Thanks Clive! I change out CryDom HD4890 SSR (480Volt 90Amp) Solid State Relays that run 7K watt heater bands on a PVC extrusion lines on a pretty regular basis. About 30 of them are used per line to heat "zones" controlled by Telemachique controllers. Always wanted to see inside one but the Crydom's are potted! Great video! Hate that the Crydom's sometimes fail in a closed gate state and burn up heater bands!
The shorted failure is a common semiconductor failure mode, particularly in the event of the load short circuiting. It's a good idea to way over-rate the solid state relays and make sure they are mounted on proper heatsinks with extra cooling if needed to make them survive longer. Another good idea is to consider a generously rated high temperature thermal fuse inline with the heating elements to kill the power to them if they start running too hot.
Bought a few "FOTEK" SSR-25DA a few years ago, used them to control approx. 1kW heaters at 230V. A very light load, only 4-5amps and I have had no problems. Fortunately they are also mounted on a decent heatsink, enclosed in a metal box and controlled by a PLC so there is no human contact with the input. Had some leftovers so after watching this video I cut one open and found a single BTA20-600 inside. A bit bigger than the one in the video but still only rated for 20A, not the full 25A. The input circuit is different and actually quite nice, it has a LM337 regulator connected as 12mA current limiter, which supplies constant current to the optocoupler for the full range of 3-30V. However the PCB is horribly badly designed, input and output traces very close (1-2mm) for the whole width of the board, there is quite enough space to do this properly, really makes one wonder why they had to screw it up. After some thought I got more curious and two weeks ago I ordered a few more SSR-25DA and SSR-40DA from different China ebay sellers just to check them out. Shipping was unusually quick, they all arrived here in Iceland 7-10 days after ordering. First I noticed that none of the items looked exactly like the pictures in the listings, minor difference like "Taiwan made" changed to "Made in Taiwan" and "Solid State Relay" becomes "Solid State Module". Next I split them open and found a single BTA12-600 inside the SSR-25DA type as in the video. So only a 12A TRIAC in a 25A rated module, not good! The listing pictures for the 40A show a module with two screws on the bottom, indicating it might have two power-components, one could even dream about two antiparallel SCRs which is the proper way to make a 40A SSR. Unfortunately they sent a different type with just one screw and a single BTA16-600 in the SSR-40DA. 16A instead of 40A, even worse than the 25A one!! Also the RC filter across the TRIAC was completely missing, no components, just empty holes on the PCB!!! The PCB of the 25A type is exactly like the one in the video, but the 40A has a much better PCB, with SMD components, good clearance and a milled slots between input/output traces. The 40A is also filled with more resin, the space below the PCB was almost completely filled on the output side and the components on top of the board well covered. A real shame they then ruined the whole thing with a 16A TRIAC. Some used heat conducting paste when mounting the TRIACs, some had no paste. It is possible to disassemble those SSRs without much damage, first pry the top cover with sticker up using a small screwdriver. Next use a fine tipped soldering iron to melt/burn the resin from the head of the TRIAC-fixing screw. Then remove the screw and finally the metal heatsink. The TRIAC will stay in place and hopefully not tear any traces on the PCB. Reassembly is also possible, but hardly recommended. After this experience I will try to avoid those SSRs. They are only good for appox. 25-30% of rated load. They should not be used where humans (or any other live creatures) might come in contact with the input wires, the insulation between input/output is not to be trusted.
i got one too, im using it with a PID temperature controller and a 2kW water heater. works fine, i dont have heatsink mounted. few minutes of use and it was slightly warm.
I have considered these relays but avoided them because of safety concerns. I have no problem piling on heaps of dubious low-voltage components that are practically free, but it is a different situation when a component is to be connected to the mains. I don't want to burn down the house or electrocute someone. Even though yours seems reasonably good, I think I'll pass. When certification markings are false and no one is accountable, it is hard to have confidence in a part. This is the first video of yours that I've watched Mr. Clive, and it is simply fabulous. First class. Thanks for doing this.
+pillsnotbills I agree with you. Something that fails with no disastrous consequences is sometimes worth the risk. If I fitted one of those SSR's to some kiln or printing press and then it set fire to the factory, well I'd be to blame. £15 will get you a proper american or european approved unit, so it's not even worth the risk.
Thanks for this tear down Clive. I have the exact same part from an ebay seller that I intended to use in a Kiln controller (~3KW max output). So after watching this I stripped mine down to reveal that it also contains the BTA12-600B, obviosuly not quite sufficient for my application. I have now ordered a BTA24-600B to uprate this to its original spec of 25A, at least I will now be operating its safe area. Thanks again for your entertaining and interesting videos!
I took apart an old 1980's vintage USA made 10Amp 140V ssr and it is indeed fully filled with dense black compound. The lid is a full cover and they have little legs that extend into the potting compound to hold the lid on. Much more robust construcutib than what is shown above.
SSRs are really intended for fast switching that would wear out a contactor (controlling temperature on a heater, switching in a VFD) If you want something to be left on for long periods your usually better off with just a contactor. Best for resistive loads, they don't handle inductive surges well.
I messed around with some triac circuits some years ago which were basically schematic equivalent to this thing. Had no difficulties powering incandescent lights, occasionally some difficulties with compact fluorescents (the holding current you mentioned). One thing that I think would be extremely good advice is to be VERY CAREFUL about powering heavy loads and/or heater loads with these devices. When the triacs got warm, they tended to self trigger, so that even when the opto-isolator circuit was turned off, they would still pass current. (negative temperature coefficient of solid-state devices) If you are thinking of controlling a heating device with one of these, I strongly recommend using an electromagnetic relay instead -- especially given clive's teardown results showing you can't trust the printed current rating.
Or, yanno, you could use a proper SSR instead of a fake one. Real SSRs are not built this way, and they use back to back SCRs not triacs, so there's fewer junctions in the current path, the heat sink plate isn't a joke, the cases are backfilled properly, and so on. You do realize that you CAN'T use a mechanical relay for PID controled systems, right? You knew that, right? RIght?
I cant get my head around the concept of faking a 25A SSR and selling it for a couple of pounds on ebay when they could sell it for the same price rated and labelled as an 8A device instead. It would still be a very competitive price but legitimately rated.
as far as i know they all agree this fakery, yes, competition kicks in and they have ridiculous prices but... i saw that 60% of what they sell is just fake, fake chips, fake transistors, only modules work ok, because 99% of who buys this stuff can't tell the difference and a chunky transistor will turn on a small motor even if there's a 1\2 current transistor actually inside, novices are happy, modules work... i'm not è_é... i couldn't find a BC639 or compatible except on super expensive stores outsice the china cheap (just as en example, don't buy op amps eighter)... yes instead of 1€ i'd like to spend 1.5€ for a bag of 100 of those transistors but LEGIT ones, not caring of branded ones much
Welcome to China - Manufacturing philosophy: why waste money using unnecessary resin... you get what you pay for, but this is dangerous and a fire risk
In general, it seems like the Chinese tend to prioritize things working and being cheap over reliability. So if they give you a current rating, that's not going to be a guarantee of steady-state functionality at that load, it's going to be "this will probably work for all these applications"
@Alweg Fan , as a long time engineer working for a company that supplies industrial automation system to Chinese heavy industries, including power generation, practically everything is sourced either out of the USA, Europe or Japan. I go over to a plant and I am rubbing elbows with representatives/engineers from Germany, Japan and the States primarily. I don't see ANY Chinese sourced parts/assemblies in those critical industries. So I think the top guys who are in charge of critical industries have a good handle on the quality and reliability of their own lesser industries. Now I suppose that over time, this is gradually changing, and more in some industries that in others.
+Adam Perry Very easy. ul.com/newsroom/publicnotices/ul-warns-of-solid-state-relay-with-counterfeit-ul-recognition-mark-release-13pn-52/ Anything that doesn't match the image of the legit one is fake. The one shown in this video is missing the maximum amperage rating.
It's fascinating to see what I've been using for the last six years! I'm glad now that I acted on impulse and used them conservatively, and only for personal projects. But they have worked without complaint.
Years ago we used to build home control boards and in one of them we used a similar arrangement but simpler as the input voltage was known - hence resistor into the zero-crossing opto-triac then the same components on the right. We used 15A triacs on heatsinks and controlled 2KW heaters no problem.
Thank you! You've been very helpful. I've been eyeing some of those for a project. Every time I hovered over the checkout button I would get a little voice telling me, "You get what you pay for."
Wow! Thanks for such a quick response. I am pretty much still in the electro-mechanical era so this type of thing is new to me, but still interesting.Pity most schools can't make subjects as interesting as you do.Fantastic work!
Just got one of these to control the heated bed on a 3d printer I'm building. It was a relief to hear your verdict that it may be up to the job. Fingers crossed.
Have used 25 amp solid state relays but when load was more then 3 amps would mount it on a heat sink and use heat conducting paste. Have just used a 6 by 6" aluminum plate that was 1/4" thick for loads up to 50% of label. Nice vid.
Stumbled across this whilst going through your older videos - and I just happen to have one of these sat on my desk! Very useful teardown and glad to see they aren't all that bad.
You can sit here ranting nonsense, but you have never created a product in your life. BTW the gap there is around 5mm which would require a voltage of 15000v to jump across that. There is no design flaw here, only a flaw in your thinking.
in theory and ideal conditions that's the isolation ( remember, if it ever did cross over that would be VERY VERY bad ; it would go straight to a likely low voltage control network) ... and like clive said there is literally no reason they made it so much less then it could have been.
