedge trimmer or drop the socket in a puddle of water? This is very common.

[@bradley3549]

@@samTollefson I think that is changing though with these companies allowing you to link two or more systems for 240v output. So having a device that blows itself up if it is connected to a neutral bonded panel is a real pain. Of course, maybe the linked 240v system doesn't suffer from the same neutral bonding issue when linked. Then the problem sort of solves itself since most people feeding a panel from a generator input are doing 240v. Still curious if they warn of this in the documentation or not. But not quite curious enough to check myself. lol.

[@killer2600]

I liked watching the Rigol in action. This video boils down to generators and inverters have surge limits and there's a reason why they have them.

[@johntate5284]

Have you tried this with an inverter, such as a Victron, which has a large transformer to improve the performance with inductive loads? You may be able to improve the performance by plugging a good large isolation transformer between the inverter output and your inductive load. There's a good 120 VAC soft starter on Amazon for $130 made for table saws, routers, ect. (which i just ordered thanks to your video!). As a side note, the batteries used in some portable power tools, are 30C or even 60C devices, meaning that that 4 AHr battery can supply 120 to 240 amps of inrush current to start a motor, which in many cases, makes them far superior to their plug in counterparts. Great and in depth information!

[@konzetsu6068]

Like you said about the large smoothing capacitor, its not optimal, but several ways to get around it. having an external cap like you did, just in its own casing that you put between the power tool and the battbox. or integrating it by putting a dedicated outlet with extra smoothing. in both cases you can certainly toggle it with a switch, or automatically on load detection.

[@Benoit-Pierre]

its all good for me : - your manufacturer follows you and provides technical response - you make a video response to yourself 👍🏻❤

[@MCsCreations]

Fantastic testing, Matthias! 😃 That's a really impressive device indeed!!! Stay safe there with your family! 🖖😊

[@wehtjns]

Great video, thanks for the update. I personally like ECOFLOW and this video only improves my feelings about them.

[@Norman_Fleming]

Wonder why the neutral to ground is a unique problem for them? Awesome video. Nice they didn't just blow you off but responded with details.

[@pennyroyal3813]

Ecoflow can be certain that for the average person their device would be excellent. I like how Matthias pushes it beyond its limitations thereby making a very good advertisement for Ecoflow.

[@Nono-hk3is]

I'm so glad you showed the soft start case.

[@JosiahK555]

great info, the ecoflow is at the top of the list for me for potential battery back up and solar, i just haven't pulled the trigger on one yet

[@mckenziekeith7434]

This is very excellent and top notch. Thank you for doing this follow-up.

[@foobarables]

Excellent follow up. The inverter protection does work properly. I wonder why you didn't setup a test like this in the first video.

[@niske]

9:58 : "plugging in my many heaters" sounds like the start to a poem dedicated to your channel

[@sirajhussain791]

Very good way of explaining with practical wave shape explaining. My request please explain and distinguish between the Neutral and Ground, how the inverters get damaged by that ? pl explain in practical manner as you had explained this video. Thanks

[@justDIY]

Very nice detailed technical response they sent! I have an "old school" Magnasine PSW interverter, rated for 2800w continuous and some thousands more peak... it weighs maybe 80 lbs, without batteries. I've run my circular saw, air compressor and other inductive loads off it, I wonder what the waveform looks like during an overload condition.

[@jimzeleny7213]

Probably much better. I run an old Trace DR inverter and guess that the transformer makes all the difference.

[@D2O2]

Ignore that other YT'r. Glad you showed it is possible to filter the HF content, but does come at a penalty. Could employ an active strategy to bring in the filter circuit when needed and pull it out when not to remove the phantom power.

[@JessicaFEREM]

this is a great advert for ecoflow. it's great that they're happy that you're doing some testing and not just "don't buy this product"

[@aaronalquiza9680]

I've been eyeing Ecoflow Delta Pro since last year but Canada didn't have stock until this past summer when I didn't have cash to buy. Looks like I will definitely be buying 2 Delta Pros + 240V adapter this January.

[@sybergoosejr]

eco flows response was nice. because of that i will consider their products in the future where they may fit my needs.

[@daw162]

Literally was going to ask about the units getting a capacitor installed....you nailed it. Unlike other people here, I don't know much about electricity aside from the usefulness of a capacitor for fast discharge but low total wattage needs.

[@jeffweber8244]

The soft-start feature on the power tools seems to be key, if you want to power them from an inverter.

