My completely uneducated guess would be that the different metal is the culprit? it would sorta make sense if they made the newer ones with stronger metal to be more durable, but then didn't consider that it would have sharper edges (as im sure this is stamped out of a sheet) and those sharper edges end up scraping away the carbon stuff? And then those 'flakes of scraped off carbon' get in between the seams of the parts that have grease inside, causing that to leak out? and then the grease + carbon-specks mix causes unintentional connections wich results in the drift? And about the test you did, since you put way to much power on it, could it simply be that the grease is expanding because of heat? Idk though, my understanding of electronics is basic at best and my understanding of physics is even lower :P
I don't know how those potentiometers are manufactured, but could the fantom juice be moisture (water/solvent/other runny thing) from the bonding agent that got liberated when the carbon track heated up? They've got to stick that carbon in place somehow and carbon powder is pretty dry and dusty, so they've got to mix it with something, right?
What I'd like to know is why we never had stick drift with the old PS2 Dual Shocks if they used this kind of potentiometer, and what they changed to make them wear out faster. Maybe a comparison to those old ones would shed light on the issue.
I replaced the pots for 6 PS3 controllers for a friend last year. I am convinced the pots are made much more cheaply and intentionally. Not just to save a few cents per thousand but because they know it will be the leading cause of controller replacement. There's no profit to be had in making a controller that lasts for years on end without having to be repaired. And most people don't bother getting them fixed. They just go and buy new ones. That's why my friend had 6 of them. He'd wear one out, go buy another and put the old one in a box. It was literally what Einstein defined insanity as.
There center bit, although carbon coated, will act as a bias resistor similar to a series resistor. Wearing through there film will cause that point to show lower to that one leg causing the signal to appear to the controller as the stick being pushed slightly up.
Resistance centered and calibrated before 5k plus 120 ohms. Once the center ring is worn through 5K plus 2 ohms hence the drift in one direction due to the lower resistance.
They’ve been using the same shit since analog controllers first became a thing (n64 not withstanding, dear god), and yet older controllers were much, much more reliable. I’ve got a few gamecube controllers that have been beaten to HELL and back, but still work perfectly nearly 20 years on (F-Zero even has a controller calibration menu so you can see the stick movement “profile”). In contrast, my PS5 controller started drifting after three months, and of course the Switch is hella notorious for drift. Hell, stick drift didn’t seem to be a very widespread issue until the PS4/XBone era, it seems. So yeah, something is up. I’d be interested in seeing a comparison autopsy between a current-gen pot and an old Gamecube/PS2 pot. I keep meaning to some burner controllers and try it eventually, but you know… lazy. Thanks for takin’ a look. Entertaining and informative as always!
@@logan_cadfgs gamecube controllers arent optical only the n64 controllers were optical. the gamecube controllers used potentiometers that were produced not by alps but by a manufacturer called noble, hence why they were of different quality than other pots(but still had their own issues).
No no no, you don't understand how obsolescence work. On another subject, I don't know how empty the joystick is but maybe you could kinda so a DIY version? I mean, I end up (almost) finished a N64 mod to use a potentiometer jostick instead.
Not really, but the AH8502/03 hall sensors operate in the XBOX Controller's voltage, so it's technically possible to mod the controller. Bought some of them but they are packaged as SMD with no leads, so i'll have to get a reflow station to solder leads into it. Also check if putting magnets on the joystick axis interferes with the left trigger sensor and maybe find a solution for that. It's not really some that would make sales tho because 1st you rely on ALPS design and 2nd most people arent comfortable soldering stuff into their controllers lol.
there are, thesedays! they're not in Production at anywhere near the scale of Potentiometers, though. and it makes the Product less Consumerism E-waste, so there's the cliff of convincing Companies to make less Money. but, they are available, so you could solder in replacements for yourself. waiting time for Shipping could be a while though due to low-ish Production scale. and pretty much a single source. Et Cetera.
You probably worked this out, but the inner carbon ring will always wear faster because the stick travels over less of the carbon for a particular angle compared to the outer ring.
