@@obviouslytwo4u What he means is that if you don't care what the ongoing cost is, and you set it to exactly £1.00 - then you can use the COST display as a second KWH display (as they will be identical at £1.00 per KWH). So you don't have to use the FUNCTION button to load the next display screen. Clever idea :)
We used to use ¼ watt resistors as substitutes for pyrotechnic fuses because you need a licence for the latter. They worked well in theatre pyrotechnic applications.
Thanks for the video! I bought two of these about 2 years ago. They both failed recently with blank displays. First look inside saw corrosion of the batteries in each so I ordered replacements. No improvement. Had to delve deeper, and eventually found your video, thanks! First device I tested proved CX1 was low value (0.3uF), so I ordered a bag of 10. When they arrived, I fitted them in both units. One unit burst into life again. The other didn't. Now found that D7 in one device has failed. So I am waiting for yet another delivery now...
I've found if you get your power factor low enough the active power is so delayed, it gets pushed into the next billing cycle, I switch power companies every month and haven't paid a cent in years.
Amazingly they've actually changed the board design over the years. I have a much older unit. I have to hold Function and Cost buttons down to get the date code to show up, which is from Dec 18, 2009 although the board says 2014.Still working fine, although the NiMH battery is corroding, so I'll have to remove or replace it eventually. And yes, I had to resolder the ribbon cable as it broke off. :) My model only has one blob-type microcontroller on the LCD board, which might just be an LCD driver. And then the AC board has a small power analyzer chip, which given the lesser number of pins, might mean a lower model. And it also has the main microcontroller chip, 32 pin or so, on the AC board. Seems they've definitely upgraded this cheap meter over the years. Highly recommend getting one! Especially the models with a backlight
Lusting after the HOPI, I found and bought one of these cheapies on ebay a year or three ago. I am pleased to see that it gains the seal of approval. Some reviewer or critic at the time (I forget who or where) gave me a moment's pause by a snarky comment about accuracy - but the price was irresistible and I have been very happy with it.
Same experience on reddit after I asked a question about my experience with it. Only bought one about two months ago. I’m unsure what to do with it now. I’d like to measure my washer/dryer but the plug is behind it and the thing is just so bloody heavy
I've had about 10 of these things over the last few years, I keep lending them out to people who end up asking if they can keep them! They normally cost about GBP8.00 each and simply work. The 8051 was a bit of a surprise, and a blast from the past as I spent about 10 years of my life programming them in PLM51 and ASM51. That said, they are cheap enough and adequately powerful for this sort of job.
@purringChaos Depends on the specific processor. Some 8051 chips disable letting you read out the firmware with a debugger or accessing the flash some other way. Of course there are methods to get around that (power glitching, clock glitching, etc.) but that stuff requires specialized tools (usually fpgas) and can be a lot more work than it is worth. It's also of course very forbidden to reverse engineer firmware like that, so if you plan on doing anything commercial/public with it you should stay far away from doing that. If you are interested in these methods in general, there are a lot of talks about it from security conferences. I haven't tried it, but the Chipwhisperer seems like a great tool if you have about 300 to 500 bucks to burn.
@@drjekyllmrheil1001 To be fair, at that point (since I don't have any of those tools), I'd probably just write my own (assuming the full datasheet is available), all the heavy lifting is done by the power monitor chip the display is by the Holtek (that I might use a logic analyser on if I don't want to muck around too much figuring it out). Not sure how long it would take with the tools you mentioned. Then again, I'm cheap and I tend to over-estimate my actual skills :P
I've thought about putting together a modification to integrate something like a Nordic nrf52 (Bluetooth / 6LoWPAN + ARM) MCU (commonly found in cheap fitness trackers, and related to the chip in the BBC microbit) into one of these types of power meters. Looks like that would actually be pretty easy on this one.
I bought a couple of these years ago that use a standard set of button cells..they seemed to go flat between 3 to 6 months...these newer ones look so much better...I will check Aldi again to get the new ones!
@@3Dparallax Did you tear it apart? Does it have the same chip? I found on Ebay/Ali this one with blue buttons but also a similar one with green buttons (and different display) - wonder if the green one also works well and has the same chip.
I have 3, the one I thought was an ALDI was an older (2013 ish) Kogan and pretty much identical inside, same chip, yes the battery was leaking slightly. The rellies have the ALDI one (2015) but I couldn't take it apart, appearance wise, identical as the Kogan one.
I've got the same model and it died a while back, a surface mount resistor in the power chain had cooked, so I replaced it with 1/2W wire-ended device. works fine now, doesn't get warm either.
My favorite applications for these devices is for getting a rough idea of what you can run on a given current budget, or for making a really nice 'dashboard' for my old school generator. At this point I know I can run one fan, one LED light, and one USB charger per bedroom off our relatively small generator.
I've got a slightly older model of the same meter with North American plugs (the main board is V0.9 with a date code 6 months earlier than yours). My tests agree with your findings - surprisingly accurate, especially when you consider the price.
I designed something similar a couple of years ago based around an Arduino and some op-amp input circuitry with voltage and current transformers for sampling house mains power. Every second it calculated and stored RMS current, RMS voltage, real power and power factor to a USB flash store for later download and analysis. The complex bit was not the digital calculations which are actually very straightforward once you have accurate waveform samples, but rather the analogue input circuitry for the current sampling. The problem is the vast range that this has to deal with, and how to maintain sampling accuracy for a milliamp scale waveform versus up to a 50 amp waveform. I wanted it to be able to monitor power consumption for the whole house, including accurate real power measurement of standby loads of single figure wattage. I ended up using an auto-ranging approach, changing the analogue input amplification for different current ranges. I wonder if this chip does the same internally. As you say - interesting stuff.
These watt meters have 2 issues, #1, they crash A LOT from any noise inducted in their proximity. #2, the frequency display is 50/60 so its useless for say setting a generators speed
It's for mains voltage so and frequency so it's to be expected. I mean when I hook it up to my inverter it tells me 61hz if that makes you fell any better.
I'm a power meter designer/developer for 20yrs. Those "metering ICs" are intended for low-end, fast time-to-market, cookie-cutter products. Each time a new product design is started in my division, they are quickly ruled out because of various imposed hardware limitations...
@@dl5244 What kind of chip would you use for high precision in a device like this? Also, what kind of hardware limitations are there with the cookie cutter chips?
Smart meters, ordinary dumb energy meters for measuring what is coming out of a Solar or wind inverter, public electric car charging sites, power tools hired out to customers ..... There are a lot of applications for a chip like this, besides plug-in energy monitors. So there will be plenty of them in existence. And the chip doesn't rely on many external components, which could introduce errors due to manufacturing tolerances. There are the resistors in the divider chain, but ±1% resistors are common enough nowadays and better are available. All the analogue circuitry will be on one tiny area of silicon, so it's all at the same temperature, and everything is digital and sent to the μC as binary values. It's probably even picking up accurate timing from the incoming serial data (since the analyser chip is crystal-controlled). Notice also how those socket inserts and plug pins could easily be substituted with a Schuko or French/Czech socket and matching plug pins, for use on the Continent?
