Otis was the first company to produce "safety" lifts which were still safe even when the cable broke. Mr Otis put on a show in an exhibition hall: in front of the public he would be hauled up, then the cable was cut and the lift cabin fell briefly and stopped. That was the foundation of their fortune. The cabin runs on rails in the shaft, and there are brakes against the rails. The clever part is that the brakes are naturally ON, and they are connected to the suspension point of the cabin so that the tension in the cable pulls them into the OFF state. If the cable breaks the brakes immediately come on.
I'm in the elevator industry, and have worked for Otis. If you hold down one of the Otis guys that I worked with and tickle them until they cry uncle, they will confess that the best elevators are Mitsubishi. I heard my boss say the same the other day, and he's a born and bred Kone guy. One example of the different philosophies: Mitsubishi Japan decides what capacity you have to install new elevators in your area, and if you sell more elevators than that in a year, they say "Sorry, nope, we're not supplying you because you're gonna rush the job!" Otis is definitely good kit, especially the German stuff, but a lot of the Gen 2 stuff come from China from factories that Otis only own 35% of, and any independent elevator supplier can walk in there and buy Otis kit with their own brand name on. The standard blue SVT service tool is plugged in to do parameters on the standard software. Crazy!
Those two empty footprints next to the WAGO connector at 9:52 are for a LEM current sensor module (CAS 6-NP type) to monitor motor current and the IGBT module is most likely a Mitsubishi PS21563. How I know this is because I work with servicing these type of things and are also used in ThyssenKrupp Universal Door Operator 6300PAx series. I have seen that Infineon processor used in a Schindler (LUST LTI) IDD door operator too. Although this particular OTIS drive is not familiar to me, here in America a variant of the type uses an STM32F103 for the processor. OTIS Germany makes really good stuff, I serviced an OVF20 lift drive made in Berlin a while back and the lifetime counter showed on the tool that it was in operation for over 30 years and made over 2 million run cycles.
Building I used to live at has the original lift in it, with an install date for the motor, gearbox, cables and car of January 1939. Originally FIAM control gear, later on replaced with Express cam switches, which got a little crotchety with time, as the shaft guides were never correctly aligned all the way in the shaft. Eventually we paid Otis to put in the upgrade kit, which replaced the crotchety tappet switches with a tape and magnets, and put in a new controller. Otis underquoted, turned out the existing trailing cables did both not have enough cores, but also failed insulation test, not a problem with 110VDC relays and lamps, but a tiny issue with 24VDC running next to 230VAC power for the car lighting and fan. So in went a nice new 60m of flat trailing cable, 50 cores of 1.5mm copper, and it was commissioned. week later the flat cable relaxed, and moved on the bar, catching on the bottom of the car, and slicing through it. Controller did not survive having 230VAC applied to the 24VDC power rail, though the power transformer, the original FIAM one from 1939, was totally happy, seeing as it only had 2A load on the 24VAC winding, while it was rated for 20A, and the 110VAC winding still had a 2A load from either the brake coil or the retiring cam. New controller and cable under warranty, ouch. There is however a piece of cardboard with a crude drawing I made on it, laminated into the wiring diagram book, with a backing page containing a sign off on the modification by a Pr Eng, where I solved the one issue, that it would trip the breaker for the retiring cam and brake intermittently. Only if you went from top to ground, then called immediately back to top, where it would reliably trip exactly half a floor short. My mod was to move one wire one position, and install one jumper wire, so that the brake coil and the retiring cam each had their own section of the 3 phase breaker, used there with one position being integrated into the safety circuit, like all the other breakers, with a safety link for breaker fail. Also had to train the techs as to how to use MOV devices, they put 110VAC ones on 230VAC, and they exploded. Had to train them on what those numbers mean, and that using a 275 or 430 MOV on 24VAC, instead of the 33 MOV, was not going to provide any protection. Also correct application of them for the job, as they were not really trained much for retrofits, more dealing with installs of complete kits. If you order a lift complete from any of the manufacturers you get a kit, complete down to the last detail, and containing every single fastener for every part, all bagged and marked on a label which floor they are for, and taped to the dorr for there. Shaft as well, all sections bagged, and fasteners all taped to that section. 10 ton kit of parts for a 6 floor building, excluding the container that had the car and MRL drives.
Yeah, definitely looks like that igtb module, doesn't contain any intelligence or isolation, which would explain all the optocouplers on the inputs and outputs
I've been using Wago connectors for decades in many different applications & never had an issue - especially when compared to the competing screw terminals that get over/under tightened etc.
Oh yes, the serial report is still in use. RS-232/DB9 connector can still be purchased on a new laptop. Dell’s “Rugged” line of laptops have RS-232 as standard, and you can have a second port added, during or after purchase. It’s a slotted module. Good old COM1 port! Super versatile and nice to have without a finicky driver for USB-Serial adapter. COM1 just always works and there’s a ton of legacy equipment out there where RS-232 is the only interface available.
Yes and also having real serial helps, especially for things that need that proper 12V drive, and which draw power off the status lines as well. FTDI chips do not do that well, and nobody wants to pay the price for the Max232 either, to do the job properly.
Oh nonononono. I'm friggin' DONE adding those damned 323/3232 level shifters to everything I wish to communicate with. I'd much rather install a thousand USB serial drivers (not as if there are more than 3 or 4 different ones, at most, ever, in practice) then bother with the level shifter, the four caps, and a DB-9 ever again.
Yea I build industrial machinery and RS232 is still the more reliable and preferred methode of programming frequency convertrrs and engine controllers.
@@NinoJoel The word you're looking for is "legacy". Same reason you can still buy a 741 opamp, yet no one in their right fucking mind should ever use one in anything designed in the last few DECADES.