There is a lot to be said for truth in advertising, or at least truth in specs. No matter what your skill level, there is no excuse for lies like a 12A triac in a device that claims the ability to switch 25A.
I'm using a pair of DC-type SSRs to control the electromechanical spring brakes on a 12.8m satellite dish (former NATO, we're restoring it for scientific use). The control system is 24VDC, as are the brakes, but the brakes are higher current than the controller can sink, so, SSRs. They're notionally rated for 40A, but I'm guessing the truth is much less. They came with heatsinks and the brakes are only putting a roughly 4-5AMP load through the relay output--rather more than the controller can directly sink. I've been an electronics tinkerer since 1973, and this is my first experience with SSRs. I also picked up some 24V contactors, in case the SSRs didn't work out, but, so far, they're fine. I placed large (30A) diodes across the brake coils, because it wasn't clear whether there was any kind of snubber network inside the SSR.
I bought a couple of these several years ago for a brew fermentation controller (BrewPi). Hooked it up to my old fridge and when the fridge should have turned on I heard the sound of something from the compressor but it didn't start up. I assumed the compressor must have seized up after being idle for months so gave it a few whacks with a rubber hammer but still the same, Something then made me think to check the SSR and turned out it had a crazy on resistance so when hooked to the fridge, the output was something like under 100V rather than 220ish. Swapped it out for the other SSR I had and it worked fine... Invested in regular relays for that project instead. I do understand there are no shortage of fake FOTEK SSRs that likely you and I found for bargain basement prices, but the genuine FOTEKs are more reliable. What about ordering one from CPC or the like and seeing if that's any different?
Thank you for sacrificing this relay to show us how it works. I am concerned that when there is 240V AC on the relay it is showing about 10V AC between the backplate and ground or neutral. I wanted to bolt this to a grounded chassis but with a potential between them it sounds like a bad idea.
I have bought 4 of these. one of the first two would not operate at all. I've had no issues with the other three. I have two operating in my cnc setup, switching on my spindle and vacuum system with no issues at all. My little spindle pulls maybe 8 amps, and my vacuum about the same.
We actually use similar relays. Allen bradley drives switch it to power a 120 volt brake on sew eurodrive motor. Works great. Great video as usual Big Clive ⚡️🏴⚡️ Our relays are allen bradley or omron.
"Our relays are allen bradley or omron." -- which means they aren't built like this trash at all, that's why they work fine. People in comments seem to think this video is an indictment of solid state relays, when it isn't - it's an indictment of FAKE CHINESE "solid state relays", built completely different on the inside from the real thing.
Thanks for posting, it's good to see the insides. I used an SS40DD (to switch a DC 3A load) controlled by an arduino GPIO pin. I was using it well within spec, but it got very hot indeed and all the case melted and twisted. I built my own after that :) :)
I prefer the Use of Crydom SSR's. I have a bunch of Crydom SMR 4850-6 which can handle 96-480VAC at 50Amps. To control these thing you need a Supply Voltage (8-32VDC) to power the internal Logic, and then you can use the inverting or non-inverting control input to turn this thing on and off. It also has a N.O. and N.C. Alarm output and a integrated red Alarm Status LED to show abnormal conditions. A green LED shows normal operation when switched on. The Encapsulation is made with thermally conductive Epoxy.
PodeCoet thats insane? what is with the dopey chinese engineers that approve of this? After watching dozens of clive's vids, you get the idea they arent taught to actually understand electronics.
Two RC constants in the circuity is using two time constants or double time constants is to make full use of the triac for more efficacy. Thus making use of the negative and positive parts of the sine wave
I love your videos, big guy (I am a big guy, also). This one in particular is motivating. I have long wondered why solid state relays rated at 25 amps do not perform as advertised. After a really looooooong research, I found that the amperage rating is bullshit. It is all about the wattage. At 5 volts, sure the relay will switch 25 amps, but at 280 volts, it will only switch 1 amp. So, people that want to switch mains power (120 v in the US) are tricked into thinking that they can switch 25 amps, when they can only switch about 300 watts (2.5 amps)
Thanks for doing this video, I have used this in my arduino controlled espresso machine, I didn't pop it (rare for me!) I figured it would be encapsulated, there are a few brands on the bay, identical except the red 'fotek' brand. Price was the same, so I went with the Fotek, although mine is also lightweight and airy and I also expected it to be potted full. Learned a little and a lot here today, I am running mine at 5v switching and only using it for a 1300w-ish heater under supervision so for me I am sure it will be fine. Great video!
I believe one can never have enough Arduino controlled espressomachines. Thanks 😄 And who is supervising ? Not another Arduino ,I mean why an Arduino in the first place then ,of have you totally hacked your espressomachine ?
I don't know what is the norm, but I tore open a US brand SSR once, and it was basically a chip and wire hybrid, bonded to the copper base plate. No PCB, no packaged components, just chips, bus wires, covered with clear silicone. The circuit was stupid simple, too, like 3 chips and no passives.
I just bought one of these. After watching this I stated wondering if I fill it with resin it might be better than it is now. The load that it needs to handle would be around 8 amps so it should handle that well.
Just received my second batch of these for some small arduino projects. They've all withstood a couple of months of switching 5A every few seconds. Very happy with them. Obviously not a fully Professional part though.
Here is a reply I received from an Amazon seller after leaving a review of their fake SSR. It's a pretty sad excuse. "I'm very sorry. I know our mistake made you a little disappointed, but please give me a few minutes to explain that，this product is from OEM and delivered within the period of novel coronavirus pneumonia ,at that time China is lack of raw materials and lack employee .Most of the worker, including engineers, was on vacation. Only a few people stayed in the factory. (originally, SSR-25AA used BAT24 Chip,SSR-40AA used BTA41A ) Due to the fact that engineers can't get to the company in time, many employees can't go to work in time because of pneumonia in China. Pneumonia affects the closure of the city) which leads to the following mistakes. Originally, our brand is eqiek, but due to the negligence of the printing staff, they make the label FOTEK instead of EQIEK . But we confirmed that there is only a batch of part labeled printed wrong. In fact, the main drawing of our products is eqiek(Pls check OEM brand EQIEK as attachment ) We have confirmed with the factory that we have ordered 1000 pieces of ssr25da and ssr25aa respectively. More than 900 pieces are produced with bat24a. Only a few dozen pieces are produced by BTA12A within a period of novel coronavirus pneumonia , engineer can't back to office immediately ,one worker mistaked BTA12 instead by bta24a, When we found out, more than 1000 goods had been delivered to Amazon's warehouse. We contacted Amazon to return the goods. However, due to the impact of pneumonia in the United States, there was no reply. Last , we have no idea to deal with returning the goods. .AMZ service answered :Because 1000pcs SSR-25DA with SSR-25AA have been diverted by AMZ warehouse .we feel very helpless about this matter. But we will be responsible for our every customers.Please Believe us, we only do honest business, we will not deceive every customer. On behalf of our company, I apologize to you. Now our company proposes two solutions: we will refund the money to you, or we will send you replenishment from China (this time we promise SSR-25AA 100% function and with BTA24A chip ) ,However, the US postal service is very slow and may not be able to deliver it to you in time as the spread of pneumonia in USA ,so hope your advice to us . Hope you can forgive us and hope you can give us a chance to service for you .We will cherish this chance .we are willing to return payments for you .Thanks for your time to read this long email . All best wishes for you and your family Happy everyday JESSICA"
Hi, I've just been dealing with Amazon support over the fact that when I went looking for SSRs on their site, I still get advertised by them fakes which I already reported to them over a year ago, left bad reviews and so have many other people - but enough people seem to either not realise or go 'oh well it's cheap and does work for the lower level job I have for it' that they drown out the many reviews clearly calling them out for being fakes or just not working. As a recipient of fakes, do you fancy teaming up to at least try to get some proper action from Amazon?
@@edumaker-alexgibson Amazon deleted my negative review. I just sent this email to FOTEK along with loads of photos. "To whom it may concern, In September of 2020 I purchased a FOTEK branded 25A SSR from a seller on Amazon. After receiving and installing the SSR it quickly failed. I was very suspicious and decided to open the SSR and found the TRIAC was a BTA12 rated for half the amperage stated on the label. I sent the information to Amazon customer service and received the response shown below from the seller. My review on Amazon showing I received a counterfeit was originally posted but has since been removed. I'm not too sure what FOTEK can do, but Amazon and the seller are selling dangerous counterfeit SSR's with the FOTEK name on them."
@@edumaker-alexgibson Here is the lackluster reply from FOTEK... [Dear Sir/Madam, Thank you for your information We know that there are a lot of counterfeit from China because of counterfeit is from China FOTEK is from TAIWAN I know it's problem so We are not sell on any Website to be buyer's protection and we have provided "where to buy" on our FOTEK's Website]
Pretty sad excuse indeed. I do business with China on a daily basis, and you need to plough through these kind of BS excuses every time things go wrong. You ask yourself why they even keep trying, stacking lies on top of lies, wasting your and their time, instead of admitting they screwed up and use that same time to fix it. One time we received bicycle carriers with only half of the parts in the boxes, "because factory didn't have parts". Took months to receive the missing parts (on their costs) and open up every box to add them. Big chunk of their and our revenue lost. Sigh. Simply admit you can't accept the order because of parts shortage. It's a strange business culture over there.
