I am a developer for frequency inverters and I fully agree that this behavior is normal. The reason is that with that massive overload, the output inductance saturates - this could already lead to oscillation. Additionally, controlling a saturated inductance and a dropping battery voltage is extremely difficult. To prevent an overload in this case, the physical size and cost of the inverter would probably be increased by factors (assuming this device is already at a good quantity). So there are only two possible reactions that a device with that size on this load can do: Either shut off or accept some oscillations. On the other hand: the input circuit with the capacitive voltage dividing on the power meter is a very bad design practice, although this is done a lot especially in cheap products. When testet according to standards that are mandatory for sale in the EU, most of these products fail. All products must withstand some harmonics/flicker/emi/...
EcoFlow just got back to me with this explanation here: link.medium.com/tDYDK9mTjvb Apparently, there is a protection circuit that forces the inverter transistors off when a certain current is exceeded, which appears to be independent of the normal control loop. This protection circuit leads to the oscillations.
Adding to that, powering devices with two diodes and a capacitor is cheap but TOO DANGEROUS to use. I consider it the cheapo way to power anything, so I strongly recommend to not buy/use anything powered in that way ;)
I mean I see a lot of people praising matthias integrity but I think there was a misunderstanding here ..as you said desining this kind of products it's really difficult so when overloading the inverters with such large currents, transients are to be expected ....ecoflow isn't a bad company ..much better that other portable batteries manufacturer .if not the best
This is a revised version of the video. The original version inadvertently made the oscillations look 4x bigger than they were. Explanation at 9:05 into the video.
I thought there was a glitch in the matrix, since I was sure I'd watched this hours before. Thank goodness I'm not crazy. Well I am, but not because of this!
TYVM for being honest about this problem. I'd considered getting one of these to run my power tools AND charge my phone at the same time. That woulda been a very costly mistake.
I would love to see what happen when you have already some load(200+W), and then start circular saw. This can make problem go away or it can make it 10 times worse. And it would be closer approximation of full house backup scenario.
I’ve been following Matthias for about 10 years now, and I remember the first time he did a sponsored video I was pretty upset. I thought he had sold himself out. I subsequently have watched all of his videos, especially the ones where he talks about sponsorship. I really appreciate his integrity. And this is a great example of a class act on his part.
Test like this actually make me more confident about buying one of this boxes, I prefer to know the limitations of the equipment I use without having to blow up stuff myself. I love when companies listen to feedback and use it to improve, it's a shame they weren't willing to do so this time, I hope you can get a sponsor who does.
Well, the problem here is - there is not much they can technically do about the problem. Other than create absolutely new inverter design, that would not fit to the enclosure and would be horribly inefficient for loads that are below its rating. So, it's always a trade-off... In general, the business case here is - people that use circular saw will not use it in parallel with AC powered sensitive devices - because they are somewhere at construction site, etc. And small devices (laptops, phones, etc.) have separate DC output (100W USB-C, for example) on the front panel that will not be impacted by start of the saw at all.
@@rklauco I think it would be reasonable for the company to add a warning about this specific use case rather than ignoring it so people know what precautions they should be taking.
@@rklauco Can you explain why they can't filter their way out of the problem? A simplistic look at things is that they have to produce one frequency, so everything higher would be filterable, wouldn't it? (I hope to learn something new in the answer, it's not a know-it-all question)
@@rklaucohow do you know that? You don’t. Unless you are assuming that all small devices have a BMS that provides overcurrent protection, which they don’t. Predicting the behaviors of construction workers is best left to Sociologists.
Pro tip: don't try to run massive inductive loads( saws and power tool start ups) on a inverter that has a High frequency like these ones. This isn't a defect, it's a limitation of the tech. Low frequency inverters are the ones that can handle these type of loads but they are WAY more expensive.
There were some good comments on the prior version of this video. 1) The neutral fault: as mentioned elsewhere here, you must use a differential or isolated probe for this kind of testing. Otherwise you're tying mains ground to generator neutral which is a terribly bad idea. Best case that causes a ground loop and will trip an AFCI breaker. Worst case your scope is now a bypass and/or ground reference for the generator and you'll have unknown amounts of current running through the scope ground plane. The test was done incorrectly. I've inadvertently made similar mistakes when checking the output of my gensets and working too quickly, resulting in exciting sparks and breakers throwing, but I blamed myself and not the products. Neutral is not the same as ground!!!! 2) The line noise is the result if the inductive load of the motor. This happens any time an A/C motor starts. The effect on mains is mitigated because the power distribution system acts as giant capacitors. But this is why lights flicker or dim when a large air conditioning unit kicks on. As noted elsewhere, when sitting on a generator, there's nowhere for this inductive load to go and the generator can't spit out enough power instantly to manage it. The table saw is better because the motor is probably not all that much larger than the circular saw, but you have dramatically larger output buffering with those two units than the one small unit, plus the table saw starts slowly. The real point of the 240V on the table saw is for motor torque, not for higher power output. You'll see this same noise on *any* generator - battery or gas. It's not a product fault, it's a use case fault. This is one big reason why large (i.e. >1/4hp) A/C motors should run from *traditional gas* generators. There it's just a spinning rotor - no electronics except for the voltage regulator - so the inductive load gets (partially) sunk into the magnetic fields and there's no electronics to damage. You can hear the two motors (gas and the tool/appliance) fight each other at startup, then run in sync. Generator companies, especially consumer grade, and double especially this new wave of battery ""generator"", are built for common household use cases - refrigerator, lights, TV, not much else. A wood shop is not the designed use case! Sadly their advertising does say "power tools", but their example is a hand drill. And the inductive load of a hand drill startup is far, far less than a circular saw. So sure, they need to be clearer in their marketing, but running a wood shop off of a battery power pack is a highly unusual use case. And construction crews are *not* going to be buying this kind of thing - they're buying plain old gas generators. Even in that use case, the only thing plugged in will be motors - saws, compressors, drills, mixers, etc - which don't care about noise very much. So as long as the genset output is sufficiently protected so it doesn't get killed by the noise (which appears to be the case), then they did things correctly. You've got good material, but you need to better understand what you're testing and what your results actually show before drawing conclusions.
Totally agree. I was elated the first time I witnessed the signal of a hairdryer on an o-scope. Didn't have a differential probe either. One thing I didn't do is make a video and blame Tektronics for my results!
I stopped this video less than halfway through to see if anybody else caught this. Of course, they did, and Jason did a very excellent job of explaining things. I would have probably gone way overboard. The only thing I would add is if you want battery solution to use power tools, get a heavy-duty transformer inverter that is made for these types of loads. Any nontransformer inverter will have the same characteristics, if not worse. The issue with Ground vs Neutral is well documented all over the internet, separate problem. As always, there is a disconnect between design/engineering and marketing. Unfortunately, we see this all too often.
I'm with you most of the way, and I understand your point about using non-inverter generators. However I will say I'm seeing quite a lot of stir in the industrial use space for inverter generators. So I think the focus should be on solving the problem vs saying 'you don't understand and use case, that wouldn't happen'. I think things like noise constraints and sites that have to deal with emissions limitations (I'm thinking of confined spaces, low grades (holes, pits, ect...), Flammability limitations, ect.... not really the miniscule environmental impact.). I think we are going to see battery 'generators' start becoming more common as battery tech/capabilities improve.
People make mistakes. It happens and there is nothing that can be done to prevent it. What a person does when they make a mistake defines their character. My hat is off to you, Matthias, for admitting your error, explaining it, and then correcting it. That is what integrity IS. Most people do not have the strength to admit when they are wrong and that is a sad commentary on society.
It never going to be "perfect" but Oversize the inverters and add inductance. It would drive up the cost and size of the box and becomes uncompetitive.
"Pure Sine" inverters like this (and like UPS' and the like), along with the inverters for the 3-phase motors in EVs all work the same way. They use a microcontroller to make a high-frequency PWM signal to switch a pair of MOSFETS (or IGBTs). This is because FETs are almost completely efficient when on and off, but can't handle analog voltages in-between. So the signal is switching on and off at a value somewhere in the multiple KHz range, and it's period is adjusted to form a 60Hz sine wave, which approximates a sine wave. This is then smoothed with an inductor to fill in any switching gaps and make it a full sine. With massive inrush currents like your circular saw is introducing, it's overwhelming the inductor's capability to smooth the PWM signal, as the current is pulled out of the inductor before the next cycle can introduce more. Because of this, the output sees (momentarily) the raw PWM signal. The only way to "fix" this is to massively increase the size of the inductor to cope with this one-in-a-million inrush. Sensitive electronics usually use a capacitor on the input to smooth the DC ripple from their rectification anyway, which would entirely negate this "noise", but because the Kill-A-Watt (and devices like it) require interacting directly with line voltage to measure the throughput, it can't use smoothing components on the line. Incidentally, the other inverter strategies used by UPS' are "Stepped Approximation of a Sine", which uses capacitor and resistor ladders to create a very coarse sine-line signal, and the cheapest units omit the ladders and just outputs raw 60Hz square waves, letting the cabling act as an inductor to smooth the power a little.
Some of what you are saying is not right. The inverters used for three phase motor control cannot be used on normal three phase loads. Three phase inverters for motor control do not have any inductor between the switches (IGBT or MOSFET) and the motor. They rely on the motor inductance to smooth the current waveform. The voltage output of a three-phase motor drive is the raw PWM.
As you mentioned, it is an extremely minor issue, I'm surprised they weren't happy with your proposed warning of not running power tools and sensitive electronics simultaneously... It's still a fairly strong recommendation overall.