That was a nice exploration, and not your average everyday kind of device. Lifts have always fascinated me since as far back as I can remember. And, scared me in equal measure. I can still remember the little old man who used to operate the lift in one of the big department stores in York - They've certainly come a long way since then! That would have been a crafted job in its time, and now anyone can step into a lift, push a button, and off you go. As the old man said "It has its ups and downs, but it's a steady job. You meet all sorts in a lift".
@@TrondBørgeKrokli Indeed, and the old man was a bit of a wise guy! It would have been in the early '60s and about the time I started at school, but it was his regular line, and given the chance he'd either, get that line in, or, something like it. He was a popular chap with parents and adults alike, and given the torterly pace of that lift, it was just about possible to have a short conversation with him, see depending on how many floors you had to go.
I lived in Reading in the 90’s and occasionally I’d see one of their company vans with their office address on the side. Yes, “Otis, Reading” - always raised a smile :-)
As for the resistors (R1, R2) and the relay: The resistors limit the in-rush current from the mains when you plug it in. Once the bulk of the capacitors are charged the relay kicks in and shorts the resistor. Helps to reduce the stress to all the components on the board AND your fuses dont blow if you switch on the whole system at once, like after a power outage.
Clive I have to say I have been watching you for quite awhile now, I have no understanding really of any of the things you talk about but you make it so lovely to watch. Best tech mumbo jumbo youtuber :)
Date of construction on label says 2007, so the software is most likely the original from factory then. The pcb is also labeled 2007 in silkscreen label and in copper.
Dang, I came here for "Exploding an OTIS door controller" and only later realized that I misread based on the usual channel expectations. Can we please have a follow-up?
I love seeing inside well designed industrial stuff! Controlling power electronics digitally with microcontrollers/DSPs/etc is becoming increasingly common in well-engineered, complex bits of hardware. Gives you a lot of flexibility and configurability of control algorithms (since it's all done in firmware) - plus you get street cred for doing things the hard way ;) Of course it also introduces the blessing and the curse of being able to "fix" things in software. In the last few years, with specialty power designs, I usually look for microcontrollers rather than dedicated power chips.
4:30 Canadian here, the Wago thing is just north-american electricians getting mad that you insult our oh so beloved wire nuts. Even though Wago style connectors have been ratified into code for permanent installations. It's mostly the electricians over say, 30-35 that decided they were done learning after they got their journeyman cert. all of the failures i've seen, weren't from material failure but rather installation mishaps or perhaps carelessness.
In Finland I usually see either Otis or Kone elevators. The Group Control computers of Kone have a AMD Geode processor (from the early 2000) still in the new models. While the motherboard has had some changes in the last 20 years due to component availability it still looks pretty much the same. Obviously much more reliable than current consumer tech.
@@alexbrown1050 new stuff is cheaper altho you are correct moving to new stuff would cost a lot as well. However current motherboards for example can't be expected to survive reliably for more than 5 years while the 20 year old motherboards are still going. Mostly this is due to electrolytic condensators and can be worked around with proper components. Higher frequencies cause their own issues with lifetime as well. One of the main issues is much thinner tracks which are more suspectible to corrosion. Greater performance has it's downsides.
I think the shaving thing was more kind of a prevention of getting caught by any rotating mechanisms which elevators have plenty of. Just like with long hair things can go bad and ugly sometimes.
Looks like R112 and R113 discharge the extra-spicy capacitors. Only thing better than a good MELF is two good MELFs! Properly engineered with adequate voltage rating across the resistors, low impedance copper pours with good separation, teeny tiny trace buggering off to the discharge rail. From the via in the pour between the two resistors, it looks like they're also providing an elevated voltage reference for something or other. I love that the brain box has an "ESD sensitive - donut touch" warning on the PCB and Clive keeps poking it with his finger.
Otis Elevator manufactured the first major frame assemblies and several other sub assemblies for the Brunswick Blake Collender company. Which would eventually become Brunswick Bowling. An industry leader in automatic pinsetting equipment for over 65 years!
A good friend of mine is an elevator tech here in the city. Their field has got to the most in-demand out of all the trades right now, personally I think it's overlooked. Overtime gets thrown at them like crazy, not enough guys to handle the load. Can be a lot of work with all the troubleshooting.
I have always preferred Schindler to Otis for the simple reason that there is something special about getting into Schindler's Lift and being reminded of one of the greatest movies of all time.
That's some beefy components there, kind of reminds me of the old door controllers we used to use on the buses that had a similar programming system and options using a PC and link up system.
I've heard that spring terminals can better than screw terminals in the long run. When temperature changes, materials expand and contract. That can cause a screw terminals to loosen over time. While spring terminals remain springy.
The service tool exists in two variations. Black and blue. The blue tool has a connection counter that goes down every time it's connected to a unit, and has to be above 0 to allow to perform certain actions on new systems. The black tool does not have this but it is used on older control equipment. There also exist a interface for connecting to a computer, but those are used in very niche cases.
I was doubting non-screw connectors until I met a retired guy, who was once building test gear for ABB circuit breakers. He told me they had such connotations in their 10kA test station w/o issues. Since then I stopped using the screwed ones. To the MCU: even a STM32F103 from a bluepill board can generate a 3-phase PWM signal with dead time by using a timer and a DMA channel. As long as the generated AC frequency needs no change, the CPU needs to do exactly zero. And the IGBT-module contains 3 half bridges and current sense. It’s capable of generating 400V 3-ohase AC up to ~5..10A. Quiet powerful, but remember the 50€ Chinese variable speed drive units. They are based on such modules. The SMPS with the vipers are good stuff, if you protect them against surges from the grid. But as you can generate revenue/GDP worth a new dishwasher/washing machine, just by saving 50cts of components, they do it. And the vipers/TYnnn‘s/… occasionally disclose their magic smoke finally.
You mentioned that the doors needed plenty of power for weights and friction and such, but left out that they also need enough power to crush baby's heads. Hence the large capacitors to provide that power reserve.