The work done by friction and the amount of carbon are both proportional to the radius, so I don't think the amount of carbon in the inner ring should matter, unless it's also thinner. I reckon this is more to do with the pressure applied by the inner springs. They are shorter, so I'd expect them to be stiffer as well.
Ben I've gotta say, watching your videos with the different and funny voices has now rubbed off on me, I work in customer service on phones all day long and I nail a really good Sean Connery and Harrison Ford impression, love watching your videos Ben! Edit;the gun metal look is a blueing process to protect the steel and do it withstands heat more, the gunk is like you predicted imo as "carbon lithium grease"
It's not necessarily just that the carbon film is wearing away faster. Controllers from the PS2/GC era calibrated their pot center value when first powered on, meaning any bias would be accounted for and if it did drift all you had to do was unplug it and plug it back in to stop it.
@@kadupse They don't. You can test it by holding the stick in one direction or another while powering on. If they auto calibrate like this then it will induce drift in the opposite direction since the center point will be set at the edge of the stick. Modern cotrollers don't do this.
@@meatsafemurderer7743 I definitely did that on accident more than a few times when booting some of those old consoles back in the day. I had to train myself not to touch the analog sticks until I was at least at some kind of menu (a habit I still practice today.)
Making hall effect joysticks from worn out pot joysticks isn't too bad, but you'd also need an extra MCU to either convert the voltages or just act as the controller itself. Converting the voltages is less complicated in theory, but since most common MCUs come with ADCs but no DACs it's usually more practical to just use them to replace the stock MCU. Doing a voltage conversion is viable if you need the controller to remain as is logically, but it does mean getting a discrete DAC and figuring out what voltages the pots should give you nominally.
@@guycxz or design the hall effect chip to output the correct values, most analog hall effect sensors output an analog voltage, they could just design them to output correct voltages for specific controllers.
18:00 I'm not sure if it's the lighting/angle, but it also looks like the newer pot on the right hand side has bulges in the common wear zones, which may be an indicator that they know they're using a carbon that wears more quickly, but to compensate for that they increase the thickness to try and achieve similar longevity.
The leaky grease soaked into the printed carbon film resistor and made it weaker. Now it is less resistant to the mechanical wear of the rotating wipers, and the carbon was scraped away. Would be my guess as the failure mode of the first pot we saw. As to the results of the second experiment.... To silkscreen print the carbon, it must (might) be conductive particles suspended in a medium during printing. When you heated the pot with the 24v test - the magic part of the magic smoke - made a skid mark as it cooked off the now dry but not totally dry printing medium? A kinda glue I guess, in a solvent that dries off?
I've rebuilt sticks for 360 controllers by using new china sticks, taking off the crappy pots, then swapping them with the original (black) 360 pots. The old 360 pot tracks are different looking and are almost always in good shape. They seem to be harder and smoother looking.
The grease could carry abrasive particles, and might also dissolve whatever the carbon is suspended in. The inner tines are also much thicker and a bit shorter, so they are likely stiffer, so I'd expect them to apply more pressure.
Upstate New Yorker here, and I can confirm that winter does indeed suck. Having to unbury your car and defrost the windows before driving, plow and/or shovel the driveway and/or sidewalks, drive slower on slushy and/or icy streets, and laugh at the folks in pickup trucks and/or SUVs who think that the physics of icy surfaces doesn't apply to them... ok that last one doesn't suck.
removing the carbon from the centre track may make it stable again. The carbon on the centre track is acting like a resistor on the centre leg, when it's worn out this resistance is changing enough to scew wiff the adc reading it. is there a resistor on the board from the centre pin to the adc.
The inner tines may have more force pushing into the carbon ring increasing its wear. The moment arm is shorter and the arm geometry is very different as well.
tI'm thinking the wipers on both old and new potentiometers are nickel-plated spring steel. The funky colors on the 2016 pot are screaming "bad heat treat" when viewed from the back. I wonder if the contact points are plated with something like silver for conductivity.... also, you're thinking cycle life, not shelf life.