@@sugarbooty this VF9261F metering IC chip is probably perfect for a cheap low end device like this with maybe a dozen possible measurements. the PGA design likely precludes the ability to make continuous (aka "gapless") calibrated measurements over a wide dynamic range (not suitable for meteorology certifications like MID, ANSIC12.20, IEC62052-11, etc). I like that the bandpass filters are somewhat soft configurable, but gain stability looks like a risk... maybe it could be compensated for, but usually not (because of part to part variation or drift over time, external influence factors, etc). The bandwidth is also quite poor. I'd like to see at least 3kHz for a low end meter and 10kHz for higher end (with some oversampling of course, so sampling rate would be >2x this) The biggest problem with all these metering chips I've looked at is they do not make (continuous) digital waveforms available for power quality algorithms and/or validation.
Maybe that transistor on the input of the power supply is being fed to the processor to get the frequency of the mains. Should get a nice square wave off that using an internal pull-up.
Since the interface is a UART, it might be possible to swap the display board for a bluetooth serial interface and get a much more complete display on a remote monitor, or even a phone.
@@bigclivedotcom if you power the Bluetooth module from the mains, the Bluetooth is your isolation. Development would be tricky, but could be done by plugging the module I to an isolation transformer
@@Senshikaji He's welcome to come and do that in our house. Our electrical systems are far more robust than his. Even our 5A lighting circuits could blow 1 or even 2 'woodworm' holes in light bulb caps without us needing to replace the 5A fuse wire. Well, at least I think it's 5A that's the fuse marking.
Beautiful review. I just bought two of these. Yes at 14:47, you query if the pin is for a backlight. Indeed, some units have a blue LED backlight, for a couple of quid more.
I bought one of those from Tesco. It was defective. One of the terminals inside had some really dodgy soldering and was hanging on by a thread. I was going to use it on a high current device (a "large water purifier for the purposes of purifying ethanol-containing water"). Good job I opened it up first and resoldered the connection!
It wasn't always. Originally it was licensed from Intel. That's going back decades though. Atmel did some very nice 20 pin variants. Philips did some excellent ones too. I still have the manuals for them as there are some tricks that are stull useful these days EG for running serial ports over multiple KM links.
@Mai Mariarti Granted they don't really care about IP. But look at it this way, if you have two options sitting in front of you, which are functionally equivalent, barring the fact that one of them is royalty free which are you going to choose to start manufacturing in the billions? Makes business sense to go with the option that has zero chance of getting you sued. FWIW I've yet to see any Chinese uC's ripping off other architectures that aren't royalty free, doesn't mean it's not happening, but I've not seen it. Certainly there's counterfeiting happening but that's a different discussion, stealing somebody's lithography masks to produce verbatim copies of their chips is easier to disavow (also easier to do) than taking their core and integrating it into your own silicon.
I was on the fence about buying this and you convinced me to take the plunge so thank you! Then I realised I was only 4 mins in and you completely lost me with the rest, perhaps I will be ready for such information one day, very impressive. Great review!
The layout of the display circuit board is beautiful. I wonder if it was designed expressly for this application, or if it was reused from another application.
I have a older version of this. Less surface mount stuff. The screen was dim as hell making it Basically unreadable. I now opened it up cleaned the contacts for the screen. Works great now. Thanks BigClive I can use it now.
I have the V2.0 of the version of the microcontroller board and its only a 8 pin chip with more welcome improvements. Including a cr2032 replaceable battery and a simpler pcb layout with less components to save cost. Brilliant in my opinion.
Always feels weird when I see a product I actually own being reviewed. I have the US version of this and I'm pleasantly surprised to learn how well made it actually is inside. I got mine off Amazon for cheap (as I was writing that I got a notification from my bank app on my phone through my PC that my Amazon Prime subscription just renewed).
@@KK-hy6sx It just charges off the mains. I'm sure there's some circuit in there that ensures whatever power it's drawing for both the display and the battery isn't included in the stats.
I have one provided by my power company for joining them years ago. Works great, but my one gripe is that the price per unit is something like 29.35 cents, but since I can enter only two decimal places (i.e. .29 in this case), it's perhaps not as accurate over a long period of time. Actually, a second gripe is that the switches for the power sockets are typically above the socket, and that the device hides the switch, so one has to switch the socket on first before plugging the device in. That's in AU/NZ, anyway.
I have a couple of these I bought a few years ago to read wattage coming in from my solar panels. But the new ones have an overload protection and will not allow inrush current from panels and reads overload.
Or even esp8266. Was thinking the same. You could leave the existing 8051 in place to keep the calibration and then tap into the tx/rx to sniff the data. I'm wondering if the power sense chip sends all the data simultaneously to the 8051 for selective display. If that's the case then you would be able to show all values remotely. Be careful of the live coupled ground and dirty 3.3v power tho.
Ordered one to do just that. :) Might be a bigger job that I initially thought though... EDIT: I wonder if powering up the display board without the metering board connected would still let the MCU send all the calibration data anyway...
I bought two of these from Amazon. Both stopped working after 30 minutes of use, just zeros on the display. Sent them back got two more. Both did exactly the same. Replaced with Smart Plugs that give wattage etc on phone app. These are working fine.
Love these meters . Use them as to determine how much someone on a spur has used and with a pwm controller for distilling to adjust in real time. They are cheap and effective and although I havent even got to the disassembly yet I'm very pleased you chose to tear these down. Great!
When it comes to power meter chips for devices like this I figure it's a ton easier and cheaper to just get one someone else has made in bulk of metric craploads than to attempt a DIY that would have to be done from scratch since no off-the-shelf parts would likely be a 1/10th as accurate or resilient, especially since the one you found has a manual of 100 pages that I'd bet has instructions on support component circuit design for how to properly use it like LED example circuits but more complex and specific (yes you have to leave that cap of x value in...and resistor y).
But existing displays are never exactly what you want. So building your own (or modifying one like this to run your own firmware) offers the possibility of having all the functions you want.
I assume you tried changing time and date while in 'calibration mode' I always think of the triangular earth symbol as being mains ground (connected to the planet), and the one with the diagonal lines as chassis ground.
I've had one sitting in a drawer for a couple of years. Din't think mine did displayed the power factor, so got it out and check. Yep, it does all that your one does. Now it's plugged in to charge the battery.
i bought one in Brazil. Surprisingly, it can handle the 220V between 2 127V phases that's used in my city and the 127V between phase and neutral in my friends city. And yes, there's a model with backlight.
I used one to measure power consumption of an old freezer in my garage. Expected poor efficiency but actually got the opposite. Turned out the freezer wasn't suited to outdoor use and so wasn't working at all when the ambient temp dropped below 12c. Replacing that freezer before winter and saved us from potential food poisoning. Glad I have one now
I've got a similar one branded "Energenie" (green buttons). I opened it for comparison, and has a roughly similar board. I'm glad I did that - one of the electrolytics had leaked, so I will be re-capping it!
Do you know if it has the same chip Clive showed in this video? I'm not sure which one to buy, the blue one Clive showed or the green one, so any more information would be nice. Yours has (from left to right) History, Energy, Cost, Set buttons, yes?
I have been using the exact same bodied EU version of the same energy monitor. Its plugged in 24/7/365 for about 2-3 years now, with constant load (100w-2kw). I even tried the oven as a load, and it maxes at around 4.4kw
I bought the North American version of this one. It stopped working after only using it a handful of times, however that was over the course of a year so the warranty had run out. Hope yours holds up better than mine did. :)
You've been reading my mind! I was just searching for a device like this and was comparing with the hopi meter. Eventually I ordered a similar one, not exactly yours, instead of the hopi. The hopi was too expensive and the cable was not convenient for me. So i ordered the other power meter from China and a few hours later i am watching this video! Thanks a lot Clive. Really. I've been enjoying your videos and learning a lot.