“An extremely minor issue” lol. Since when is a current control loop going unstable when it’s doing what it’s supposed to do (limiting current) an “extremely minor issue”. Don’t be an apologist for them. That’s the sort of a fail EE students would get lots of points taken off if it was a class project. What you claim as minor is a very basic engineering fail that anyone competent would be ashamed of, not apologizing for.
I am so happy that I watched this through. I am slowly setting up a fully off grid wood shop. But I may have some more sensitive tools in there. I had no idea that some tools could cause others to blow out. I wonder if those surge protector power strips would be harmed by this? My thinking is to have everything off unless needed to prevent such issues.
Your review made me more impressed with the Ecoflow. I'm ok with that limitation and would gladly buy one if the need arises in the future. It seems like a very well built device. As you said, a simple warning in the manual would probably suffice, especially for the whole home model.
If I needed one of those, I'd still buy one. Better to know the limitations. Which in this case are acceptable in my opinion. Good and honest videos, as per usual.
@@matthiasrandomstuff2221 It just occurs to me that you were testing with the circular saw without putting a load on it (i.e. resistance, cutting wood) so the power current would have been higher to maintain the saw's rotations against cutting something, leading to worse oscillation.
4:14 - so you've built an FM radio transmitter! The drop off in response is a little disappointing, though using a circular saw on a portable battery bank is quite a bit of an outlier compared to the typical case of plugging in fridge, a laptop, and a couple other smaller appliances. Next time I talk to Ecoflow (I might be working on a solar project where they could be a sponsor), I'll be sure to bring this up and see if I also get the silent treatment. Hopefully not!
@@matthiasrandomstuff2221 This may be the reason I find that brushed ebike motors have way more instant "grunt" than any brushless motors I have tried 🤔.
Ha as soon as he mentioned the switch I thought oh lol it's coupling over isn't it... Gotta love the auto setting, I really need to learn to use mine better.
Not so, I was watching the Skill Builder's video this morning where he was explaining these battery boxes are great for on site, especially if the electrics are off. He was pushing the Ecoflow Delta Pro.
Great video. I really respect you interrogating your sponsors like this and holding them accountable. The reason that sponsorship have become so popular is that they work. people tend to trust the words of people that they like and well trust.
I use a lot of power tools and sensitive electronics (PC overclocking carpenter...) and I have been looking into a couple different brands of these power stations for off-grid use. Thanks for going in-depth on these units, and thank you for holding to your principals, your content and character speak for themselves. Keep up the awesome work!
I've always been way of running power tools on the same plug as electronics, and this confirms that they can cause issues. Good to know with certainty.
Thanks for this PSA! I have one of their power stations and great information to know! So don't plug sensitive electronics in while using power tools at the same time. Good stuff. I didn't watch through till the end to see if you mentioned it, but I would imagine most portable power stations like this probably have the same issue
The problem might not the generator but the saw itself because of how the commutation in those motors work. What you are seeing on the scope are probably resonances in the output filters of the generator. You can probably observe the same messy waveform when the saw is connected to mains. The frequencies will be different but the resonances caused by stray inductance in your house wiring will be visible. It can help to connect a capacitor directly across the motor. (they have special capacitors for those applications with three wires where one wire connects to the motor stator.)
I wish he gave a comparison waveform what happens at mains. From experience I'm 90% sure it will happen at mains voltage. That's why you don't want to plug your oscilloscope and power tool to the same mains tap.
I'm not convinced either way right now, but I had a similar thought right when the video started. Would definitely like to see it done on mains. Which, perhaps still a fair warning about the device's limitation, but, if the same limitation applies to mains outlets then I couldn't really blame the product.
Seriously, the Ecoflow looks like a well built product as it is. So what if it has limitations? Everything in the world has limits. The important thing is to document it well, and inform the users. Ecoflow needs to get on board with that.
This is a typical case of over-promissing and under-delivery... I can imagine the discussions between the Marketing dept and Engineering... One of them is to blame and we all know who has been overpromissing :-). These companies hype up their products and I see sponsorship deals with influencers where they push the demo's beyond 'intended use'. There is a reason why established companies selling off grid inverters sell heavier and more expensive inverters...
@@Dextermorga”Do not use it on a saw”… man, that thing is a power source. If you can plug a load into the mains socket and have it behave decently, then this thing that is a portable mains socket should do it too. You’re excusing poor engineering.
I very much appreciate your general approach to issues; you give each problem a thorough effort, as a result we all learn a lot. Keep the videos coming. senior from Ajax
I respect what he is trying to do, but unfortunately, in this case, this is actually completly normal behaviour of inverter boards and their necessary protection circuits. In fact, the oscillation he shows in this video by the newer version of the devices is actually impressive.
The title of the video is clickbait as fk and he owns a scope but is not able to research enough to understand the details... Video is relative misleading ion the beginning. Goes to show how unusable UA-cam is as a scientific sourcebook :-)
4:37 the setup they show is a simple full bridge rectifier circuit with a capacitor to filter out the high frequency stuff, and a zener diode to cut off the voltage. I dont know about with higher wattage stuff, but when I was looking into making my own rectifier, but zener diode regulators were frowned upon. along with that, they are super noisy under load. but that is only the charging circuit, so it was probably some response about how the breaker flipped when you plugged it in and not the AC generation stuff.
yeah, high-inductive loads will be hard on any inverter-based supply controlled with semiconductors. The instantaneous demand they create is really outside the scope of normal loads. That being said, that can be considered a conducted emission, and (IMO) UL should consider testing for this. These huge battery banks have a real potential use case as jobsite power for construction projects without their own supply, so it's definitely relevant info.
Double respect for 1) not having "should your soul" for ad money and 2) owning up to your own mistakes. Integrity, honesty, and humility are nt popular virtues these days... Thank you, Matthias.
This is why I love the sponsorships you take. The fact that you won't stand behind something you have an issue with not only builds trust with you, but also with the brands who choose to work with you. If a company is willing to let you take their product apart and do tests on it, I respect their product far more than if they just paid someone to say it's neat. I'd expect that behavior from an overloaded inverter, so I suppose that isn't necessarily abnormal, but I would expect overcurrent protection to kick in before the waveform gets distorted enough to potentially damage other devices. I suppose that's why theirs can power much higher load stuff like power tools, where other inverters would just trip when you try to start one up. Maybe a 'safe' and 'high power' mode would work, enabling or disabling the OCP, or at least changing the duration it will trip on, and a simple warning about mixing sensitive and high-power devices on the same battery.
I have never seen your videos before, but after watching this, you have gained my subscription! love to see someone willing to stand behind their word and what they believe in! keep on!
My company designs and builds inverters from 125 watt 12 volt in to 100KW 108 DC (and everything in between/many customizations availale). We use a triple DC stage conversion and microprocessor controlled monitoring to insure the output responds to any load the inverter is rated for (up to X% surge for X-seconds, through to and including direct short, depending on the model). Our best line product responds to your example within a half a cycle to 3 cycles, and there will never be a high frequency result of inductive or capacitive loads. The old adage is still true - you get what you pay for. Awesome review. You've earned my subscription on the first view of your videos. BTW, their revised device is admirable for what it is. You are correct that most people will never have sensitive electronics loaded at the same time as a power tool.
Though lots of people comment that it's "normal" for it to go funny, I think a smarter control circuit / algorithm would have made this high frequency oscillation problem significantly less bad.
@@matthiasrandomstuff2221 exactly - not an issue with our inverters. I was the engineering tech running the scope captures for our inverters during development, that even start powerful air compressors (1 to 3 phase). Battery condition needs to be processor controlled as well as the output. Their next version will probably be an improvement. But as long as it's in a compact form factor, the batteries seriously won't help. They would be better off just collapsing the waveform for a cycle (restart at the projected/computed current zero cross from the last overload condition). Most smart electronics will recover easily from a single missing cycle. I was impressed with their version II improvement and your analysis. You should keep in touch with them and see if they're working on a mod (even if they don't get back with you). Might be an exciting chance for you to reignite the relationship for a sweeter deal. Best wishes.
The sine wave getting chopped to bits when you massively overload the inverter seems absolutely normal. Impressive the small thing even managed to start it. I suggest a brushless saw for anyone using aux power, they should have soft start circuits.
@@matthiasrandomstuff2221 Issue is the saw is causing the ringing and voltage spikes due to inductance leakage in the motor. Its possible the motor starting capacitor is dead or out of spec. or just undersized which is making the issue worse. Try running the same test using grid power. That said the company you should be complaining about is the Kill-a-watt with horrible power supply design.
“Massively overload the inverter”… uh, like the entire point of an inverter is that it controls the output current, and there is no case in which it lets itself be overloaded. It will just keep the output current within limits, down to a very low voltage. If an inverter is overloaded then it’s the inverter’s problem. I’m literal here. You should be able to short the output and the inverter should keep the current to a slightly distorted sine wave, limited to the RMS current on the nameplate of the inverter long-term, and to some peak rating typically 4-10x the continuous rating for short durations like for a couple mains cycles. It’s not hard, and it works very well for starting hard loads cleanly.
I have a Tripp Lite power conditioner unit that I put inline between electronics and a sketchy power source when using one is necessary. The efficiency is only 80% so that's a big downside for a battery or generator fuel. I do appreciate the head's up about inrush vs. electronics on the same source. That never crossed my mind.
This kind of thing is very rare on UA-cam I think when it comes to product sponsers. Thank you for the insight and integrity. Got my subscription. Looking forward to seeing more.
