Great video! Based on this video I did a HACK by cutting a hole in the top to access the pressure adjustment screw and now we have the proper amount of water (about 20% more than factory) in our machine 🙂
Keep in mind that modern front loaders deliberately use very little water at the start of the cycle as they only intend to make the clothing wet. That increases the activity of the detergent and also increases the friction as the drum tumbles the saturated items so they rub against each other as opposed to just floating about in water. After that the machine uses much more water in the rinse stages. The machine will initially keep topping up the water and tumbling the clothes until it detects that no more water is being absorbed by them.
Did you turn it clockwise and if so, how much? Trying to figure the same thing out myself after seeing endless videos of other people adjusting switches where the set screw is accessible.
Before Big Clive (BBC?) I couldn't have even imagined being excited by the prospect of watching a video called "Inside a washing machine water level transducer. (schematic)" :-)
I have no technical know-how, nor can I do pretty much any of this stuff, yet here I am, on a Sunday morning enjoying to learn how things work. The only problem I have with these videos of Clive's is that I wish I could actually use a soldering iron and do all the things it would enable. You Sir are fighing neck and neck with Richard Patterson for the "Best Scot in the world"
Soldering is piss easy. Just get a cheapy (but safe) soldering iron, some solder, desoldering wick and get some old device with circuitry and desolder stuff. When you have burned stuff out and ripped enough tracks off of a PCB, you'll learn how much heat can be applied and for how long. Soldering components to a PCB will be childs play compared to that - de-soldering can be a proper pain. You literally do what Clive demonstrates when assembling something, heat the pad and the components lead with the tip of the iron, simultaneously add solder and it'll flow, not much solder, just a bit till the pad and the lead are in unison with no gaps. Practice, it's not hard.
Buy yourself inexpensive kits for practise. Use leaded solder. Don't buy the cheapest iron you can find, take a branded one. Cut copper wire to hand lenght, lay them together in a right angle and solder all joints for practise untill you get a nice mesh with nice looking, shiny soldering joints.
Get yourself an okay soldering iron (Clive has a couple reviews on cheap ones) and some cheap kit off eBay! You'll be surprised how easy it is to solder through hole components! If you're afraid of messing up, you can just practice on some random dead electronic device you have at home. De-soldering components is a major pain (for me), but you'll get a feel for it. I started off with random things and my first actual kit was the clock kit BigClive made a video on (and then the astable multivibrator), you can "build a long". It's really great when you build something and it actually works. After getting started, you can attempt to troubleshoot and fix random things that break. I must warn you though, when you start being able to fix random things, your wallet will suffer. You'll start wanting to stock some components so you can fix things right away, then a nicer multimeter, then a nicer soldering station... Then you need a little electronics storage cabinet but half of the drawers will be empty, so you buy more components to put in them. Oh and leds, so many leds! Go for it. It's fun, and eBay stuff is cheap!
I have absolutely no idea what almost anything that was said in this video means, nor do I really desire to know what any of it means. And yet I watched the whole thing. Big Clive is really something to behold.
This is why I'm here. Love this stuff! I had no idea something so simple was being done like that. I'd love to see the bit of programming that is being used to convert the pulse frequency to water level.
I would love to talk the guy that came up with this idea and listening to his thought process that lead to coming up with this idea. Impressive and unique use for that chip.
I expect the man was looking at the ready-made MEMS sensor on his desk he couldn't use for cost reasons and shaking his head. ;-) It's not all fun and games you know. Figuring out that circuit is only a tiny part of the process and by the time you have sat through months of meetings on the more mundane aspects of this sensor you are really looking forward to working on something different.
Oh I am so glad I saw this video. This is exactly the type of housing over the water level swtich on my machine. I hope that plastic screw works exactly like the older models. Thus, turning the screw will allow my machine to fill with more water. I'll soon find out. I'm sick of smelly clothes!!! I'm not going bother prying it open. I'll just use a heated knife and burn/cut a hole on top to access the screw.
The modern machines only make the clothing damp for the first part of the wash to concentrate the detergent and friction between the clothing. If you adjust it, it will probably just compensate for it, as it is monitoring how much water the clothes absorb.
@@bigclivedotcom I'll see. I already exposed the screw. In the older models, turning the screw modified the pressure valve/sensor- so more water was allowed into the drum. These machines are useless, they put a small amount of water- enough to dampen the clothes and then they sit there closed up for 2 hours- smelling like mold.
Jammit Timmaj This is not an NTSC colorburst crystal. Those run at 3.579545 MHz. The item in question was specifically designed for digital clock circuits. Virtually EVERY electronic clock/watch circuit uses this type crystal. They are practically all the same 16.384 KHz frequency, and in a similar tubular metal package. Look them up! Some colorburst trivia: IBM used a 14.31818 KHz crystal in the original PCs. Why such weird value? Divide by 4 and you get the 3.58 MHz signal for the analog CGA video output, AND divide by 3 to get the famous 4.77 MHz which underclocked the 5 MHz CPU, OR divide by 2 to get the odd 7.15909 MHz which they underclocked the 8 MHz CPU in the PC/XT. IBM got cheap, and reused the clock circuit outputs. That's where the old weird CPU clocks came from. Hardware hackers of the day replaced (soldered) their crystals to be amongst the first overclockers and get their best CPU performance.
After seeing your video I wanted to do the same with my sensor. Looks nearly identical. However, the LED was constantly on. And I laughed after opening the case. From the outside there are three contacts: S+, GND, S-. But inside there were no electronics. The winding is connected directly to S+ and S-. Only a small capacitor goes from S+ to GND inside. The manufacturers have saved even more costs and moved the IC to the mainboard. 😂
Interesting, i notice modern washers seem to run the empty pump and pause just before they begin a wash. Could this be to make sure the drum is empty and then perform a calibration/zero on the level gauge?
True, and same for dishwashers. To get a higher water level for rinsing they allow the water to 'run on' some after the water level switch is closed. However this doesn't work so well if the supply flow is constantly changing due to other large water consumers eg baths and showers, hence the improved sensor. If you take the old 2 level sensors (the second level contact is to protect from overfilling!) apart for 'investigation', they are the very devil to recalibrate properly, just don't ask me how I know this... >:-|
They're called "Linear Pressure Switches". Must admit, I thought the 'Out' might , have been serial data , but it's way more basic than that! Thanks Clive very interesting.
I'm kinda glad that the new wave of washing machines are taking this approach,as opposed to the really expensive timed device. It means now we have what is essentially simple microprocessor with "simple" software reading simplified sensors. A person with brains could now set about and replace all the gubbins with an arduino and get sensors to make a washer work again, albeit without the "finesse" of the research in washing cycles. YMMV, but still, it's a good thing.
The new machine seems to be doing clever things. It fills up with a small amount of water and then tumbles the clothing for a while. If pauses and sees if the water level dropped and if it has then it adds more water. If the water level did not drop then it knows that the clothing is now fully saturated and that results in more efficient detergent, water and heater use. Then for the rinse cycles it uses a lot more cold water.
Yeah I have a horrid Hotpoint washer drier (I never use the drier, slow expensive self condenser)and that does all the fancy self balancing stuff too. It could have been good but it's destroyed the bearings, and fails to saturate clothes properly. I have to put it on a full load for a half load else I end up with tacky clothes. I blame the device you took to bits in this vid, but I dare say that mine is a momentary version, it's really high up, with a really soft silicone hose. And it's running from a timer with a controller, bad combo. I've had it in bits many times replacing rabbit warren split filler hoses and alsorts.
Neffers I was thinking the same, code I'm using to measure rpm from a reed switch would probably work. Arduino controlled washing machine sounds like a nice little challenge. Could get really fancy and have Bluetooth connection to smartphone or maybe thats a little to far.
Imagine being able to hack your oven to give it PID control for temperature fluctuations using an arduino and say goodbye to +/- 20 degree temperature swings.
Watching this video brought back memories which i didn't want to remember. Oh boy, running all around my city repairing WMs was not a fun job. Clients tried their hardest to skimp on payment(or even forgo it completely), boss was quite loud drug addict who thought that a day not spent yelling at employees is a day lived without purpose. Also he demanded that all of us will be available 24/7. Not fun at all.
Nice one. Cheap, reliable design. The winding should be fairly stable and those caps look like C0G. All they need to do is calibrate the various thresholds. They might even get away without calibration, but if you're going to test it you might as well calibrate it. The low frequency is perfect for stuffing into a capture pin on the main processor, which will be running at several MHz giving plenty of "bits" of resolution. The other beauty of this sort of method is it's a doddle to galvanically isolate should the need arise.
Clive, what about putting a small dot onto your bench to give you a reference for where the center of shot is. When you zoom in you would be able to tell where you are. Or just keep doing what you're doing. Great video!
A little suggestiong if I may Clive. You say you keep your house cool, so I wonder if it may help to have plastic devices you wish to dismantle near a gentle heat source for a while prior to prying? Or at the very least use your solder station's heat gun to gently warm things up a tad.