I have a suspicion about the voltage & current sensing chip: It might be one of the 24-bit sigma/delta converters. That way you would get a trimendous sensitivity and range. On the other hand, as you show differential (or maybe instrumentation) amplifier input stages, maybe that would not be necessary. In any case, it is obvious that the key to accuracy without any trimpots is some calibration table, which then requires really good reference(s). Interesting, as you keep saying!
Looks like a clone of HAMA 47770 energy meter made by Xavax, later branded as HAMA 111908 XAVAX, LOGILINK EM0003, ENERGENIE EG-SSM-01. The first two made by Xavax are using Cirrus Logic CS5460 chip and there is an eeprom & blob chip on the lcd board. Same battery, similar power design with a few extra protection options like a thermal fuse
I was looking for this on Ali and just before I was going to order the cheapest one available I noticed there's also a model with backlight. Only $2 more and for me well worth the extra cost. So if you want/need backlight, be aware there's 2 models.
I have the 120 volt version of the same meter. It works well. However, the rechargeable backup battery in it gives out if the power outage lasts for than a couple of hours. It's memory then forgets all the last readings and goes back to 00.00.
It's nice to see you review a design that's actually *good* for a change, instead of "here are all the shortcuts they took that dragged it well into *unsafe* and *we have regulations for a reason*" :-)
I've got one of these, thing that didn't last long in it was the battery, wouldn't hold a charge after a year of on and off using to measure different devices. What about using a small lithium cell in it? I measured the voltage and it wouldn't charge over 3.77v no matter how long it was connected.
Your 3.77v measurement was probably with the battery connected. With the battery disconnected I would expect around 4.3 volts output. You could probably get away using a protected lithium cell, but certainly don't use a bare unprotected cell.
I just bought one of these from Amazon, but the "calibrate" function doesn't work. I looked inside (of course) and the chip is a BU6523GX, possibly a different method to calibrate, but the voltage is 1 volt lower than my Fluke 289, so I cannot complain! Also, the time display does not increment if the load is less than 1 watt, along with the power factor display... interesting! I thought it had locked up, but as soon as I gave it a load, the counter started and the PF display updated.
I've had 2 of those for measuring the 24h cost of my diy cannabis lamps,they've both broke current passes through them but they don't measure anything anymore
@@Ulven3829 240 is not used in North America for anything the average person knows can be unplugged, between that and the half dozen different plugs in use I doubt there's any market for a kill a watt type product. Anybody that curious likely knows how to rig up an amp clamp anyhow.
i have a "paget trading 9149" for maybe 20years, and i've noticed that it 'seems' to accurately integrate over very long times. so for a very small load, if you wait, you eventually get an energy value, seems the accuracy is better than the display resolution. (you can also hack this a bit by using a 'silly' monetary conversion that scales the display up a bit.)
Looks like with that power supply circuitry, it should be okay to use with square wave or modified sine wave inverters that can cause issues with the power supplies in the very popular Kill-A-Watt power meters...
I have a similar unit. The Ni-Mh battery failed in reverse voltage and backfed about -1V. After desoldering it, the meter turned on again, but all the measurements are completely out of spec. Any idea what could be the cause?
@@bigclivedotcom Thanks for the reply, The model I have actually does have an EEPROM on the board next to the battery. But the EEPROM itself seems to be powered from the 5V Regulator. It's a standard 24LC02 jobby.
Nice curvy display board. And a great chip. Now we just need a BigClive series where you build your own including schematic, and have a non flicking screen showing all the pertinent information!
I actually have one of these with a Schuko socket, purchased from Aliexpress for about 7 or 8 euros. I have not checked, but I bet the cost calculator cannot handle separate daytime and nighttime tariffs if they differ.
Thanks Clive for all your great videos! I was wondering about the labeling of the inputs L and N in the circuit. I'm trying to understand and follow along with the explanation of the circuit but I'm seemingly unable to grasp Line and Neutral. I thought Neutral was 0v and not Line as specified in the drawing. If someone could clear that up I would really appreciate it.
The 'Peacefair' branded power monitoring solutions on AliExpress are pretty good... you can get ones that have all the info on the display at once like the Hopi, ranging between four and six difference bits of info you can access. Plus there are versions that are TTL serial and even RS485 ready... making them PC or microcontroller project ready.
I love mine. During an energy survey at work (school building) I did some nifty math to use the electricity supply meter (the type with spinning disc) to measure the building base load. To do this I turn off everything that is normally switched off, count the number of disc spins in say 30 seconds. The meter says how many spins equals 1kwh. Using those numbers I was able to estimate the base load. It would also be possible to disconnect circuits to try to use the electricity company meter to measure each circuit. However that looked like it would cause hassle for my IT background equipment and who knows possibly other timers running 24/7. I had suspected that a streetlight on a lane outside the building supplied through the school. No one knows who owns the lane but the light has electricity co signs on it. There is a local streetlight LED program ongoing. They changed sodium for LED and the next time I measured the base load it had gone from 18 to 16kwh per day. Of course it is possible that reduction is linked to some other change. On my meter I got annoyed that it wouldn't remember the cost price, it needs to be inputted each time. A notebook is handier. At 25 mins that might explain it. It is best used with an extension cable for sockets in hard access places
I've repaired one of these that died not too long ago, I think it was a resistor that burned up surprisingly enough, replaced it with 4 1/4W resistors (2s2p) which was the best I had on hand... I had previously had to repair a Kill-A-Watt where the X2 capacitor had died (foil burnt up and capacitance had dropped to the point of the LCD barely being visible, and the measurements being off by a factor of about 2).
I have this in Australia and when measuring one of my Philips hue lights in red vs blue, there was quite a difference. I posted on Reddit about it and was told that it’s garbage. Anyway, I’m still unsure about the different colours at the same “brightness” according to the app. Anyway, small story.
The power up delay could be the noise floor adjust. Is so, it MOST be carried at NO LOAD. First chip is just the input stage AD and MUX. The other chip carries the processing. Some metering solutions, in POLYPHASE circuits, NEED some means of INSULATION (depending on the design). So the use chips like this ("remotes"). They are feed from a basic capa dropper on the hot side and use optocoupler to send the data strean to the main micro who is receiving the other 2 phases information as well. So, if you are not using CTs, you need isolation of some sort. Thery took the AFE part from the main chip, put in a separate chip and send data over isolated lines. Advantage is shunt are LINEAR and inexpensive.
There are some statistical tricks like oversampling, to produce 12 bit resolution with a cheap general purpose micro controller, that has a 10 bit A/D converter. Some µC also have fancy stuff inside for such applications.
usually the problem is that cheap ADCs have a wide tolerance for linearity. The effective number of bits (ENOB) in most cheap ADC's is several bits lower than the marked resolution. The result is that regardless of precision, the accuracy for signal levels will be high at some levels and low at others... Personally, (20 years designing power meters) I've never seen the oversampling design idea get past the napkin & speadsheet stage. It always makes more sense to buy the right part. And of course with that integrated metering IC, there's not even a way to read the digital samples... only the aggregated/calculated measurements.
What meter is this? I notice you did not put an affiliate markings to say what brand it is. It would be nice to buy one that has been thought out rather than a clone that could catch fire.