Imagine you are a battery company. You intend promotion, and instead you get some of the most thorough product testing anywhere. I can't imagine being mad about that! I would love having that kind of feed back. You can't improve what you don't know isn't working!
If this is a big company the sponsorships are probably handled by the marketing department. Technical questions like Matthias' would require the responsible ad-guy to contact another department. That is a lot of tedious e-mail communication between two people with a very different expertise. Guy probably just gave up lol
The company I work for makes controls for lots of industrial applications. The noise you are seeing in the output is (as you say at the end of the video) not that uncommon. One of the products we developed had a capacitive dropper power supply originally, but when (in this application) a motorized door was opened on the same circuit it would reset the controller almost every single time. Changed the design over to a switch mode transformer based power supply and the issues went away. In short, the noise you are seeing in the generated voltage is "normal" noise, and it is very much up to the other devices on the circuit to be designed with good enough protection to save themselves from these types of transients. On a positive note, laptops, desktops, phone chargers, etc should all be immune to this type of transient due to being SELV compliant (safety extra low voltage), which requires isolation from mains for end user protection. All devices that use the capacitive dropper power supply shouldn't have end user touchable connections because there's potential for phase reversal (line/neutral swapped).
Thanks for the explanation. Indeed, these capacitive dropping power supplies may have to go away due to inverter issues like this. If I hadn't already pinned my explanatory comment, I would pin this one (maybe I'll pin it later)
You hooked up a massive brushed motor to an inverter - and are somehow surprised that you're noticing noise from the commutator. The inverter boxes are fine. Even a generator will have the same issues with your saw.
Its normal in my opinion, the same issues occur on most switching power supplies when faced with an inductive load that will try to pull all the current in the universe for a split second. I think you're expecting too much if you expect to not see any change in the sine wave during initial power up. Maybe if you had some iron, like a big 110v to 110v transformer and a nice capacitor you could mitigate it somewhat.
I am glad someone on youtube has finally taken a proper look at what HF inverters can get up to under these types of loads. Highly distorted waveforms are, at least in every HF inverter I've tested so far, the norm. The nature of the distortion is different between designs but it's always there. Don't be fooled into thinking this isn't a known problem by the manufacturer. They know exactly what is going on. As for sensitive electronic devices... it's not just having to worry about the device being damaged, a router power supply (switchmode wall plug pack) can cause a HF inverter to produce a highly distorted waveform just by being connected to the AC output when the waveform is near the peak and the input capacitor is fully discharged. I have another inverter yet to be tested, I'm not optimistic.
If anyone actually reads this, the latest inverter I'm testing passed the router power supply test, and even more impressively (to me at least) passed the old laptop power supply test. Waveform didn't distort, voltage didn't drop and slowly build back up (overcurrent protection scheme in the inverter). Will it survive the power tool test...
Matthias, man, what a class act you are!!! This is why i trust you before i'll trust others. You've proven your worth, Sir, and it is of the highest value.
Ok, when you measure AC voltage, it's always a good idea to NOT connect the ground from your scope to the device under test (be it neutral or ground). A safer method is to take two scope probes, connect one probe to the HOT, and another probe to the signal ground (btw, this is not necessarily earth ground and can still have the potential to kill), then use the scope built-in subtraction function to get the desired waveform.
Use a differential probe, that is even better, since otherwise you rely on the ground connection of the scope as a common return and pick up a lot of noise. Or even better: power the equipment under test via an isolation transformer, that way it's completely isolated from the signal ground.
Yeah when I was learning to use an oscilloscope what I pretty much got pounded into my head is that everything will probably be ok as long as you don't screw up the grounds
While a differential probe is great, you can't beat using a battery powered scope when testing mains issues. It just keeps everything isolated so no worries about where signals are coming from.
Never seen using the other channel as a reference in videos/docs that cover this, good idea! Every time I've needed to do this I've just put the DUT on my isolation transformer, but it's only good for 100VA and I don't have a bigger one or a diff probe.
I bought that inverter AC unit after his review. I will consider his opinion if I ever buy the eco flow generator. I appreciate how honest he is. This is great content!!
Very interesting. Thank you for being so honest as to come back and correct the mistake you discovered. It will be interesting to see if Ecoflow will now follow up with you. It was very worrying that they stopped contacting you, it's OK if there was a difference between you and Ecoflow on the severity of the issue you diagnosed. They should not have broke off contact with you, that is a worrying sign about the companies approach to problem solving/customer care. Regarding the other unnamed company that reached out to you to do a review of a solar power generator, you should let us know who they are. I am now more interested in seeing a review of their product than ever, because you do a thorough critical review should not be a reason for them not to follow through with to review there product.
Repeating what has been said, but we need more folks who stand up for what is right. THANK YOU. I would think in a small shop that has a CNC or LASER, this could be an issue. Glad you pointed it out.
So it's not a big deal because like you say it is common for machines like that to cause big spikes/instability but it's definitely something to be aware of. Still thankyou for being honest and even re-uploading whilst admitting your own mistake.
The Big Deal is the damage on the other circuits attached. The saw motor can handle the dirty power being applied. It is the EXPENSIVE b.s. that gets plugged in next to it that you have to worry about. Get those two issues strait.
Pertaining to plugging into the house if there's a blackout- those 2 powerbanks will only run a certain amount of items for a certain time. Especially because the time frame is unknown with blackouts, nobody is about to start running high drain power tools when their house is running on portable battery packs. And people will naturally turn off all but necessities such as fridges, lights and a stray small appliance such as a kettle to save the batteries.
Wow the turnaround time between the two videos was very fast. I just watched the first one this morning, and here's a new revision in just a couple hours.
I can’t tell you how long I’ve followed you and I also can’t tell you why, but I view almost every one of your videos you put out. I dig your style and the integrity you have is almost unmatched. You are the safe for work version of AvE!
I wonder if the fact that the circular saw has a brushed motor has something to do with triggering the oscillations since some brushed motors make a lot of noise. I wonder if that noise, which would be loudest when it’s drawing the most current at start up, messes with the ups’s feedback system and causes it to oscillate.
Matthias is a guy that puts value on his integrity. Ecoflow should feel honored to, even with a minor flaw that he notice, still doesn't bash it, and is honest about how it would only affect costumers in a few corner cases. No product is perfect and I think it is still a good battery, and being aware of this limitation, IMHO, doesn't affect the product reputation.
Thank you for your honest assessment of this inverter's flaws and weaknesses. This is exactly why American inverter manufacturers use only Low frequency inverter tech in their inverter designs. Only uses cheap high frequency inverter tech in their designs. Hint, high frequency inverters offer a dismal 2X surge capacity, a shorter life expectancy, and they provide no galvanic isolation between the DC boost stage of the inverter and it's AC output which can lead to damage to your AC appliances during a catastrophic failure of the inverter's H-Bridge circuit. Only the Chinese use this poorly engineered high frequency, output transformerless design in their inverters.
You (and others) need to do more videos like this and hold the boardrooms accountable for the money making products they peddle. Thanks for sharing. Cheers
Good video :) . This is the main reason that I run 3 invertors here. A sinewave one for the sensitive appliances, and a pair of modified squarewave ones for the heavy tools / machines. In all these 28 years, there has never been a problem :) .
would be interestered to see how this would operate on ham radio antenna rotator - can see a usecase for 'prepping' when main electrics are off and want to rotate a yagi and drive the transceiver also.
Wouldn’t it be great if product makers communicated with their customers like this. I would feel so much better about buying something with some transparency into the testing processes.
Integrity. What is more interesting is the company in question did not challenge your results or have someone review your process to ensure they could replicate your results. Or if they were aware at least compare results; at least I didn't get the impression they did. It is still an interesting issue to be aware of.
@@matthiasrandomstuff2221 I know enough to understand the issues and what happened. I agree with other poster in that Ecoflow should have problem solved with you rather than cutting you loose. I will not post to YT unless I'm an expert about content. That precludes all posts about electronics!
I view this as an application issue. Circular saws are tough on generators and products like these. But I'm investigating this product for an RV power backup. It seems to be well designed for that application. I enjoy your videos. Thank you.
I love this. Ecoflow had no idea they were sending their products to someone who actually understands electronics. Maybe their marketing team learned a thing or two about what they are trying to sell.
I'd say they sent it to him because he did understand electronics. There's no reason that they'd send it to a woodworker. But my guess is they were hoping he'd endorse it without question.
@@stoojinator I've seen woodworkers in the UK being sent them (and using them) - @Badgerworkshop for instance. I've wondered why they are sending these things out in quite such large numbers around the world - I guess they simply want people to see the content and connect Ecoflow with mobile power supplies.
If you weren't a regular viewer, it's very possible to not know that Matthias is a trained and working Electrical Engineer, not a full time UA-camr. I don't think a marketing department would feel comfortable sending a product like this to ElecroBoom or Dave Jones of the EEVBlog, not that I think they would take it.
@@ArkanoidZero Check out EEVblog 1499, although whether he was sponsored is up to interpretation, as neither his forum post or the video itself provides an answer to that. Perhaps it was a bit of independent investigation on his part.
Your bench oscilloscope can be used to make the measurement by using both channels in add and invert mode to make a differential high voltage probe. That is what the add and invert functions are for. The same applies to your Rigol. This does require two probes, and the grounds of the probes should be connected to each other but nothing else.
Thanks for your double integrity. Even if the saw is drawing out of spec, surely the least the company can do is modify their design so that even if they don’t clamp the excess voltage, it fails gracefully. Just because other inverters also failed to fix the problem doesn’t make it OK. I’d also be curious what the startup waveform looks like on a wall outlet.