If a brit says he keeps his house cool, that's near freezing for the rest of us :) They consider 18C a "heat wave" in the summer LOL. I am currently setting the A/C on 30C to get the room to 24, cause it's cold AF.
I seem to remember Clive mentioning that his home was a "toasty" 13C . . . . . Sorry, but our idea of "toasty" is somewhere warmer than 30C! Whilst we regard "hot" as 40C+ - I suspect Clive regards 25C as "quite THE Heatwave!" :-D
That was one of the reasons I'm glad I left the UK (among other ones that will not be discussed here, because I don't want to piss anyone off). Outside, a chilling 13C with wind and rain, and the trains and my workplace were blasting A/C at full power. Don't know about you, but this is beyond ridiculous. Discussing whether it's warm at 28 or 34 is a thing, but blasting AC full-power at 13C is a sick joke.
I see Clive's place more as a Wallace & Grommit type lab, only with more tartan and a robot butler. He pulls a string above his bed which catapults him to his work station & camera.
Thanks for that Clive, very interesting, I suppose using this method the manufacturer can control and adjust the water level through software rather then mechanical methods depending on the market it is being shipped to.
Hi Clive - do you have an EBay link or anything for that transducer? I'd like to muck around with one as well, but the results I'm getting all look like the mechanical variety.
That decision is made by the control system. These machines initially just make the load damp to maximize detergent strength and friction as the drum rotates. It adds a lot more water for the rinses.
Very interesting little design! Dynamic variable inductance. One IC that one would not have designed like that before the HC version came out. In fact why do they even need the HC version and not just the C version? HC is High speed Cmos and it's really not going all that fast.
C versions? You mean HCT or LS, or not? You don't pick a certain type for speed in a setup like this. You pick it for the logic level voltages and/or whatever logic family can drive the "weird" inductive oscillator.
Right. Could be the rest of the washer is TTL and they want to stick with these "standardized" TTL voltages unlike the cd4060. I've taken apart my share of washing machines and never seen anything this complex used. Last washer I took apart had an extra side tank with a little float that lifted a plunger that turned off the water level off. That was it. No oscillator, no frills. ( I put the motors to good use as buffing/grinding wheels and another as a Wolverine rip-off sharpener for my wood lathe tools. They're super for that & run very quietly.)
Theoretically, yes, the CPU could do a "count the pulse length" and directly control the valve. But that could lead to a flooded room in case of a software bug or (only slightly less bad) a completely overfilled drum which would damage the bearings and possibly overheat the motor. You could do a rising edge-triggered monoflop and something like a couple of xor/and gates - so if the monoflop still outputs a signal when the transducer doesn't, it enables a second monoflop (or capacitor) for a couple of seconds, which in turn enables enables a transistor between the CPU output pin and the water inflow valve. This way, you make sure that the CPU requests more water AND the water level is low enough.
Why this guy eliminated the very good switch for anti-flooding that there was inside the previous pressure switch? No complicated software, no hardware required, only on-off switchs. One for the first level, one for the anti-flooding. Only a 110/230 VAC sensing on one microcontrollor pin.
I guess mine is blown or something? The machine began to just stop, and I know it's an electronic issue, cause it mechanically runs perfectly fine. It just randomly stops, and when it does, it's a PITA to get it back moving. Yesterday it filled up almost to the top and then began full-speed dry-spinning without spitting the water out first. That was a delight to watch, got myself a little fountain here :)
That could be the water sensor not being detected or failed. If used in other modes does the motor run? Another possibility is very worn brushes causing intermittent operation and possibly glitching the processor too.
Out of the question, it runs perfectly, sometimes a full cycle. I figure it must be the small tube going to the sensor or a bad connection or something, I have to open it up. That's the thing, it's not even close to being dead yet, it just randomly stops working once in a while. I already took out the PCB, washed it (it was full of dust) and resoldered all the connections, as some appeared to be bad, but it didn't change much.
About the same look, the DC96-01703G only have the two pF capacitors on the PCB, nothing else. There is a 23ohm resistance between the two signal pins or the module and it does not vary when I move the ferite log. Possibly the reason why my washer does not detect water level and produce an OE code. I suppose the role of the IC as been moved in the main controller board.
Yes, I also took a DC96-01703G apart, and can confirm that there are just 2 capacitors on the board; no IC. And yes, moving the core of the coil in or out will NOT change the resistance; but it WILL affect the inductance of the coil, and the resonance of the tuned circuit.
Hey we use these on our dishwashers Hobart LXe, AM16, CLeR, FT1000E (with dual rinse) That said, ours are true analog. The output is generally 400-2000mV
It's not easy to test it because of the pulsed output. If there's a water level issue, try blowing down the tube in case it's blocked, and if in doubt, get a replacement water level detector.
So if I wanted to make the switch tell the computer to add more water, would I make the spring stronger or add a spacer??? Add resistor or something to slow pulse??
The water change is controlled in software. The modern front loader machines deliberately only add enough water to saturate the clothing in the first part of the cleaning cycle. That way the detergent is most concentrated and there is stronger agitation against the drum and other items as it rotates. More water is used in subsequent rinse cycles.
I've done some work for appliance manufactures and their method is the K.I.S.S. method you'll find in nearly every shop in the world. Simple keeps cost low, complexity low and in effect failure rates low. Using this configuration probably lets them use the same sensor in a sophisticated IoT machine with fully integrated feedback on the HMI and let's them use it in the "old fashioned" simplified models as well. Oddly enough, if you can salvage these sensors they work great for years and cost a fraction of what an industrial robust equivalent would cost. They just have a lot less documentation available to work with which makes integration much more difficult, unless you take one to bits and make a schematic. Ofcourse, that said it does explain why replacement sensors are so expensive to buy new. Robust industrial sensors are very expensive to buy, and they wouldn't want to flood the market putting themselves out of business or putting precision sensor makers out of business with simple cheap devices. There are huge differences in application for both. Thanks for the video Clive.
@@dariovicenzo8139 I know a little about a lot of things. If you want good understanding of sensors, how they work, and what they actually detect I suggest reading Grob and Malvino when you are ready for solid state. They cover the base physics and cover the base chemistry of how devices function and are good first year books for any EET or hobbyist.
Main micro calibrates to sensor at start of cycle? And micro may have a go-no go code if frequency is out of tolerance at start. Hobart LX E undercounter warewashers have up to 3 of these pressure sensors. We can see the values change in real time,in Service mode.
Can the frequency of the output signal be read using an ESP32 module? I have a pump in a rain barrel to water my plants, the sensor could signal the barrel is empty and keep the pump from working.
Hi Big Clive. How do you test this pressure switch to see if it's working. My washing machine doesn't spin. It will fill and empty but refuses to spin. Checked the motor brushes and they are low but look ok, getting continuity from brush connector to brush connector 4ohms approx. Stripped of the hose going to the pressure switch and was free of debris along with all other pipes. I have heard you shouldn't blow into some pressure switches because it can damage them, I think I have the same one as you (linear). Any help would be most appreciated. Thanks in advance!
Nice vid. funnily enough this is called an analog water level sensor or pressostat. The old one which has 2 modes: On / off is a digital pressostat, because on=1/ off=0. It also clicks when you suck on it.
Here is a snippet from a service manual of a frigidaire washer/dryer (laundry center) that touches on the water pressure/level sensor. " 9.11 Pressure Sensor - FFLE3911QW and FFLG4033QW 9.11.1 General Characteristics The Pressure Sensor is an analogue device that controls the water level in the tub, used in models with electronic control system, and it is directly connected to the Main Control Board. A tube is connected between the bottom of the Tub and the Pressure sensor. When the tub is filled with water, it creates a pressure inside the hydraulic circuit that causes the membrane to change position. This in turn modifies the position of the core inside the coil, thus changing the inductance and the frequency of the oscillating circuit. The Main Control Board recognizes how much water has been filled into the tub according to the frequency. Operating frequency varies according to the quantity of water in the tub " Specifications Pressure Frequency Diagram ( At Water Level “0” mm (empty tub) Frequency will be 45.4 Hz ) Mm H20* - Hz* : Points with 100% control of frequency output on assembly line Pressure (mm H2O) Nominal frequency (Hz) Water Level at 0* 45.400* Tub empty, 0 pressure and frequency is 45.4 Hz 150* 43.140* 200* 42.350* 350* 39.990* 500* 37.772* 600 36.560 Tub if full and frequency 36.560. Mm H20* - Hz* : Points with 100% control of frequency output on assembly line
I was once required to repair a relative's washing machine, after it had turned into an oven, turning the clothes inside to carbon. The fault was lime scale that had blocked the air pipe of a similar device to that shown in the video. The effect was that a 2kW water heater had switched on when the drum did not have any water in it. As there wasn't any water to heat the water temperature sensor did not detect the temperature in the mechine until the clothes had burned. I had considered using one of these devices to monitor the water level in a garden pond, however there are all sorts of problems doing this. Not only is the lime scale problem still an issue, but biological active water tends to block the pipe, and oxygen bubbles change the amount of air trapped in the pipe. The volume also changes with temperature. In the washing machine application the fill empty cycle reloads the pipe with air on each cycle. In my pond application, I have considered fitting a plastic bulb to the end of the tube making it a closed water pressure monitor, but this too has its problems. The volume of air changes with ambient temperature and the plastic of the bulb hardens in the active water environment and becomes brittle. I am currently experimenting with a capacitive water level sensor, as this again has no moving parts, the only problem is its concentric sensor pipe dimensions are relatively large to reduce the chance of fouling.