Thanks for the review. I have always used Voltcraft energy meters as they were the accurate ones. But The energy meter 3000 failed after 3 years, then a new energy monitor 4000 failed just after warranty period and today I found out that another EM4000 I have bought failed exactly the same way. They started to show random data and after some time totally wrong measurements.
there are power measuring modules in aliexpress that are probably based on the same chip. Search words:AC Digital Multifunction Meter Watt Volt Power Amp Current Test Module PZEM-004T G08
I'm going to ask a silly question. Current shunt - big bit of wire with low resistance. Why wire above the circuit board - why not just use a track - space for a track to carry that much A? The shunt is connected to what its measuring by circuit tracks though?
I would presume the resistance of PCB trace would be much more variable between units due to manufacturing tolerances, possibly more than calibration can account for easily. The measurement traces don't carry any current so resistance doesn't matter there.
In addition to less variation between boards, shunts are also often designed to have a better temperature coefficient. And probably cools better as well.
Speaking of discharging capacitors, the worst I've ever seen in consumer electronics was when I was taking apart my DSLR. I knew I should take great care about the capacitor for the internal flash, but the first time I've done it, I just shorted the leads. In the middle of the night, no less. Yeah, what a horrible idea :D That was a hell of a bang still. The next time I made it drain more slowly, that's for sure. I see people taking apart their cameras on the internet all the time, and it's upsetting how many seem completely oblivious to that capacitor and/or don't even bother with a disclaimer to warn the viewers/readers. That capacitor can be dangerous.
I got four power meters. A newer one is smaller w/ LED white back light. Founded many electrical appliances with high standby current, I used Wifi plugs to solve the issues, even though it is a Wifi UV air conditioner.
I wish these had two different reset options, one to reset everything (like it does currently) but another where it would only reset the electricity usage figures, *not* the cost you set! If you reset it then you have to put the cost in _every_ single time you reset it. Tedious when you're going around the house trying it on 30 different things.
Hi, I have a Vodafone P3 Sure Signal unit which stopped working. It stopped giving mobile signal then got hot and the lights went out. It is 3 years old. Would you like to have it to see what went wrong?
I bought four of these excellent meters (under the Status brand) from a clearance sale at Dunelm for £2.49 each. I already had a couple of the Nikkai meters, sourced from Maplin (which you also have used on this channel). The Vangotech V9261F is an interesting device. Thanks for the teardown. 👍
uK8cvPAq Ho yus. I think the normal price of the Nikkai N67FUs was around £14 but I picked up a couple when they were priced at 50% off. 😁 I remember getting a couple of UNI-T UT30B multimeters in a similar sale for about the same price (£6 ea, iirc). I don’t think I ever paid full price for the more expensive gear. I only bought the bench PSUs because I had a discount voucher from previous purchases. Maplin got ridiculously expensive towards the end. I paid £80 or so for the PSUs which finished up just shy of £120. I remember a couple of their more expensive Middlesex electronics kits from their range actually doubling in price over a few years. They were a marker for me and I would often pop in to the local store just to see what the price had climbed to. I do miss their presence on the high street though.
When a manufacturer showes the energy rating of an appliance ( B,B+ etc) is the power factor taken into account? I ask because as you have mentioned that smart meters being reprogrammed to read this at some point in the future. Keep up the good work with the channel.
I normally use these metres to detect a current leak through the Earth. you can make up a wired device that allows you to monitor any current that is being leaked through the earth that might interrupt a RCD module in your consumer unit.
This meter is sold on Amazon under the brand "Energenie". If you go to the customer reviews there and search for "fire", you can find various reviews and pictures of it where the live of the socket has burnt out.
@@bigclivedotcom Just bought one that is externally identical with a V9240 8 pin chip made by the same manufacturer & the live pin going into the wall was pretty warm (not hot) after boiling a kettle of water at just below 3KW for around 5 minutes. Intertek/GS certification stamp is missing on mine...though were they really certified at all anyway?
i saw that... actually i 'm 99% sure you can do that, the decoupling capacitor on the input may just serve as noise suppressor and the circuit is too small to catch noise, also the resistor is just a resistor, i think there's no reason to self oscillate... i've seen a schematic like that somewhere to reduce dissipated power on the regulator, apart the case of making a crude battery charger to limit the current (i've done that with a LM317 for LiIon)
Ok so not clear on the operation on the battery inside. I have a Bluetti EB70 and after a year of using it i want to use a power plug to test how much output in KW/h it has from fully charged before it shuts down . My problem is once the power bank shuts down . The power meter plug turns off. And i lose the KW/H given as the socket is shut off and the meter is blank. And if i plug it back in , is the memory lost?
set the cost to £1.00 per kWh, and then you can read kWh and instant watts at the same time.
You need to find out how much your provider is charging you per kilowatt hour.
@@obviouslytwo4u aparently you didnt get the point here
@@urugulu1656 cocobolo told me so
jamie Webb you still don’t get it.
@@obviouslytwo4u What he means is that if you don't care what the ongoing cost is, and you set it to exactly £1.00 - then you can use the COST display as a second KWH display (as they will be identical at £1.00 per KWH). So you don't have to use the FUNCTION button to load the next display screen.
Clever idea :)
We used to use ¼ watt resistors as substitutes for pyrotechnic fuses because you need a licence for the latter. They worked well in theatre pyrotechnic applications.
I have videos showing 10 ohm quarter watt carbon film resistors igniting stuff.
Thanks for the video! I bought two of these about 2 years ago. They both failed recently with blank displays. First look inside saw corrosion of the batteries in each so I ordered replacements. No improvement. Had to delve deeper, and eventually found your video, thanks! First device I tested proved CX1 was low value (0.3uF), so I ordered a bag of 10. When they arrived, I fitted them in both units. One unit burst into life again. The other didn't. Now found that D7 in one device has failed. So I am waiting for yet another delivery now...
I've found if you get your power factor low enough the active power is so delayed, it gets pushed into the next billing cycle, I switch power companies every month and haven't paid a cent in years.
Amazingly they've actually changed the board design over the years. I have a much older unit. I have to hold Function and Cost buttons down to get the date code to show up, which is from Dec 18, 2009 although the board says 2014.Still working fine, although the NiMH battery is corroding, so I'll have to remove or replace it eventually. And yes, I had to resolder the ribbon cable as it broke off. :)
My model only has one blob-type microcontroller on the LCD board, which might just be an LCD driver. And then the AC board has a small power analyzer chip, which given the lesser number of pins, might mean a lower model. And it also has the main microcontroller chip, 32 pin or so, on the AC board.
Seems they've definitely upgraded this cheap meter over the years. Highly recommend getting one! Especially the models with a backlight
Lusting after the HOPI, I found and bought one of these cheapies on ebay a year or three ago. I am pleased to see that it gains the seal of approval. Some reviewer or critic at the time (I forget who or where) gave me a moment's pause by a snarky comment about accuracy - but the price was irresistible and I have been very happy with it.
Same experience on reddit after I asked a question about my experience with it. Only bought one about two months ago. I’m unsure what to do with it now. I’d like to measure my washer/dryer but the plug is behind it and the thing is just so bloody heavy
I've had about 10 of these things over the last few years, I keep lending them out to people who end up asking if they can keep them! They normally cost about GBP8.00 each and simply work. The 8051 was a bit of a surprise, and a blast from the past as I spent about 10 years of my life programming them in PLM51 and ASM51. That said, they are cheap enough and adequately powerful for this sort of job.
@purringChaos Depends on the specific processor. Some 8051 chips disable letting you read out the firmware with a debugger or accessing the flash some other way.