[@davidcross890]

Will work with soft start but ultimately masks design flaws per normal consumer expectations

[@beyondfossil]

I agree. Adding a soft start reduces both mechanical stress on the power tool and on the backend power source which is not always grid. I notice some table saws have such a hard start that it messes up the blade height adjustment that I carefully adjusted prior. A soft start could also reduce damage to a body in the terrible case of an accidental startup at the wrong time. Even a generator with its higher inductive buffer could get bogged down depending on its size and multiple hard starting power tools that might be starting at the same time. I've seen gas generators advertised as 2000W instantly shut off with just a single miter saw on it. The bottom line is adding a soft start device to the power tools costs the tool manufacturers money upfront and introduces another failure point for their warranty program which is a backend cost to both their financials and reputation.

[@car9167]

I'm running a 5.5KW 3 phase motor with a 220V inverter single phase powered. This motor runs as an idler and smooths the 3 phases. With it I'm 3 phase powering my surface grinder and other tolls in my garage. Never had any issues with the inverter. The idler motor coils acts as low pass filter and dumps any current surges

[@mckenziekeith7434]

Do you have any extra inertia on the idler motor? Or is it just the rotor and shaft that is spinning?

[@car9167]

@@mckenziekeith7434 Just the rotor spinning.

[@ches74]

Sounds like Bluetti are admitting defeat, they must be admitting they can't match the performance of Ecoflow! 😂 You've put the Ecoflow through its paces and it has stood up really well. Ecoflow also conducted themselves well from your account.

[@BeyondHomeCooking]

Mattias, thanks for the more in-depth looking in on this. I have one of their Delta flow max guys, and since I paid my own hard earned dollars for it, I would hate to blow it up. That said, I do quite like it. But I think, on balance, after this video, I am going to stop plugging my table saw into it. Lol. Cheers.

[@sidewalkcinema]

Just watched your 2 EcoFlow videos and thought they were terrific. Really appreciate the care and depth you did in the review and bringing us all along in your experimentations. 🎉 I now wonder what is going on in my Honda 2200i generator and its’ inverter.

[@Fabsenet]

How about a long extension cord to the saw? The long extension cord should limit the starting current and should limit the protection circuit requirements?

[@HyperactiveNeuron]

Great video! This reminds me of my electrical engineering and physics lectures and labs. You'd make a very good teacher Mattias This would only give me more confidence in this product.

[@flat-earther]

jj_crank I suggest watch a 13 part series called _What on earth happened_ by Ewaranon to learn that the earth is not a globe. I got it in my about tab.

[@christopherdekonstrukt444]

I used to work on switch mode power supplies and used to use a probe across a shunt resistor to see a visual representation of current on an oscilloscope. Usually the shunt resistor is usually 1/10th an ohm value range.

[@mckenziekeith7434]

If you have an induction motor spinning with a large inertia load, it can kind of buffer the AC voltage and help smooth it out. If you haven't exhausted the topic, try starting up the skilsaw while the bandsaw or a tablesaw is already running on the same circuit.

[@zeropointmx2552]

I wonder if the "chopping" of the inductive load by the FETs (during protection) is causing a flyback voltage spike, similar to a ignition coil.

[@Sevalecan]

More or less, with the inductance of your motor being the coil that is generating the voltage spikes. The transformer itself is an inductor, but most of the current it produces is through operating as a transformer, so when the FETs shut off, the current goes to nearly zero through that transformer. For 60Hz it's relatively close to an open circuit, at the OCP frequency it's even closer to being an open circuit. For the sake of simplicity, pretend the FETs are on primary side and there is no transformer. So, the energy stored in the windings of the motor has only one path if charge is to keep flowing: through the filter cap. Charges it up fairly quickly, so we see high voltage spikes. If you increase that capacitance like Matthias did, it takes more time for the voltage to rise and fall when the FETs are off, since the current in the motor is the same, and your capacitance has increased.

[@wardprocter2371]

Looking forward to the follow up video on making a diy soft start circuit 😁

[@flat-earther]

wardprocter I suggest watch a 13 part series called _What on earth happened_ by Ewaranon to learn that the earth is not a globe. I got it in my about tab.

[@googacct]

A quick search on amazon shows some external softstarters that can be added inline with the load. Maybe it would be worth getting one and testing it with the ecoflow to see if that helps with the inverter instability.

[@EpicHardware]

low frequency inverters are dealing with high starting currents quite well, but they are too heavy for portable applications like this. I have a oscilloscope and a low frequency inverter, i will upload a short video soon

[@jlh2119]

Than you for the information and video! I appreciate the effort you put into your channel !! Question… While battery generators are able to run refrigerators and forced air blower furnaces I assume doing so often enough would greatly reduce the lifespan of the artery generator. Would you have a battery / inverter combination recommendation better for refrigerators and furnaces? A 12v / 100AH battery paired with what kind of inverter is best for a fridge or furnace? Thank you again !!