I agree the design was very poor for a very well known brand. This was some time ago, but given the fires being attributed to washing machines and tumble driers today, it can only be assumed that manufactures are not paying sufficient attention to safety and failure mode effect analysis. During my days as an electronics designer, the most serious cardinal sin was to design a product that was unsafe.
Just go with some floating material attached to a potentiometer, like I do in my pond. Yes its a moving part and its prone to wear over the time. But I have mine for more than two years now and it never failed on me. I use it to maintain the water level and turn off the main pump in case of leaks. And its very precise, in a 1700L (2x2 meter) tank I can measure +- 1L. Btw, ultrasonic sensors are very inaccurate. Don't know about capacitive sensors.
VMFRD I have considered this, but it fails if the pond freezes over, and the water level falls under the ice, for example if the ice cracks the side of the pond. I am not a great fan of devices that require moving parts operating in a hostile environment. The potentiometer in a float arrangement can be replaced by an inductor and coil, much like the device in the video. This at least isolates the water from the electronics and unlike a pot, the inductor is not subject to contact wear. FYI. The capacitive sensor uses two concentric tubes open to the air at both ends. The tubes are conductors, covered in a layer of plastic. The concentric conductors act as a capacitor, with the outer plate being connected to zero volts to act as a screen, much like a coax cable. As water fills the tube, it changes the permittivity of the dielectric, replacing air with water. By putting the capacitor into an oscillator circuit, the oscillator frequency falls linearly as the rising water displaces the air in the tube. An alternative circuit, which amounts to the same thing, is to measure the capacitor charge discharge voltage. This can be done very simply using a pic micro. In all cases it is necessary to compensate for stray and fixed offset capacitance. This solution works well as the probe can extend blow any ice level. The only problem is that the tubes need to be of a relatively large diameter to avoid algae blocking the pipes at the water surface. There are few nuances that can be used to improve the arrangement.
Does it really need to be that complicated? I mean, I guess so, as it works, but I feel like this is a bit overkill. I wonder if you could use the moving parts as a surface vibration speaker?
complicated? Sure, the IC is complex, but that is made, in it's millions under very controlled conditions. That means the rest of the device is actually very simple, with just a couple of moving parts, and only one "wearing" part (the diaphragm itself), unlike the old mechanical pressure switches that were levers, pivots, springs, and on which the contacts wore everytime it operated
I'm a appliance repair man. And I can tell you. For the consumers, the low water level is a issue. Would you think adding one or two inches of hose, might increase the water level? I would be a hero if I could add water to the washers fill level.
Hardee prepper The sensor would have to be moved up or down to change the head pressure,or recalibrate the preload springs,as seen in the video. My guess.
I've noticed that my new machine fills with water in bursts and seems to tumble the clothes briefly before sensing again and adding water if the level has gone down. My guess is that it will keep filling and tumbling until the clothes are saturated and can't absorb any more water. That way the minimum water is used to maximise the efficiency of the detergent and heating. Then in the rinse cycles it will add a lot more to remove the loosened dirt.
I once washed my clothes in a HE machines, then washed them again and again and again, after "4 washes" they were almost clean :)) yes, I tend to become "Pig-Pen" at times ( charlie brown cartoon character), little rolling around in grease, oil, road grime, dirt, etc. in the garage or lawn
My machine “calibrates” by filling the drum until the max level and/or overflow sensor trips when the switch is set to a position labeled “super/reset”. Presumably the brain box stores the reading associated with overflow and chooses some intermediate value for the other settings.
The range of 0-0.3m of water quoted on the casing corresponds to about half a psi, which is well within the typical range that people can produce from their lungs. Might be interesting to blow in it if you ever run into another?
It's not easy to test this type without making up a dedicated tester. It will be easier to just buy a replacement and install it. If your machine seems to be using very little water in the main wash cycle then it's normal. It only uses a lot during the rinses. Initially it just aims to get the clothing soaked in a minimal quantity of water to increase the detergent concentration and give better friction when tumbling.
How do they prevent the connector to get clogged up ? There needs to be a small flow of liquid / gas in and out, but it seems as if everything on a washing machine gets clogged up with micro fiber and detergent residues over time... - so why not that small intake for the pressure ?
As an added benefit the CPU can self zero the sensor. When you first start a load the CPU knows the drum is empty. It then samples the sensor and uses the reading to use as a reference for zero. Short term accuracy would be close enough and you don't really need long term accuracy (plus you can self zero again after the dirty, sudsy water is removed and right before it fills again for the first and optional second rinse).
Hi Clive This looks the same as on an Electrolux washer and trying to adjust the water level. My one has no screw on top. Is there any way to adjust this for more water in tub
Modern washing machines will initially just make the clothing damp initially to maximise detergent strength, increase agitation and reduce foaming. The machine usually adds water, tumbles the clothes and then checks to see if the level has gone down. When it detects the level is the same it will stop adding water. You can't make it add more water during the detergent cycle, it only uses more water during rinses.
Just my 5 cents here. On Bosch series 4 ecosilence washer this screw setting can not be altered as firmware is looking for specific frequency - in my case 625Hz for idle and goes down with pressure ( tested with multi-meter) If it senses discrepancy, two things happen, will not stop draining or will add more water ( I was happy to see that ) but if you happen to change program or pause machine - then door lock unlocks even there is full drum with water... Some people over internet reported heating element not working ... So after 3x days playing around, conclusion is, looks like some engineers made sure that consumers do not tinkle with this setting or it is simple firmware glitch where if parameters are not met, then some functions may fail or may work differently from originally programmed...
The screw doesn't really set a liquid level. The machine starts the cycle with a very low amount of water deliberately. It puts water in to a fixed level and tumbles the clothes then looks to see if the level has gone down. If it has, the clothing isn't saturated yet and it adds more water. Once it sees the level remain the same it knows the clothing is fully saturated. That means that during the first part of the wash cycle the detergent is at its strongest and the friction between tumbling clothing is highest, giving a more thorough wash. The rinse cycles use much more water.
I'm wondering why it is necessary to divide the 214kHz signal at the sensor, would that be because a lower frequency is more able to be transmitted to the controller over low grade wires?
I realise pressure is applied to the diaphragm from the water in the in drainage sump. However in my Samsung machine there is a similar device which sits in the sump at the bottom of the drum next to the drainage hose which leads to the pump. As our Samsung is a EcoBubble machine there is very little water in the machine. I think the last time I had to manually drain it due to a blockage in the pump I only counted about 7 litres of water. Which leads me to think this device is calibrated directly in relation to the actual amount of water in the machine rather than just any old number being generated.
No. The machines only look for a change in water height to see if the clothing is still soaking up water. The point is to only make the clothing wet initially so that the detergent is at its most concentrated, and so that the clothes rub together as they are tumbled instead of floating in the water. It results in a much better cleaning action. After the detergent cycle the machine will use a lot more water during rinsing.
It't literally just a transformer and sometimes some fuses too. Usually potted into the case with the terminals and fuses protruding through the potting compound.
That's probably not a Schmidt trigger gate. That would be used for an RC oscillator. For an LC or crystal oscillator, it would be a standard gate forced into its linear region and acting as an amplifier.
Give me an old mechanical washer & drier any day. They are simply the correct technology for the application. Our last pair were Sears Kenmore (Whirlpool) bought c. late 60's and were still going strong when we moved a couple of years ago. Yea, had to do a couple of timers, a door switch or two, a belt, and once a toothpick got jammed in the water pump... but thats about it; any & all parts are reasonable & available. With a manual I did 100% of the repairs. The new washer & drier have already required several service calls, and the 'technicians' were saying (the very expensive) parts are already getting scarce; they seem to require proprietary diagnostic equipment, and the build quality appears abyssal... I could go on, but it gets my dander up. Think we'll be lucky to get another 2 or 3 years out of them.
rwbishop their bread and butter is laundromat machines, but they also make a version with a normal consumer mechanical timer. If you are familiar with the old-style Maytag washers, you will feel right at home.
Thanks will definitely look into them. Not really familiar with Maytag... other than have a neighbor with an old one who swears by it. However, remember seeing/hearing that their newer generation stuff may be problematic.
rwbishop I have a lg front loader, only thing done is bearings , and drain pump,,, drain pump is like a fish tank water pump.. In over ten years... Have the matching dryer, but it is that same as all dryers
could one of these being faulty trip a circuit breaker? ive checked the heating element by disconnecting it, the water drains out of the machine when it starts up,, so not the pump, the water runs into the prewash section of the detergent tray and then once the water for the main section of the wash starts the breaker trips!! ive replaced the water inlet valve and it still trips at this point so not the solenoid in the water inlet valve, next down the line is this bad boy and mine looks just like this one. washer is a bush A127QW. thanks!