That Otis Blue Box is so rare. Been looking for one for years as I have a Otis 411 lift I look after and we have to program it by hand :( Otis make some of the best gear out there with safety.
I like how even though you have no use for that device, you still don't want to damage it, like the respect someone gives to fine artwork. Also, why no circuit diagram??? ; p
Good connector, on a device operating since 2007 and only responded to a command 1000 times! (open 1000, close 1000,up or down, open 1000,close 1000)...
I had the pleasure for the first time this year to work with Germans at Siemens. I loved every minute. They knew what I was thinking before I had to explain. It was an utter pleasure. What was even more of a pleasure was they laughed their tits off when I mentioned Boris. I think that says a lot.
BigClive at 1.5 speed is hilarious. I was previously listening to David Vizard Performance on YT who speaks slooooooooowly and forgot to change it. Oh well, back to the start.
Nice bit of kit. I can relate to your feelings about electronics vs electromechanical. In my early days the pinnacle of what I worked with were the Ampex VPR2 1” C format VT machines. Loved those. Really well engineered precise mechanisms along with elegant electronics that you could actually repair down to component level. Things started going to pot when the cassette formats like Betacam started to appear. The mechanics changed from precision machining to bent tin and the electronics got more and more complex, in part to allow for/compensate for the shite mechanicals. The final straw was when they moved to storing the signals on tape in digits rather than analogue. By that time, there was nothing field repairable left in the darn things. Pretty much reduced to board swapping. The machines still cost the same as an average house, but they didn’t look like very good value! Ah, nostalgia…. :)
A building I worked in for a while had a reliable but very slow old Otis elevator. Someone (necessarily clumsily) altered the name OTIS above the controls to OTIOSE…
Twisted pair RS485 comms cable, used by these terminals, is reliable and cost effective. If you used wiring such as ethernet then the surge spikes surrounding elevator electrical noise will make higher speed comms problematic. The cable quality would have to increase, it would need more shielding and would be stiffer and less flexible to install. All the practicalities of the hard real world conditions are in this design. IMHO, you are looking at the best quality of industrial designs.
Still use for the newest stuff in NA. Module 1 is ocss, 2 is mcss and 4 is dbss. The whole Elevator. They only show when communicating. We call that a glide controller.
In early 2000 I was a computer consultant on a project at UPS. They had a requirement that all male employees be clean shaven unless they had a religious reason for having facial hair. I have a beard, and did not shave it because I was not an employee. So, I can totally see Otis being the same way.
I remember recruiting an engineer only to be told by the boss “How can you be so stupid to hire someone who is too lazy to shave !” That was in the eighties though. It’s hard to imagine such nonsense would still prevail.
@@anonymous.youtuber It depends. If the beard is well kept, that's one thing. But there's also the "I haven't bothered to shave, trim, or anything else". Don Johnson being the poster child.
Used to be in construction and was always envious of the Otis guys. Always sitting around aha. I will admit the weights they use to calibrate or set the elevators with are heavy though. They were always the same guys on the job, and important enough that they were “allowed” to smoke in the shaft even 420 at times. One time my tile company broke a panel of glass beside an escalator that was turned off for the night, covering it completely. Mr Otis was not happy, not for the mess, but because it was off and we didn’t destroy it.
Was it Otis that built the elevator test shaft in Germany and had the grace to put an observation tower at the top? One of the companies did, and the building is tastefully done on the outside.
@@almostanengineer yes, but I think back then the technical stuff was more interesting. They remodeled the thing and now it genuinely looks like a wizards tower.
infineon is relatively well known within the pc component industry as vrm (voltage regulator module)/pwm controller manufacturer, also makes mosfets to go with the vrm controllers, would make sense they also makes a processor like the one used in this door controller
I used to be quite suspicious of Wagos and similar (from an early bad experience with the terminals on Grundfos pumps) but love them now and wish the 221 style were available in larger sizes up to maybe 25mm2. It seems silly that they do DIN rail mount large ones but not inline ones. Seen so many melted cables and burned devices due to poor terminations in cage clamps. In the other hand, I was trying to deal with a faulty solar installation recently where it turned out the clips in a Schneider KQ board weren't clamping in properly. It had been installed correctly, worked correctly for a while then just decided that it didn't want to clamp in properly. Not all the phases were present on the MCB output. Took the MCB out, everything was clean and happy, tested the breaker and it worked, replaced the breaker and then the inverter had all the phases back again as it should have. There was nothing obviously wrong with the MCB but it was definitely the spring clip things at fault.
Because synchronous motor control is such a common thing these days a great many general purpose micro-controllers can do the job. The capability is built into the PWM and ADC modules. The cleverness is being able to configure them to be synchronised to each other and sample currents and voltage at precise times.
I think you might have a dual motor drive board. The outer doors being separate from the inner doors. If one set of doors fail, the others can still close. Always hated elevators, but got stuck with them in our factory for 14 years. I redrew all the schematics into standard 'electrician's' type. I don't know why AC, Elevator, Automobile and Phone/computer all have to insist on their own proprietary schematics, but use the same parts. WHY CAN'T WE ALL JUST GET ALONG!!!!!!!!!!
because corporations have "sustainable income" and "diversity" quotas to meet, while the CEO's enjoy watching peasants do grunt work in the heat waves that simple cheap robots could've been doing.
At least for HVAC, the schematics have to resemble/model the real wire runs and connections that a technician would see... Computer schematics I never recall as being nonstandard?
The big elevator companies do this on purpose. Without a handbook for their random abbreviations and hidden functions you are pretty much lost. This helps them to keep their expensive maintenance contracts going. There are many small companies offering the same or better service at lower cost. But sadly there is just so much you can do when a fault occurs in these "locked" systems.