@@resspls10k They get to keep your dollars by accepting the order regardless of ability to fulfill it. The lies are to weary you out/delay for time - in the hope you will give up or run out of time for refund
I use this one for my brewing temprature controller. It works well in that application with a 12v signal from a PID to trigger it. If I was to market the unit I would probably use better relays and pid's though.
It is interesting to see how one of those commercial modules does it. I don't know about pulling multiple amperes through it with only that bottom plate for heatsinking, though I guess that depends on what you have it mounted to. And I'm assuming that the tab on that triac is isolated? I have a bunch of 4A/400V triacs on hand and plans to build one or more of these. I suspect that I'd probably use somewhat different parts for the snubber network, though. I plan to use the MOC3010 opto, I think it's the MOC3020 that's the one with the zero-crossing detector in it. Been a while since I looked at those datasheets.
Zero crossing is not great for inductive loads. I've used these FOTEK relays quite a bit, however I generally use them @ 1/10 spec. Doing that I have never had any trouble and they are cheap enough that such over-specing is very affordable.
I let the smoke out of one of these, yesterday. On a temp. Controller, for my homebrew. It worked fine for the first 20' or so. Then that cooking component smell, it failed open.
Opened up an eBay Fotek SSR-40 VA (phase angle controller) you'll never guess what! It's (the triac) not rated for 40A but 25A. At the price it's worth buying two to tare one apart before using the other. I would advise a PSR-25 if you are looking for something a little more tried and tested for phase angle control.
So if i understand this correctly these relays would as you say be used to switch power on and off on say a AC single phase or AC 3 phase big foxtrot motor/pumps etc and then that would mean that the other side of the relay could use a much smaller dc or even ac voltage, switch and wires negating the need for big Foxtrot cables, wires running through to the switch and cabinets etc. Great idea really, I guess you just leave the switch in the on position until you want to turn it off, way safer way to handle all the switches inside cabinets on low voltages and smaller cabling... always wondered how they worked Thanks Clive...
I am trying to do a similar thing with a 2kW (~8-9A - 200/230V) heater for my house. What did you use to solve this problem? Did you buy a better quality SSR that has a higher current rating?
Her's a thought - could the metal backing need to be mounted on a surface that can radiate heat? if you mount this thing loose in a box it might cause over heating.
+Ed Rundle Well, if the damned thing melted into a blob from 8 amps of purely resistive load, then I don't think even a heatsink with forced cooling (i.e. a fan) would help things much (except if it's a CPU cooler perhaps). Even at 8 amps. If the thing would run "only" hot to the touch, then maybe. But melting (=temperatures of 200 °C and more) means that 8 amps is WAY above the rating of the semiconductor part inside.
Real ones have about the same inside, but better clearance and filled generally with a glass bead filled resin to save the cost of the expensive resin. This one is filled with a powder filled resin ( probably chalk powder or some milled dry clay) so keep it dry, it probably is hygroscopic. The 2n ceramic snubber is probably the same 10n 2kV Class Y ceramic used in SMPS use, where it has quite a good pulse handling and also will fail open circuit. Not a bad choice, cheap but will do the job better than a film unit. Anyway, standard appnote, and the zener does double duty as reverse polarity protection for the opto, though the LED indicator will die with reverse voltage, but who will ever see the red inside a cabinet in any case. As to the triac, I doubt it has any compound, relying instead on the resin seeping under to fill gaps from the screw bending the tab, and if the screw is loose it will keep it in place as well. Isolated tab, so not really meant for more than 5A of load, unless you really use a decent heatsink, which that zincalume casting is not. Rating 220VAC 2.5A with a peak of 5A is more appropriate for it.
Only just come across this, and I imagine that at time of watching (Dec 2023) any SS relay on eBay priced at under £5,- isn't even going to have a 10A TRIAC in it. 100mA or even 1A perhaps (Whatever will just about withstand a 30s inbound test on a 20A load) but definitely not 10A! 📦💸😉 That aside; Living in an all-electric flat I'm struggling a lot with cost of heating and I keep considering ideas like running TRIACs and ICs close to their upper tolerances and benefitting from the heat this will give off as a biproduct. Even if I have to swap in new TRIACs every week it's probably still going to work out a lot cheaper than running my 3kW convection heater at the present 36p/kWh for a few hours each day! ⚡🔥💡
We used a 75A SSR in a safety analyzer (bit like a PAT tester). It needed to switch 20A inductive loads and pop a 20A fuse on fault without popping itself. So ... yeah - do it right. If you've spent $4000 on a big brand safety analyzer, it had better "bounce" and not break when you abuse it a little.
I have several of those SSR. You need 5V to switch them on; 3.3V doesn't gonna make it. I made my first SSR with the same components before buying them and have it working for more than a year in a Sous Vide (1 KW heater), so, component wise, them seems reliable to me.
Great Vid, especially the drawn diagram. The one thing that has always confused me about most Asian products is that they will do something about 99% right, but manage to pull the defeat outta the jaws of victory at the very last moment every time.
I have seen 100 AMP 60 volt DC SSRs on eBay I highly doubt they can sink 100 AMP heat dissipation in the same form factor package that you have FOTEC without liquid cooling. My interest in the DC SSRs was for a golf cart speed controller. The limit however is the PWM response. Others have said the max PWM response was 800 to 900 cps. Krydon brand SSRs are better and can have a slightly higher PWM from 1000-1200 cps but still is within audible hearing range. I would like to see you test and tear down a 100 Amp DC SSR. Thanks for the video.
After watching this video earlier this week, we had a system come in from a university that had one of the counterfeit SSRs, Luckily it hasn't been run at all.
You didn't show the actual teardown, so I can't tell if the thing was damaged in the process. I'm wondering if the rather nice case could simply be upgraded with a better triac, along with some attention to improving the isolation. Another concern not mentioned is that unless earthed, the bottom plate could become live in a failure. I'm going to be using one of these for a small kiln, so thanks for the warnings on these. I'll bolt it to a grounded heatsink, and make a voltage detector circuit that will enable the Arduino controller to check that the SSR has mains supply voltage across it when it is supposed to be off.
Hi, slightly late to the party, my company use SSRs to switch 240v 10A motors with 12v DC, we are talking literally thousands of motors, we use SSRs from a certain UK supplier and they're very reliable, except when the SSR fails, they go open circuit across the switch, that then obviously causes the motor to run continuously until the thermal cutout detects that it's glowing and about to catch fire, or it actually catches fire. So, we got a few of these to test, not based on cost, and sure enough, we managed to get one to fail and it just died and didn't go open circuit, brilliant, or so we thought, we then passed these SSRs to our tech guys who threw them in the bin and said that 'if we ever put that shit on one of our units without consulting them first then there'd be trouble', they said that these things were 'lethal' and they'd rather kill motors than engineers.
The triacs going short circuit is a common failure mode. Your tech guys were probably concerned about the possibility of poor isolation resulting in low voltage circuitry becoming live at mains voltage.
Today my 40a fotek ssr burnt out after a good few years of switch a 2kw heater so I decide to lake a look, the triac was a bta24-600b rated for 25A. Thought I would share as this may be useful in relation to Clive's video.
"ekoenig" is actually german for roughly translated "e" for electronics and "koenig" or "König" ("oe" is spelled in germany as "ö") for "king". So the seller seems to be a king in electronics.
Great tear-down, really interesting! I've been using one of these Fotek branded SSR 25 DA (not sure if the one I have is genuine or fake) for about 1.5+ years to switch a 1.3kW purely resistive load at 240V AC from a 3.3V input. It was ridiculously cheap on ebay, so I guess mine is the fake one. Probably still working because I'm well within the "safe" operating envelope. Would no doubt have exploded by now if I was on 110V mains :-)
GreatScott also did a review of a 40A version of these, which also turned out to have a 12A rated triac inside. So maybe we should consider them all as 12A SSRs regardless of their marked amperage... at least until we have destructively dissected one to confirm a higher rating of its components.
One thing you didn't test is the turn on voltage. Triacs are not the same gate turn on in all quadrants. You will see that below 3V (2.2 - 2.7V) the SSR will only be half wave. That could damage some AC devices which are inductive and that may be very close for 3.3V devices. More expensive SSR like CRYDOM use a constant current circuit and LED drive current will be a constant 2ma from 3-32V. You get what you pay for.
This is dangerous stuff. By most electronic safety regulations you;d expect 6.4 mm CL/CR between the live circuit and the 30V circuit on PCB material. that means for a DIL package you need all of it!
if you use one of these make sure you derate it by 50% especially if you don't use a heat sink. (ebay sells them) I have had quite a few fry spectacularly (fire smoke total met down) controlling a 1500 watt heater 120 v. swapped out for 40A rated ones (almost same price) and stopped the fire hazard. BTW they work very well connected to output pins on arduino boards
i use these in pinball machines that have ac motors controlling backbox animations they are easy to drive off the 24v solenoid lines and don't have woefully undersized traces on pcbs to burn up without blowing an 8A fuse
The Crydom unit is fully packed with resin and with an admixture of whst appear to be very fine fibres.The unit failed quite expressively by blowing out part of the top on the a.c. side. Looks as if it may have been the resistor -capacitor network. Another blow out on the upper rh side looking at the unit from above.Further dissection looks like it will destroy any components left.
I was wiring one of these up at work today and this video showed up in my feed when i got home. How odd. It's only switching 3 Amps but controlling a rather critical temperature. Thinking tomorrow i'll change it for one of the £25 ones from R.S.