Its the circular saw that is generating the voltage spikes & ringing from inductor leakage from the motor. Really the issue is with cheap electronics that don't include any transient voltage protection. Its like blaming the car manufacture because you didn't use your seat belt during an accident.
I appreciate your honesty and fairness. One question though: what happens to the waveforms when the saw is on normal line power without the battery box? Is it possible that the oscillations are due to the sudden high inductive load?
This also happens in wall regular wall outlets. As an AV integrator I've seen products die because customers have connected powerful electric panel heaters or fans to the same wall outlet or breaker circuit. Connecting a psu with a backup battery and surge protector usually solves stops this from happening, unless they connect powerful electronics to this also :D
Matthias, I guess they didn't realize you were an actual engineer. Their loss. I guess that's the price you pay for honesty and integrity. Excellent video by the way.
Absolutely! Not just ecoflow. Brands trying to get on the "owning this makes you a worthwhile person" bandwagon. Fine for the brainwashed, but no thanks.
@@dustysparks Established Titles? It smells like a scam, I mean I doubt they are actually planting trees and/or making any kind of charitable donation. Of course the title itself is fake but I guess people buy it thinking they're helping to reforest Scotland, because if they really think they are buying a piece of land and becoming members of the nobility then they should go back to school or maybe even see a psychiatrist.
Do you have arc fault breakers? AFCI will immediately trip as soon as it sees any continuity between neutral and ground. As an electrician, we see this a lot when an apprentice installs receptacles and the ground wire gets up against the neutral screws on the receptacle. Won't even let you set the breaker, but thankfully easy fix. Also makes locating circuit breakers very easy as you just touch ground to neutral, no spark, and breaker cuts off. But, can be a pain if you have an older house and neutrals are shared somewhere. Or if a drywaller makes contact with a screw between ground and neutral. Had that happen on one job and the breaker would hold for a few hours, so it caused some headaches until we tore the wall back out and reran the trouble circuit.
50 amp rms inrush current at 120V rms is 3000 watts. Would oversizing your ecoflow system to facilitate this inrush power mitigate the problem? Also is the OCP on the ecoflow not tripping the 50 amp inrush because its relay curve response time is too slow?
I had to do a little bit with power supply design in the past. Most probably a simple solution is not possible. A simple passive filter would most probably be large but still not very effective. And it might be neccessary to redirect the high frequency energy into some resistor via a capacitor or it could lead to even higher oscillations on the input side which could even kill the power supply. The only clean solution here would probably be to choose a much stronger power supply that can easily cope with the high start up current of the circular saw. Or alternatively use a saw that has a built in soft start feature.
That is normal circumstances for transient noise back into the mains supply which superimposed on the mains sinewave during the inrush current especially with motors. This phenomenon is acceptable within EMC EMI regulation if product complied during the conducted or radiated emissions. This even also happen on when you connect directly on your mains supply depends how far you are from the power distribution or added line impedance e.g. used on Flickers and Harmonics test. Unfortunately this product is under reviewed with no EMC EMI aspect to take into consideration. I cannot see any problem with this product. One thing to consider also is the bandwidth limitation of the scope. I have 500MHz Tektronix scope which is ideal for measuring transient noise. In fact, that noise is nothing compared to surge and EFT spikes injected into the mains from lightning or other source of noise. Also when measuring mains power, it's adviseable to use a high frequency bandwidth and high voltage differential probe which I have one, 100MHz 14kV. One day I would like to accept sponsor and happy to review this product on an EMC EMI perspective. #ecoflow
It's not often YT'ers have enough integrity to call out a sponsor when they see a potential flaw in their product(s). Thank you very much for sharing all your findings, you just gained a lot of respect from at least one viewer! 😎
May not be a lack of Integrity as much as a lack of knowledge. You see many YT channels hawking products that clearly are not in their area of knowledge. Maybe they don't understand the issues at his level.
most likely the voltage ringing you are measuring is a function of load transient response. When the source goes from 0% to >100% in a few milli-seconds, it is a common occurrence for output stability to be at it's worst. To mitigate (like the power an entire house scenario you mention), simply have other loads already powered up. The true fault is likely with the kill-a-watt. It almost certainly underperforms on EMC compliance
This is not integrity, it's ignorance. Of course this will happen with an HF inverter. What you need is a high quality LF inverter that can handle power tools. I'm surprised that little hf inverter circuit even powered that saw! That's insane. And I have shown how to actually break these units and some of the ground neutral bonding issues present when using the double voltage hub. But putting a massive inductive load on such a small hf circuit is going to cause this issue regardless of who the manufacturer is. Use a nice victron with a surge capacity design for your application and you won't have an issue. I try to critique these products and break them on my channel every single day, but this "critique" just shows a lack of understanding of these circuits. Trying to filter that massive surge is just unrealistic. Try an LF inverter instead. And one large enough for the inductive load you wish to power.
I love that you gave that company the benefit of the doubt by saying they “ gave this power supply to some random UA-camr.” And not like “a major player and former EE at RIM.” Even as a non-engineer myself I know who I’ve been watching for the last many years. Never doubted your integrity!
Matthias, Is probably transient high voltage spikes that took out your Watt meter. Try some high power varisters to clamp the transients. It won't solve the onset frequency instability but I don't think that's your problem here. I would look at Bowthorpe surge protectors as they specialise in this area (I'm not affiliated with them). The reason the new higher power devices are less prone to this start up behaviour is that they present a far lower driving impedance to the load. Btw a step down high voltage capacitor is a dangerous way to go. When the capacitor breaks down in a short circuit failure mode, your equipment is then connected live to the mains. As for your scope, try a capacitive step down voltage divider to reduce the voltage presented to usb scope. As for the neutral / earth short issue, to get around that, use a dual channel scope and run the two inputs in differential mode, tie the grounds of the scope to equipment ground. This avoids the shorting problem you experienced with a floating neutral.
If that was powering your house, the inductive ringing likely would less due to all the LC circuits (from power supplies) in your house, and it's distance from the inductive load. Also a larger cap on the in the inductive load would help.
I am a developer for frequency inverters and I fully agree that this behavior is normal.
The reason is that with that massive overload, the output inductance saturates - this could already lead to oscillation.
Additionally, controlling a saturated inductance and a dropping battery voltage is extremely difficult.
To prevent an overload in this case, the physical size and cost of the inverter would probably be increased by factors (assuming this device is already at a good quantity).
So there are only two possible reactions that a device with that size on this load can do: Either shut off or accept some oscillations.
On the other hand: the input circuit with the capacitive voltage dividing on the power meter is a very bad design practice, although this is done a lot especially in cheap products. When testet according to standards that are mandatory for sale in the EU, most of these products fail. All products must withstand some harmonics/flicker/emi/...
EcoFlow just got back to me with this explanation here:
link.medium.com/tDYDK9mTjvb
Apparently, there is a protection circuit that forces the inverter transistors off when a certain current is exceeded, which appears to be independent of the normal control loop. This protection circuit leads to the oscillations.
Adding to that, powering devices with two diodes and a capacitor is cheap but TOO DANGEROUS to use. I consider it the cheapo way to power anything, so I strongly recommend to not buy/use anything powered in that way ;)
@@sucotronic what's the recommended way for things to be powered with such devices?
@@testingoyay4648 batteries or transformers
I mean I see a lot of people praising matthias integrity but I think there was a misunderstanding here ..as you said desining this kind of products it's really difficult so when overloading the inverters with such large currents, transients are to be expected ....ecoflow isn't a bad company ..much better that other portable batteries manufacturer
.if not the best
This is a revised version of the video. The original version inadvertently made the oscillations look 4x bigger than they were. Explanation at 9:05 into the video.
no worries, thanks Mattias :)
I thought there was a glitch in the matrix, since I was sure I'd watched this hours before. Thank goodness I'm not crazy. Well I am, but not because of this!
To be fair, the old waveform is what the KillaWatt was getting!
TYVM for being honest about this problem. I'd considered getting one of these to run my power tools AND charge my phone at the same time. That woulda been a very costly mistake.
I would love to see what happen when you have already some load(200+W), and then start circular saw.
This can make problem go away or it can make it 10 times worse.
And it would be closer approximation of full house backup scenario.
I’ve been following Matthias for about 10 years now, and I remember the first time he did a sponsored video I was pretty upset. I thought he had sold himself out. I subsequently have watched all of his videos, especially the ones where he talks about sponsorship. I really appreciate his integrity. And this is a great example of a class act on his part.
Yeap. It is difficult to findi any reviews of anything out there anymore, that aren't full of bias and neglecting to mention the bad things.
Yep the Dewalt sponsorship, John Heitz got the same sponsorship.
I’m a similar time line, he’s a good fella
IIRC, he was as upset as you. Ha
You were upset about a stranger making money off a sponsorship?
Test like this actually make me more confident about buying one of this boxes, I prefer to know the limitations of the equipment I use without having to blow up stuff myself. I love when companies listen to feedback and use it to improve, it's a shame they weren't willing to do so this time, I hope you can get a sponsor who does.
Well, the problem here is - there is not much they can technically do about the problem. Other than create absolutely new inverter design, that would not fit to the enclosure and would be horribly inefficient for loads that are below its rating. So, it's always a trade-off...
In general, the business case here is - people that use circular saw will not use it in parallel with AC powered sensitive devices - because they are somewhere at construction site, etc. And small devices (laptops, phones, etc.) have separate DC output (100W USB-C, for example) on the front panel that will not be impacted by start of the saw at all.