I've actually drawn dual sets of schematics, one showing the logical connections and the second showing physical layout, including terminal blocks, node numbers, etc. Both are extremely useful
Infineon does make a lot of very nice microcontrollers with integrated motor controls and FET / IGBT driver logic. It's likely the IGBT module is also an Infineon device and it and the micro designed to work well together. Might even be a reference design available :)
Bahco isn't Swedish these days. Bought by SnapOn if my memory serves and production moved from Sweden (initially Spain but there's a vague voice in my head saying that they moved on since to further reduce the payment to workers).
I didn't know Infineon made MCUs! I've used some of their H-bridges. I think that IGBT package is a combination of individual IGBT transistors, because of its designation of Q8, instead of U8.
I found it moderately funny to look at the struggle to get the circuit board loose from its base fixtures, where I thought there was still a screw or 2 in the center of the board where the L2 and L3 components are somewhat hidden in shadow from C29, C30, C33, and C34. Interesting stuff, indeed.
That was fascinating, thank you Clive. I've always been kinda afraid of lifts, even though I know the amount of safety interlocks they have, I used to always take the stairs, although now I'm a wheelchair user that's not really an option :p Your information about how they work over the years, even just the stuff in passing, has made me feel a little more comfortable using them :) Also, what's the death beams thing about with the capacitors? I saw that one you did on the mobile one, but don't remember it saying anything about capacitors... I'm just getting it's another squirrel-y video that came out?
The lift to my sister in laws apartment in Turkey had a lift with outer doors only and no inner door. After a few years they retro fitted some inner doors.
I'm with you on mechanical lifts, nothing like the sound of contactors going in and out when it operates One of the lifts where i worked was upgraded to electronic, though it retained contactors (although replaced with new ones), so we still had the sound I was trained to rellease passengers from lifts and have accompanied lift engineers, it's the safest way to ride on top of the car.
A friend works for Otis..20 yrs.. he makes excellent money.. I see this stuff all over his garage..At 215k a year a service fuel included and you get to use it personal use.. I'd shave...
Exactly the same processor is used by certain german carwash manufacturer. It seems very capable beeing able to control washtunnels and every other carwash type there is. Washtunnels have tens of lightbarriers, proximity switches, valves, motors, VFD-s and so on. There might be several cars inside at the same time. So lots of controlling to do at the same time by one processor.
If I'm not mistaken, the infineon chip is an fpga. Think of it as a programmable logic array specializing in one task, in this case, elevator operation. It's different from a general purpose cpu found on arduinos and raspberry pis
I bought one of those IGBT modules once for a project that never happened, but it had a copper heatsink pad, and i did have to design my own part in EAGLE to make a board and thats as far as i got. I believe i still have the module somewhere.
Otis was the first company to produce "safety" lifts which were still safe even when the cable broke. Mr Otis put on a show in an exhibition hall: in front of the public he would be hauled up, then the cable was cut and the lift cabin fell briefly and stopped. That was the foundation of their fortune. The cabin runs on rails in the shaft, and there are brakes against the rails. The clever part is that the brakes are naturally ON, and they are connected to the suspension point of the cabin so that the tension in the cable pulls them into the OFF state. If the cable breaks the brakes immediately come on.
Tbh he should've stood in the elevator.
It's scary. But damn it would show people they got balls.
Recommend Tim Hunkin's "The Secret Life of the Lift", available on TH-cam (now at a sharper resolution, and with an epilogue)
@@RedRingOfDead - Er, that's exactly what he did.
@@RFC-3514 i think he meant standing in the elevator shaft
@@RedRingOfDead What part of "he would be hauled up and the cable cut" did you fail to understand?
I'm in the elevator industry, and have worked for Otis. If you hold down one of the Otis guys that I worked with and tickle them until they cry uncle, they will confess that the best elevators are Mitsubishi. I heard my boss say the same the other day, and he's a born and bred Kone guy. One example of the different philosophies: Mitsubishi Japan decides what capacity you have to install new elevators in your area, and if you sell more elevators than that in a year, they say "Sorry, nope, we're not supplying you because you're gonna rush the job!" Otis is definitely good kit, especially the German stuff, but a lot of the Gen 2 stuff come from China from factories that Otis only own 35% of, and any independent elevator supplier can walk in there and buy Otis kit with their own brand name on. The standard blue SVT service tool is plugged in to do parameters on the standard software. Crazy!
Those two empty footprints next to the WAGO connector at 9:52 are for a LEM current sensor module (CAS 6-NP type) to monitor motor current and the IGBT module is most likely a Mitsubishi PS21563. How I know this is because I work with servicing these type of things and are also used in ThyssenKrupp Universal Door Operator 6300PAx series. I have seen that Infineon processor used in a Schindler (LUST LTI) IDD door operator too. Although this particular OTIS drive is not familiar to me, here in America a variant of the type uses an STM32F103 for the processor. OTIS Germany makes really good stuff, I serviced an OVF20 lift drive made in Berlin a while back and the lifetime counter showed on the tool that it was in operation for over 30 years and made over 2 million run cycles.
Building I used to live at has the original lift in it, with an install date for the motor, gearbox, cables and car of January 1939. Originally FIAM control gear, later on replaced with Express cam switches, which got a little crotchety with time, as the shaft guides were never correctly aligned all the way in the shaft. Eventually we paid Otis to put in the upgrade kit, which replaced the crotchety tappet switches with a tape and magnets, and put in a new controller.
Otis underquoted, turned out the existing trailing cables did both not have enough cores, but also failed insulation test, not a problem with 110VDC relays and lamps, but a tiny issue with 24VDC running next to 230VAC power for the car lighting and fan. So in went a nice new 60m of flat trailing cable, 50 cores of 1.5mm copper, and it was commissioned. week later the flat cable relaxed, and moved on the bar, catching on the bottom of the car, and slicing through it. Controller did not survive having 230VAC applied to the 24VDC power rail, though the power transformer, the original FIAM one from 1939, was totally happy, seeing as it only had 2A load on the 24VAC winding, while it was rated for 20A, and the 110VAC winding still had a 2A load from either the brake coil or the retiring cam. New controller and cable under warranty, ouch.