Be careful with this: I used a 25 A of this type to switch a 2000 W 230 V water boiler , so less than 10 A. However, after a few weeks it completely burnt. (It was hanging on its wires open in the air, so it could cool on the air (20 degrees C). I am happy my house was not set on fire.
IIRC real ssrs have the die on the back and everything wirebonded and then there are the mosfet based ones, which are also mounted on the back but they use back to back mosfets instead.
I know that this video is old, but there is one thing about this video that (I felt) should have been discussed. I noticed that the Triac was mounted on the Rear Alloy Casting, (as a Heat Sink) but the Triac had mains voltage on it! In the design of the Fotek SSR, do you think that NOT enough emphasis or consideration has been given by the Designer of this Unit, by the inclusion of an Earth?? If the underrated Triac becomes destroyed, there is a high potential for the rear alloy casting to become "live". This could be hot, dripping solder from the Triac itself - making contact with the alloy rear casing. BTW I always like your informative content, a regular viewer who likes the Ayrshire accent!
The better high power SSR's use back to back SCR's rather than TRIACS because SCR's are more reliable when driving large inductors like motors. I talked to Motarola years ago and they told me you have to overdrive TRIACS to reliable switch high inductance loads in all four quadrants. This is easy to do if you know what voltage your going to be driving from but can be more problematic if you really do have to have a wide input control voltage. We were buiding our own SSR's so we controlled the input switching voltage and could set the gate driving voltage and thus the current.
Brilliant! I've been concerned about this unit that I bought recently, it's for a 240v floodlight being connected to my 12v alarm, you've showed it's properties well. Thanks! Subscribed!
It's not OK, someone I know got one and it burned up switching a 1kW resistive load. (At 120V, so about 8.334 amps ... not even 10 amps!) (The load is a a heating element. The inrush is very low compared to an incandescent bulb load!)
+Benjamin “Ozias” Esposti A heating element is similar to an incandescent bulb load for the same reason, the resistance is lower when cold. A long wire between the the load and the relay also turns it into an inductive load, which causes most of the inrush problems in domestic light circuits.
the crydom genuine ones are solid inside i have tried to open them but they are full to the brim with resin and very reliable the failed one was on a transformer for a sealing bar that failed and shorted the 300v ac to earth very melty very very stinky
One thing that was omitted in this teardown is a review of the triac to baseplate isolation. It is critical that the line voltage is isolated from the heatsink.
I thought I was gonna be really clever and put a few of these in my off road rig. I was all sorts of proud, of how pretty the wiring looked, show stopper... any how, I was having all sorts of troubles, and of course the last things I was to check were these batch of SSR's.... I am quite sure that these are fine in a clean power supply environment, but, these absolutely do not like choppy supply voltages. Please note, the charging system is extremely clean, by automotive standards, quite powerful too. I went as far as adding a few stiffening caps to level out the chop from the rectifier set up in the alternator, that did help lots, but still had issues still left an unpredictable temperment with these, I really wanted to keep them, as per the amount of work that went into wiring them in an art form, never mind the amount of work that would be required to yank'em out, and completely rewire the entire set up, for typical mechanical relays, of which is fine, 40 hours later...LOL... Like I mentioned, I am sure these perform great with a clean power supply, I do not recommend these for automotive use. I thought about trying to build some sort of input anti-spike stabilizer system, but, then, it is kinda defeating the point of the cleanliness and cool factor at that point.... I yanked'em.. Cool video, again!!!
Stuff like on board air compressors (twins), all of the lighting that points forwards, backwards, underside. Electric fan, etc... any location that would need a relay, I had these set up, instead...
I really don't know way I had trouble with them... I used to use these in some of the golf cart mod's and never had a single issue with them, heck, with as much as my off road rig has given me over the years, it is likely just my rig...lol. More over, I could recommend the idea, if you desire to use these in automotive application(s), I would wire up a simple test circuit and run it for a week, even if it is simply wired through the cigarette lighter / 12 volt power supply port... maybe to just turn on a single light bulb or a few bulbs... I have used the ones that came off the rig for other projects and they seemed to be just fine, as far as I know.. They do look cool, all wired up, especially if there is several of them... Good luck.. (test'em before going nuts with'em)
jumping back 7 yrs in time now. im guessing the yt algo chose this due to the words resin used so many times so that would be a logical grouping as well as triac which would then bump up the relevance.
I tore down a Fotek SSR 100DA after it failed and it had 2 x BTA24-800 Triacs in it which are rated at 25A each. It seems that it is common that the manufacturer claims double the actual rating. Another issue is that the heat dissipation is not good as they use an alloy for the heat plate instead of aluminium, this further reduces the efficiency and capability of the Triacs. Very dangerous, be cautions, attach a very large heat sink and don't go beyond 30% of the rating if using these cheap type relays.
Purchased a few of these not so long ago. Didn't take them up the the marked ratings. Guess I won't be now. LOL I had intended on using one to switch a MOT (MircroWaveOvenTransformer) Spot Welder. I'm quite sure it would had pulled more then 10A. May have worked for some time. Would only be on for microseconds at a shot.
While to great in design for general use this looks quite decent. I've always been afraid of buying these on Ebay but I think I'll grab a couple now for personal projects.
Thanks clive :-D, i sort of expected it to be more heavy duty inside, that ssr looks more like on par with a lamp dimmer. I had something like that in a draw, it had a thick base and solid like a block of resin, its a black device apparently made in america, never found a use for it.
I pulled the DC version apart, it had a BJT transistor, with the part number on it, good for about 8A, nowhere near its 25A nameplate. Fraud from China. At 8A it dropped 3V.
borrowed off the Fotek website about them some what long winded . Established 1995 FOTEK CONTROLS CO., LTD. Established in 1985, the company dedicated itself to the research, development and manufacture of industrial automation hardware in order to support the automation of Taiwan’s machine industries. As all sectors of the industry face fierce global competition,FOTEK continues to provide the guaranteed best quality through advanced technologies and strict quality control ( certified ); At the same time, the company has continue to pushed to increase production, standardization and international standards compliance of electronic controllers in order to reduce costs and increase competitiveness. FOTEK products are in use in many Taiwanese industries such as rubber/plastic machinery packaging, food, printing, parking, transportation, many major machinery producers. The company’s products are sold electronics and cabling. The company is alsoa designated supplier to all around the world, and many internationally renowned companies in Italy, Germany, the USA, the UK and more use FOTEK’s OEM services. With nearly ten thousand products, the customer’s requirements can always be met, making FOTEK best partner for the manufacturing and machining industries. Management Philosophy : → To assert quality excellent。 → To pursue company lasting operation。 → To treat Customers as forever partners。 → To research and develop over-valued product constantly。 → Innovation、Responsibility、Reliability and Performance。
Now we are talking as this is used (or even better the 40A version) for 3d printer heated beds that uses mains instead of a PSU so it will heat up faster.
I find hard to believe that it only have a single triac I build a very complex pulsing array for a custom power inverter... and I used 20 fet's of 45A each for a load of only 250 Amps... so it will be interesting to open one from a reputable brand "Expensive" and see if they only use one of this devices also!!!
Thanks Clive! I change out CryDom HD4890 SSR (480Volt 90Amp) Solid State Relays that run 7K watt heater bands on a PVC extrusion lines on a pretty regular basis. About 30 of them are used per line to heat "zones" controlled by Telemachique controllers. Always wanted to see inside one but the Crydom's are potted! Great video! Hate that the Crydom's sometimes fail in a closed gate state and burn up heater bands!
The shorted failure is a common semiconductor failure mode, particularly in the event of the load short circuiting. It's a good idea to way over-rate the solid state relays and make sure they are mounted on proper heatsinks with extra cooling if needed to make them survive longer. Another good idea is to consider a generously rated high temperature thermal fuse inline with the heating elements to kill the power to them if they start running too hot.
Bought a few "FOTEK" SSR-25DA a few years ago, used them to control approx. 1kW heaters at 230V. A very light load, only 4-5amps and I have had no problems. Fortunately they are also mounted on a decent heatsink, enclosed in a metal box and controlled by a PLC so there is no human contact with the input.
Had some leftovers so after watching this video I cut one open and found a single BTA20-600 inside. A bit bigger than the one in the video but still only rated for 20A, not the full 25A.
The input circuit is different and actually quite nice, it has a LM337 regulator connected as 12mA current limiter, which supplies constant current to the optocoupler for the full range of 3-30V.
However the PCB is horribly badly designed, input and output traces very close (1-2mm) for the whole width of the board, there is quite enough space to do this properly, really makes one wonder why they had to screw it up.
After some thought I got more curious and two weeks ago I ordered a few more SSR-25DA and SSR-40DA from different China ebay sellers just to check them out. Shipping was unusually quick, they all arrived here in Iceland 7-10 days after ordering.
First I noticed that none of the items looked exactly like the pictures in the listings, minor difference like "Taiwan made" changed to "Made in Taiwan" and "Solid State Relay" becomes "Solid State Module".
Next I split them open and found a single BTA12-600 inside the SSR-25DA type as in the video. So only a 12A TRIAC in a 25A rated module, not good!
The listing pictures for the 40A show a module with two screws on the bottom, indicating it might have two power-components, one could even dream about two antiparallel SCRs which is the proper way to make a 40A SSR.
Unfortunately they sent a different type with just one screw and a single BTA16-600 in the SSR-40DA. 16A instead of 40A, even worse than the 25A one!!