Of course there are methods to get around that (power glitching, clock glitching, etc.) but that stuff requires specialized tools (usually fpgas) and can be a lot more work than it is worth.
It's also of course very forbidden to reverse engineer firmware like that, so if you plan on doing anything commercial/public with it you should stay far away from doing that.
If you are interested in these methods in general, there are a lot of talks about it from security conferences. I haven't tried it, but the Chipwhisperer seems like a great tool if you have about 300 to 500 bucks to burn.
@@drjekyllmrheil1001 To be fair, at that point (since I don't have any of those tools), I'd probably just write my own (assuming the full datasheet is available), all the heavy lifting is done by the power monitor chip the display is by the Holtek (that I might use a logic analyser on if I don't want to muck around too much figuring it out). Not sure how long it would take with the tools you mentioned.
Then again, I'm cheap and I tend to over-estimate my actual skills :P
I've thought about putting together a modification to integrate something like a Nordic nrf52 (Bluetooth / 6LoWPAN + ARM) MCU (commonly found in cheap fitness trackers, and related to the chip in the BBC microbit) into one of these types of power meters. Looks like that would actually be pretty easy on this one.
And it's probably fast version of classic 8051 (in datasheet 1T=1clk/cycle instead of original 12 clks/cycle construction) with max 24MHz xtal
@@Tim_Small I've ordered one, I'll see how far I can get, probably going to use an ESP32
Looks identical to the one I bought from Aldi in Australia (sold under one of their brand names). I find it works well and has a nice big display.
I have the same one, I'm gonna tear mine apart today (if only to change that leak-prone NiMh battery)
I bought a couple of these years ago that use a standard set of button cells..they seemed to go flat between 3 to 6 months...these newer ones look so much better...I will check Aldi again to get the new ones!
Any way to have the display flip through readings at about 2 seconds per 'page'?
@@3Dparallax Did you tear it apart? Does it have the same chip? I found on Ebay/Ali this one with blue buttons but also a similar one with green buttons (and different display) - wonder if the green one also works well and has the same chip.
I have 3, the one I thought was an ALDI was an older (2013 ish) Kogan and pretty much identical inside, same chip, yes the battery was leaking slightly. The rellies have the ALDI one (2015) but I couldn't take it apart, appearance wise, identical as the Kogan one.
Power meter useful for measuring devices that use a crappy capacitive dropper supply uses a crappy capacitive dropper supply.
I've got the same model and it died a while back, a surface mount resistor in the power chain had cooked, so I replaced it with 1/2W wire-ended device. works fine now, doesn't get warm either.
Hmm, I wonder if this is what happened to mine. I don’t have the skills (or tools) to repair surface mount components unfortunately.
My favorite applications for these devices is for getting a rough idea of what you can run on a given current budget, or for making a really nice 'dashboard' for my old school generator. At this point I know I can run one fan, one LED light, and one USB charger per bedroom off our relatively small generator.
I've got a slightly older model of the same meter with North American plugs (the main board is V0.9 with a date code 6 months earlier than yours).
My tests agree with your findings - surprisingly accurate, especially when you consider the price.
I designed something similar a couple of years ago based around an Arduino and some op-amp input circuitry with voltage and current transformers for sampling house mains power. Every second it calculated and stored RMS current, RMS voltage, real power and power factor to a USB flash store for later download and analysis. The complex bit was not the digital calculations which are actually very straightforward once you have accurate waveform samples, but rather the analogue input circuitry for the current sampling. The problem is the vast range that this has to deal with, and how to maintain sampling accuracy for a milliamp scale waveform versus up to a 50 amp waveform. I wanted it to be able to monitor power consumption for the whole house, including accurate real power measurement of standby loads of single figure wattage. I ended up using an auto-ranging approach, changing the analogue input amplification for different current ranges. I wonder if this chip does the same internally. As you say - interesting stuff.
These watt meters have 2 issues, #1, they crash A LOT from any noise inducted in their proximity. #2, the frequency display is 50/60 so its useless for say setting a generators speed
It's for mains voltage so and frequency so it's to be expected. I mean when I hook it up to my inverter it tells me 61hz if that makes you fell any better.
Could that fancy chip be intended for use in smart electricity meters? If so it would need to be sophisticated and cheap, which it obviously is.
And the power company will still make you pay shitloads of money every month for a smartmeter.
I'm a power meter designer/developer for 20yrs.
Those "metering ICs" are intended for low-end, fast time-to-market, cookie-cutter products.
Each time a new product design is started in my division, they are quickly ruled out because of various imposed hardware limitations...
@@dl5244 What kind of chip would you use for high precision in a device like this? Also, what kind of hardware limitations are there with the cookie cutter chips?
Smart meters, ordinary dumb energy meters for measuring what is coming out of a Solar or wind inverter, public electric car charging sites, power tools hired out to customers ..... There are a lot of applications for a chip like this, besides plug-in energy monitors. So there will be plenty of them in existence. And the chip doesn't rely on many external components, which could introduce errors due to manufacturing tolerances. There are the resistors in the divider chain, but ±1% resistors are common enough nowadays and better are available. All the analogue circuitry will be on one tiny area of silicon, so it's all at the same temperature, and everything is digital and sent to the μC as binary values.
It's probably even picking up accurate timing from the incoming serial data (since the analyser chip is crystal-controlled).
Notice also how those socket inserts and plug pins could easily be substituted with a Schuko or French/Czech socket and matching plug pins, for use on the Continent?
@@sugarbooty this VF9261F metering IC chip is probably perfect for a cheap low end device like this with maybe a dozen possible measurements.
the PGA design likely precludes the ability to make continuous (aka "gapless") calibrated measurements over a wide dynamic range (not suitable for meteorology certifications like MID, ANSIC12.20, IEC62052-11, etc).
I like that the bandpass filters are somewhat soft configurable, but gain stability looks like a risk... maybe it could be compensated for, but usually not (because of part to part variation or drift over time, external influence factors, etc). The bandwidth is also quite poor. I'd like to see at least 3kHz for a low end meter and 10kHz for higher end (with some oversampling of course, so sampling rate would be >2x this)
The biggest problem with all these metering chips I've looked at is they do not make (continuous) digital waveforms available for power quality algorithms and/or validation.
I have a strange looking variable power supply, 11.5v, if youd like to take a look at it, it uses two bul129d's
Maybe that transistor on the input of the power supply is being fed to the processor to get the frequency of the mains. Should get a nice square wave off that using an internal pull-up.
I thought that too, but the power monitor chip can also provide accurate frequency information.
Since the interface is a UART, it might be possible to swap the display board for a bluetooth serial interface and get a much more complete display on a remote monitor, or even a phone.
It's very viable, but would need isolated circuitry and the ability to send data back for calibration. An opto isolator arrangement could work.
@@bigclivedotcom if you power the Bluetooth module from the mains, the Bluetooth is your isolation.
Development would be tricky, but could be done by plugging the module I to an isolation transformer
Stuff a PSU and a wemos / esp8266 in the unit, fire up Espurna and you are good to go - tinkerman.cat/post/connected-power-meter/
@@dbranagh146 PSU for 12, 5 and 3.3V already on main board and included in ribbon cable. Just replace display board by BT serial board and go.
@2:19.... "and then they exploded..." Keep up the good work, great channel!
@@Senshikaji He's welcome to come and do that in our house. Our electrical systems are far more robust than his. Even our 5A lighting circuits could blow 1 or even 2 'woodworm' holes in light bulb caps without us needing to replace the 5A fuse wire. Well, at least I think it's 5A that's the fuse marking.