@@bigclivedotcom thanks for the reply. ive disconnected the heating element so its not that and ill look at he eclectic filter, i dont know what it is yet but ill check that out thank you. but why would the machine start, run through the prewash section of the cycle then cut out every time right when the solenoid switches to the main wash water inlet, some water comes through but it switches off/trips the breaker within 1 second.
@@bigclivedotcom ok so heres an update, ive swapped out the mains interference electrical filter and things have improved but now its doing something else. the machine now makes it past the prewash and fills the drum with water for the main wash, then it stops seemingly ready to wash the clothes, and then the stop cycle light flashes a bit then the machine drains the water, stopping the cycle :(. man im so close, what do you think?? is this now a motherboard issue? thank you!
That 200R resistor should be on the output of the gate, not the input. Edit: And looking at the PCB the coil appears to be connected directly to pin 11, which is the input to the first gate in the oscillator cct.
That may increase rinse level, but the initial detergent fill will use just enough to the point that the clothes are saturated by detecting a change in the water level.
Yes, I also wanted to know how to test it in the field. Now that I know how it works, I think you might be able to use a multimeter with a frequency function, or use a graphing multi-meter (which I have) and find which terminals (signal + ground) that have a "frequency" on them. First verify that you have approximately 5 VDC going into the device. Now you've identified ground and the Vcc lines. Then, connect your meter to the 3rd terminal and Look for a frequency of around 8 to 16 Hz. Then if you carry around a small piece of tubing with you, you might be able to attach it to the sensor, blow into it gently, and see if the frequency changes. (I have not tried this method yet, since I just saw this video, but I will try it the next time I work on a Samsung washer.)
3:26 How it calibrates: Two opposing springs of equal strength and a cross head threaded insert bonded with anti-tamper marks. Allows for dialing in the 0 position at factory. =^-^=
..... I wonder if there is a program calculation detecting load with this, in conjunction with the 'post spin up' g-sensor input (or whatever they use).
The pre-spin balancing seems to use an algorithm to monitor the feedback from the motors tacho. The speed of a balanced load is consistent, but an unbalanced load wavers significantly on each revolution.
@@bigclivedotcom this is a problem with our machine, its sometimes used just to spin a small but occasionally heavy load, the thing throws a wobbly and just wont spin it, you have to wring it as much as possible by hand before spinning! previous machine was far less of a bitch!
I failed to get if this one just tells on/off (like full, not full) or it shows the level (10%, 50% etc) inside the washing machine? If its level, I am confused how it reads level with that small travel!!!
Keith Neubauer it’s under-engineered. Over engineered would be a ARM based computer reading a load cell or piezoelectric pressure sensor, all tucked into a CNC machined case. From a mechanical standpoint, this device is very similar to the electromechanical water level switches of days gone by. The electronics are very simple; no microcontroller, no ASIC, just a bog standard off the shelf part, a few passives and some wire.
Makes me glad my pre-indesit buyout Hotpoint washer is all switches and dials, seems wrong to slap an IC into the water level switch when a couple of microswitches does the job perfectly well... :) But, as Scotty said, "The more they overthink the plumbing, the easier it is to stop up the drains", overcomplication of simple things is just a wasted effort in my opinion... :)
Switches stick, especially with age.. and that's quite a common failure of the old type "pressostats". This device will never stick in that way. So it's a far better design.
So, let me get this straight, you want to use something which has more moving parts, i.e. up to 3 switches for example, which has terrible resolution (3 positions) over something that potentially can tell the washer EXACTLY how much water is in the drum because of it's near infinite resolution? This switch is a brilliant idea and probably costs a LOT less to make. Switches are expensive because of the exotic materials used on the contacts, such as gold, silver, platinum etc. They tarnish over time, springs become weak, yadda yadda yadda. I know what I'd rather have.
davegsm82 these are very reliable. I’ve never seen one break. The pressure dome/hose clogs with detergent powder at times, but the pressure transducer itself is reliable. How do I know? I’m a WASHING MACHINE ENGINEER. *high fives Chris and Peter*
Interesting design. I assume the cost of the ones with physical switch contacts is now considered prohibitive by the manufacturer. I expect in the really old machines with mechanical controllers needed the contacts to be rated for 250v and they would eventually fail due to contact wear, whereas this design should last indefinitely - or as long as the diaphragm is intact. Clever use of cheap off the shelf components.
The mechanical water level switch contacts didn't fail, they only passed a low current for the switch module's clock. They're less mechanical, more electrical now, a sign of the times.
Advantage of doing it this way is you can have any number of fill levels. With the older diaphragm and microswitch type you need one for every fill level - the one here has two.
Easiest way would be to swap in a new one. Keep in mind that modern machines initially just make the clothing wet with no excess water, to maximise detergent strength and friction between garments with lower risk of excess foam. Subsequent rinse cycles would use a lot more water.
@@bigclivedotcom Less water makes sense. If it fails how would you know if not filling, over filling or not spinning was not due the the pressure switch? I always test them when facing these issues.
my washing machine empties all the water every time I change the program. How can I trick the pressure switch into believing for a few moments before starting the new program that there is no water?
Great video! Based on this video I did a HACK by cutting a hole in the top to access the pressure adjustment screw and now we have the proper amount of water (about 20% more than factory) in our machine 🙂
Keep in mind that modern front loaders deliberately use very little water at the start of the cycle as they only intend to make the clothing wet. That increases the activity of the detergent and also increases the friction as the drum tumbles the saturated items so they rub against each other as opposed to just floating about in water. After that the machine uses much more water in the rinse stages.
The machine will initially keep topping up the water and tumbling the clothes until it detects that no more water is being absorbed by them.
Did you turn it clockwise and if so, how much? Trying to figure the same thing out myself after seeing endless videos of other people adjusting switches where the set screw is accessible.
Before Big Clive (BBC?) I couldn't have even imagined being excited by the prospect of watching a video called "Inside a washing machine water level transducer. (schematic)" :-)
next week 'clive does the washing up'
Same - it's gotta be down to his presentation style!
+Jusb1066 In this video he would share some simple washing-up advice you would never imagine worked and say it's pretty common among his circles
I have no technical know-how, nor can I do pretty much any of this stuff, yet here I am, on a Sunday morning enjoying to learn how things work. The only problem I have with these videos of Clive's is that I wish I could actually use a soldering iron and do all the things it would enable.
You Sir are fighing neck and neck with Richard Patterson for the "Best Scot in the world"
Soldering is piss easy. Just get a cheapy (but safe) soldering iron, some solder, desoldering wick and get some old device with circuitry and desolder stuff. When you have burned stuff out and ripped enough tracks off of a PCB, you'll learn how much heat can be applied and for how long.
Soldering components to a PCB will be childs play compared to that - de-soldering can be a proper pain. You literally do what Clive demonstrates when assembling something, heat the pad and the components lead with the tip of the iron, simultaneously add solder and it'll flow, not much solder, just a bit till the pad and the lead are in unison with no gaps.
Practice, it's not hard.
yep break some junk before you work on fixing stuff
Buy yourself inexpensive kits for practise. Use leaded solder.
Don't buy the cheapest iron you can find, take a branded one.
Cut copper wire to hand lenght, lay them together in a right angle and solder all joints for practise untill you get a nice mesh with nice looking, shiny soldering joints.
Get yourself an okay soldering iron (Clive has a couple reviews on cheap ones) and some cheap kit off eBay! You'll be surprised how easy it is to solder through hole components! If you're afraid of messing up, you can just practice on some random dead electronic device you have at home. De-soldering components is a major pain (for me), but you'll get a feel for it.
I started off with random things and my first actual kit was the clock kit BigClive made a video on (and then the astable multivibrator), you can "build a long". It's really great when you build something and it actually works. After getting started, you can attempt to troubleshoot and fix random things that break.
I must warn you though, when you start being able to fix random things, your wallet will suffer. You'll start wanting to stock some components so you can fix things right away, then a nicer multimeter, then a nicer soldering station... Then you need a little electronics storage cabinet but half of the drawers will be empty, so you buy more components to put in them. Oh and leds, so many leds!
Go for it. It's fun, and eBay stuff is cheap!
It's ok with the cheapy stations on eBay. Clive uses one himself. Just make sure you have a chisel tip rather than the conical.
I have absolutely no idea what almost anything that was said in this video means, nor do I really desire to know what any of it means. And yet I watched the whole thing. Big Clive is really something to behold.
This was one of the most enlightening videos for me on your channel. Thanks, Clive!
This is why I'm here. Love this stuff! I had no idea something so simple was being done like that. I'd love to see the bit of programming that is being used to convert the pulse frequency to water level.
Just count the number of pulses seen over five seconds. Very simple to code.
I would love to talk the guy that came up with this idea and listening to his thought process that lead to coming up with this idea. Impressive and unique use for that chip.
I expect the man was looking at the ready-made MEMS sensor on his desk he couldn't use for cost reasons and shaking his head. ;-)
It's not all fun and games you know. Figuring out that circuit is only a tiny part of the process and by the time you have sat through months of meetings on the more mundane aspects of this sensor you are really looking forward to working on something different.