@@rklauco I think it would be reasonable for the company to add a warning about this specific use case rather than ignoring it so people know what precautions they should be taking.
@@rklauco Can you explain why they can't filter their way out of the problem? A simplistic look at things is that they have to produce one frequency, so everything higher would be filterable, wouldn't it? (I hope to learn something new in the answer, it's not a know-it-all question)
@@rklaucohow do you know that? You don’t. Unless you are assuming that all small devices have a BMS that provides overcurrent protection, which they don’t. Predicting the behaviors of construction workers is best left to Sociologists.
Pro tip: don't try to run massive inductive loads( saws and power tool start ups) on a inverter that has a High frequency like these ones. This isn't a defect, it's a limitation of the tech. Low frequency inverters are the ones that can handle these type of loads but they are WAY more expensive.
There were some good comments on the prior version of this video.
1) The neutral fault: as mentioned elsewhere here, you must use a differential or isolated probe for this kind of testing. Otherwise you're tying mains ground to generator neutral which is a terribly bad idea. Best case that causes a ground loop and will trip an AFCI breaker. Worst case your scope is now a bypass and/or ground reference for the generator and you'll have unknown amounts of current running through the scope ground plane. The test was done incorrectly. I've inadvertently made similar mistakes when checking the output of my gensets and working too quickly, resulting in exciting sparks and breakers throwing, but I blamed myself and not the products. Neutral is not the same as ground!!!!
2) The line noise is the result if the inductive load of the motor. This happens any time an A/C motor starts. The effect on mains is mitigated because the power distribution system acts as giant capacitors. But this is why lights flicker or dim when a large air conditioning unit kicks on. As noted elsewhere, when sitting on a generator, there's nowhere for this inductive load to go and the generator can't spit out enough power instantly to manage it. The table saw is better because the motor is probably not all that much larger than the circular saw, but you have dramatically larger output buffering with those two units than the one small unit, plus the table saw starts slowly. The real point of the 240V on the table saw is for motor torque, not for higher power output. You'll see this same noise on *any* generator - battery or gas. It's not a product fault, it's a use case fault. This is one big reason why large (i.e. >1/4hp) A/C motors should run from *traditional gas* generators. There it's just a spinning rotor - no electronics except for the voltage regulator - so the inductive load gets (partially) sunk into the magnetic fields and there's no electronics to damage. You can hear the two motors (gas and the tool/appliance) fight each other at startup, then run in sync. Generator companies, especially consumer grade, and double especially this new wave of battery ""generator"", are built for common household use cases - refrigerator, lights, TV, not much else. A wood shop is not the designed use case! Sadly their advertising does say "power tools", but their example is a hand drill. And the inductive load of a hand drill startup is far, far less than a circular saw. So sure, they need to be clearer in their marketing, but running a wood shop off of a battery power pack is a highly unusual use case. And construction crews are *not* going to be buying this kind of thing - they're buying plain old gas generators. Even in that use case, the only thing plugged in will be motors - saws, compressors, drills, mixers, etc - which don't care about noise very much. So as long as the genset output is sufficiently protected so it doesn't get killed by the noise (which appears to be the case), then they did things correctly.
You've got good material, but you need to better understand what you're testing and what your results actually show before drawing conclusions.
Totally agree. I was elated the first time I witnessed the signal of a hairdryer on an o-scope. Didn't have a differential probe either. One thing I didn't do is make a video and blame Tektronics for my results!
I stopped this video less than halfway through to see if anybody else caught this. Of course, they did, and Jason did a very excellent job of explaining things. I would have probably gone way overboard. The only thing I would add is if you want battery solution to use power tools, get a heavy-duty transformer inverter that is made for these types of loads. Any nontransformer inverter will have the same characteristics, if not worse.
The issue with Ground vs Neutral is well documented all over the internet, separate problem.
As always, there is a disconnect between design/engineering and marketing. Unfortunately, we see this all too often.
I'm with you most of the way, and I understand your point about using non-inverter generators. However I will say I'm seeing quite a lot of stir in the industrial use space for inverter generators. So I think the focus should be on solving the problem vs saying 'you don't understand and use case, that wouldn't happen'.
I think things like noise constraints and sites that have to deal with emissions limitations (I'm thinking of confined spaces, low grades (holes, pits, ect...), Flammability limitations, ect.... not really the miniscule environmental impact.). I think we are going to see battery 'generators' start becoming more common as battery tech/capabilities improve.
@@calebpurvis6195 Everything will be affected by a huge current spike like that. Motors don't care.
My hobby is retro computers. Since computers tend to have their 0v rail tied to chassis, and thus to earth ground, this isn't really an issue for me.
About VPN sponsors, Tom Scott made a great video detailing how a lot of their claims are dishonest.
He still ends up doing spots for Nord, but I can see how they changed the talking points, so they've learned something at least...
I will forever fondly remember gay pirate assassins.
Look up Naomi Brockwell's video about VPNs if you want to know just how unscrupulous most of them are
It doesn't take an obnoxious Tom Scott video to understand that VPN ads are lies
@@ewicky what do they lie about?
Matthias proved his integrity twice in one day!
3 times, if you count that he made the video to send to the prospective sponsor.
People make mistakes. It happens and there is nothing that can be done to prevent it. What a person does when they make a mistake defines their character. My hat is off to you, Matthias, for admitting your error, explaining it, and then correcting it. That is what integrity IS. Most people do not have the strength to admit when they are wrong and that is a sad commentary on society.
Not really
You need to look up the meaning of integrity! 🥴
This is the opposite of integrity
I'm unsure how one would design an inverter to make the waveform perfect during motor startup.
It never going to be "perfect" but Oversize the inverters and add inductance. It would drive up the cost and size of the box and becomes uncompetitive.
The internet needs more people like Matthias. Thank you for all your honest content.
"Pure Sine" inverters like this (and like UPS' and the like), along with the inverters for the 3-phase motors in EVs all work the same way. They use a microcontroller to make a high-frequency PWM signal to switch a pair of MOSFETS (or IGBTs). This is because FETs are almost completely efficient when on and off, but can't handle analog voltages in-between. So the signal is switching on and off at a value somewhere in the multiple KHz range, and it's period is adjusted to form a 60Hz sine wave, which approximates a sine wave. This is then smoothed with an inductor to fill in any switching gaps and make it a full sine.
With massive inrush currents like your circular saw is introducing, it's overwhelming the inductor's capability to smooth the PWM signal, as the current is pulled out of the inductor before the next cycle can introduce more. Because of this, the output sees (momentarily) the raw PWM signal. The only way to "fix" this is to massively increase the size of the inductor to cope with this one-in-a-million inrush.
Sensitive electronics usually use a capacitor on the input to smooth the DC ripple from their rectification anyway, which would entirely negate this "noise", but because the Kill-A-Watt (and devices like it) require interacting directly with line voltage to measure the throughput, it can't use smoothing components on the line.
Incidentally, the other inverter strategies used by UPS' are "Stepped Approximation of a Sine", which uses capacitor and resistor ladders to create a very coarse sine-line signal, and the cheapest units omit the ladders and just outputs raw 60Hz square waves, letting the cabling act as an inductor to smooth the power a little.
Some of what you are saying is not right. The inverters used for three phase motor control cannot be used on normal three phase loads. Three phase inverters for motor control do not have any inductor between the switches (IGBT or MOSFET) and the motor. They rely on the motor inductance to smooth the current waveform. The voltage output of a three-phase motor drive is the raw PWM.
@@mckenziekeith7434 Well yes. In the case of 3-phase motor control, the inductor(s) are the motor windings.
As you mentioned, it is an extremely minor issue, I'm surprised they weren't happy with your proposed warning of not running power tools and sensitive electronics simultaneously... It's still a fairly strong recommendation overall.
“An extremely minor issue” lol. Since when is a current control loop going unstable when it’s doing what it’s supposed to do (limiting current) an “extremely minor issue”. Don’t be an apologist for them. That’s the sort of a fail EE students would get lots of points taken off if it was a class project. What you claim as minor is a very basic engineering fail that anyone competent would be ashamed of, not apologizing for.
@@absurdengineering What scale was the scope in? How wide was the current clamping window?
I am so happy that I watched this through. I am slowly setting up a fully off grid wood shop. But I may have some more sensitive tools in there. I had no idea that some tools could cause others to blow out. I wonder if those surge protector power strips would be harmed by this? My thinking is to have everything off unless needed to prevent such issues.
Your review made me more impressed with the Ecoflow. I'm ok with that limitation and would gladly buy one if the need arises in the future. It seems like a very well built device.
As you said, a simple warning in the manual would probably suffice, especially for the whole home model.
I respect your honesty and desire to do the right thing. You are a good person. Thanks for the clarity.
If I needed one of those, I'd still buy one. Better to know the limitations. Which in this case are acceptable in my opinion.
Good and honest videos, as per usual.
I would not be dissuading you from getting one. Use it with power tool or with tech, but not both at the same time and you will be fine.
@@matthiasrandomstuff2221 I think that's a fair assessment, it looks like a really SOLID unit all tolled.
@@dustysparks "Ask not for whom the bell tolls" - you will be told that it has tolled for your iPhone! 😱😜🙄
@@matthiasrandomstuff2221 It just occurs to me that you were testing with the circular saw without putting a load on it (i.e. resistance, cutting wood) so the power current would have been higher to maintain the saw's rotations against cutting something, leading to worse oscillation.