[@Jasonoid]

Excellent follow up, great information from your testing!

[@ColbyJohnson]

Very good follow up video! Covered all the bases here

[@TgWags69]

A common work around for making a static phase converter is using a momentary switch or relay coil to disconnect the capacitor after the motor has started. I wonder if you couldn't do something similar as an add on in-line box that would hold you capacitors and some sort of switching control to help with the spikes but shut off so it doesn't draw continuous power?

[@Vykori]

honestly I'm more likely to buy this brand if I'm in the market for this because now I feel like I know how they work on the inside. Love it

[@ericrandall3539]

If you want a legitimate battery backup for power tools, you should look into a low-frequency inverter. They are made for that kind of inductive inrush. I'd be interested in a Matthias review of those.

[@simonachkar]

Love your channel man! Thanks for making these videos! Rock on! 🍾🍾

[@Loyd-d8d]

It gets even worse with well pump that is also AC motor but starting under load. I have off-grid cottage in New Brunswick where I use 12V battery with 2000W cont / 4000W peak inverter, and it runs 15A Dewalt table saw no problem, but fails to start 800W well pump 3 out of 4 times. Table saw starts without load, so inrush ends quick, but well pump is already under load so inrush takes longer and inverter overloads. You showing over 100A inrush to your saw makes now perfect sense, I was told to plan for 4x rated, i.e 4kW peak for my pump should work, but it seems it can go much higher. For now I can only have water with generator, will try soft start next time.

[@TonnyCassidy]

Just a guess but could the reason for that kill a watt failure is due to how its power supply is deaigned ? I dont have the exact same device here but power meters we have here uses capacitive dropper, and capacitor being capacitor, the “conductive” changes based on frequency, my guess would be at the high freq oscillation that capacitor essentially turns into a dead short, overvolting the microcontrollers and other parts damaging it, or maybe just shorting a zener instead of the main components

[@wouter11234]

I have the same problem you have with your big bandsaw with an old amplifier of mine. When I try to power it with my bluetti ac200max it shuts down every time, detecting a short circuit. Edit: it has a very big transformer, presumably without any inductors on the input side. When I power it on the mains I hear a loud "thump" sound (I'm on 230v)

[@m8e]

Amplifiers usually have a lot of capacitance for smoothing. The capacitors need to be charged and that can create a current spike that "look" like a short.

[@MrMersh-ts7jl]

@@m8e exactly. you're amplifier needs a soft start unit that allows the capacitors to charge slowly before the soft start relay shorts and allows full power. xraytonyb has a great video on this.

[@wouter11234]

That makes more sense, I'll look into that! Thanks guys

[@apollorobb]

A lot of it has to do with the noise thats induced onto the ADC's that sense the voltage and current .And oscillations on Resistive overload are caused by the drive voltage on the gates of the Output FETS . Theres actually 2 sections in the inverter. 1 a boost circuit that brings the voltage up to around 300V or so DC then section 2 switches that high voltage DC into AC . I in 30 years of playing with H-Bridges and inverters have never seen one that will accept this kind of overload gracefully. I think by the look of the waveforms on the scope the ECOFLOW does a pretty good job

[@conradfjetland4970]

Matthias, I like how you push the limits of the product. How else are you going to know what those limits are unless you test them. Thanks.

[@thomgt4]

This hardware current limitation isn't endless though. Based on the article on medium and what you've said the gate drivers are detecting IGBT desat and pulling it off to prevent runaway. It does however mean that the IGBT is subjected to desat. That does damage the transistor a little each time (and it happens hundreds of times in a single start). Desat survivability is typically in the thousands of times range, so I wouldn't recommend you do keep overloading it

[@flat-earther]

thomgt I suggest watch a 13 part series called _What on earth happened_ by Ewaranon to learn that the earth is not a globe. I got it in my about tab.

[@Ramog1000]

Switch mode power supplies also work at high frequencies because the transformers can be smaller that way. (since the transformer doesn't get saturated due to the high frequency and also a less windings per side due to higher impedance)

[@T3sl4]

I'm surprised they don't operate much higher; MOSFETs, and ferrite-core magnetics, can get small enough to save significantly on size and therefore cost. Maybe not an overall cost savings, but the reduction in size can be worth a minor cost premium. IGBTs are usually chosen to reduce cost, without much concern over efficiency (because they always drop a volt or two, whereas MOSFETs look like resistors: voltage drop proportional to current). Presumably they could save some efficiency too. Perhaps it wasn't worth the engineering time -- faster, more compact designs can take a lot of time to optimize, unfortunately.