Another BigClive "OH NO" moment, with smoke & flames, lost to the ages, because the camera wasn't running. ( ._.)
Clive-lectroBoom?
Beautiful review. I just bought two of these. Yes at 14:47, you query if the pin is for a backlight. Indeed, some units have a blue LED backlight, for a couple of quid more.
I bought one of those from Tesco. It was defective. One of the terminals inside had some really dodgy soldering and was hanging on by a thread. I was going to use it on a high current device (a "large water purifier for the purposes of purifying ethanol-containing water"). Good job I opened it up first and resoldered the connection!
I'd argue that the popularity of the 8051 architecture is more likely to do with the fact that it's royalty free :)
We ARE talking about our chinese friends here, right? I don't think they care :)
It wasn't always. Originally it was licensed from Intel. That's going back decades though. Atmel did some very nice 20 pin variants. Philips did some excellent ones too. I still have the manuals for them as there are some tricks that are stull useful these days EG for running serial ports over multiple KM links.
@Mai Mariarti Granted they don't really care about IP.
But look at it this way, if you have two options sitting in front of you, which are functionally equivalent, barring the fact that one of them is royalty free which are you going to choose to start manufacturing in the billions? Makes business sense to go with the option that has zero chance of getting you sued.
FWIW I've yet to see any Chinese uC's ripping off other architectures that aren't royalty free, doesn't mean it's not happening, but I've not seen it.
Certainly there's counterfeiting happening but that's a different discussion, stealing somebody's lithography masks to produce verbatim copies of their chips is easier to disavow (also easier to do) than taking their core and integrating it into your own silicon.
I was on the fence about buying this and you convinced me to take the plunge so thank you! Then I realised I was only 4 mins in and you completely lost me with the rest, perhaps I will be ready for such information one day, very impressive. Great review!
re the 120 vs 120k resistors, its good to know that professionals sometimes have those brain switch-off moments too
By pure chance I had bought one that arrived the other day not used it yet, so I was so pleased to note your positive comments. Thanks Clive.
Hi Clive! I have the "PR10 power recorder" and I like it
The layout of the display circuit board is beautiful. I wonder if it was designed expressly for this application, or if it was reused from another application.
Such displays are often used in those cheap weather modules that show temp and other basic calendar stuff i bet it's the same module 🤔
It looks like it was hand played out with tape on mylar instead of by using PCB layout software. I used to do this year's and years ago ;-)
I have a older version of this. Less surface mount stuff. The screen was dim as hell making it Basically unreadable. I now opened it up cleaned the contacts for the screen. Works great now. Thanks BigClive I can use it now.
I have the V2.0 of the version of the microcontroller board and its only a 8 pin chip with more welcome improvements. Including a cr2032 replaceable battery and a simpler pcb layout with less components to save cost. Brilliant in my opinion.
Always feels weird when I see a product I actually own being reviewed. I have the US version of this and I'm pleasantly surprised to learn how well made it actually is inside. I got mine off Amazon for cheap (as I was writing that I got a notification from my bank app on my phone through my PC that my Amazon Prime subscription just renewed).
@@KK-hy6sx It just charges off the mains. I'm sure there's some circuit in there that ensures whatever power it's drawing for both the display and the battery isn't included in the stats.
I've had one of these around for a while, glad to see an analysis here.
Same
I have one provided by my power company for joining them years ago. Works great, but my one gripe is that the price per unit is something like 29.35 cents, but since I can enter only two decimal places (i.e. .29 in this case), it's perhaps not as accurate over a long period of time.
Actually, a second gripe is that the switches for the power sockets are typically above the socket, and that the device hides the switch, so one has to switch the socket on first before plugging the device in. That's in AU/NZ, anyway.
I think I learn more here than in my uni. Great video clive!
I could use a new power meter, it would be nice if they used that huge screen and showed all the stats like the hopi
I have the exact same one, only with a North American plug. It's really handy and startlingly accurate.
Where did you get it? I would like to buy one.
Amazon has them. It works well.
@@RobertLeyland Name/model please?
Jose Pointero try this... www.amazon.com/Electricity-Monitor-Voltage-Overload-Protection/dp/B07DPJ3RGB
I have a couple of these I bought a few years ago to read wattage coming in from my solar panels. But the new ones have an overload protection and will not allow inrush current from panels and reads overload.
Now we replace the uC in there with a esp32 and we can look them up remotely
Or even esp8266. Was thinking the same. You could leave the existing 8051 in place to keep the calibration and then tap into the tx/rx to sniff the data. I'm wondering if the power sense chip sends all the data simultaneously to the 8051 for selective display. If that's the case then you would be able to show all values remotely.
Be careful of the live coupled ground and dirty 3.3v power tho.
Ordered one to do just that. :)
Might be a bigger job that I initially thought though...
EDIT: I wonder if powering up the display board without the metering board connected would still let the MCU send all the calibration data anyway...
Thanks for the breakdown. I have a couple of these I've used for years and always wondered how accurate they are.
I bought two of these from Amazon. Both stopped working after 30 minutes of use, just zeros on the display. Sent them back got two more. Both did exactly the same. Replaced with Smart Plugs that give wattage etc on phone app. These are working fine.
Love these meters . Use them as to determine how much someone on a spur has used and with a pwm controller for distilling to adjust in real time. They are cheap and effective and although I havent even got to the disassembly yet I'm very pleased you chose to tear these down. Great!
6:33
Why are there two brown live wires? Shouldn't the ground wire in the meter be yellow and green per EU electrical conventions?
When it comes to power meter chips for devices like this I figure it's a ton easier and cheaper to just get one someone else has made in bulk of metric craploads than to attempt a DIY that would have to be done from scratch since no off-the-shelf parts would likely be a 1/10th as accurate or resilient, especially since the one you found has a manual of 100 pages that I'd bet has instructions on support component circuit design for how to properly use it like LED example circuits but more complex and specific (yes you have to leave that cap of x value in...and resistor y).
But existing displays are never exactly what you want. So building your own (or modifying one like this to run your own firmware) offers the possibility of having all the functions you want.
Yes, I was just thinking it'd be great fun to get one of these just for that metering board. All kinds of exciting experiments to play with!
I assume you tried changing time and date while in 'calibration mode'
I always think of the triangular earth symbol as being mains ground (connected to the planet), and the one with the diagonal lines as chassis ground.
That's correct from the British standard for electrical symbols.
I've had one sitting in a drawer for a couple of years. Din't think mine did displayed the power factor, so got it out and check. Yep, it does all that your one does. Now it's plugged in to charge the battery.
i bought one in Brazil. Surprisingly, it can handle the 220V between 2 127V phases that's used in my city and the 127V between phase and neutral in my friends city. And yes, there's a model with backlight.
I actually got this 2 days ago and just found out you dissected this appliance just now
I used one to measure power consumption of an old freezer in my garage. Expected poor efficiency but actually got the opposite. Turned out the freezer wasn't suited to outdoor use and so wasn't working at all when the ambient temp dropped below 12c. Replacing that freezer before winter and saved us from potential food poisoning. Glad I have one now
The best part of these is they are cheap to make smart by adding an esp8266. They can come with 2 different serial chips, both work
I've got a similar one branded "Energenie" (green buttons). I opened it for comparison, and has a roughly similar board. I'm glad I did that - one of the electrolytics had leaked, so I will be re-capping it!