6:00 i like how you were swapping them, to test i fits different, then you just straight up change the moisture levels on the contact. Did you check the place its attached for more "grease" ?
Gulikit is a company that makes hall effect sensors for controllers. The king Kong pro 2 controller is a good one. They have hall effect stick replacement for steam deck and are making some for the switch joycons that I'm really excited about.
@@YoureUsingWordsIncorrectly I'm saying an LDR is far cheaper than an "encoder" or hall effect sensor, like the trim pots on these sticks only it doesn't involve two surfaces grinding against each other. I don't design controllers but as long as you can tune your ADC to match the range it could be a direct replacement with the addition on one or two (also cheap) LEDs. They're also readily available. It's probably been tried in some design lab somewhere...
i think its the bonding agent in the carbon that was attacked by the greases PH. or other chemical terminology.. etc etc. it does seem easier to remove in that one instance. but at were the carbon was worn was the resting position of the controller where the grease was left to fester. just my 2 cents.
All carbon resistors have a published lifetime in their datasheet. Those look closer to trim pots that don't last long vs. the large pots in guitar or AV equipment. I assume the center is carbon coated for lubrication and that's not a great idea in this case.
hi ben, why don't you try to make a photo actuated analog? you could use one led and a photo transistor/diode, it would be an interesting option over the hall sensor (it would probably be more expensive but I believe it would need less changes to the analog circuit)
Would it be possible to add carbon back to the potentiometer case? I'm guessing no and that it probably requires very expensive industrial equipment, but you never know until you pose the question.
The leaking grease combined with the flaking carbon would be more abrasive than the carbon alone, right? With the grease it's basically scotch brighting the inside of the carbon layer.
I dunno if this is crazy or not. I have carpal tunnel. I take out the rumble motors by desoldering and just replace the place with some lead sinkers for fishing that I hot glue in place. Same nice weighty feeling no numb hands after a while of playing games.
I wish these big companies would move away from potentiometers. Hall Effect joysticks are a much better option Edit: look at that, you brought it up at the end lol.
the two tines create a voltage divider between the two as one gets lower in resistance, the other gets higher, but the total of the two is ALWAYS equal to the reference of direct center unless, as you saw, it was damaged electricity always takes the path of least resistance.... but that doesnt mean that it never takes the path of most resistance, it merely means that it will put less current through the path of most resistance, and both end up being a lower resistance than one
How hard would it be to retrofit a hall effect sensor in place of the pot? It's still hard to believe, after decades of pots wearing out, that we still depend on them for these controllers.
@@Phos9 I think its barking at the wrong issue Pots are Tried and true same with hall effect and Optical. something else is efffing up. Dont have stick drift on anything except new controllers.
I had a POS offbrand Xbox360 controller... and it had a slight stick drift. It was due to the return spring not returning to true, and had nothing to do with the pots at all.
Btw, any variable resistance on the "follower" in middle of the pot will be common on readout and not matter. The voltage divider is outer section and it must be what is causing the issue. Center tap that is read is feeding the adc at high impedance. a little more or less will not be noticed.
I think it's always the carbon film pot. I have a Switch and all the controllers develop drift after 6 months and become unusable after a couple of years. I have PS2 and PS4 controllers and an original XBOX, never had any drift with quality old school pots. I did have a cheap controller which didn't recentre well but I think that was spring imbalance. I haven't pulled apart my Switch controllers yet, I have at least 3 joycons with significant drift now and several more with minor drift. Previously I just replaced them under warranty. I really think it's designed obselescence and they should go back to the old style. Edit: Can you pull apart a PS1 pot, I'd love to see the difference because I heard only the latest Gen controllers used the new graphite and the last gen XBOX pro controller. I do have a PSOne somewhere too, no issues with it's controllers except the plug rusted on one. Edit 2: Loving your tests. If you could get to the bottom of this mystery I'm sure all the people that use controllers would be mighty impressed, the manufacturers might not though :P Edit 3: Awesome vid, I'd love to see more on this. Could use a little more singing though :P
If the carbon is so easy to wear off, why do they keep using it... Oh yeah, because money. It's cheap to make, and if it breaks you spend $50 on a new one.