[@ProbeLem]

I am learning a lot in this channel. Thanks great videos.

[@ctrlaltdebug]

Would be interesting to see if your bandsaw running could act as a flywheel and buffer the inrush current from the circular saw.

[@smsn13act85]

Me too. Poor mans single phase synchronous condenser. I'll have to think about how this would work with a brushed motor and a probably fixed frequency inverter. Might be easier to test than think.

[@deonmo1963]

South Africa has a 240V N E connected grid supply, to prevent N E floating voltages. I am keen to know how Ecoflow manages to prevent N E floating voltages. My assumption is that Ecoflow will not be a suitable inverter for South Africa and any other country with similar grid topology as South Africa. This is an excellent video demonstrating inrush currents of motors (dc motors have the same inrush current properties as AC motors)..

[@flat-earther]

deonmo I suggest watch a 13 part series called _What on earth happened_ by Ewaranon to learn that the earth is not a globe. I got it in my about tab.

[@rpavlik1]

Thanks for showing the even more detailed scope data. It does seem like this oscillating protection behavior should be opt-in with warnings because of the possibility of damage. (It should default to turning off with a longer period, put some hysteresis in it) Especially since there's movement to ban gas generators so more folks will start using these things for more and heavier loads. Those chargers might be ok, but led lights won't be...

[@fourfortyroadrunner6701]

So called "universal" motors, AKA older AC/DC brush type motors in vac cleaners, shop saws, etc, have always had a tremendous surge / start load. Not surprised that the overwhelm an inverter Largest UPS I was ever around was back when installing E911 telco/ data/ communications systems and upgrades in dispatch and radio sites. One dispatch had a UPS the size of two large coke machines. Most of course, was batteries.

[@FabioRossiniSluzala]

Can you test with something like a Pi filter (capacitor, inductor in series, capacitor) to see if the oscillations pass to the other side? Maybe will be safer to use the electronics with the filter between them and the UPS.

[@T3sl4]

Tricky: you need to make sure the components don't resonate with each other at the switching (or faulting) frequency, or else things get *much* worse! You need to mind both the cutoff frequency Fo = 1 / (2 pi sqrt(L C)), and the characteristic impedance Zo = sqrt(L/C), and preferably have some damping or termination resistance so that, even if hit with frequencies near resonance, it doesn't spiral out of control. (The inverter itself acts like a very low resistance, we can't count on it for damping. We can put a resistor in parallel with the output to do the job. Now, a whole ass resistor would dissipate a massive amount of power at 60Hz, but we can connect a capacitor in series with it so it mostly doesn't affect the circuit, except near Fo where it's needed. So, instead of C-L-C, it can be C-L-(C || (R+C)).) Just putting the 24uF out there, does a good enough job illustrating the concept, though: all the current ripple was confined between it, and the inverter's internal circuitry (mainly the big fat filter inductor I would think), and much less ripple voltage appeared at the output. This is safer (in terms of ease of avoiding unlucky values that might make it more prone to oscillation) because the capacitor more or less acts in parallel with the internal capacitors. So it's more or less the same LC circuit, without extra LCs tacked on. We can infer that, given proper dimensions, damping etc., it could indeed have less of both current and voltage ripple; and less while using less total capacitance than the bulk approach (the 24uF) -- the tradeoff is having to add bulky inductors, which also reduce regulation (they drop some voltage, at 60Hz, under normal load conditions). And these are not small inductors, we're talking 100s of uH and they have to be rated for full peak load current, so, saturation over 20 or 30A. About as big, each, as what's on the inverter.

[@Vigo327]

I appreciate the time spent quantifying why the capacitor is not to be LEFT in the circuit. But, isn't it frequently done that a motor device removes the 'start capacitor' from the circuit shortly after startup? This can be done externally pretty simply, so my thinking is not to throw the baby out with the bathwater as far as the homebrew 'soft start'. Just that you should not expect to leave the capacitor in the circuit full time with no consequence?

[@matthiasrandomstuff2221]

Because you don't know when the motor starts, you would have to leave it on all the time in anticipation

[@ccoder4953]

You should try doing a collab with Dave Jones from EEVBlog. He's been doing experiments with these battery powered inverters for powering his house. He has the equipment and knowledge to do a deeper dive on this. Would probably make a really interesting video.