There is however a piece of cardboard with a crude drawing I made on it, laminated into the wiring diagram book, with a backing page containing a sign off on the modification by a Pr Eng, where I solved the one issue, that it would trip the breaker for the retiring cam and brake intermittently. Only if you went from top to ground, then called immediately back to top, where it would reliably trip exactly half a floor short. My mod was to move one wire one position, and install one jumper wire, so that the brake coil and the retiring cam each had their own section of the 3 phase breaker, used there with one position being integrated into the safety circuit, like all the other breakers, with a safety link for breaker fail.
Also had to train the techs as to how to use MOV devices, they put 110VAC ones on 230VAC, and they exploded. Had to train them on what those numbers mean, and that using a 275 or 430 MOV on 24VAC, instead of the 33 MOV, was not going to provide any protection. Also correct application of them for the job, as they were not really trained much for retrofits, more dealing with installs of complete kits. If you order a lift complete from any of the manufacturers you get a kit, complete down to the last detail, and containing every single fastener for every part, all bagged and marked on a label which floor they are for, and taped to the dorr for there. Shaft as well, all sections bagged, and fasteners all taped to that section. 10 ton kit of parts for a 6 floor building, excluding the container that had the car and MRL drives.
Also there is a 68mΩ current sense resistor (marked R14 on pcb) at the input of the igbt module
Yeah, definitely looks like that igtb module, doesn't contain any intelligence or isolation, which would explain all the optocouplers on the inputs and outputs
I love these kind of replies.
Good to know, I've made a game of finding STM32F103s in the field. They're *everywhere*.
A nice change to see well designed electronics instead of 'how cheap can we make this?'
Nice 1 Clive
I've been using Wago connectors for decades in many different applications & never had an issue - especially when compared to the competing screw terminals that get over/under tightened etc.
It's the knockoff green Chinese ones that can cause trouble (never have for me though)
Aren't those technically Wago-to-Phoenix couplers?
@@petersage5157 no, they're Wago proprietary connectors.
I've been using screw terminals for decades in many different applications & never had an issue
Screw terminals can be ok if you use ferrules or solid wires. In saying that probably my least preferred termination type.
Oh yes, the serial report is still in use. RS-232/DB9 connector can still be purchased on a new laptop. Dell’s “Rugged” line of laptops have RS-232 as standard, and you can have a second port added, during or after purchase. It’s a slotted module. Good old COM1 port! Super versatile and nice to have without a finicky driver for USB-Serial adapter. COM1 just always works and there’s a ton of legacy equipment out there where RS-232 is the only interface available.
YES! COM1 232 Is still Alive!!!
Yes and also having real serial helps, especially for things that need that proper 12V drive, and which draw power off the status lines as well. FTDI chips do not do that well, and nobody wants to pay the price for the Max232 either, to do the job properly.
Oh nonononono. I'm friggin' DONE adding those damned 323/3232 level shifters to everything I wish to communicate with. I'd much rather install a thousand USB serial drivers (not as if there are more than 3 or 4 different ones, at most, ever, in practice) then bother with the level shifter, the four caps, and a DB-9 ever again.
Yea I build industrial machinery and RS232 is still the more reliable and preferred methode of programming frequency convertrrs and engine controllers.
@@NinoJoel The word you're looking for is "legacy". Same reason you can still buy a 741 opamp, yet no one in their right fucking mind should ever use one in anything designed in the last few DECADES.
That was a nice exploration, and not your average everyday kind of device. Lifts have always fascinated me since as far back as I can remember. And, scared me in equal measure.
I can still remember the little old man who used to operate the lift in one of the big department stores in York - They've certainly come a long way since then! That would have been a crafted job in its time, and now anyone can step into a lift, push a button, and off you go.
As the old man said "It has its ups and downs, but it's a steady job. You meet all sorts in a lift".
Nice pun, "It has its ups and downs" - surely that one was intentional, given that an elevator/lift is the talking point. 😄
@@TrondBørgeKrokli Indeed, and the old man was a bit of a wise guy!
It would have been in the early '60s and about the time I started at school, but it was his regular line, and given the chance he'd either, get that line in, or, something like it. He was a popular chap with parents and adults alike, and given the torterly pace of that lift, it was just about possible to have a short conversation with him, see depending on how many floors you had to go.
@@whitesapphire5865 Nice. Good old quality times. More than enough to make me nostalgic.
When I was in San Francisco around 2014 I met a lift operator in a old records store.
oh hey, another Sapphire o/
I lived in Reading in the 90’s and occasionally I’d see one of their company vans with their office address on the side. Yes, “Otis, Reading” - always raised a smile :-)
"I'd rather have the surprise when I'm expecting it" - best quote
Thank god you made this, I've been trapped in here for so long.
As for the resistors (R1, R2) and the relay: The resistors limit the in-rush current from the mains when you plug it in. Once the bulk of the capacitors are charged the relay kicks in and shorts the resistor. Helps to reduce the stress to all the components on the board AND your fuses dont blow if you switch on the whole system at once, like after a power outage.
I saw something similar done in the Velleman K4040 valve amplifier. It's a neat trick!
Clive I have to say I have been watching you for quite awhile now, I have no understanding really of any of the things you talk about but you make it so lovely to watch. Best tech mumbo jumbo youtuber :)
Date of construction on label says 2007, so the software is most likely the original from factory then. The pcb is also labeled 2007 in silkscreen label and in copper.
Dang, I came here for "Exploding an OTIS door controller" and only later realized that I misread based on the usual channel expectations. Can we please have a follow-up?
I love seeing inside well designed industrial stuff! Controlling power electronics digitally with microcontrollers/DSPs/etc is becoming increasingly common in well-engineered, complex bits of hardware. Gives you a lot of flexibility and configurability of control algorithms (since it's all done in firmware) - plus you get street cred for doing things the hard way ;) Of course it also introduces the blessing and the curse of being able to "fix" things in software. In the last few years, with specialty power designs, I usually look for microcontrollers rather than dedicated power chips.