seems like over a long period the grease breaks down whatever adhesive is used to bond the carbon powder together and to the surface. Bring back hall effect sensors.
the greese may have either caused an insolating barrier or if it is conducting maybe bridging the center and outter though im not and expert at this stuff
15:07 - Oh, I saw Melora Hardin in a movie this week. Reckless Kelly, with Yahoo Serious! Remember Yahoo Serious? He did like three movies and that was it. There's a very young Hugo Weaving in it too. He's Agent Smithing hard in that movie.
You know, it's interesting that the outer track has TWO wipers (the bump), but the inner is single. I wonder if each inner side had two small, individual, wipers (feet? heh), if it would be less susceptible since it wouldn't have as much spring tension...?
Hey Ben, are there any fixes for unreliable A button. Every Xbox controller I have had since the XBOneS has had this issue including the series S and X and the Elite series 2 controller. It’s like it only registers like 9 out of 10 presses. Seems to be mostly the A and Y buttons.
@@guycxz they are like this brand new out of the box. I'm starting to think it's not actually missing. You can see this behavior when entering text using the onscreen keyboard and it will double enter letters when you press the button once. Some games may see this as too fast of an input and ignore it?
I've lost a ps5 controller because the usbc connector died in it and now the firmware is too old. I have another ps5 controller with the same hardware fault with newer firmware so its a matter of time. If you want them to play around with you can have them. The pads are tore on the port because the solder flood doesn't make a good mechanical connection for the report.
Based on evidence presented, it looks like the carbon film is getting melted, causing the grease seen. And then further friction easily erodes it in its softened state. When Ben scraped the carbon film off manually, it produced a lot of clingy debris. However: There was none of this debris when first opening up the module. Just the mysterious oil. When Ben overcharged the stick, the oil became present, and some of the Carbon looked as though it were misshapen. If I wanted to confirm it, I'd ask Ben, when he was overcharging them did he match positive/negative as it would be in an unmodified controller? Because it's strange that it appeared on the outer ring. But I suppose, past of least resistance.
I have heard of a hal effect joysticks. That would solve the issue with pots wearing out. And I'm sure the pots are being made cheaper. My oh Xbox controllers still work flawlessly but my 360 controllers all get stick drift in about a couple years of play. I'm sure the joycon sticks are cheapest, it's like they get drift out of the box
At the least they should be giving us adjustment pots, if not going full hall-effect for the prices they charge these days. They've known they've been selling us bad sticks for 15+ years. great vid as always, but this topic always makes me a bit angry. 🤣
Ben, we met at Maker Faire. I'm a longtime friend of Jeri's. Anyway, I have invented a way to make driftless joysticks, NOT using Hall effect [which is the current best way]. My invention is something you could scratch build in a couple of days. It has a lower BOM. 50/50 everything when we talk about it. PS: I named Darlington.
There are basically 3 technologies for potentiometers: Carbon film, wirewound and conductive plastic film. Wirewound would be real hard to do in this size and expensive, but I'm surprised they don't use plastic film, as it used in high reliability applications as it has much more wear resistance. I guess carbon film is "good enough" and cheaper than the plastic film pots.
If the only difrence between the pots on the controller and the one in your parts bin is max resistance but they only use such a limited airea in the center of the pot couldn't you just add a serface mount resistor to both the outer leads to coppy the larger total resistance.
I got a couple of the steam deck modules to poke at. The magnet holders have a similar shape to a joystick potentiometer, except the casing has no traces and the bit that has the swiper on it just hold a magnet at the bottom. The actual sensor is soldered to the board directly under the magnet holder casing, and isn't very visible. It outputs between 1.3-2.3v and rests and 1.6-1.7(on my units). Ifixit have made a video on the modules if you want any more details.