Also the RC filter across the TRIAC was completely missing, no components, just empty holes on the PCB!!!
The PCB of the 25A type is exactly like the one in the video, but the 40A has a much better PCB, with SMD components, good clearance and a milled slots between input/output traces. The 40A is also filled with more resin, the space below the PCB was almost completely filled on the output side and the components on top of the board well covered. A real shame they then ruined the whole thing with a 16A TRIAC.
Some used heat conducting paste when mounting the TRIACs, some had no paste.
It is possible to disassemble those SSRs without much damage, first pry the top cover with sticker up using a small screwdriver. Next use a fine tipped soldering iron to melt/burn the resin from the head of the TRIAC-fixing screw. Then remove the screw and finally the metal heatsink. The TRIAC will stay in place and hopefully not tear any traces on the PCB.
Reassembly is also possible, but hardly recommended.
After this experience I will try to avoid those SSRs.
They are only good for appox. 25-30% of rated load.
They should not be used where humans (or any other live creatures) might come in contact with the input wires, the insulation between input/output is not to be trusted.
i got one too, im using it with a PID temperature controller and a 2kW water heater. works fine, i dont have heatsink mounted. few minutes of use and it was slightly warm.
No problems with them at 6 amps or so. Varistor protection and additional RC snubber is highly recommended.
I have considered these relays but avoided them because of safety concerns. I have no problem piling on heaps of dubious low-voltage components that are practically free, but it is a different situation when a component is to be connected to the mains. I don't want to burn down the house or electrocute someone. Even though yours seems reasonably good, I think I'll pass. When certification markings are false and no one is accountable, it is hard to have confidence in a part.
This is the first video of yours that I've watched Mr. Clive, and it is simply fabulous. First class. Thanks for doing this.
+pillsnotbills I agree with you. Something that fails with no disastrous consequences is sometimes worth the risk. If I fitted one of those SSR's to some kiln or printing press and then it set fire to the factory, well I'd be to blame. £15 will get you a proper american or european approved unit, so it's not even worth the risk.
www.rapidonline.com/kudom-ksi240d25-l-panel-mount-ssr-4-32vdc-48-280vac-25a-load-with-led-zero-cross-60-1579
Thanks for this tear down Clive.
I have the exact same part from an ebay seller that I intended to use in a Kiln controller (~3KW max output).
So after watching this I stripped mine down to reveal that it also contains the BTA12-600B, obviosuly not quite sufficient for my application.
I have now ordered a BTA24-600B to uprate this to its original spec of 25A, at least I will now be operating its safe area.
Thanks again for your entertaining and interesting videos!
I took apart an old 1980's vintage USA made 10Amp 140V ssr and it is indeed fully filled with dense black compound. The lid is a full cover and they have little legs that extend into the potting compound to hold the lid on. Much more robust construcutib than what is shown above.
SSRs are really intended for fast switching that would wear out a contactor (controlling temperature on a heater, switching in a VFD) If you want something to be left on for long periods your usually better off with just a contactor. Best for resistive loads, they don't handle inductive surges well.
I messed around with some triac circuits some years ago which were basically schematic equivalent to this thing. Had no difficulties powering incandescent lights, occasionally some difficulties with compact fluorescents (the holding current you mentioned).
One thing that I think would be extremely good advice is to be VERY CAREFUL about powering heavy loads and/or heater loads with these devices. When the triacs got warm, they tended to self trigger, so that even when the opto-isolator circuit was turned off, they would still pass current. (negative temperature coefficient of solid-state devices)
If you are thinking of controlling a heating device with one of these, I strongly recommend using an electromagnetic relay instead -- especially given clive's teardown results showing you can't trust the printed current rating.
Or, yanno, you could use a proper SSR instead of a fake one. Real SSRs are not built this way, and they use back to back SCRs not triacs, so there's fewer junctions in the current path, the heat sink plate isn't a joke, the cases are backfilled properly, and so on. You do realize that you CAN'T use a mechanical relay for PID controled systems, right? You knew that, right? RIght?
I cant get my head around the concept of faking a 25A SSR and selling it for a couple of pounds on ebay when they could sell it for the same price rated and labelled as an 8A device instead. It would still be a very competitive price but legitimately rated.
as far as i know they all agree this fakery, yes, competition kicks in and they have ridiculous prices but... i saw that 60% of what they sell is just fake, fake chips, fake transistors, only modules work ok, because 99% of who buys this stuff can't tell the difference and a chunky transistor will turn on a small motor even if there's a 1\2 current transistor actually inside, novices are happy, modules work... i'm not è_é... i couldn't find a BC639 or compatible except on super expensive stores outsice the china cheap (just as en example, don't buy op amps eighter)... yes instead of 1€ i'd like to spend 1.5€ for a bag of 100 of those transistors but LEGIT ones, not caring of branded ones much
Welcome to China - Manufacturing philosophy: why waste money using unnecessary resin... you get what you pay for, but this is dangerous and a fire risk
It’s Chinese.
In general, it seems like the Chinese tend to prioritize things working and being cheap over reliability. So if they give you a current rating, that's not going to be a guarantee of steady-state functionality at that load, it's going to be "this will probably work for all these applications"
@Alweg Fan , as a long time engineer working for a company that supplies industrial automation system to Chinese heavy industries, including power generation, practically everything is sourced either out of the USA, Europe or Japan. I go over to a plant and I am rubbing elbows with representatives/engineers from Germany, Japan and the States primarily. I don't see ANY Chinese sourced parts/assemblies in those critical industries. So I think the top guys who are in charge of critical industries have a good handle on the quality and reliability of their own lesser industries. Now I suppose that over time, this is gradually changing, and more in some industries that in others.
I have a few of those in my shop. Thanks for the helpful teardown.
+Applied Science Well, you wer`nt to know were you. Hope you sell them and make a profit at the end of the day.
He is tearing down a fake one, just saying.
+Bradman175 With as many fakes as real ones: How can you tell?
+Adam Perry Very easy.
ul.com/newsroom/publicnotices/ul-warns-of-solid-state-relay-with-counterfeit-ul-recognition-mark-release-13pn-52/
Anything that doesn't match the image of the legit one is fake. The one shown in this video is missing the maximum amperage rating.
+Adam Perry and holy crap how scary it is that most of these fotek ssr are fake!! They're meant to cost like $40!
It's fascinating to see what I've been using for the last six years! I'm glad now that I acted on impulse and used them conservatively, and only for personal projects. But they have worked without complaint.
Years ago we used to build home control boards and in one of them we used a similar arrangement but simpler as the input voltage was known - hence resistor into the zero-crossing opto-triac then the same components on the right. We used 15A triacs on heatsinks and controlled 2KW heaters no problem.
Thank you! You've been very helpful.
I've been eyeing some of those for a project. Every time I hovered over the checkout button I would get a little voice telling me, "You get what you pay for."
Wow! Thanks for such a quick response. I am pretty much still in the electro-mechanical era so this type of thing is new to me, but still interesting.Pity most schools can't make subjects as interesting as you do.Fantastic work!
OMRON tops there SSRs all the way with sealant. Very seldom have any problems with their components, but they are quite pricey.
Just got one of these to control the heated bed on a 3d printer I'm building. It was a relief to hear your verdict that it may be up to the job. Fingers crossed.
Have used 25 amp solid state relays but when load was more then 3 amps would mount it on a heat sink and use heat conducting paste. Have just used a 6 by 6" aluminum plate that was 1/4" thick for loads up to 50% of label. Nice vid.
Thanks for warning us about this kind of devices !
I tested mine today, the led lights up, but it doesn't work at all.
Stumbled across this whilst going through your older videos - and I just happen to have one of these sat on my desk! Very useful teardown and glad to see they aren't all that bad.
'with a little bit more design effort' a true saying for most bargain Chinese products!
There are a few that are exceptional. They seem to have two levels of experience - highly skilled and rookie with nothing in between.
You can sit here ranting nonsense, but you have never created a product in your life. BTW the gap there is around 5mm which would require a voltage of 15000v to jump across that. There is no design flaw here, only a flaw in your thinking.
Who?
in theory and ideal conditions that's the isolation ( remember, if it ever did cross over that would be VERY VERY bad ; it would go straight to a likely low voltage control network) ... and like clive said there is literally no reason they made it so much less then it could have been.
There is a lot to be said for truth in advertising, or at least truth in specs. No matter what your skill level, there is no excuse for lies like a 12A triac in a device that claims the ability to switch 25A.
I'm using a pair of DC-type SSRs to control the electromechanical spring brakes on a 12.8m satellite dish (former NATO, we're restoring it for scientific use). The control system is 24VDC, as are the brakes, but the brakes are higher current than the controller can sink, so, SSRs. They're notionally rated for 40A, but I'm guessing the truth is much less. They came with heatsinks and the brakes are only putting a roughly 4-5AMP load through the relay output--rather more than the controller can directly sink. I've been an electronics tinkerer since 1973, and this is my first experience with SSRs. I also picked up some 24V contactors, in case the SSRs didn't work out, but, so far, they're fine. I placed large (30A) diodes across the brake coils, because it wasn't clear whether there was any kind of snubber network inside the SSR.
I bought a couple of these several years ago for a brew fermentation controller (BrewPi). Hooked it up to my old fridge and when the fridge should have turned on I heard the sound of something from the compressor but it didn't start up. I assumed the compressor must have seized up after being idle for months so gave it a few whacks with a rubber hammer but still the same,
Something then made me think to check the SSR and turned out it had a crazy on resistance so when hooked to the fridge, the output was something like under 100V rather than 220ish. Swapped it out for the other SSR I had and it worked fine... Invested in regular relays for that project instead.