I love it when you break things open to release the knowledge inside :)~
Just finished fixing my washing machine, this was invaluable during diagnostics. Thanks!
Oh I am so glad I saw this video. This is exactly the type of housing over the water level swtich on my machine. I hope that plastic screw works exactly like the older models. Thus, turning the screw will allow my machine to fill with more water. I'll soon find out. I'm sick of smelly clothes!!! I'm not going bother prying it open. I'll just use a heated knife and burn/cut a hole on top to access the screw.
The modern machines only make the clothing damp for the first part of the wash to concentrate the detergent and friction between the clothing. If you adjust it, it will probably just compensate for it, as it is monitoring how much water the clothes absorb.
@@bigclivedotcom I'll see. I already exposed the screw. In the older models, turning the screw modified the pressure valve/sensor- so more water was allowed into the drum. These machines are useless, they put a small amount of water- enough to dampen the clothes and then they sit there closed up for 2 hours- smelling like mold.
@@gfairbanks1653 Hey! Did you manage to increase water level by adjusting screw or washing machine compensated water level? Thanks!
@@majso-profile yes I'd like to know too
7:39 . . . Little spring: "I'm outta here!!"
Genius design with a variable inductor and divider!
LakeNipissing
I put a capacitor dropper over my tablet speaker and I could hear it gasping for air.
The chip was optimized to produce a stable 1Hz output from a cheap 16.384KHz tuning fork crystal. Very interesting use in your application.
Ah! I never spotted that. How useful!
I think the colorburst crystal required was only found in American TV's but found their way into digital watches.
Jammit Timmaj This is not an NTSC colorburst crystal. Those run at 3.579545 MHz.
The item in question was specifically designed for digital clock circuits. Virtually EVERY electronic clock/watch circuit uses this type crystal. They are practically all the same 16.384 KHz frequency, and in a similar tubular metal package. Look them up!
Some colorburst trivia: IBM used a 14.31818 KHz crystal in the original PCs. Why such weird value? Divide by 4 and you get the 3.58 MHz signal for the analog CGA video output, AND divide by 3 to get the famous 4.77 MHz which underclocked the 5 MHz CPU, OR divide by 2 to get the odd 7.15909 MHz which they underclocked the 8 MHz CPU in the PC/XT. IBM got cheap, and reused the clock circuit outputs. That's where the old weird CPU clocks came from.
Hardware hackers of the day replaced (soldered) their crystals to be amongst the first overclockers and get their best CPU performance.
Watch crystals are 32.768kHz.
Graham Langley Yep. Mentioned above.
Thanks Clive! No high tech CPU sensor here just good old CMOS 4000 logic! The design has elegant simplicity. Cool teardown.
After seeing your video I wanted to do the same with my sensor.
Looks nearly identical. However, the LED was constantly on. And I laughed after opening the case.
From the outside there are three contacts: S+, GND, S-. But inside there were no electronics. The winding is connected directly to S+ and S-. Only a small capacitor goes from S+ to GND inside.
The manufacturers have saved even more costs and moved the IC to the mainboard. 😂
Neat, clever, simple, cheap and reliable. Thanks Clive.
Interesting, i notice modern washers seem to run the empty pump and pause just before they begin a wash. Could this be to make sure the drum is empty and then perform a calibration/zero on the level gauge?
True, and same for dishwashers. To get a higher water level for rinsing they allow the water to 'run on' some after the water level switch is closed.
However this doesn't work so well if the supply flow is constantly changing due to other large water consumers eg baths and showers, hence the improved sensor.
If you take the old 2 level sensors (the second level contact is to protect from overfilling!) apart for 'investigation', they are the very devil to recalibrate properly, just don't ask me how I know this... >:-|
"It can be opened without breaking it."
"Oh! I broke it!"
"This is totally okay."
: D
Sometimes you must break it to release the delicious knowledge inside :)
They're called "Linear Pressure Switches". Must admit, I thought the 'Out' might , have been serial data , but it's way more basic than that! Thanks Clive very interesting.
I'm kinda glad that the new wave of washing machines are taking this approach,as opposed to the really expensive timed device. It means now we have what is essentially simple microprocessor with "simple" software reading simplified sensors. A person with brains could now set about and replace all the gubbins with an arduino and get sensors to make a washer work again, albeit without the "finesse" of the research in washing cycles. YMMV, but still, it's a good thing.
The new machine seems to be doing clever things. It fills up with a small amount of water and then tumbles the clothing for a while. If pauses and sees if the water level dropped and if it has then it adds more water. If the water level did not drop then it knows that the clothing is now fully saturated and that results in more efficient detergent, water and heater use. Then for the rinse cycles it uses a lot more cold water.
yes the old days of 1 or 2 water levels isnt enough, it has to be per load and how absorbent they are for efficiency
Yeah I have a horrid Hotpoint washer drier (I never use the drier, slow expensive self condenser)and that does all the fancy self balancing stuff too. It could have been good but it's destroyed the bearings, and fails to saturate clothes properly. I have to put it on a full load for a half load else I end up with tacky clothes. I blame the device you took to bits in this vid, but I dare say that mine is a momentary version, it's really high up, with a really soft silicone hose. And it's running from a timer with a controller, bad combo. I've had it in bits many times replacing rabbit warren split filler hoses and alsorts.
Neffers I was thinking the same, code I'm using to measure rpm from a reed switch would probably work. Arduino controlled washing machine sounds like a nice little challenge. Could get really fancy and have Bluetooth connection to smartphone or maybe thats a little to far.
Imagine being able to hack your oven to give it PID control for temperature fluctuations using an arduino and say goodbye to +/- 20 degree temperature swings.
Well done Clive as always, very well explained! Thank you.
Thank you for all your effort and time.
Watching this video brought back memories which i didn't want to remember. Oh boy, running all around my city repairing WMs was not a fun job. Clients tried their hardest to skimp on payment(or even forgo it completely), boss was quite loud drug addict who thought that a day not spent yelling at employees is a day lived without purpose. Also he demanded that all of us will be available 24/7. Not fun at all.
Very interesting and unusual compared to the normal pressure switch found in many washing machines.
It’s so fascinating watching your videos 😂 love your content! ✌🏻
Nice one. Cheap, reliable design. The winding should be fairly stable and those caps look like C0G. All they need to do is calibrate the various thresholds. They might even get away without calibration, but if you're going to test it you might as well calibrate it. The low frequency is perfect for stuffing into a capture pin on the main processor, which will be running at several MHz giving plenty of "bits" of resolution. The other beauty of this sort of method is it's a doddle to galvanically isolate should the need arise.
Clive, what about putting a small dot onto your bench to give you a reference for where the center of shot is. When you zoom in you would be able to tell where you are. Or just keep doing what you're doing. Great video!
I usually do. I may have accidentally burnt it off or something.
A little suggestiong if I may Clive. You say you keep your house cool, so I wonder if it may help to have plastic devices you wish to dismantle near a gentle heat source for a while prior to prying? Or at the very least use your solder station's heat gun to gently warm things up a tad.
Remember that warehouse at the end Of Raiders of the Lost Ark?
Well clive has one like that that, but it's all lit by warm white LEDs. :)
If a brit says he keeps his house cool, that's near freezing for the rest of us :)
They consider 18C a "heat wave" in the summer LOL. I am currently setting the A/C on 30C to get the room to 24, cause it's cold AF.
I seem to remember Clive mentioning that his home was a "toasty" 13C . . . . . Sorry, but our idea of "toasty" is somewhere warmer than 30C! Whilst we regard "hot" as 40C+ - I suspect Clive regards 25C as "quite THE Heatwave!" :-D
That was one of the reasons I'm glad I left the UK (among other ones that will not be discussed here, because I don't want to piss anyone off). Outside, a chilling 13C with wind and rain, and the trains and my workplace were blasting A/C at full power.
Don't know about you, but this is beyond ridiculous. Discussing whether it's warm at 28 or 34 is a thing, but blasting AC full-power at 13C is a sick joke.
I see Clive's place more as a Wallace & Grommit type lab, only with more tartan and a robot butler. He pulls a string above his bed which catapults him to his work station & camera.
I watch a lot of your videos on mute so I can work out the mechanisms myself sometimes, thanks to you and Ave I'm pretty good now ;P
Great video. helped me with my mechatronics assignment. Thank you!
Loved the video. Very educational. Thanks
That's really cool - I love good engineering!
Thanks for that Clive, very interesting, I suppose using this method the manufacturer can control and adjust the water level through software rather then mechanical methods depending on the market it is being shipped to.
Hi Clive - do you have an EBay link or anything for that transducer? I'd like to muck around with one as well, but the results I'm getting all look like the mechanical variety.
they have moved on, the one i took apart had a 2 level pressure switch, very simple. but that was an old 70s machine with mechanical cam timer
Very nice and useful. How would I trick it to fill with more water?
That decision is made by the control system. These machines initially just make the load damp to maximize detergent strength and friction as the drum rotates. It adds a lot more water for the rinses.