Do you think there would be a way to integrate touch-controller (capacitive) sensor into a joystick? In other words, the joystick would just move a conductive probe over a capactive surface, thus eliminating electrical contact? The Steam Controller tried a capacitive stick, but it wasn't popular - maybe an actual 'stick' that just moves a small conductive 'finger' over an internal surface would do the trick? Love the videos and Bud! Cheers!!
Please discuss what ya'll think the issue is. That phantom juice was weird, especially when it appeared after I cooked the one pot on purpose.
The problem is you confusing Upstate NY with Western NY, we get 6 feet of snow out here in Western NY but Upstate never gets it that bad
My completely uneducated guess would be that the different metal is the culprit? it would sorta make sense if they made the newer ones with stronger metal to be more durable, but then didn't consider that it would have sharper edges (as im sure this is stamped out of a sheet) and those sharper edges end up scraping away the carbon stuff?
And then those 'flakes of scraped off carbon' get in between the seams of the parts that have grease inside, causing that to leak out? and then the grease + carbon-specks mix causes unintentional connections wich results in the drift? And about the test you did, since you put way to much power on it, could it simply be that the grease is expanding because of heat?
Idk though, my understanding of electronics is basic at best and my understanding of physics is even lower :P
We gave it our best shot, Ben. Well, I did my part perfectly well anyway 😁
I don't know how those potentiometers are manufactured, but could the fantom juice be moisture (water/solvent/other runny thing) from the bonding agent that got liberated when the carbon track heated up? They've got to stick that carbon in place somehow and carbon powder is pretty dry and dusty, so they've got to mix it with something, right?
What I'd like to know is why we never had stick drift with the old PS2 Dual Shocks if they used this kind of potentiometer, and what they changed to make them wear out faster. Maybe a comparison to those old ones would shed light on the issue.
The best thing Ben has ever done was when he decided to get a cat. Always entertaining to find out what Bud is up to.
That cat has more personality than Felix.
@@retrotony4119 And that's even considering Felix is named after a cat.
I replaced the pots for 6 PS3 controllers for a friend last year. I am convinced the pots are made much more cheaply and intentionally. Not just to save a few cents per thousand but because they know it will be the leading cause of controller replacement. There's no profit to be had in making a controller that lasts for years on end without having to be repaired. And most people don't bother getting them fixed. They just go and buy new ones. That's why my friend had 6 of them. He'd wear one out, go buy another and put the old one in a box. It was literally what Einstein defined insanity as.
The definition of insanity is doing the exact same thing over & over & expecting different results.
@@DTW-bx2vy Duh.. if you read the original comment that is what he is discribing. Mr well actually. 🤓
The microscope thingy gives a pretty great image, you should remember about it more often.
There center bit, although carbon coated, will act as a bias resistor similar to a series resistor. Wearing through there film will cause that point to show lower to that one leg causing the signal to appear to the controller as the stick being pushed slightly up.
Resistance centered and calibrated before 5k plus 120 ohms. Once the center ring is worn through 5K plus 2 ohms hence the drift in one direction due to the lower resistance.
They’ve been using the same shit since analog controllers first became a thing (n64 not withstanding, dear god), and yet older controllers were much, much more reliable. I’ve got a few gamecube controllers that have been beaten to HELL and back, but still work perfectly nearly 20 years on (F-Zero even has a controller calibration menu so you can see the stick movement “profile”). In contrast, my PS5 controller started drifting after three months, and of course the Switch is hella notorious for drift.
Hell, stick drift didn’t seem to be a very widespread issue until the PS4/XBone era, it seems.
So yeah, something is up. I’d be interested in seeing a comparison autopsy between a current-gen pot and an old Gamecube/PS2 pot. I keep meaning to some burner controllers and try it eventually, but you know… lazy.
Thanks for takin’ a look. Entertaining and informative as always!
GameCube controllers are also optical sensors, at least the main stick. I'm not sure about the C stick.