I do understand there are no shortage of fake FOTEK SSRs that likely you and I found for bargain basement prices, but the genuine FOTEKs are more reliable. What about ordering one from CPC or the like and seeing if that's any different?
Thank you for sacrificing this relay to show us how it works. I am concerned that when there is 240V AC on the relay it is showing about 10V AC between the backplate and ground or neutral. I wanted to bolt this to a grounded chassis but with a potential between them it sounds like a bad idea.
I have bought 4 of these. one of the first two would not operate at all. I've had no issues with the other three.
I have two operating in my cnc setup, switching on my spindle and vacuum system with no issues at all.
My little spindle pulls maybe 8 amps, and my vacuum about the same.
We actually use similar relays. Allen bradley drives switch it to power a 120 volt brake on sew eurodrive motor. Works great. Great video as usual Big Clive ⚡️🏴⚡️
Our relays are allen bradley or omron.
"Our relays are allen bradley or omron." -- which means they aren't built like this trash at all, that's why they work fine. People in comments seem to think this video is an indictment of solid state relays, when it isn't - it's an indictment of FAKE CHINESE "solid state relays", built completely different on the inside from the real thing.
@@railgap You are right allen bradley solid state relays we use have 100% reliability. So far 26 years and I have never seen one fail.
Thanks for posting, it's good to see the insides. I used an SS40DD (to switch a DC 3A load) controlled by an arduino GPIO pin. I was using it well within spec, but it got very hot indeed and all the case melted and twisted. I built my own after that :) :)
I prefer the Use of Crydom SSR's. I have a bunch of Crydom SMR 4850-6 which can handle 96-480VAC at 50Amps. To control these thing you need a Supply Voltage (8-32VDC) to power the internal Logic, and then you can use the inverting or non-inverting control input to turn this thing on and off. It also has a N.O. and N.C. Alarm output and a integrated red Alarm Status LED to show abnormal conditions. A green LED shows normal operation when switched on. The Encapsulation is made with thermally conductive Epoxy.
Always nice to see what you're really getting in these modules.
Christ... I had one of these explode on me today, an SSR-40 DA unit rated to 40A. Opened it up to find the same bloody triac as the one in the video!
PodeCoet lmao, that sucks dude
PodeCoet thats insane? what is with the dopey chinese engineers that approve of this? After watching dozens of clive's vids, you get the idea they arent taught to actually understand electronics.
Hint: It's not the engineers, it's the marketing people...
Two RC constants in the circuity is using two time constants or double time constants is to make full use of the triac for more efficacy. Thus making use of the negative and positive parts of the sine wave
I love your videos, big guy (I am a big guy, also). This one in particular is motivating. I have long wondered why solid state relays rated at 25 amps do not perform as advertised. After a really looooooong research, I found that the amperage rating is bullshit. It is all about the wattage. At 5 volts, sure the relay will switch 25 amps, but at 280 volts, it will only switch 1 amp. So, people that want to switch mains power (120 v in the US) are tricked into thinking that they can switch 25 amps, when they can only switch about 300 watts (2.5 amps)
awesome. that makes sense. im gonna remember that
Thanks for doing this video, I have used this in my arduino controlled espresso machine, I didn't pop it (rare for me!) I figured it would be encapsulated, there are a few brands on the bay, identical except the red 'fotek' brand. Price was the same, so I went with the Fotek, although mine is also lightweight and airy and I also expected it to be potted full. Learned a little and a lot here today, I am running mine at 5v switching and only using it for a 1300w-ish heater under supervision so for me I am sure it will be fine. Great video!
I believe one can never have enough Arduino controlled espressomachines.
Thanks 😄
And who is supervising ? Not another Arduino ,I mean why an Arduino in the first place then ,of have you totally hacked your espressomachine ?
,
It is always interesting when you take things to bits. More spacing! More potting compound! Ya, for limited usage, looks OK.
This is one I needed to see, but I didn't know that till I saw it. This is amazing to see the insides of something I have used so many of!
I don't know what is the norm, but I tore open a US brand SSR once, and it was basically a chip and wire hybrid, bonded to the copper base plate. No PCB, no packaged components, just chips, bus wires, covered with clear silicone. The circuit was stupid simple, too, like 3 chips and no passives.
I just bought one of these. After watching this I stated wondering if I fill it with resin it might be better than it is now. The load that it needs to handle would be around 8 amps so it should handle that well.
Thanks for that...I haven't used triacs in my ex-line of work so would not know but I do know now. I will be on the look out if I ever do.
Just received my second batch of these for some small arduino projects. They've all withstood a couple of months of switching 5A every few seconds. Very happy with them. Obviously not a fully Professional part though.
Here is a reply I received from an Amazon seller after leaving a review of their fake SSR. It's a pretty sad excuse.
"I'm very sorry. I know our mistake made you a little disappointed, but please give me a few minutes to explain that，this product is from OEM and delivered within the period of novel coronavirus pneumonia ,at that time China is lack of raw materials and lack employee .Most of the worker, including engineers, was on vacation. Only a few people stayed in the factory. (originally, SSR-25AA used BAT24 Chip,SSR-40AA used BTA41A ) Due to the fact that engineers can't get to the company in time, many employees can't go to work in time because of pneumonia in China. Pneumonia affects the closure of the city) which leads to the following mistakes. Originally, our brand is eqiek, but due to the negligence of the printing staff, they make the label FOTEK instead of EQIEK . But we confirmed that there is only a batch of part labeled printed wrong. In fact, the main drawing of our products is eqiek(Pls check OEM brand EQIEK as attachment ) We have confirmed with the factory that we have ordered 1000 pieces of ssr25da and ssr25aa respectively. More than 900 pieces are produced with bat24a. Only a few dozen pieces are produced by BTA12A within a period of novel coronavirus pneumonia , engineer can't back to office immediately ,one worker mistaked BTA12 instead by bta24a, When we found out, more than 1000 goods had been delivered to Amazon's warehouse. We contacted Amazon to return the goods. However, due to the impact of pneumonia in the United States, there was no reply. Last , we have no idea to deal with returning the goods. .AMZ service answered :Because 1000pcs SSR-25DA with SSR-25AA have been diverted by AMZ warehouse .we feel very helpless about this matter. But we will be responsible for our every customers.Please Believe us, we only do honest business, we will not deceive every customer. On behalf of our company, I apologize to you. Now our company proposes two solutions: we will refund the money to you, or we will send you replenishment from China (this time we promise SSR-25AA 100% function and with BTA24A chip ) ,However, the US postal service is very slow and may not be able to deliver it to you in time as the spread of pneumonia in USA ,so hope your advice to us . Hope you can forgive us and hope you can give us a chance to service for you .We will cherish this chance .we are willing to return payments for you .Thanks for your time to read this long email .
All best wishes for you and your family
Happy everyday
JESSICA"
Hi, I've just been dealing with Amazon support over the fact that when I went looking for SSRs on their site, I still get advertised by them fakes which I already reported to them over a year ago, left bad reviews and so have many other people - but enough people seem to either not realise or go 'oh well it's cheap and does work for the lower level job I have for it' that they drown out the many reviews clearly calling them out for being fakes or just not working. As a recipient of fakes, do you fancy teaming up to at least try to get some proper action from Amazon?
@@edumaker-alexgibson
Amazon deleted my negative review.
I just sent this email to FOTEK along with loads of photos.
"To whom it may concern,
In September of 2020 I purchased a FOTEK branded 25A SSR from a seller on Amazon. After receiving and installing the SSR it quickly failed. I was very suspicious and decided to open the SSR and found the TRIAC was a BTA12 rated for half the amperage stated on the label. I sent the information to Amazon customer service and received the response shown below from the seller. My review on Amazon showing I received a counterfeit was originally posted but has since been removed. I'm not too sure what FOTEK can do, but Amazon and the seller are selling dangerous counterfeit SSR's with the FOTEK name on them."
@@edumaker-alexgibson Here is the lackluster reply from FOTEK...
[Dear Sir/Madam,
Thank you for your information
We know that there are a lot of counterfeit from China
because of counterfeit is from China
FOTEK is from TAIWAN
I know it's problem
so We are not sell on any Website to be buyer's protection
and we have provided "where to buy" on our FOTEK's Website]
Pretty sad excuse indeed. I do business with China on a daily basis, and you need to plough through these kind of BS excuses every time things go wrong. You ask yourself why they even keep trying, stacking lies on top of lies, wasting your and their time, instead of admitting they screwed up and use that same time to fix it. One time we received bicycle carriers with only half of the parts in the boxes, "because factory didn't have parts". Took months to receive the missing parts (on their costs) and open up every box to add them. Big chunk of their and our revenue lost. Sigh. Simply admit you can't accept the order because of parts shortage. It's a strange business culture over there.
@@resspls10k They get to keep your dollars by accepting the order regardless of ability to fulfill it. The lies are to weary you out/delay for time - in the hope you will give up or run out of time for refund
I use this one for my brewing temprature controller. It works well in that application with a 12v signal from a PID to trigger it. If I was to market the unit I would probably use better relays and pid's though.