Do you know if it has the same chip Clive showed in this video? I'm not sure which one to buy, the blue one Clive showed or the green one, so any more information would be nice. Yours has (from left to right) History, Energy, Cost, Set buttons, yes?
12:18 - Ah, I see you signed the artwork on the bottom right corner. ;)
I have been using the exact same bodied EU version of the same energy monitor. Its plugged in 24/7/365 for about 2-3 years now, with constant load (100w-2kw). I even tried the oven as a load, and it maxes at around 4.4kw
I bought the North American version of this one. It stopped working after only using it a handful of times, however that was over the course of a year so the warranty had run out. Hope yours holds up better than mine did. :)
You've been reading my mind!
I was just searching for a device like this and was comparing with the hopi meter. Eventually I ordered a similar one, not exactly yours, instead of the hopi. The hopi was too expensive and the cable was not convenient for me. So i ordered the other power meter from China and a few hours later i am watching this video!
Thanks a lot Clive. Really. I've been enjoying your videos and learning a lot.
I have a suspicion about the voltage & current sensing chip: It might be one of the 24-bit sigma/delta converters. That way you would get a trimendous sensitivity and range. On the other hand, as you show differential (or maybe instrumentation) amplifier input stages, maybe that would not be necessary. In any case, it is obvious that the key to accuracy without any trimpots is some calibration table, which then requires really good reference(s). Interesting, as you keep saying!
Looks like a clone of HAMA 47770 energy meter made by Xavax, later branded as HAMA 111908 XAVAX, LOGILINK EM0003, ENERGENIE EG-SSM-01. The first two made by Xavax are using Cirrus Logic CS5460 chip and there is an eeprom & blob chip on the lcd board. Same battery, similar power design with a few extra protection options like a thermal fuse
I was looking for this on Ali and just before I was going to order the cheapest one available I noticed there's also a model with backlight. Only $2 more and for me well worth the extra cost.
So if you want/need backlight, be aware there's 2 models.
I have the 120 volt version of the same meter. It works well. However, the rechargeable backup battery in it gives out if the power outage lasts for than a couple of hours. It's memory then forgets all the last readings and goes back to 00.00.
It's nice to see you review a design that's actually *good* for a change, instead of "here are all the shortcuts they took that dragged it well into *unsafe* and *we have regulations for a reason*" :-)
I got an American version from Amazon, the key difference being that it is backlit. I’m glad to know that it’s not a fire hazard.
Aldi sold an identical one in Australia. Nice piece of kit on the bench.
15:23 "Quite Complex" - Big Clive "WTF?" - Me
the definition of simple pretty much is "could you make/replicate it on a breadboard or two in a shed?" and yes, yes you can
I've got one of these, thing that didn't last long in it was the battery, wouldn't hold a charge after a year of on and off using to measure different devices. What about using a small lithium cell in it? I measured the voltage and it wouldn't charge over 3.77v no matter how long it was connected.
Your 3.77v measurement was probably with the battery connected. With the battery disconnected I would expect around 4.3 volts output. You could probably get away using a protected lithium cell, but certainly don't use a bare unprotected cell.
I just bought one of these from Amazon, but the "calibrate" function doesn't work.
I looked inside (of course) and the chip is a BU6523GX, possibly a different method to calibrate, but the voltage is 1 volt lower than my Fluke 289, so I cannot complain!
Also, the time display does not increment if the load is less than 1 watt, along with the power factor display... interesting!
I thought it had locked up, but as soon as I gave it a load, the counter started and the PF display updated.
UK version has slightly different Vango V9240 measurment chip. Board says JK-PM01-A(V2.1) 2022.02.18.
I've had 2 of those for measuring the 24h cost of my diy cannabis lamps,they've both broke current passes through them but they don't measure anything anymore
Where they left of the "C8" capacitor might be used only for lower voltage areas like the US & Japan.
Gregory Thomas Lower voltage? The US have 240V as well.
I’m not sure they make one of the power monitors with a US 240v plug though.
@@Ulven3829 240 is not used in North America for anything the average person knows can be unplugged, between that and the half dozen different plugs in use I doubt there's any market for a kill a watt type product. Anybody that curious likely knows how to rig up an amp clamp anyhow.
I have one running for 2341 days now to monitor my PC consumption. Its at 3028kWh used. Had 0 issues with it, only clean the dust few times a year.
i have a "paget trading 9149" for maybe 20years, and i've noticed that it 'seems' to accurately integrate over very long times.
so
for a very small load, if you wait, you eventually get an energy value, seems the accuracy is better than the display resolution.
(you can also hack this a bit by using a 'silly' monetary conversion that scales the display up a bit.)
Looks like with that power supply circuitry, it should be okay to use with square wave or modified sine wave inverters that can cause issues with the power supplies in the very popular Kill-A-Watt power meters...
I have a similar unit. The Ni-Mh battery failed in reverse voltage and backfed about -1V. After desoldering it, the meter turned on again, but all the measurements are completely out of spec. Any idea what could be the cause?
It sounds like it dumped its calibration. I thought that might have been stored in the processor's non volatile memory.
@@bigclivedotcom Thanks for the reply, The model I have actually does have an EEPROM on the board next to the battery. But the EEPROM itself seems to be powered from the 5V Regulator. It's a standard 24LC02 jobby.
Damn, so this thing becomes useless if I just keep in around for some years without using it? Meh, then it's kinda useless. Shame.
Nice curvy display board. And a great chip.
Now we just need a BigClive series where you build your own including schematic, and have a non flicking screen showing all the pertinent information!
I actually have one of these with a Schuko socket, purchased from Aliexpress for about 7 or 8 euros. I have not checked, but I bet the cost calculator cannot handle separate daytime and nighttime tariffs if they differ.
Thanks Clive for all your great videos! I was wondering about the labeling of the inputs L and N in the circuit. I'm trying to understand and follow along with the explanation of the circuit but I'm seemingly unable to grasp Line and Neutral. I thought Neutral was 0v and not Line as specified in the drawing. If someone could clear that up I would really appreciate it.
L is at 240volts AC, but is treated as the 0v DC line for the electronics inside the unit as all DC voltages are referenced to that connection.
The 'Peacefair' branded power monitoring solutions on AliExpress are pretty good... you can get ones that have all the info on the display at once like the Hopi, ranging between four and six difference bits of info you can access. Plus there are versions that are TTL serial and even RS485 ready... making them PC or microcontroller project ready.
I love mine. During an energy survey at work (school building) I did some nifty math to use the electricity supply meter (the type with spinning disc) to measure the building base load.
To do this I turn off everything that is normally switched off, count the number of disc spins in say 30 seconds. The meter says how many spins equals 1kwh. Using those numbers I was able to estimate the base load. It would also be possible to disconnect circuits to try to use the electricity company meter to measure each circuit. However that looked like it would cause hassle for my IT background equipment and who knows possibly other timers running 24/7. I had suspected that a streetlight on a lane outside the building supplied through the school. No one knows who owns the lane but the light has electricity co signs on it. There is a local streetlight LED program ongoing. They changed sodium for LED and the next time I measured the base load it had gone from 18 to 16kwh per day. Of course it is possible that reduction is linked to some other change.
On my meter I got annoyed that it wouldn't remember the cost price, it needs to be inputted each time. A notebook is handier. At 25 mins that might explain it. It is best used with an extension cable for sockets in hard access places
an excellent in-depth description of this energy monitor.