How is it calibrated??? That's what the springs are for! Neat!!!
Very interesting little design! Dynamic variable inductance. One IC that one would not have designed like that before the HC version came out. In fact why do they even need the HC version and not just the C version? HC is High speed Cmos and it's really not going all that fast.
C versions? You mean HCT or LS, or not? You don't pick a certain type for speed in a setup like this. You pick it for the logic level voltages and/or whatever logic family can drive the "weird" inductive oscillator.
It might be the easiest to source. I was pondering why they didn't use the CD4060 base CMOS version. Maybe this version has more robust outputs.
Right. Could be the rest of the washer is TTL and they want to stick with these "standardized" TTL voltages unlike the cd4060. I've taken apart my share of washing machines and never seen anything this complex used. Last washer I took apart had an extra side tank with a little float that lifted a plunger that turned off the water level off. That was it. No oscillator, no frills. ( I put the motors to good use as buffing/grinding wheels and another as a Wolverine rip-off sharpener for my wood lathe tools. They're super for that & run very quietly.)
The HCMOS version is probably the cheapest version in quantity at this point.
I picked up a few tips and tricks from this clever design. Thanks.
Excellent school physics project, thank you!
Back at it again
Theoretically, yes, the CPU could do a "count the pulse length" and directly control the valve. But that could lead to a flooded room in case of a software bug or (only slightly less bad) a completely overfilled drum which would damage the bearings and possibly overheat the motor.
You could do a rising edge-triggered monoflop and something like a couple of xor/and gates - so if the monoflop still outputs a signal when the transducer doesn't, it enables a second monoflop (or capacitor) for a couple of seconds, which in turn enables enables a transistor between the CPU output pin and the water inflow valve. This way, you make sure that the CPU requests more water AND the water level is low enough.
Why this guy eliminated the very good switch for anti-flooding that there was inside the previous pressure switch? No complicated software, no hardware required, only on-off switchs. One for the first level, one for the anti-flooding. Only a 110/230 VAC sensing on one microcontrollor pin.
Thanks a lot, this video is what i was looking for👍
I guess mine is blown or something? The machine began to just stop, and I know it's an electronic issue, cause it mechanically runs perfectly fine. It just randomly stops, and when it does, it's a PITA to get it back moving. Yesterday it filled up almost to the top and then began full-speed dry-spinning without spitting the water out first. That was a delight to watch, got myself a little fountain here :)
That could be the water sensor not being detected or failed. If used in other modes does the motor run? Another possibility is very worn brushes causing intermittent operation and possibly glitching the processor too.
Out of the question, it runs perfectly, sometimes a full cycle. I figure it must be the small tube going to the sensor or a bad connection or something, I have to open it up.
That's the thing, it's not even close to being dead yet, it just randomly stops working once in a while.
I already took out the PCB, washed it (it was full of dust) and resoldered all the connections, as some appeared to be bad, but it didn't change much.
The hose(s) from the bottom of the tub to the switch(es) often get blocked.
Analog pressure switch, can remember these appearing about 15 years ago on Zanussi machines
So this was the appliance of science then.
Just be thankful Zanussi don't make toilet cisterns.
About the same look, the DC96-01703G only have the two pF capacitors on the PCB, nothing else. There is a 23ohm resistance between the two signal pins or the module and it does not vary when I move the ferite log. Possibly the reason why my washer does not detect water level and produce an OE code. I suppose the role of the IC as been moved in the main controller board.
What you suppose it's corret about the chip.
Yes, I also took a DC96-01703G apart, and can confirm that there are just 2 capacitors on the board; no IC. And yes, moving the core of the coil in or out will NOT change the resistance; but it WILL affect the inductance of the coil, and the resonance of the tuned circuit.
Hey we use these on our dishwashers
Hobart LXe, AM16, CLeR, FT1000E (with dual rinse)
That said, ours are true analog. The output is generally 400-2000mV
Thank you
Your lessons is very very useful
Nice video , so how can I test this pressure switch with multimeter?
It's not easy to test it because of the pulsed output. If there's a water level issue, try blowing down the tube in case it's blocked, and if in doubt, get a replacement water level detector.
So if I wanted to make the switch tell the computer to add more water, would I make the spring stronger or add a spacer??? Add resistor or something to slow pulse??
The water change is controlled in software. The modern front loader machines deliberately only add enough water to saturate the clothing in the first part of the cleaning cycle. That way the detergent is most concentrated and there is stronger agitation against the drum and other items as it rotates.
More water is used in subsequent rinse cycles.
Very good explained, thumbs up!
I've done some work for appliance manufactures and their method is the K.I.S.S. method you'll find in nearly every shop in the world. Simple keeps cost low, complexity low and in effect failure rates low. Using this configuration probably lets them use the same sensor in a sophisticated IoT machine with fully integrated feedback on the HMI and let's them use it in the "old fashioned" simplified models as well. Oddly enough, if you can salvage these sensors they work great for years and cost a fraction of what an industrial robust equivalent would cost. They just have a lot less documentation available to work with which makes integration much more difficult, unless you take one to bits and make a schematic. Ofcourse, that said it does explain why replacement sensors are so expensive to buy new. Robust industrial sensors are very expensive to buy, and they wouldn't want to flood the market putting themselves out of business or putting precision sensor makers out of business with simple cheap devices. There are huge differences in application for both. Thanks for the video Clive.
Hi Jason, do you know somethings about this sensor? Like I have to read the frequency by a micro?
@@dariovicenzo8139 I know a little about a lot of things. If you want good understanding of sensors, how they work, and what they actually detect I suggest reading Grob and Malvino when you are ready for solid state. They cover the base physics and cover the base chemistry of how devices function and are good first year books for any EET or hobbyist.
Main micro calibrates to sensor at start of cycle? And micro may have a go-no go code if frequency is out of tolerance at start. Hobart LX E undercounter warewashers have up to 3 of these pressure sensors. We can see the values change in real time,in Service mode.
Can the frequency of the output signal be read using an ESP32 module? I have a pump in a rain barrel to water my plants, the sensor could signal the barrel is empty and keep the pump from working.
You can get water level sensing relay modules.
Hi Big Clive. How do you test this pressure switch to see if it's working. My washing machine doesn't spin. It will fill and empty but refuses to spin. Checked the motor brushes and they are low but look ok, getting continuity from brush connector to brush connector 4ohms approx. Stripped of the hose going to the pressure switch and was free of debris along with all other pipes. I have heard you shouldn't blow into some pressure switches because it can damage them, I think I have the same one as you (linear). Any help would be most appreciated. Thanks in advance!
Nice vid. funnily enough this is called an analog water level sensor or pressostat. The old one which has 2 modes: On / off is a digital pressostat, because on=1/ off=0. It also clicks when you suck on it.
Here is a snippet from a service manual of a frigidaire washer/dryer (laundry center) that touches on the water pressure/level sensor.
"
9.11 Pressure Sensor
- FFLE3911QW and
FFLG4033QW
9.11.1 General Characteristics
The Pressure Sensor is an analogue device that controls the
water level in the tub, used in models with electronic control
system, and it is directly connected to the Main Control
Board.
A tube is connected between the bottom of the Tub and the
Pressure sensor.
When the tub is filled with water,
it creates a pressure inside the
hydraulic circuit that causes the
membrane to change position. This
in turn modifies the position of the
core inside the coil, thus changing
the inductance and the frequency of
the oscillating circuit.
The Main Control Board recognizes how much water
has been filled into the tub according to the frequency.
Operating frequency varies according to the quantity of
water in the tub
"
Specifications
Pressure Frequency Diagram
( At Water Level “0” mm (empty tub) Frequency will be 45.4 Hz )
Mm H20* - Hz* : Points with 100% control of frequency output on assembly line
Pressure (mm H2O) Nominal frequency (Hz)
Water Level at
0* 45.400* Tub empty, 0 pressure and frequency is 45.4 Hz
150* 43.140*
200* 42.350*
350* 39.990*
500* 37.772*
600 36.560 Tub if full and frequency 36.560.
Mm H20* - Hz* : Points with 100% control of frequency output on assembly line
I was once required to repair a relative's washing machine, after it had turned into an oven, turning the clothes inside to carbon.
The fault was lime scale that had blocked the air pipe of a similar device to that shown in the video. The effect was that a 2kW water heater had switched on when the drum did not have any water in it. As there wasn't any water to heat the water temperature sensor did not detect the temperature in the mechine until the clothes had burned.
I had considered using one of these devices to monitor the water level in a garden pond, however there are all sorts of problems doing this. Not only is the lime scale problem still an issue, but biological active water tends to block the pipe, and oxygen bubbles change the amount of air trapped in the pipe. The volume also changes with temperature. In the washing machine application the fill empty cycle reloads the pipe with air on each cycle. In my pond application, I have considered fitting a plastic bulb to the end of the tube making it a closed water pressure monitor, but this too has its problems. The volume of air changes with ambient temperature and the plastic of the bulb hardens in the active water environment and becomes brittle. I am currently experimenting with a capacitive water level sensor, as this again has no moving parts, the only problem is its concentric sensor pipe dimensions are relatively large to reduce the chance of fouling.