It is interesting to see how one of those commercial modules does it. I don't know about pulling multiple amperes through it with only that bottom plate for heatsinking, though I guess that depends on what you have it mounted to. And I'm assuming that the tab on that triac is isolated? I have a bunch of 4A/400V triacs on hand and plans to build one or more of these. I suspect that I'd probably use somewhat different parts for the snubber network, though. I plan to use the MOC3010 opto, I think it's the MOC3020 that's the one with the zero-crossing detector in it. Been a while since I looked at those datasheets.
Zero crossing is not great for inductive loads. I've used these FOTEK relays quite a bit, however I generally use them @ 1/10 spec. Doing that I have never had any trouble and they are cheap enough that such over-specing is very affordable.
I let the smoke out of one of these, yesterday. On a temp. Controller, for my homebrew. It worked fine for the first 20' or so. Then that cooking component smell, it failed open.
Opened up an eBay Fotek SSR-40 VA (phase angle controller) you'll never guess what! It's (the triac) not rated for 40A but 25A. At the price it's worth buying two to tare one apart before using the other.
I would advise a PSR-25 if you are looking for something a little more tried and tested for phase angle control.
So if i understand this correctly these relays would as you say be used to switch power on and off on say a AC single phase or AC 3 phase big foxtrot motor/pumps etc and then that would mean that the other side of the relay could use a much smaller dc or even ac voltage, switch and wires negating the need for big Foxtrot cables, wires running through to the switch and cabinets etc. Great idea really, I guess you just leave the switch in the on position until you want to turn it off, way safer way to handle all the switches inside cabinets on low voltages and smaller cabling... always wondered how they worked Thanks Clive...
I ran a 2kW (~8A) heater through this thing.. in about 30 minutes the room was filled with melted plastic smell. The damn thing melted into a blob
Solid state relays aren't relays or contactors, however.
I am trying to do a similar thing with a 2kW (~8-9A - 200/230V) heater for my house. What did you use to solve this problem? Did you buy a better quality SSR that has a higher current rating?
+Viljar Niit And how much was that thing "officially" rated for?
Her's a thought - could the metal backing need to be mounted on a surface that can radiate heat? if you mount this thing loose in a box it might cause over heating.
+Ed Rundle Well, if the damned thing melted into a blob from 8 amps of purely resistive load, then I don't think even a heatsink with forced cooling (i.e. a fan) would help things much (except if it's a CPU cooler perhaps). Even at 8 amps. If the thing would run "only" hot to the touch, then maybe. But melting (=temperatures of 200 °C and more) means that 8 amps is WAY above the rating of the semiconductor part inside.
Real ones have about the same inside, but better clearance and filled generally with a glass bead filled resin to save the cost of the expensive resin. This one is filled with a powder filled resin ( probably chalk powder or some milled dry clay) so keep it dry, it probably is hygroscopic.
The 2n ceramic snubber is probably the same 10n 2kV Class Y ceramic used in SMPS use, where it has quite a good pulse handling and also will fail open circuit. Not a bad choice, cheap but will do the job better than a film unit.
Anyway, standard appnote, and the zener does double duty as reverse polarity protection for the opto, though the LED indicator will die with reverse voltage, but who will ever see the red inside a cabinet in any case. As to the triac, I doubt it has any compound, relying instead on the resin seeping under to fill gaps from the screw bending the tab, and if the screw is loose it will keep it in place as well. Isolated tab, so not really meant for more than 5A of load, unless you really use a decent heatsink, which that zincalume casting is not. Rating 220VAC 2.5A with a peak of 5A is more appropriate for it.
Only just come across this, and I imagine that at time of watching (Dec 2023) any SS relay on eBay priced at under £5,- isn't even going to have a 10A TRIAC in it. 100mA or even 1A perhaps (Whatever will just about withstand a 30s inbound test on a 20A load) but definitely not 10A! 📦💸😉
That aside; Living in an all-electric flat I'm struggling a lot with cost of heating and I keep considering ideas like running TRIACs and ICs close to their upper tolerances and benefitting from the heat this will give off as a biproduct. Even if I have to swap in new TRIACs every week it's probably still going to work out a lot cheaper than running my 3kW convection heater at the present 36p/kWh for a few hours each day! ⚡🔥💡
As somebody who has torn apart high quality solid state relays... Yes, they are typically fully potted basically to the brim.
I did not expect even that quality for that price.
We used a 75A SSR in a safety analyzer (bit like a PAT tester). It needed to switch 20A inductive loads and pop a 20A fuse on fault without popping itself. So ... yeah - do it right. If you've spent $4000 on a big brand safety analyzer, it had better "bounce" and not break when you abuse it a little.
I have several of those SSR. You need 5V to switch them on; 3.3V doesn't gonna make it. I made my first SSR with the same components before buying them and have it working for more than a year in a Sous Vide (1 KW heater), so, component wise, them seems reliable to me.
Great Vid, especially the drawn diagram. The one thing that has always confused me about most Asian products is that they will do something about 99% right, but manage to pull the defeat outta the jaws of victory at the very last moment every time.
Well done Clive. Good investigation.
I have seen 100 AMP 60 volt DC SSRs on eBay I highly doubt they can sink 100 AMP heat dissipation in the same form factor package that you have FOTEC without liquid cooling. My interest in the DC SSRs was for a golf cart speed controller. The limit however is the PWM response. Others have said the max PWM response was 800 to 900 cps. Krydon brand SSRs are better and can have a slightly higher PWM from 1000-1200 cps but still is within audible hearing range. I would like to see you test and tear down a 100 Amp DC SSR. Thanks for the video.
After watching this video earlier this week, we had a system come in from a university that had one of the counterfeit SSRs, Luckily it hasn't been run at all.
You didn't show the actual teardown, so I can't tell if the thing was damaged in the process. I'm wondering if the rather nice case could simply be upgraded with a better triac, along with some attention to improving the isolation.
Another concern not mentioned is that unless earthed, the bottom plate could become live in a failure.
I'm going to be using one of these for a small kiln, so thanks for the warnings on these.
I'll bolt it to a grounded heatsink, and make a voltage detector circuit that will enable the Arduino controller to check that the SSR has mains supply voltage across it when it is supposed to be off.
I love how you have teardown videos of the random stuff I buy from ebay.
Hi, slightly late to the party, my company use SSRs to switch 240v 10A motors with 12v DC, we are talking literally thousands of motors, we use SSRs from a certain UK supplier and they're very reliable, except when the SSR fails, they go open circuit across the switch, that then obviously causes the motor to run continuously until the thermal cutout detects that it's glowing and about to catch fire, or it actually catches fire. So, we got a few of these to test, not based on cost, and sure enough, we managed to get one to fail and it just died and didn't go open circuit, brilliant, or so we thought, we then passed these SSRs to our tech guys who threw them in the bin and said that 'if we ever put that shit on one of our units without consulting them first then there'd be trouble', they said that these things were 'lethal' and they'd rather kill motors than engineers.
The triacs going short circuit is a common failure mode. Your tech guys were probably concerned about the possibility of poor isolation resulting in low voltage circuitry becoming live at mains voltage.
Today my 40a fotek ssr burnt out after a good few years of switch a 2kw heater so I decide to lake a look, the triac was a bta24-600b rated for 25A. Thought I would share as this may be useful in relation to Clive's video.
"ekoenig" is actually german for roughly translated "e" for electronics and "koenig" or "König" ("oe" is spelled in germany as "ö") for "king". So the seller seems to be a king in electronics.
99.99% of these Foteks on eBay are counterfit.
Great stuff,i would like to see inside a proper non clonned one,just for reference.Cheers Clive,love your vids.
Great tear-down, really interesting!
I've been using one of these Fotek branded SSR 25 DA (not sure if the one I have is genuine or fake) for about 1.5+ years to switch a 1.3kW purely resistive load at 240V AC from a 3.3V input. It was ridiculously cheap on ebay, so I guess mine is the fake one. Probably still working because I'm well within the "safe" operating envelope. Would no doubt have exploded by now if I was on 110V mains :-)
+James Ward For a modest load like that it should actually be fine. I guess these are probably widely used in real-life Chinese factory equipment.
Wow. I’m glad I went with the 40 amp version for a heating application recently, Same brand. bought from amazon.
GreatScott also did a review of a 40A version of these, which also turned out to have a 12A rated triac inside. So maybe we should consider them all as 12A SSRs regardless of their marked amperage... at least until we have destructively dissected one to confirm a higher rating of its components.
Ben Hetland I’m wondering it it’s rated 12a with no heat sink. But maybe with a heat sink it can handle more amperage
Hi Clive. I really like your video's. keep up the good work! greetings from The Netherlands
One thing you didn't test is the turn on voltage. Triacs are not the same gate turn on in all quadrants. You will see that below 3V (2.2 - 2.7V) the SSR will only be half wave. That could damage some AC devices which are inductive and that may be very close for 3.3V devices. More expensive SSR like CRYDOM use a constant current circuit and LED drive current will be a constant 2ma from 3-32V. You get what you pay for.
This is dangerous stuff. By most electronic safety regulations you;d expect 6.4 mm CL/CR between the live circuit and the 30V circuit on PCB material. that means for a DIL package you need all of it!
if you use one of these make sure you derate it by 50% especially if you don't use a heat sink. (ebay sells them) I have had quite a few fry spectacularly (fire smoke total met down) controlling a 1500 watt heater 120 v. swapped out for 40A rated ones (almost same price) and stopped the fire hazard. BTW they work very well connected to output pins on arduino boards
i use these in pinball machines that have ac motors controlling backbox animations
they are easy to drive off the 24v solenoid lines and don't have woefully undersized traces on pcbs to burn up without blowing an 8A fuse
I had trouble driving one of these cheap ones with a raspberry pi (3.3V). Had to add an opto isolator. Works well though.