I've repaired one of these that died not too long ago, I think it was a resistor that burned up surprisingly enough, replaced it with 4 1/4W resistors (2s2p) which was the best I had on hand... I had previously had to repair a Kill-A-Watt where the X2 capacitor had died (foil burnt up and capacitance had dropped to the point of the LCD barely being visible, and the measurements being off by a factor of about 2).
You just helped me decide which one on eBay to buy. Thank you, Big Clive.
Mine is branded Energenie, green buttons, no screw in middle of plug pins and battery compartment on back at top with 2 chunky button sells.
I have this in Australia and when measuring one of my Philips hue lights in red vs blue, there was quite a difference. I posted on Reddit about it and was told that it’s garbage. Anyway, I’m still unsure about the different colours at the same “brightness” according to the app. Anyway, small story.
The power up delay could be the noise floor adjust. Is so, it MOST be carried at NO LOAD. First chip is just the input stage AD and MUX. The other chip carries the processing. Some metering solutions, in POLYPHASE circuits, NEED some means of INSULATION (depending on the design). So the use chips like this ("remotes"). They are feed from a basic capa dropper on the hot side and use optocoupler to send the data strean to the main micro who is receiving the other 2 phases information as well. So, if you are not using CTs, you need isolation of some sort. Thery took the AFE part from the main chip, put in a separate chip and send data over isolated lines. Advantage is shunt are LINEAR and inexpensive.
Just what I was going to say,😂😂
There are some statistical tricks like oversampling, to produce 12 bit resolution with a cheap general purpose micro controller, that has a 10 bit A/D converter.
Some µC also have fancy stuff inside for such applications.
usually the problem is that cheap ADCs have a wide tolerance for linearity. The effective number of bits (ENOB) in most cheap ADC's is several bits lower than the marked resolution. The result is that regardless of precision, the accuracy for signal levels will be high at some levels and low at others...
Personally, (20 years designing power meters) I've never seen the oversampling design idea get past the napkin & speadsheet stage. It always makes more sense to buy the right part.
And of course with that integrated metering IC, there's not even a way to read the digital samples... only the aggregated/calculated measurements.
What meter is this? I notice you did not put an affiliate markings to say what brand it is. It would be nice to buy one that has been thought out rather than a clone that could catch fire.
it never fails.. the camera is never rollin when things go boom
Thanks for the review. I have always used Voltcraft energy meters as they were the accurate ones. But The energy meter 3000 failed after 3 years, then a new energy monitor 4000 failed just after warranty period and today I found out that another EM4000 I have bought failed exactly the same way. They started to show random data and after some time totally wrong measurements.
there are power measuring modules in aliexpress that are probably based on the same chip.
Search words:AC Digital Multifunction Meter Watt Volt Power Amp Current Test Module PZEM-004T G08
I'm going to ask a silly question.
Current shunt - big bit of wire with low resistance.
Why wire above the circuit board - why not just use a track - space for a track to carry that much A?
The shunt is connected to what its measuring by circuit tracks though?
I would presume the resistance of PCB trace would be much more variable between units due to manufacturing tolerances, possibly more than calibration can account for easily. The measurement traces don't carry any current so resistance doesn't matter there.
In addition to less variation between boards, shunts are also often designed to have a better temperature coefficient. And probably cools better as well.
Thank you
the "wire above" is a precision milli-ohm resistor to develop a small voltage signal proportional to the current
Speaking of discharging capacitors, the worst I've ever seen in consumer electronics was when I was taking apart my DSLR. I knew I should take great care about the capacitor for the internal flash, but the first time I've done it, I just shorted the leads. In the middle of the night, no less. Yeah, what a horrible idea :D That was a hell of a bang still. The next time I made it drain more slowly, that's for sure.
I see people taking apart their cameras on the internet all the time, and it's upsetting how many seem completely oblivious to that capacitor and/or don't even bother with a disclaimer to warn the viewers/readers. That capacitor can be dangerous.
i was curious about these because i run a small home server an im curious how much power it uses and i saw this model American version for $13
I got four power meters. A newer one is smaller w/ LED white back light.
Founded many electrical appliances with high standby current, I used Wifi plugs to solve the issues, even though it is a Wifi UV air conditioner.
I wish these had two different reset options, one to reset everything (like it does currently) but another where it would only reset the electricity usage figures, *not* the cost you set! If you reset it then you have to put the cost in _every_ single time you reset it. Tedious when you're going around the house trying it on 30 different things.
Hi, I have a Vodafone P3 Sure Signal unit which stopped working. It stopped giving mobile signal then got hot and the lights went out. It is 3 years old. Would you like to have it to see what went wrong?
Instant operational mitigation operational advice is good idea for field technician
I'm having a tough time with the "L" being also the "0V" rail ....nice job !
I bought four of these excellent meters (under the Status brand) from a clearance sale at Dunelm for £2.49 each. I already had a couple of the Nikkai meters, sourced from Maplin (which you also have used on this channel). The Vangotech V9261F is an interesting device. Thanks for the teardown. 👍
I bet the Maplin one had considerable price markup.
uK8cvPAq Ho yus. I think the normal price of the Nikkai N67FUs was around £14 but I picked up a couple when they were priced at 50% off. 😁
I remember getting a couple of UNI-T UT30B multimeters in a similar sale for about the same price (£6 ea, iirc). I don’t think I ever paid full price for the more expensive gear. I only bought the bench PSUs because I had a discount voucher from previous purchases. Maplin got ridiculously expensive towards the end. I paid £80 or so for the PSUs which finished up just shy of £120. I remember a couple of their more expensive Middlesex electronics kits from their range actually doubling in price over a few years. They were a marker for me and I would often pop in to the local store just to see what the price had climbed to. I do miss their presence on the high street though.
When a manufacturer showes the energy rating of an appliance ( B,B+ etc) is the power factor taken into account? I ask because as you have mentioned that smart meters being reprogrammed to read this at some point in the future.
Keep up the good work with the channel.
I normally use these metres to detect a current leak through the Earth.
you can make up a wired device that allows you to monitor any current that is being leaked through the earth that might interrupt a RCD module in your consumer unit.
This meter is sold on Amazon under the brand "Energenie". If you go to the customer reviews there and search for "fire", you can find various reviews and pictures of it where the live of the socket has burnt out.
I'd guess they were using it with very high loads like 3kW heaters continuously.
@@bigclivedotcom Just bought one that is externally identical with a V9240 8 pin chip made by the same manufacturer & the live pin going into the wall was pretty warm (not hot) after boiling a kettle of water at just below 3KW for around 5 minutes.
Intertek/GS certification stamp is missing on mine...though were they really certified at all anyway?
A little surprised to see no decoupling capacitors on the 5v regulator.
i saw that... actually i 'm 99% sure you can do that, the decoupling capacitor on the input may just serve as noise suppressor and the circuit is too small to catch noise, also the resistor is just a resistor, i think there's no reason to self oscillate... i've seen a schematic like that somewhere to reduce dissipated power on the regulator, apart the case of making a crude battery charger to limit the current (i've done that with a LM317 for LiIon)
Ok so not clear on the operation on the battery inside. I have a Bluetti EB70 and after a year of using it i want to use a power plug to test how much output in KW/h it has from fully charged before it shuts down . My problem is once the power bank shuts down . The power meter plug turns off. And i lose the KW/H given as the socket is shut off and the meter is blank. And if i plug it back in , is the memory lost?