@@eDoc2020 oh shit I never knew! No wonder. Time to go beat the hell out of 'em some more then
@@logan_cadfgs gamecube controllers arent optical only the n64 controllers were optical. the gamecube controllers used potentiometers that were produced not by alps but by a manufacturer called noble, hence why they were of different quality than other pots(but still had their own issues).
Love your videos Ben, its always fun to hang out and watch you work. Your commentary is the best. Thanks for sharing!
If there aren’t drop-in replacement analogue hall rotary sensors, it sounds like something you could get a lot of business by making.
No no no, you don't understand how obsolescence work. On another subject, I don't know how empty the joystick is but maybe you could kinda so a DIY version? I mean, I end up (almost) finished a N64 mod to use a potentiometer jostick instead.
Not really, but the AH8502/03 hall sensors operate in the XBOX Controller's voltage, so it's technically possible to mod the controller. Bought some of them but they are packaged as SMD with no leads, so i'll have to get a reflow station to solder leads into it. Also check if putting magnets on the joystick axis interferes with the left trigger sensor and maybe find a solution for that.
It's not really some that would make sales tho because 1st you rely on ALPS design and 2nd most people arent comfortable soldering stuff into their controllers lol.
there are, thesedays! they're not in Production at anywhere near the scale of Potentiometers, though. and it makes the Product less Consumerism E-waste, so there's the cliff of convincing Companies to make less Money.
but, they are available, so you could solder in replacements for yourself. waiting time for Shipping could be a while though due to low-ish Production scale. and pretty much a single source. Et Cetera.
You probably worked this out, but the inner carbon ring will always wear faster because the stick travels over less of the carbon for a particular angle compared to the outer ring.
The work done by friction and the amount of carbon are both proportional to the radius, so I don't think the amount of carbon in the inner ring should matter, unless it's also thinner. I reckon this is more to do with the pressure applied by the inner springs. They are shorter, so I'd expect them to be stiffer as well.
The rainbow metal is tempered steel (aka heat-treated). Tempering makes it more springy and also less magnetic.
I really want to know now what the potentiometers on old ps1and ps2 era controller s looked like since those seem to last alot longer.
Ben I've gotta say, watching your videos with the different and funny voices has now rubbed off on me, I work in customer service on phones all day long and I nail a really good Sean Connery and Harrison Ford impression, love watching your videos Ben!
Edit;the gun metal look is a blueing process to protect the steel and do it withstands heat more, the gunk is like you predicted imo as "carbon lithium grease"
that would make me laugh in such a good way if i heard it while getting customer service.
I like the 2018 timestamp suspect it's not real but thanks for the time travel :)
They also seem to add extra goo on the parts which where worse.
It's not necessarily just that the carbon film is wearing away faster. Controllers from the PS2/GC era calibrated their pot center value when first powered on, meaning any bias would be accounted for and if it did drift all you had to do was unplug it and plug it back in to stop it.
@@kadupse They don't. You can test it by holding the stick in one direction or another while powering on. If they auto calibrate like this then it will induce drift in the opposite direction since the center point will be set at the edge of the stick. Modern cotrollers don't do this.
@@meatsafemurderer7743 I definitely did that on accident more than a few times when booting some of those old consoles back in the day. I had to train myself not to touch the analog sticks until I was at least at some kind of menu (a habit I still practice today.)
Glad to see someone is making a Hall Effect stick mod for the Deck, maybe they’ll do it for regular controllers some day.
Its already been made! I have a set in my deck and its great to just never have to worry about this becoming a problem!
Making hall effect joysticks from worn out pot joysticks isn't too bad, but you'd also need an extra MCU to either convert the voltages or just act as the controller itself. Converting the voltages is less complicated in theory, but since most common MCUs come with ADCs but no DACs it's usually more practical to just use them to replace the stock MCU.
Doing a voltage conversion is viable if you need the controller to remain as is logically, but it does mean getting a discrete DAC and figuring out what voltages the pots should give you nominally.