@@luminatrixfanfiction problem was just in the startup 39ms and no one starts a saw with load.
4:14 - so you've built an FM radio transmitter! The drop off in response is a little disappointing, though using a circular saw on a portable battery bank is quite a bit of an outlier compared to the typical case of plugging in fridge, a laptop, and a couple other smaller appliances.
Next time I talk to Ecoflow (I might be working on a solar project where they could be a sponsor), I'll be sure to bring this up and see if I also get the silent treatment. Hopefully not!
yes, the cicrular saw draws an insane amount of current for the first fraction of a second.
@@matthiasrandomstuff2221 This may be the reason I find that brushed ebike motors have way more instant "grunt" than any brushless motors I have tried 🤔.
Ha as soon as he mentioned the switch I thought oh lol it's coupling over isn't it... Gotta love the auto setting, I really need to learn to use mine better.
@@Lizlodude I do not trust the auto-button on my scope anymore. Once made me chase ghosts before I noticed I was looking at 50ns.
Not so, I was watching the Skill Builder's video this morning where he was explaining these battery boxes are great for on site, especially if the electrics are off. He was pushing the Ecoflow Delta Pro.
Great video. I really respect you interrogating your sponsors like this and holding them accountable. The reason that sponsorship have become so popular is that they work. people tend to trust the words of people that they like and well trust.
I use a lot of power tools and sensitive electronics (PC overclocking carpenter...) and I have been looking into a couple different brands of these power stations for off-grid use. Thanks for going in-depth on these units, and thank you for holding to your principals, your content and character speak for themselves. Keep up the awesome work!
I've always been way of running power tools on the same plug as electronics, and this confirms that they can cause issues. Good to know with certainty.
Thanks for this PSA! I have one of their power stations and great information to know! So don't plug sensitive electronics in while using power tools at the same time. Good stuff. I didn't watch through till the end to see if you mentioned it, but I would imagine most portable power stations like this probably have the same issue
The problem might not the generator but the saw itself because of how the commutation in those motors work. What you are seeing on the scope are probably resonances in the output filters of the generator. You can probably observe the same messy waveform when the saw is connected to mains. The frequencies will be different but the resonances caused by stray inductance in your house wiring will be visible. It can help to connect a capacitor directly across the motor. (they have special capacitors for those applications with three wires where one wire connects to the motor stator.)
Yeah I really don’t think it’s the generator. Probably happens on mains too.
I wish he gave a comparison waveform what happens at mains. From experience I'm 90% sure it will happen at mains voltage. That's why you don't want to plug your oscilloscope and power tool to the same mains tap.
I'm not convinced either way right now, but I had a similar thought right when the video started.
Would definitely like to see it done on mains.
Which, perhaps still a fair warning about the device's limitation, but, if the same limitation applies to mains outlets then I couldn't really blame the product.
On mains the power has other places to dissipate. On the battery it's a closed system so it needs to be accounted for
@@onurjp You can try measuring the mains waveform. But a saw like the one shown can easily create spikes which can blow up the oscilloscope.
This is literally a non-issue
Seriously, the Ecoflow looks like a well built product as it is. So what if it has limitations? Everything in the world has limits. The important thing is to document it well, and inform the users. Ecoflow needs to get on board with that.
I do not see any issue in my case.
Thank you for letting us know.
So do not use a saw on that thing and you will be ok 😊
This is a typical case of over-promissing and under-delivery... I can imagine the discussions between the Marketing dept and Engineering... One of them is to blame and we all know who has been overpromissing :-). These companies hype up their products and I see sponsorship deals with influencers where they push the demo's beyond 'intended use'. There is a reason why established companies selling off grid inverters sell heavier and more expensive inverters...
@@Dextermorga”Do not use it on a saw”… man, that thing is a power source. If you can plug a load into the mains socket and have it behave decently, then this thing that is a portable mains socket should do it too. You’re excusing poor engineering.
I very much appreciate your general approach to issues; you give each problem a thorough effort, as a result we all learn a lot. Keep the videos coming. senior from Ajax
Matthias is brutally honest, even if it hurts himself. A breath of fresh air to integrity.
He is definitely blunt, but do we know that other brands don't behave the same way? step it up Matthias! lets see you compare them
I respect what he is trying to do, but unfortunately, in this case, this is actually completly normal behaviour of inverter boards and their necessary protection circuits. In fact, the oscillation he shows in this video by the newer version of the devices is actually impressive.
The title of the video is clickbait as fk and he owns a scope but is not able to research enough to understand the details... Video is relative misleading ion the beginning.
Goes to show how unusable UA-cam is as a scientific sourcebook :-)
Agreed
@@stevewilson6193 Lol yeah the other brands would have burned down his shop
The 8 bit guy needs to watch this
Matthias, you are a rare being these days: A trustworthy man.
4:37 the setup they show is a simple full bridge rectifier circuit with a capacitor to filter out the high frequency stuff, and a zener diode to cut off the voltage. I dont know about with higher wattage stuff, but when I was looking into making my own rectifier, but zener diode regulators were frowned upon. along with that, they are super noisy under load. but that is only the charging circuit, so it was probably some response about how the breaker flipped when you plugged it in and not the AC generation stuff.
I like that you called it a battery powered station. Not a generator as many channels do. It drives me nuts lol
@ 8:17 OMG! Thank you for being honest about NORDVPN and not shilling for them.
*Thats a +1 and a sub!*
yeah, high-inductive loads will be hard on any inverter-based supply controlled with semiconductors. The instantaneous demand they create is really outside the scope of normal loads. That being said, that can be considered a conducted emission, and (IMO) UL should consider testing for this. These huge battery banks have a real potential use case as jobsite power for construction projects without their own supply, so it's definitely relevant info.
This video summarizes why I been following this channel for so long. Continue with integrity
Double respect for 1) not having "should your soul" for ad money and 2) owning up to your own mistakes. Integrity, honesty, and humility are nt popular virtues these days... Thank you, Matthias.
This is why I love the sponsorships you take. The fact that you won't stand behind something you have an issue with not only builds trust with you, but also with the brands who choose to work with you. If a company is willing to let you take their product apart and do tests on it, I respect their product far more than if they just paid someone to say it's neat.
I'd expect that behavior from an overloaded inverter, so I suppose that isn't necessarily abnormal, but I would expect overcurrent protection to kick in before the waveform gets distorted enough to potentially damage other devices. I suppose that's why theirs can power much higher load stuff like power tools, where other inverters would just trip when you try to start one up. Maybe a 'safe' and 'high power' mode would work, enabling or disabling the OCP, or at least changing the duration it will trip on, and a simple warning about mixing sensitive and high-power devices on the same battery.
I have never seen your videos before, but after watching this, you have gained my subscription! love to see someone willing to stand behind their word and what they believe in! keep on!
Good on you for having a backbone! You’re one of the good ones!
Yep, that's pretty typical for high frequency inverters... best to keep large inductive loads on low frequency inverters if you can.
My company designs and builds inverters from 125 watt 12 volt in to 100KW 108 DC (and everything in between/many customizations availale). We use a triple DC stage conversion and microprocessor controlled monitoring to insure the output responds to any load the inverter is rated for (up to X% surge for X-seconds, through to and including direct short, depending on the model). Our best line product responds to your example within a half a cycle to 3 cycles, and there will never be a high frequency result of inductive or capacitive loads.
The old adage is still true - you get what you pay for. Awesome review. You've earned my subscription on the first view of your videos.
BTW, their revised device is admirable for what it is. You are correct that most people will never have sensitive electronics loaded at the same time as a power tool.
Though lots of people comment that it's "normal" for it to go funny, I think a smarter control circuit / algorithm would have made this high frequency oscillation problem significantly less bad.
@@matthiasrandomstuff2221 exactly - not an issue with our inverters. I was the engineering tech running the scope captures for our inverters during development, that even start powerful air compressors (1 to 3 phase). Battery condition needs to be processor controlled as well as the output. Their next version will probably be an improvement. But as long as it's in a compact form factor, the batteries seriously won't help. They would be better off just collapsing the waveform for a cycle (restart at the projected/computed current zero cross from the last overload condition). Most smart electronics will recover easily from a single missing cycle. I was impressed with their version II improvement and your analysis. You should keep in touch with them and see if they're working on a mod (even if they don't get back with you). Might be an exciting chance for you to reignite the relationship for a sweeter deal. Best wishes.
Integrity cannot be bought.
It can, I've got some in a box if you want to buy ;)
Special offer 3 for 2
@@Margarinetaylorgrease lol I said integrity not tegridy.
The sine wave getting chopped to bits when you massively overload the inverter seems absolutely normal. Impressive the small thing even managed to start it.
I suggest a brushless saw for anyone using aux power, they should have soft start circuits.
the brushless tools are all battery operated tools.
@@matthiasrandomstuff2221 Issue is the saw is causing the ringing and voltage spikes due to inductance leakage in the motor. Its possible the motor starting capacitor is dead or out of spec. or just undersized which is making the issue worse. Try running the same test using grid power.
That said the company you should be complaining about is the Kill-a-watt with horrible power supply design.
“Massively overload the inverter”… uh, like the entire point of an inverter is that it controls the output current, and there is no case in which it lets itself be overloaded. It will just keep the output current within limits, down to a very low voltage. If an inverter is overloaded then it’s the inverter’s problem. I’m literal here. You should be able to short the output and the inverter should keep the current to a slightly distorted sine wave, limited to the RMS current on the nameplate of the inverter long-term, and to some peak rating typically 4-10x the continuous rating for short durations like for a couple mains cycles. It’s not hard, and it works very well for starting hard loads cleanly.