The machine was poorly designed... the element should have had its own overheat thermal cutout on it which would have prevented this from happening.
I agree the design was very poor for a very well known brand. This was some time ago, but given the fires being attributed to washing machines and tumble driers today, it can only be assumed that manufactures are not paying sufficient attention to safety and failure mode effect analysis.
During my days as an electronics designer, the most serious cardinal sin was to design a product that was unsafe.
You can use an ultrasonic distance sensor to measure the water level.
Just go with some floating material attached to a potentiometer, like I do in my pond. Yes its a moving part and its prone to wear over the time. But I have mine for more than two years now and it never failed on me. I use it to maintain the water level and turn off the main pump in case of leaks. And its very precise, in a 1700L (2x2 meter) tank I can measure +- 1L. Btw, ultrasonic sensors are very inaccurate. Don't know about capacitive sensors.
VMFRD I have considered this, but it fails if the pond freezes over, and the water level falls under the ice, for example if the ice cracks the side of the pond. I am not a great fan of devices that require moving parts operating in a hostile environment. The potentiometer in a float arrangement can be replaced by an inductor and coil, much like the device in the video. This at least isolates the water from the electronics and unlike a pot, the inductor is not subject to contact wear.
FYI. The capacitive sensor uses two concentric tubes open to the air at both ends. The tubes are conductors, covered in a layer of plastic. The concentric conductors act as a capacitor, with the outer plate being connected to zero volts to act as a screen, much like a coax cable. As water fills the tube, it changes the permittivity of the dielectric, replacing air with water. By putting the capacitor into an oscillator circuit, the oscillator frequency falls linearly as the rising water displaces the air in the tube. An alternative circuit, which amounts to the same thing, is to measure the capacitor charge discharge voltage. This can be done very simply using a pic micro. In all cases it is necessary to compensate for stray and fixed offset capacitance. This solution works well as the probe can extend blow any ice level. The only problem is that the tubes need to be of a relatively large diameter to avoid algae blocking the pipes at the water surface. There are few nuances that can be used to improve the arrangement.
Thank for sharing. I wounder about its chematics. Could you share the detail schematics? Thanks a lot.
Is there away to test this type of pressure switch with multi
meter without taking apart
Not easily, since it doesn't have a simple switch. It's often easier just to put in a replacement to see if it fixes the problem.
Does it really need to be that complicated? I mean, I guess so, as it works, but I feel like this is a bit overkill. I wonder if you could use the moving parts as a surface vibration speaker?
complicated? Sure, the IC is complex, but that is made, in it's millions under very controlled conditions. That means the rest of the device is actually very simple, with just a couple of moving parts, and only one "wearing" part (the diaphragm itself), unlike the old mechanical pressure switches that were levers, pivots, springs, and on which the contacts wore everytime it operated
Max Torque Well, true, less electrical contact areas to wear out, for sure.
I'm a appliance repair man. And I can tell you. For the consumers, the low water level is a issue.
Would you think adding one or two inches of hose, might increase the water level? I would be a hero if I could add water to the washers fill level.
Hardee prepper The sensor would have to be moved up or down to change the head pressure,or recalibrate the preload springs,as seen in the video. My guess.
I've noticed that my new machine fills with water in bursts and seems to tumble the clothes briefly before sensing again and adding water if the level has gone down. My guess is that it will keep filling and tumbling until the clothes are saturated and can't absorb any more water. That way the minimum water is used to maximise the efficiency of the detergent and heating. Then in the rinse cycles it will add a lot more to remove the loosened dirt.
bigclivedotcom That is how ours works:tumble,fill,tumble,etc.
Just adjust the plastic screw inwards.
I once washed my clothes in a HE machines, then washed them again and again and again, after "4 washes" they were almost clean :))
yes, I tend to become "Pig-Pen" at times ( charlie brown cartoon character), little rolling around in grease, oil, road grime, dirt, etc. in the garage or lawn
Hi Clive. Do you atill rmemeber qhat make washing machine was this sensor for? Plan to get one of this sensor to experiment with
It's common between brands like Indesit and Hotpoint.
My machine “calibrates” by filling the drum until the max level and/or overflow sensor trips when the switch is set to a position labeled “super/reset”. Presumably the brain box stores the reading associated with overflow and chooses some intermediate value for the other settings.
The range of 0-0.3m of water quoted on the casing corresponds to about half a psi, which is well within the typical range that people can produce from their lungs. Might be interesting to blow in it if you ever run into another?
That's a good way to test the old pressostats :) Should work with this as well
So how do you test these type of two pin pressure switch to know if they are defective
It's not easy to test this type without making up a dedicated tester.
It will be easier to just buy a replacement and install it.
If your machine seems to be using very little water in the main wash cycle then it's normal. It only uses a lot during the rinses.
Initially it just aims to get the clothing soaked in a minimal quantity of water to increase the detergent concentration and give better friction when tumbling.
How do they prevent the connector to get clogged up ? There needs to be a small flow of liquid / gas in and out,
but it seems as if everything on a washing machine gets clogged up with micro fiber and detergent residues over time...
- so why not that small intake for the pressure ?
As an added benefit the CPU can self zero the sensor. When you first start a load the CPU knows the drum is empty. It then samples the sensor and uses the reading to use as a reference for zero. Short term accuracy would be close enough and you don't really need long term accuracy (plus you can self zero again after the dirty, sudsy water is removed and right before it fills again for the first and optional second rinse).
Is that a digital or optical zoom on the ipad? Never really seen the kind of loss of detail associated with digital zoom
digital zoom, its not as bad now to use a bit of digital as the sensor has plenty of spare pixels to zoom with
Digital zoom but just homing in on a smaller area of a large image array. It's a Moto G4.
bigclivedotcom it's a fair bit better than it used to be, think I'm strictly an optics guy
Hi Clive This looks the same as on an Electrolux washer and trying to adjust the water level. My one has no screw on top. Is there any way to adjust this for more water in tub
Modern washing machines will initially just make the clothing damp initially to maximise detergent strength, increase agitation and reduce foaming. The machine usually adds water, tumbles the clothes and then checks to see if the level has gone down. When it detects the level is the same it will stop adding water.
You can't make it add more water during the detergent cycle, it only uses more water during rinses.
another great video... thank you so much...
"Clever", you`re a clever guy!
Just my 5 cents here. On Bosch series 4 ecosilence washer this screw setting can not be altered as firmware is looking for specific frequency - in my case 625Hz for idle and goes down with pressure ( tested with multi-meter) If it senses discrepancy, two things happen, will not stop draining or will add more water ( I was happy to see that ) but if you happen to change program or pause machine - then door lock unlocks even there is full drum with water... Some people over internet reported heating element not working ... So after 3x days playing around, conclusion is, looks like some engineers made sure that consumers do not tinkle with this setting or it is simple firmware glitch where if parameters are not met, then some functions may fail or may work differently from originally programmed...
The screw doesn't really set a liquid level. The machine starts the cycle with a very low amount of water deliberately. It puts water in to a fixed level and tumbles the clothes then looks to see if the level has gone down. If it has, the clothing isn't saturated yet and it adds more water. Once it sees the level remain the same it knows the clothing is fully saturated.
That means that during the first part of the wash cycle the detergent is at its strongest and the friction between tumbling clothing is highest, giving a more thorough wash. The rinse cycles use much more water.
Very gratefull explanation...🙂
I'm wondering why it is necessary to divide the 214kHz signal at the sensor, would that be because a lower frequency is more able to be transmitted to the controller over low grade wires?
The lower pulse speed gives the processor more time to time the duration and could provide more accurate results.
Would the output of the Oscillator (8-12) be directly related to how much water in litres is in the machine?
The output is directly related to the height of the water column pressing on the diaphragm.
I realise pressure is applied to the diaphragm from the water in the in drainage sump. However in my Samsung machine there is a similar device which sits in the sump at the bottom of the drum next to the drainage hose which leads to the pump. As our Samsung is a EcoBubble machine there is very little water in the machine. I think the last time I had to manually drain it due to a blockage in the pump I only counted about 7 litres of water. Which leads me to think this device is calibrated directly in relation to the actual amount of water in the machine rather than just any old number being generated.
"Oh, I've already broken this."
That just might be the quickest yet.
have this with resistor 101 and 103,
replaced with other (unknown inside) and not working. how to determine it from outside label?
Can you adjust it so the machine will fill more?
No. The machines only look for a change in water height to see if the clothing is still soaking up water. The point is to only make the clothing wet initially so that the detergent is at its most concentrated, and so that the clothes rub together as they are tumbled instead of floating in the water. It results in a much better cleaning action.
After the detergent cycle the machine will use a lot more water during rinsing.
Ok. Many people are modifying the front load washers, so they use more water.
could you have a look at what is inside one of those yellow transformer boxes found on building sites ect
a transformer, nothing else, no other bits, sitting in oil in a sealed plastic case, it only drops 240 to 110 ac, very simple to do
It't literally just a transformer and sometimes some fuses too. Usually potted into the case with the terminals and fuses protruding through the potting compound.