4:30 Canadian here, the Wago thing is just north-american electricians getting mad that you insult our oh so beloved wire nuts. Even though Wago style connectors have been ratified into code for permanent installations. It's mostly the electricians over say, 30-35 that decided they were done learning after they got their journeyman cert. all of the failures i've seen, weren't from material failure but rather installation mishaps or perhaps carelessness.
In Finland I usually see either Otis or Kone elevators. The Group Control computers of Kone have a AMD Geode processor (from the early 2000) still in the new models. While the motherboard has had some changes in the last 20 years due to component availability it still looks pretty much the same. Obviously much more reliable than current consumer tech.
In the Netherlands you see Schindler quite often.
"much more reliable than current consumer tech" seems less likely than "cheaper to use the existing design and software until it cannot be sourced"
@@alexbrown1050 new stuff is cheaper altho you are correct moving to new stuff would cost a lot as well. However current motherboards for example can't be expected to survive reliably for more than 5 years while the 20 year old motherboards are still going.
Mostly this is due to electrolytic condensators and can be worked around with proper components. Higher frequencies cause their own issues with lifetime as well.
One of the main issues is much thinner tracks which are more suspectible to corrosion. Greater performance has it's downsides.
I think the shaving thing was more kind of a prevention of getting caught by any rotating mechanisms which elevators have plenty of. Just like with long hair things can go bad and ugly sometimes.
@@GeneCash Jesus… must have been a traumatizing experience.
Looks like R112 and R113 discharge the extra-spicy capacitors. Only thing better than a good MELF is two good MELFs! Properly engineered with adequate voltage rating across the resistors, low impedance copper pours with good separation, teeny tiny trace buggering off to the discharge rail. From the via in the pour between the two resistors, it looks like they're also providing an elevated voltage reference for something or other.
I love that the brain box has an "ESD sensitive - donut touch" warning on the PCB and Clive keeps poking it with his finger.
That's a thing of beauty right there Clive.
A beautiful circuit board.
Otis Elevator manufactured the first major frame assemblies and several other sub assemblies for the Brunswick Blake Collender company. Which would eventually become Brunswick Bowling. An industry leader in automatic pinsetting equipment for over 65 years!
Seems like your typical posting schedule is exactly when I'm looking to sit down and wind down for the night. These videos are like therapy for me!
A good friend of mine is an elevator tech here in the city. Their field has got to the most in-demand out of all the trades right now, personally I think it's overlooked. Overtime gets thrown at them like crazy, not enough guys to handle the load. Can be a lot of work with all the troubleshooting.
So there are currently not enough of them to handle the load? Maybe they need to rAMP up their recruitment?
I have always preferred Schindler to Otis for the simple reason that there is something special about getting into Schindler's Lift and being reminded of one of the greatest movies of all time.
Bad joke.
Personally I've always been averse to being Shindled.
That's some beefy components there, kind of reminds me of the old door controllers we used to use on the buses that had a similar programming system and options using a PC and link up system.
Still use those everyday but with a printer cable port or RJ45 port now. Vapor doors I think.
I've heard that spring terminals can better than screw terminals in the long run.
When temperature changes, materials expand and contract. That can cause a screw terminals to loosen over time. While spring terminals remain springy.
The service tool exists in two variations. Black and blue. The blue tool has a connection counter that goes down every time it's connected to a unit, and has to be above 0 to allow to perform certain actions on new systems.
The black tool does not have this but it is used on older control equipment. There also exist a interface for connecting to a computer, but those are used in very niche cases.
Otis - "shave your beard if you want a job"
Clive - "bye"
😂😂😂
I was doubting non-screw connectors until I met a retired guy, who was once building test gear for ABB circuit breakers. He told me they had such connotations in their 10kA test station w/o issues. Since then I stopped using the screwed ones.
To the MCU: even a STM32F103 from a bluepill board can generate a 3-phase PWM signal with dead time by using a timer and a DMA channel. As long as the generated AC frequency needs no change, the CPU needs to do exactly zero.
And the IGBT-module contains 3 half bridges and current sense. It’s capable of generating 400V 3-ohase AC up to ~5..10A. Quiet powerful, but remember the 50€ Chinese variable speed drive units. They are based on such modules.
The SMPS with the vipers are good stuff, if you protect them against surges from the grid. But as you can generate revenue/GDP worth a new dishwasher/washing machine, just by saving 50cts of components, they do it. And the vipers/TYnnn‘s/… occasionally disclose their magic smoke finally.
You mentioned that the doors needed plenty of power for weights and friction and such, but left out that they also need enough power to crush baby's heads. Hence the large capacitors to provide that power reserve.
And the ability to slam shut from open to increase momentum.
That Otis Blue Box is so rare. Been looking for one for years as I have a Otis 411 lift I look after and we have to program it by hand :(
Otis make some of the best gear out there with safety.
Psst! eBay and aliexpress.
@@bigclivedotcom hehehe spanks :-)
You need to make T shirts and mugs that say "one moment please .."
I like how even though you have no use for that device, you still don't want to damage it, like the respect someone gives to fine artwork. Also, why no circuit diagram??? ; p
Circuit diagram? One moment, please.
@5:25 - Big Clive - "I used to work in elevators"
Me: - I'll bet that was a crowded workspace!
I'll get the lights.
Good connector, on a device operating since 2007 and only responded to a command 1000 times! (open 1000, close 1000,up or down, open 1000,close 1000)...
I had the pleasure for the first time this year to work with Germans at Siemens. I loved every minute. They knew what I was thinking before I had to explain. It was an utter pleasure. What was even more of a pleasure was they laughed their tits off when I mentioned Boris. I think that says a lot.
13:46 I never regret watching because of gems like this.
The board says 2007 in a few places.