First video I have ever seen but this level of integrity make me want to subscribe even if I have no idea what kind of videos you make
Massive respect for being honest and having integrity.
I have a Tripp Lite power conditioner unit that I put inline between electronics and a sketchy power source when using one is necessary. The efficiency is only 80% so that's a big downside for a battery or generator fuel.
I do appreciate the head's up about inrush vs. electronics on the same source. That never crossed my mind.
This kind of thing is very rare on UA-cam I think when it comes to product sponsers. Thank you for the insight and integrity. Got my subscription. Looking forward to seeing more.
Sir you are a very rare thing to find these days, a man of integrity. My hat off to you.
Imagine you are a battery company. You intend promotion, and instead you get some of the most thorough product testing anywhere.
I can't imagine being mad about that! I would love having that kind of feed back. You can't improve what you don't know isn't working!
Fully agree, and as Matthias said, it will not affect many customers. Is a honest limitation of the product.
If this is a big company the sponsorships are probably handled by the marketing department. Technical questions like Matthias' would require the responsible ad-guy to contact another department. That is a lot of tedious e-mail communication between two people with a very different expertise. Guy probably just gave up lol
This is why I enjoy your review videos. You are thorough and honest
The company I work for makes controls for lots of industrial applications. The noise you are seeing in the output is (as you say at the end of the video) not that uncommon. One of the products we developed had a capacitive dropper power supply originally, but when (in this application) a motorized door was opened on the same circuit it would reset the controller almost every single time. Changed the design over to a switch mode transformer based power supply and the issues went away.
In short, the noise you are seeing in the generated voltage is "normal" noise, and it is very much up to the other devices on the circuit to be designed with good enough protection to save themselves from these types of transients.
On a positive note, laptops, desktops, phone chargers, etc should all be immune to this type of transient due to being SELV compliant (safety extra low voltage), which requires isolation from mains for end user protection. All devices that use the capacitive dropper power supply shouldn't have end user touchable connections because there's potential for phase reversal (line/neutral swapped).
Thanks for the explanation. Indeed, these capacitive dropping power supplies may have to go away due to inverter issues like this. If I hadn't already pinned my explanatory comment, I would pin this one (maybe I'll pin it later)
You hooked up a massive brushed motor to an inverter - and are somehow surprised that you're noticing noise from the commutator.
The inverter boxes are fine. Even a generator will have the same issues with your saw.
Its normal in my opinion, the same issues occur on most switching power supplies when faced with an inductive load that will try to pull all the current in the universe for a split second. I think you're expecting too much if you expect to not see any change in the sine wave during initial power up. Maybe if you had some iron, like a big 110v to 110v transformer and a nice capacitor you could mitigate it somewhat.
He needs a 120 V isolation transformer. 110 V is a VERY old standard and doesnt exist in Canada.
I am glad someone on youtube has finally taken a proper look at what HF inverters can get up to under these types of loads. Highly distorted waveforms are, at least in every HF inverter I've tested so far, the norm. The nature of the distortion is different between designs but it's always there. Don't be fooled into thinking this isn't a known problem by the manufacturer. They know exactly what is going on. As for sensitive electronic devices... it's not just having to worry about the device being damaged, a router power supply (switchmode wall plug pack) can cause a HF inverter to produce a highly distorted waveform just by being connected to the AC output when the waveform is near the peak and the input capacitor is fully discharged. I have another inverter yet to be tested, I'm not optimistic.
If anyone actually reads this, the latest inverter I'm testing passed the router power supply test, and even more impressively (to me at least) passed the old laptop power supply test. Waveform didn't distort, voltage didn't drop and slowly build back up (overcurrent protection scheme in the inverter). Will it survive the power tool test...
Matthias, man, what a class act you are!!! This is why i trust you before i'll trust others. You've proven your worth, Sir, and it is of the highest value.
That would require quite the inductor and capacitor on the output of the inverter to clean up the signal in that kind of current draw.
Ok, when you measure AC voltage, it's always a good idea to NOT connect the ground from your scope to the device under test (be it neutral or ground).
A safer method is to take two scope probes, connect one probe to the HOT, and another probe to the signal ground (btw, this is not necessarily earth ground and can still have the potential to kill), then use the scope built-in subtraction function to get the desired waveform.
Use a differential probe, that is even better, since otherwise you rely on the ground connection of the scope as a common return and pick up a lot of noise.
Or even better: power the equipment under test via an isolation transformer, that way it's completely isolated from the signal ground.
@@alerighi A differential probe usually costs way more than a pair of passive probes, but yes, it is the ideal probe for this type of jobs.
Yeah when I was learning to use an oscilloscope what I pretty much got pounded into my head is that everything will probably be ok as long as you don't screw up the grounds
While a differential probe is great, you can't beat using a battery powered scope when testing mains issues. It just keeps everything isolated so no worries about where signals are coming from.
Never seen using the other channel as a reference in videos/docs that cover this, good idea! Every time I've needed to do this I've just put the DUT on my isolation transformer, but it's only good for 100VA and I don't have a bigger one or a diff probe.
I bought that inverter AC unit after his review.
I will consider his opinion if I ever buy the eco flow generator.
I appreciate how honest he is. This is great content!!
Respect for you for your honest review.
Very interesting. Thank you for being so honest as to come back and correct the mistake you discovered. It will be interesting to see if Ecoflow will now follow up with you. It was very worrying that they stopped contacting you, it's OK if there was a difference between you and Ecoflow on the severity of the issue you diagnosed. They should not have broke off contact with you, that is a worrying sign about the companies approach to problem solving/customer care.
Regarding the other unnamed company that reached out to you to do a review of a solar power generator, you should let us know who they are. I am now more interested in seeing a review of their product than ever, because you do a thorough critical review should not be a reason for them not to follow through with to review there product.
Now I’m curious if more well known brands of battery boxes (Jackery? Anker?) have this problem.
Repeating what has been said, but we need more folks who stand up for what is right. THANK YOU. I would think in a small shop that has a CNC or LASER, this could be an issue. Glad you pointed it out.
So it's not a big deal because like you say it is common for machines like that to cause big spikes/instability but it's definitely something to be aware of. Still thankyou for being honest and even re-uploading whilst admitting your own mistake.
The Big Deal is the damage on the other circuits attached.
The saw motor can handle the dirty power being applied.
It is the EXPENSIVE b.s. that gets plugged in next to it that you have to worry about.
Get those two issues strait.
Pertaining to plugging into the house if there's a blackout- those 2 powerbanks will only run a certain amount of items for a certain time. Especially because the time frame is unknown with blackouts, nobody is about to start running high drain power tools when their house is running on portable battery packs. And people will naturally turn off all but necessities such as fridges, lights and a stray small appliance such as a kettle to save the batteries.
Wow the turnaround time between the two videos was very fast. I just watched the first one this morning, and here's a new revision in just a couple hours.
I hurried. If I just took the old one down, people would have all kinds of theories!
@@matthiasrandomstuff2221 good point!
Thank you Sir! Always a pleasure to enjoy your time and experiences!
I can’t tell you how long I’ve followed you and I also can’t tell you why, but I view almost every one of your videos you put out. I dig your style and the integrity you have is almost unmatched. You are the safe for work version of AvE!
I wonder if the fact that the circular saw has a brushed motor has something to do with triggering the oscillations since some brushed motors make a lot of noise. I wonder if that noise, which would be loudest when it’s drawing the most current at start up, messes with the ups’s feedback system and causes it to oscillate.
Matthias is a guy that puts value on his integrity. Ecoflow should feel honored to, even with a minor flaw that he notice, still doesn't bash it, and is honest about how it would only affect costumers in a few corner cases. No product is perfect and I think it is still a good battery, and being aware of this limitation, IMHO, doesn't affect the product reputation.
HF vs LF inverter....
Thank you for your honest assessment of this inverter's flaws and weaknesses. This is exactly why American inverter manufacturers use only Low frequency inverter tech in their inverter designs. Only uses cheap high frequency inverter tech in their designs. Hint, high frequency inverters offer a dismal 2X surge capacity, a shorter life expectancy, and they provide no galvanic isolation between the DC boost stage of the inverter and it's AC output which can lead to damage to your AC appliances during a catastrophic failure of the inverter's H-Bridge circuit. Only the Chinese use this poorly engineered high frequency, output transformerless design in their inverters.
You (and others) need to do more videos like this and hold the boardrooms accountable for the money making products they peddle.
Thanks for sharing.
Cheers
Good video :) .
This is the main reason that I run 3 invertors here.
A sinewave one for the sensitive appliances, and a pair of modified squarewave ones for the heavy tools / machines.
In all these 28 years, there has never been a problem :) .
Funny timing with the revised version, something like half an hour ago I watched the old version
You must have still had the link. I unlisted it more a few hours ago.
@@matthiasrandomstuff2221 Probably added it to watch later from the feed like I did earlier, before it was unlisted and just now got to the video.
@@matthiasrandomstuff2221 that makes sense, i opened a bunch of videos open in different tabs and came back to watch them all
would be interestered to see how this would operate on ham radio antenna rotator - can see a usecase for 'prepping' when main electrics are off and want to rotate a yagi and drive the transceiver also.
Wouldn’t it be great if product makers communicated with their customers like this. I would feel so much better about buying something with some transparency into the testing processes.
Thank you for the heads up! I just got one of these so I will make sure to avoid using power tools on it while also powering electronics.