Thanks Clive. Have the 40a version switching my home theatre as the 10a relay kept welding shut!
Make sure it's in an area that is fire-safe. These cloned SSRs are not rated anywhere near their stated rating.
@@bigclivedotcom If it's really only 10a with a heatsink, load should only be about 6a by my calculations.
The Crydom unit is fully packed with resin and with an admixture of whst appear to be very fine fibres.The unit failed quite expressively by blowing out part of the top on the a.c. side. Looks as if it may have been the resistor -capacitor network. Another blow out on the upper rh side looking at the unit from above.Further dissection looks like it will destroy any components left.
I was wiring one of these up at work today and this video showed up in my feed when i got home. How odd. It's only switching 3 Amps but controlling a rather critical temperature.
Thinking tomorrow i'll change it for one of the £25 ones from R.S.
Just bought a different make, similar spec on the outside. Might just tear it down to check what’s in it before using.
Be careful with this: I used a 25 A of this type to switch a 2000 W 230 V water boiler , so less than 10 A. However, after a few weeks it completely burnt. (It was hanging on its wires open in the air, so it could cool on the air (20 degrees C). I am happy my house was not set on fire.
Do you understand what you did wrong?
IIRC real ssrs have the die on the back and everything wirebonded and then there are the mosfet based ones, which are also mounted on the back but they use back to back mosfets instead.
I know that this video is old, but there is one thing about this video that (I felt) should have been discussed. I noticed that the Triac was mounted on the Rear Alloy Casting, (as a Heat Sink) but the Triac had mains voltage on it! In the design of the Fotek SSR, do you think that NOT enough emphasis or consideration has been given by the Designer of this Unit, by the inclusion of an Earth?? If the underrated Triac becomes destroyed, there is a high potential for the rear alloy casting to become "live". This could be hot, dripping solder from the Triac itself - making contact with the alloy rear casing. BTW I always like your informative content, a regular viewer who likes the Ayrshire accent!
In most industrial applications the heatsink surface would be earthed.
The better high power SSR's use back to back SCR's rather than TRIACS because SCR's are more reliable when driving large inductors like motors. I talked to Motarola years ago and they told me you have to overdrive TRIACS to reliable switch high inductance loads in all four quadrants. This is easy to do if you know what voltage your going to be driving from but can be more problematic if you really do have to have a wide input control voltage. We were buiding our own SSR's so we controlled the input switching voltage and could set the gate driving voltage and thus the current.
Brilliant! I've been concerned about this unit that I bought recently, it's for a 240v floodlight being connected to my 12v alarm, you've showed it's properties well. Thanks! Subscribed!
I have several of these that I use around the house, never had an issue driving just a few amps..
It's not OK, someone I know got one and it burned up switching a 1kW resistive load. (At 120V, so about 8.334 amps ... not even 10 amps!) (The load is a a heating element. The inrush is very low compared to an incandescent bulb load!)
+Benjamin “Ozias” Esposti A heating element is similar to an incandescent bulb load for the same reason, the resistance is lower when cold. A long wire between the the load and the relay also turns it into an inductive load, which causes most of the inrush problems in domestic light circuits.
Watching this was interesting I just replaced one on my T-shirt heat press, fully repaired for a few pounds.
Good video. Anyway to get rid of the lights flickering when using these for PID control?
thank you for reverse engineering and the nice schematic (which I copied) for future reference. great engineering work!
the crydom genuine ones are solid inside i have tried to open them but they are full to the brim with resin and very reliable the failed one was on a transformer for a sealing bar that failed and shorted the 300v ac to earth very melty very very stinky
One thing that was omitted in this teardown is a review of the triac to baseplate isolation.
It is critical that the line voltage is isolated from the heatsink.
+Peter Dvorak I should have done a 1kV test. The BTA12-600B is an isolated tab triac if it is genuine.
I thought I was gonna be really clever and put a few of these in my off road rig. I was all sorts of proud, of how pretty the wiring looked, show stopper... any how, I was having all sorts of troubles, and of course the last things I was to check were these batch of SSR's.... I am quite sure that these are fine in a clean power supply environment, but, these absolutely do not like choppy supply voltages. Please note, the charging system is extremely clean, by automotive standards, quite powerful too. I went as far as adding a few stiffening caps to level out the chop from the rectifier set up in the alternator, that did help lots, but still had issues still left an unpredictable temperment with these, I really wanted to keep them, as per the amount of work that went into wiring them in an art form, never mind the amount of work that would be required to yank'em out, and completely rewire the entire set up, for typical mechanical relays, of which is fine, 40 hours later...LOL... Like I mentioned, I am sure these perform great with a clean power supply, I do not recommend these for automotive use. I thought about trying to build some sort of input anti-spike stabilizer system, but, then, it is kinda defeating the point of the cleanliness and cool factor at that point.... I yanked'em.. Cool video, again!!!
What were you hooking up?
Stuff like on board air compressors (twins), all of the lighting that points forwards, backwards, underside. Electric fan, etc... any location that would need a relay, I had these set up, instead...
Just curious, never tried these in an auto application.
I really don't know way I had trouble with them... I used to use these in some of the golf cart mod's and never had a single issue with them, heck, with as much as my off road rig has given me over the years, it is likely just my rig...lol. More over, I could recommend the idea, if you desire to use these in automotive application(s), I would wire up a simple test circuit and run it for a week, even if it is simply wired through the cigarette lighter / 12 volt power supply port... maybe to just turn on a single light bulb or a few bulbs... I have used the ones that came off the rig for other projects and they seemed to be just fine, as far as I know.. They do look cool, all wired up, especially if there is several of them... Good luck.. (test'em before going nuts with'em)
FWIW, Jeep apparently used SSR as a radiator fan relay in Cherokees some time after 2000, and they're notorious for failing.
jumping back 7 yrs in time now. im guessing the yt algo chose this due to the words resin used so many times so that would be a logical grouping as well as triac which would then bump up the relevance.
I tore down a Fotek SSR 100DA after it failed and it had 2 x BTA24-800 Triacs in it which are rated at 25A each. It seems that it is common that the manufacturer claims double the actual rating. Another issue is that the heat dissipation is not good as they use an alloy for the heat plate instead of aluminium, this further reduces the efficiency and capability of the Triacs. Very dangerous, be cautions, attach a very large heat sink and don't go beyond 30% of the rating if using these cheap type relays.
Purchased a few of these not so long ago. Didn't take them up the the marked ratings. Guess I won't be now. LOL I had intended on using one to switch a MOT (MircroWaveOvenTransformer) Spot Welder. I'm quite sure it would had pulled more then 10A. May have worked for some time. Would only be on for microseconds at a shot.
While to great in design for general use this looks quite decent.
I've always been afraid of buying these on Ebay but I think I'll grab a couple now for personal projects.
This is niftiness incarnate. I've been looking for a thing like that for a while now.
iv had a few of these off ebay, the white ones last for about 5 minutes
the black ones r more expensive but they last and are the ones to get :)
Thanks clive :-D, i sort of expected it to be more heavy duty inside, that ssr looks more like on par with a lamp dimmer.
I had something like that in a draw, it had a thick base and solid like a block of resin, its a black device apparently made in america, never found a use for it.
I pulled the DC version apart, it had a BJT transistor, with the part number on it, good for about 8A, nowhere near its 25A nameplate. Fraud from China. At 8A it dropped 3V.
borrowed off the Fotek website about them some what long winded
. Established 1995
FOTEK CONTROLS CO., LTD.
Established in 1985, the company dedicated itself to
the research, development and manufacture of industrial
automation hardware in order to support the automation
of Taiwan’s machine industries. As all sectors of the
industry face fierce global competition,FOTEK
continues to provide the guaranteed best quality through
advanced technologies and strict quality control
( certified ); At the same time, the company has
continue to pushed to increase production,
standardization and international standards compliance
of electronic controllers in order to reduce costs and
increase competitiveness. FOTEK products are in use
in many Taiwanese industries such as rubber/plastic
machinery packaging, food, printing, parking,
transportation, many major machinery producers.
The company’s products are sold electronics and
cabling. The company is alsoa designated supplier to
all around the world, and many internationally renowned
companies in Italy, Germany, the USA, the UK and
more use FOTEK’s OEM services. With nearly ten
thousand products, the customer’s requirements can
always be met, making FOTEK best partner for the
manufacturing and machining industries.
Management Philosophy :
→ To assert quality excellent。
→ To pursue company lasting operation。
→ To treat Customers as forever partners。
→ To research and develop over-valued product constantly。
→ Innovation、Responsibility、Reliability and Performance。
I used one on my off grid solar set up to turn the immersion heater on when the batteries were full.
It's a good idea to have a secondary thermal protection with SSRs as they tend to fail short circuit and jam a load on.
thank you so much for this video
i keep learning more and more from you
keep up the good work sir
Now we are talking as this is used (or even better the 40A version) for 3d printer heated beds that uses mains instead of a PSU so it will heat up faster.
I find hard to believe that it only have a single triac I build a very complex pulsing array for a custom power inverter... and I used 20 fet's of 45A each for a load of only 250 Amps... so it will be interesting to open one from a reputable brand "Expensive" and see if they only use one of this devices also!!!