So it seems that the software version might be the original one, or close to that, which is impressive.
Excellent bits of kit thanks Clive
BigClive at 1.5 speed is hilarious. I was previously listening to David Vizard Performance on YT who speaks slooooooooowly and forgot to change it. Oh well, back to the start.
Industrial stuff like this is such a nice change sometimes from the typical consumer Chinese stuff that we all tear apart / fix etc. so refreshing.
Ridiculing the "very very expert" 👍👏
I love the comment to "Benno" - nice touch!
The "final" test. Aptly identified. Clive a much more learned and brave soul than most. :-) 13:30
Yes still using this tool even with brand new ones
I can't wait for the part 2 with the schematick.
Hi Clive, there is a resistor next to that blue capacitor(?), with "68 mOhm F" written on it. 20:00 Any idea what the "F" designates?
@@Okurka. Of course. 🤦♂
I even knew this but my brain failed to work for some reason. Thanks!
"The final test.." Made me laugh more than it should have when he touched the contacts for the capacitors.
Nice bit of kit.
I can relate to your feelings about electronics vs electromechanical. In my early days the pinnacle of what I worked with were the Ampex VPR2 1” C format VT machines. Loved those. Really well engineered precise mechanisms along with elegant electronics that you could actually repair down to component level. Things started going to pot when the cassette formats like Betacam started to appear. The mechanics changed from precision machining to bent tin and the electronics got more and more complex, in part to allow for/compensate for the shite mechanicals. The final straw was when they moved to storing the signals on tape in digits rather than analogue. By that time, there was nothing field repairable left in the darn things. Pretty much reduced to board swapping. The machines still cost the same as an average house, but they didn’t look like very good value! Ah, nostalgia…. :)
Same story with everything else, everywhere. All gravy and no meat.
A building I worked in for a while had a reliable but very slow old Otis elevator. Someone (necessarily clumsily) altered the name OTIS above the controls to OTIOSE…
Was it really that useless?
Perhaps the Otisian anti-beard thing was safety-related? Insurer perhaps disliked the prospect of a beard getting caught in the works?
I just found this on reddit, love it when you talk lifts fella!
Thanks Clive I found this very interesting and uplifting.
I read the title of this as "EXPLODING an Otis door controller", and was waiting for the BANG trough the whole video! 🤣
Ah the bahco sl25 set ... Love em.
Twisted pair RS485 comms cable, used by these terminals, is reliable and cost effective. If you used wiring such as ethernet then the surge spikes surrounding elevator electrical noise will make higher speed comms problematic. The cable quality would have to increase, it would need more shielding and would be stiffer and less flexible to install. All the practicalities of the hard real world conditions are in this design.
IMHO, you are looking at the best quality of industrial designs.
OTIS and elevator techs .... I see the spark
Still use for the newest stuff in NA. Module 1 is ocss, 2 is mcss and 4 is dbss. The whole Elevator. They only show when communicating.
We call that a glide controller.
In early 2000 I was a computer consultant on a project at UPS. They had a requirement that all male employees be clean shaven unless they had a religious reason for having facial hair. I have a beard, and did not shave it because I was not an employee. So, I can totally see Otis being the same way.
Great case of "--You cant do that! Who's your supervisor!? --Eh, myself, I'm on contract"
UPS? My mailman got a beard
I remember recruiting an engineer only to be told by the boss “How can you be so stupid to hire someone who is too lazy to shave !” That was in the eighties though. It’s hard to imagine such nonsense would still prevail.
@@anonymous.youtuber It depends. If the beard is well kept, that's one thing. But there's also the "I haven't bothered to shave, trim, or anything else". Don Johnson being the poster child.
Used to be in construction and was always envious of the Otis guys. Always sitting around aha. I will admit the weights they use to calibrate or set the elevators with are heavy though. They were always the same guys on the job, and important enough that they were “allowed” to smoke in the shaft even 420 at times.
One time my tile company broke a panel of glass beside an escalator that was turned off for the night, covering it completely. Mr Otis was not happy, not for the mess, but because it was off and we didn’t destroy it.
Was it Otis that built the elevator test shaft in Germany and had the grace to put an observation tower at the top? One of the companies did, and the building is tastefully done on the outside.
that's ThyssenKrupp if I'm not mixing things up
Didn’t Tom Scott do a video on that place, the video was more on the testing of new technologies, and the test shaft than a viewing platform though.
@@almostanengineer yes, but I think back then the technical stuff was more interesting. They remodeled the thing and now it genuinely looks like a wizards tower.
infineon is relatively well known within the pc component industry as vrm (voltage regulator module)/pwm controller manufacturer, also makes mosfets to go with the vrm controllers, would make sense they also makes a processor like the one used in this door controller
I used to be quite suspicious of Wagos and similar (from an early bad experience with the terminals on Grundfos pumps) but love them now and wish the 221 style were available in larger sizes up to maybe 25mm2. It seems silly that they do DIN rail mount large ones but not inline ones.
Seen so many melted cables and burned devices due to poor terminations in cage clamps.
In the other hand, I was trying to deal with a faulty solar installation recently where it turned out the clips in a Schneider KQ board weren't clamping in properly. It had been installed correctly, worked correctly for a while then just decided that it didn't want to clamp in properly. Not all the phases were present on the MCB output. Took the MCB out, everything was clean and happy, tested the breaker and it worked, replaced the breaker and then the inverter had all the phases back again as it should have. There was nothing obviously wrong with the MCB but it was definitely the spring clip things at fault.
The building I look after used to have several old electro mechanical lifts, the control gear was amazing I will try and find a photo
Does the little symbol by the infineon chip mean 'keep your pokey fingers off'?
Wow, remove the lid and what a compacted PCB unit. Wonderful and powerful engineering at its best, defiantly an open and shut case by OTIS lifts.