Integrity. What is more interesting is the company in question did not challenge your results or have someone review your process to ensure they could replicate your results. Or if they were aware at least compare results; at least I didn't get the impression they did. It is still an interesting issue to be aware of.
They said "this is normal", which suggests they were aware of the issue.
@@matthiasrandomstuff2221 I know enough to understand the issues and what happened. I agree with other poster in that Ecoflow should have problem solved with you rather than cutting you loose.
I will not post to YT unless I'm an expert about content. That precludes all posts about electronics!
I view this as an application issue. Circular saws are tough on generators and products like these. But I'm investigating this product for an RV power backup. It seems to be well designed for that application.
I enjoy your videos. Thank you.
I love this. Ecoflow had no idea they were sending their products to someone who actually understands electronics. Maybe their marketing team learned a thing or two about what they are trying to sell.
I'd say they sent it to him because he did understand electronics. There's no reason that they'd send it to a woodworker. But my guess is they were hoping he'd endorse it without question.
@@stoojinator I've seen woodworkers in the UK being sent them (and using them) - @Badgerworkshop for instance.
I've wondered why they are sending these things out in quite such large numbers around the world - I guess they simply want people to see the content and connect Ecoflow with mobile power supplies.
@@FFVoyager They send them out because it's a very effective and cheap way to market directly to a target audience.
If you weren't a regular viewer, it's very possible to not know that Matthias is a trained and working Electrical Engineer, not a full time UA-camr. I don't think a marketing department would feel comfortable sending a product like this to ElecroBoom or Dave Jones of the EEVBlog, not that I think they would take it.
@@ArkanoidZero Check out EEVblog 1499, although whether he was sponsored is up to interpretation, as neither his forum post or the video itself provides an answer to that. Perhaps it was a bit of independent investigation on his part.
There's a 200ma fuse soldered in the Kill-A-Watt that probably opened due to the current spike.
thank you for your honesty and intelligence
Your bench oscilloscope can be used to make the measurement by using both channels in add and invert mode to make a differential high voltage probe. That is what the add and invert functions are for. The same applies to your Rigol. This does require two probes, and the grounds of the probes should be connected to each other but nothing else.
Thanks for your double integrity. Even if the saw is drawing out of spec, surely the least the company can do is modify their design so that even if they don’t clamp the excess voltage, it fails gracefully. Just because other inverters also failed to fix the problem doesn’t make it OK.
I’d also be curious what the startup waveform looks like on a wall outlet.
I'm curious about that too
Its the circular saw that is generating the voltage spikes & ringing from inductor leakage from the motor. Really the issue is with cheap electronics that don't include any transient voltage protection. Its like blaming the car manufacture because you didn't use your seat belt during an accident.
Thanks Matthias, this is the exact reason I always love ur videos.
I was on fence on buying EcoFlow now just might hold off the purchase.
Matthias is so rigorous with his methodology. It's great! Keep doing what you do, bud.
Put a isolation transformer between the box and saw to save electronics in the house
I appreciate your honesty and fairness. One question though: what happens to the waveforms when the saw is on normal line power without the battery box? Is it possible that the oscillations are due to the sudden high inductive load?
This also happens in wall regular wall outlets. As an AV integrator I've seen products die because customers have connected powerful electric panel heaters or fans to the same wall outlet or breaker circuit. Connecting a psu with a backup battery and surge protector usually solves stops this from happening, unless they connect powerful electronics to this also :D
Matthias, I guess they didn't realize you were an actual engineer. Their loss. I guess that's the price you pay for honesty and integrity. Excellent video by the way.
I feel like the algorithm knows, because I just watched another video with one of these products in and was immediately recommended this video!
I'm happy never see another ecoflow sponsored video on YT again - nothing against you Matthias they just flooded the platform
Absolutely! Not just ecoflow. Brands trying to get on the "owning this makes you a worthwhile person" bandwagon. Fine for the brainwashed, but no thanks.
The latest one I've seen is the "Become a lord or lady by claiming 1sqft of Scotland"... they're EVERYWHERE!
@@dustysparks Established Titles? It smells like a scam, I mean I doubt they are actually planting trees and/or making any kind of charitable donation.
Of course the title itself is fake but I guess people buy it thinking they're helping to reforest Scotland, because if they really think they are buying a piece of land and becoming members of the nobility then they should go back to school or maybe even see a psychiatrist.
Do you have arc fault breakers? AFCI will immediately trip as soon as it sees any continuity between neutral and ground. As an electrician, we see this a lot when an apprentice installs receptacles and the ground wire gets up against the neutral screws on the receptacle. Won't even let you set the breaker, but thankfully easy fix. Also makes locating circuit breakers very easy as you just touch ground to neutral, no spark, and breaker cuts off. But, can be a pain if you have an older house and neutrals are shared somewhere. Or if a drywaller makes contact with a screw between ground and neutral. Had that happen on one job and the breaker would hold for a few hours, so it caused some headaches until we tore the wall back out and reran the trouble circuit.
50 amp rms inrush current at 120V rms is 3000 watts. Would oversizing your ecoflow system to facilitate this inrush power mitigate the problem?
Also is the OCP on the ecoflow not tripping the 50 amp inrush because its relay curve response time is too slow?
So much of this video is over my head but it’s good to know about. Commented and liked for the algorithm.
A low-pass filter on the output cable made with a ferrite core would likely solve the problem. Simple, cheap and effective.
I had to do a little bit with power supply design in the past. Most probably a simple solution is not possible. A simple passive filter would most probably be large but still not very effective. And it might be neccessary to redirect the high frequency energy into some resistor via a capacitor or it could lead to even higher oscillations on the input side which could even kill the power supply.
The only clean solution here would probably be to choose a much stronger power supply that can easily cope with the high start up current of the circular saw. Or alternatively use a saw that has a built in soft start feature.
That is normal circumstances for transient noise back into the mains supply which superimposed on the mains sinewave during the inrush current especially with motors. This phenomenon is acceptable within EMC EMI regulation if product complied during the conducted or radiated emissions. This even also happen on when you connect directly on your mains supply depends how far you are from the power distribution or added line impedance e.g. used on Flickers and Harmonics test. Unfortunately this product is under reviewed with no EMC EMI aspect to take into consideration. I cannot see any problem with this product. One thing to consider also is the bandwidth limitation of the scope. I have 500MHz Tektronix scope which is ideal for measuring transient noise. In fact, that noise is nothing compared to surge and EFT spikes injected into the mains from lightning or other source of noise. Also when measuring mains power, it's adviseable to use a high frequency bandwidth and high voltage differential probe which I have one, 100MHz 14kV. One day I would like to accept sponsor and happy to review this product on an EMC EMI perspective. #ecoflow
It's not often YT'ers have enough integrity to call out a sponsor when they see a potential flaw in their product(s). Thank you very much for sharing all your findings, you just gained a lot of respect from at least one viewer! 😎
May not be a lack of Integrity as much as a lack of knowledge. You see many YT channels hawking products that clearly are not in their area of knowledge. Maybe they don't understand the issues at his level.
most likely the voltage ringing you are measuring is a function of load transient response. When the source goes from 0% to >100% in a few milli-seconds, it is a common occurrence for output stability to be at it's worst.
To mitigate (like the power an entire house scenario you mention), simply have other loads already powered up.
The true fault is likely with the kill-a-watt. It almost certainly underperforms on EMC compliance
This is not integrity, it's ignorance. Of course this will happen with an HF inverter. What you need is a high quality LF inverter that can handle power tools. I'm surprised that little hf inverter circuit even powered that saw! That's insane.
And I have shown how to actually break these units and some of the ground neutral bonding issues present when using the double voltage hub. But putting a massive inductive load on such a small hf circuit is going to cause this issue regardless of who the manufacturer is.
Use a nice victron with a surge capacity design for your application and you won't have an issue.
I try to critique these products and break them on my channel every single day, but this "critique" just shows a lack of understanding of these circuits. Trying to filter that massive surge is just unrealistic. Try an LF inverter instead. And one large enough for the inductive load you wish to power.
Why wouldn't it just brown out like a normal supply when faced with startup current?
I was surprised this video wasn't sponsored by established titles like every other video I have seen this week.
I love that you gave that company the benefit of the doubt by saying they “ gave this power supply to some random UA-camr.” And not like “a major player and former EE at RIM.” Even as a non-engineer myself I know who I’ve been watching for the last many years. Never doubted your integrity!
Matthias,
Is probably transient high voltage spikes that took out your Watt meter.
Try some high power varisters to clamp the transients. It won't solve the onset frequency instability but I don't think that's your problem here. I would look at Bowthorpe surge protectors as they specialise in this area (I'm not affiliated with them).
The reason the new higher power devices are less prone to this start up behaviour is that they present a far lower driving impedance to the load.
Btw a step down high voltage capacitor is a dangerous way to go. When the capacitor breaks down in a short circuit failure mode, your equipment is then connected live to the mains.
As for your scope, try a capacitive step down voltage divider to reduce the voltage presented to usb scope.
As for the neutral / earth short issue, to get around that, use a dual channel scope and run the two inputs in differential mode, tie the grounds of the scope to equipment ground. This avoids the shorting problem you experienced with a floating neutral.
You pissed off a sponsor again?
ha ha
@@matthiasrandomstuff2221 it could be worse - you could have dropped it off of a scaffold...
If that was powering your house, the inductive ringing likely would less due to all the LC circuits (from power supplies) in your house, and it's distance from the inductive load. Also a larger cap on the in the inductive load would help.