Arnt the newer ones digital?
How exactly would a transformer be "digital" ?
Ugh, they do inverter based ones? Completely missing the point of a site transformer in the process i'll bet... :/
would it have been simpler to just use the resistance of the coil as a level indicator?
It is using a coil's resistance, which is inductive resistance, the resistance to AC flow.
Just out of curiosity could it be used in conjunction with an arduino to monitor a fish tank water level?
Yes it could, but a float switch would also be a simple solution.
That's probably not a Schmidt trigger gate. That would be used for an RC oscillator. For an LC or crystal oscillator, it would be a standard gate forced into its linear region and acting as an amplifier.
Give me an old mechanical washer & drier any day. They are simply the correct technology for the application. Our last pair were Sears Kenmore (Whirlpool) bought c. late 60's and were still going strong when we moved a couple of years ago. Yea, had to do a couple of timers, a door switch or two, a belt, and once a toothpick got jammed in the water pump... but thats about it; any & all parts are reasonable & available. With a manual I did 100% of the repairs. The new washer & drier have already required several service calls, and the 'technicians' were saying (the very expensive) parts are already getting scarce; they seem to require proprietary diagnostic equipment, and the build quality appears abyssal... I could go on, but it gets my dander up. Think we'll be lucky to get another 2 or 3 years out of them.
In the US, Speed Queen still makes all-mechanical top loaders.
Thanks Ray... I've heard of Speed Queen, but am not familiar with them. Do they produce (and support) quality products?
rwbishop their bread and butter is laundromat machines, but they also make a version with a normal consumer mechanical timer. If you are familiar with the old-style Maytag washers, you will feel right at home.
Thanks will definitely look into them. Not really familiar with Maytag... other than have a neighbor with an old one who swears by it. However, remember seeing/hearing that their newer generation stuff may be problematic.
rwbishop I have a lg front loader, only thing done is bearings , and drain pump,,, drain pump is like a fish tank water pump.. In over ten years... Have the matching dryer, but it is that same as all dryers
could one of these being faulty trip a circuit breaker? ive checked the heating element by disconnecting it, the water drains out of the machine when it starts up,, so not the pump, the water runs into the prewash section of the detergent tray and then once the water for the main section of the wash starts the breaker trips!! ive replaced the water inlet valve and it still trips at this point so not the solenoid in the water inlet valve, next down the line is this bad boy and mine looks just like this one. washer is a bush A127QW. thanks!
It's unlikely to cause rogure tripping. Two common reasons are the heating element and sometimes the incoming supply electrical filter.
@@bigclivedotcom thanks for the reply. ive disconnected the heating element so its not that and ill look at he eclectic filter, i dont know what it is yet but ill check that out thank you. but why would the machine start, run through the prewash section of the cycle then cut out every time right when the solenoid switches to the main wash water inlet, some water comes through but it switches off/trips the breaker within 1 second.
@@bigclivedotcom incoming supply electrical filter roger that, ill try that tomorrow! youre a top G big clive!
@@bigclivedotcom ok so heres an update, ive swapped out the mains interference electrical filter and things have improved but now its doing something else. the machine now makes it past the prewash and fills the drum with water for the main wash, then it stops seemingly ready to wash the clothes, and then the stop cycle light flashes a bit then the machine drains the water, stopping the cycle :(. man im so close, what do you think?? is this now a motherboard issue? thank you!
Haven't watched the full video yet, I'm guessing it's a strain gauge?
+AbdulAminGani Good guess, but it's a lot cheaper than that.
That 200R resistor should be on the output of the gate, not the input.
Edit: And looking at the PCB the coil appears to be connected directly to pin 11, which is the input to the first gate in the oscillator cct.
To increase water level, just adjust the plastic screw inwards!
That may increase rinse level, but the initial detergent fill will use just enough to the point that the clothes are saturated by detecting a change in the water level.
Well, it worked on mine, going from "save the Earth dry-wash" to a proper wet wash. Draining the Earth's resources one liter at at time! :)
@@RolfRBakke What model washer and sensor do you have? Thanks.
How do you test this?
Yes, I also wanted to know how to test it in the field. Now that I know how it works, I think you might be able to use a multimeter with a frequency function, or use a graphing multi-meter (which I have) and find which terminals (signal + ground) that have a "frequency" on them. First verify that you have approximately 5 VDC going into the device. Now you've identified ground and the Vcc lines. Then, connect your meter to the 3rd terminal and Look for a frequency of around 8 to 16 Hz. Then if you carry around a small piece of tubing with you, you might be able to attach it to the sensor, blow into it gently, and see if the frequency changes. (I have not tried this method yet, since I just saw this video, but I will try it the next time I work on a Samsung washer.)
Is it possible to measure the value of the capacitors in the oscillator?
I just isolated the caps by cutting tracks next to them. They measured 2.2nF (2200pF).
3:26 How it calibrates: Two opposing springs of equal strength and a cross head threaded insert bonded with anti-tamper marks. Allows for dialing in the 0 position at factory. =^-^=
..... I wonder if there is a program calculation detecting load with this, in conjunction with the 'post spin up' g-sensor input (or whatever they use).
The pre-spin balancing seems to use an algorithm to monitor the feedback from the motors tacho. The speed of a balanced load is consistent, but an unbalanced load wavers significantly on each revolution.
@@bigclivedotcom this is a problem with our machine, its sometimes used just to spin a small but occasionally heavy load, the thing throws a wobbly and just wont spin it, you have to wring it as much as possible by hand before spinning! previous machine was far less of a bitch!
The word "sneaky" came to mind for that L/C oscillator to divider :-D
Very sneaky. Never seen that arrangement before.
Hi, please do a teardown on non-contact liquid level sensors. XKC-Y25 etc.
They look like capacitive sensors.
nice to see some simple circuit making good use of a classic digital logic chip
I failed to get if this one just tells on/off (like full, not full) or it shows the level (10%, 50% etc) inside the washing machine? If its level, I am confused how it reads level with that small travel!!!
thanks for the great video clive, over engineering but hey gota get your worth somewhere for the $1200 dirty draws washer
Keith Neubauer it’s under-engineered. Over engineered would be a ARM based computer reading a load cell or piezoelectric pressure sensor, all tucked into a CNC machined case. From a mechanical standpoint, this device is very similar to the electromechanical water level switches of days gone by. The electronics are very simple; no microcontroller, no ASIC, just a bog standard off the shelf part, a few passives and some wire.
agree to disagree lol
Makes me glad my pre-indesit buyout Hotpoint washer is all switches and dials, seems wrong to slap an IC into the water level switch when a couple of microswitches does the job perfectly well... :)
But, as Scotty said, "The more they overthink the plumbing, the easier it is to stop up the drains", overcomplication of simple things is just a wasted effort in my opinion... :)
But this isn't complex...it's simple, and cheap, and easy to replace. So you comment is invalid.
You bet me to the Scotty quote by an hour.
It's more complicated, costs more to make, and the same cost to replace.
Comment valid.
Switches stick, especially with age.. and that's quite a common failure of the old type "pressostats". This device will never stick in that way. So it's a far better design.
So, let me get this straight, you want to use something which has more moving parts, i.e. up to 3 switches for example, which has terrible resolution (3 positions) over something that potentially can tell the washer EXACTLY how much water is in the drum because of it's near infinite resolution? This switch is a brilliant idea and probably costs a LOT less to make.
Switches are expensive because of the exotic materials used on the contacts, such as gold, silver, platinum etc. They tarnish over time, springs become weak, yadda yadda yadda. I know what I'd rather have.
davegsm82 these are very reliable. I’ve never seen one break. The pressure dome/hose clogs with detergent powder at times, but the pressure transducer itself is reliable.
How do I know? I’m a WASHING MACHINE ENGINEER.
*high fives Chris and Peter*
Interesting design. I assume the cost of the ones with physical switch contacts is now considered prohibitive by the manufacturer. I expect in the really old machines with mechanical controllers needed the contacts to be rated for 250v and they would eventually fail due to contact wear, whereas this design should last indefinitely - or as long as the diaphragm is intact. Clever use of cheap off the shelf components.
Switches don't give you fine grained water level control.
The mechanical water level switch contacts didn't fail, they only passed a low current for the switch module's clock.
They're less mechanical, more electrical now, a sign of the times.
Advantage of doing it this way is you can have any number of fill levels. With the older diaphragm and microswitch type you need one for every fill level - the one here has two.
How would one test that to see if it was faulty?
Easiest way would be to swap in a new one. Keep in mind that modern machines initially just make the clothing wet with no excess water, to maximise detergent strength and friction between garments with lower risk of excess foam. Subsequent rinse cycles would use a lot more water.
@@bigclivedotcom Less water makes sense. If it fails how would you know if not filling, over filling or not spinning was not due the the pressure switch? I always test them when facing these issues.
my washing machine empties all the water every time I change the program. How can I trick the pressure switch into believing for a few moments before starting the new program that there is no water?
It's usually in the software that the machine will want to start from an empty drum so it knows the state of the water in it.