Because synchronous motor control is such a common thing these days a great many general purpose micro-controllers can do the job. The capability is built into the PWM and ADC modules. The cleverness is being able to configure them to be synchronised to each other and sample currents and voltage at precise times.
No wonder the capacitors were discharged. You had it upside down and all the electrons fell out.
I work as customer service for Thyssen Krupp US and I find this stuff SUPER interesting...
I honestly would not want Beno anywhere near lifts paid or not, you wont have to dig much to find out why. More death beams please 👍
I think you might have a dual motor drive board. The outer doors being separate from the inner doors. If one set of doors fail, the others can still close. Always hated elevators, but got stuck with them in our factory for 14 years. I redrew all the schematics into standard 'electrician's' type. I don't know why AC, Elevator, Automobile and Phone/computer all have to insist on their own proprietary schematics, but use the same parts. WHY CAN'T WE ALL JUST GET ALONG!!!!!!!!!!
because corporations have "sustainable income" and "diversity" quotas to meet, while the CEO's enjoy watching peasants do grunt work in the heat waves that simple cheap robots could've been doing.
At least for HVAC, the schematics have to resemble/model the real wire runs and connections that a technician would see... Computer schematics I never recall as being nonstandard?
The big elevator companies do this on purpose. Without a handbook for their random abbreviations and hidden functions you are pretty much lost.
This helps them to keep their expensive maintenance contracts going. There are many small companies offering the same or better service at lower cost. But sadly there is just so much you can do when a fault occurs in these "locked" systems.
I've actually drawn dual sets of schematics, one showing the logical connections and the second showing physical layout, including terminal blocks, node numbers, etc.
Both are extremely useful
Infineon does make a lot of very nice microcontrollers with integrated motor controls and FET / IGBT driver logic. It's likely the IGBT module is also an Infineon device and it and the micro designed to work well together. Might even be a reference design available :)
"sorry for the noises" dude half of us are here because it's like electricity asmr
That SVT is still in wide use. It can used on anything from gamma D all the way to elevonic RMH that was turned over yesterday.
I don’t think I’ve ever laughed so hard about babies being crushed between doors.
How are you not at 1M subscribers yet?!
q8 is a dedicated ipm module that makes the 3 phases UVW.
I'd rather have the surprise when I'm expecting it ..... just magical
Bahco isn't Swedish these days. Bought by SnapOn if my memory serves and production moved from Sweden (initially Spain but there's a vague voice in my head saying that they moved on since to further reduce the payment to workers).
Fascinating, as usual
The PCB itself was made by Würth Elektronik. Fantistic quality. (I've used them for my own desings at work).
I thought that looked like their logo on the back.
I didn't know Infineon made MCUs! I've used some of their H-bridges.
I think that IGBT package is a combination of individual IGBT transistors, because of its designation of Q8, instead of U8.
I found it moderately funny to look at the struggle to get the circuit board loose from its base fixtures, where I thought there was still a screw or 2 in the center of the board where the L2 and L3 components are somewhat hidden in shadow from C29, C30, C33, and C34. Interesting stuff, indeed.
Oh, and yeah, the expression *"Spicy mama voltages"* must be the funniest expression I have heard this week.
IGBT module looks like ffrom Motion SPM 3 family, comes with integrated drivers, but not sense resistors.
yeh, the pins that are close are for the low side drive and the pins there are separated in pairs are supply and input for the high side drive
That was fascinating, thank you Clive. I've always been kinda afraid of lifts, even though I know the amount of safety interlocks they have, I used to always take the stairs, although now I'm a wheelchair user that's not really an option :p Your information about how they work over the years, even just the stuff in passing, has made me feel a little more comfortable using them :)
Also, what's the death beams thing about with the capacitors? I saw that one you did on the mobile one, but don't remember it saying anything about capacitors... I'm just getting it's another squirrel-y video that came out?
The lift to my sister in laws apartment in Turkey had a lift with outer doors only and no inner door. After a few years they retro fitted some inner doors.
EGads, a well designed circuit board, I was starting to fear that they didnt exist anymore!!
I'm with you on mechanical lifts, nothing like the sound of contactors going in and out when it operates
One of the lifts where i worked was upgraded to electronic, though it retained contactors (although replaced with new ones), so we still had the sound
I was trained to rellease passengers from lifts and have accompanied lift engineers, it's the safest way to ride on top of the car.
Riding on top of a lift car can be necessary but it can only be described as safe when compared with sticking your fingers in an alligator's eyes.
I used to work in elevators.
Then I got my own office.
😎😁
I would love to see you tear apart more door controllers and remotes. How about a LiftMaster garage door opener? :)
A friend works for Otis..20 yrs.. he makes excellent money.. I see this stuff all over his garage..At 215k a year a service fuel included and you get to use it personal use.. I'd shave...
Exactly the same processor is used by certain german carwash manufacturer. It seems very capable beeing able to control washtunnels and every other carwash type there is. Washtunnels have tens of lightbarriers, proximity switches, valves, motors, VFD-s and so on. There might be several cars inside at the same time. So lots of controlling to do at the same time by one processor.
If I'm not mistaken, the infineon chip is an fpga. Think of it as a programmable logic array specializing in one task, in this case, elevator operation. It's different from a general purpose cpu found on arduinos and raspberry pis
It’s MiSTer for lifts!
Could be, or just a micro
If Otis is the Fluke in the Elevator world, what would Thyssen be? Rohde & Schwarz?
The two resistors are, most likely, pre charge of the dc bus caps.
I really enjoy elevator items
Fun fact:
0:00 Visible Sicker: "DATE OF CONSTR.: 2007"
2:47 "I don't know how old this unti is…"
¯\_(ツ)_/¯
@13:42 Hey Clive, did you see the 80's horror movie 'The Lift'?
It had a very low budget feel.
I bought one of those IGBT modules once for a project that never happened, but it had a copper heatsink pad, and i did have to design my own part in EAGLE to make a board and thats as far as i got. I believe i still have the module somewhere.