@@guycxz or design the hall effect chip to output the correct values, most analog hall effect sensors output an analog voltage, they could just design them to output correct voltages for specific controllers.
18:00 I'm not sure if it's the lighting/angle, but it also looks like the newer pot on the right hand side has bulges in the common wear zones, which may be an indicator that they know they're using a carbon that wears more quickly, but to compensate for that they increase the thickness to try and achieve similar longevity.
Bulges would increase resistance.
It looks as though they all have them, though.
Love your videos Ben, thank you!
Love the Pagan references! That was my favorite Ultima game.
The leaky grease soaked into the printed carbon film resistor and made it weaker. Now it is less resistant to the mechanical wear of the rotating wipers, and the carbon was scraped away. Would be my guess as the failure mode of the first pot we saw.
As to the results of the second experiment....
To silkscreen print the carbon, it must (might) be conductive particles suspended in a medium during printing. When you heated the pot with the 24v test - the magic part of the magic smoke - made a skid mark as it cooked off the now dry but not totally dry printing medium? A kinda glue I guess, in a solvent that dries off?
Another afficionado of that sweet, sweet lead solder
Holy Moly, an Ultima 8 reference! Careful! Bud's gonna come back having mastered the titans!
yay! An Ultima 8 reference for, ultimately, 8 people.
Not Ultima 8 - it was mostly called "Super Avatar Bros." back then!
;-)
lovin the no production november. and october, and september, and then there was august too, and july. june was also good
I've rebuilt sticks for 360 controllers by using new china sticks, taking off the crappy pots, then swapping them with the original (black) 360 pots. The old 360 pot tracks are different looking and are almost always in good shape. They seem to be harder and smoother looking.
The carbon laid down on the malleable plastic could be forming defects under the pressure and biasing the resistance?
Just for the taste of it is a permanent portion of my brain
Detritus: become human made laugh. Keep the awesome work.
The grease could be a solvent for whatever they use to glue the carbon to the substrate.
I agree with your hypothesis regarding the potentiometers. It wouldn't surprise me...
Have you looked at the hall-effect sticks used by the new gulikit controllers ?
The grease could carry abrasive particles, and might also dissolve whatever the carbon is suspended in. The inner tines are also much thicker and a bit shorter, so they are likely stiffer, so I'd expect them to apply more pressure.
Upstate New Yorker here, and I can confirm that winter does indeed suck. Having to unbury your car and defrost the windows before driving, plow and/or shovel the driveway and/or sidewalks, drive slower on slushy and/or icy streets, and laugh at the folks in pickup trucks and/or SUVs who think that the physics of icy surfaces doesn't apply to them... ok that last one doesn't suck.
can you mod one of those new magnetic analog modules into a standard controller?
removing the carbon from the centre track may make it stable again. The carbon on the centre track is acting like a resistor on the centre leg, when it's worn out this resistance is changing enough to scew wiff the adc reading it. is there a resistor on the board from the centre pin to the adc.
Nice, now i am thinking of MS doing A/B testing on choosing the best metal for planned obsolescence based on number of returns...
The question can you solder the Hall Effects module on these controllers?
The inner tines may have more force pushing into the carbon ring increasing its wear. The moment arm is shorter and the arm geometry is very different as well.
Maybe a component of the leaky grease dissolves the carbon layer. Or at least reacts with it to leading to increased west from the wiper.
That would be my guess.
We feel like that nurse that hands over the scalpel in your operations. DR. HECK
I notice when you have the 2 carbon tracks side by side the right one has the carbon thickened in the wear areas....
This channel needs more real estate.
tI'm thinking the wipers on both old and new potentiometers are nickel-plated spring steel. The funky colors on the 2016 pot are screaming "bad heat treat" when viewed from the back. I wonder if the contact points are plated with something like silver for conductivity.... also, you're thinking cycle life, not shelf life.
Hahahaha the first seconds you started the desolderingsucktion i though it was you making the sound!
6:00 i like how you were swapping them, to test i fits different, then you just straight up change the moisture levels on the contact. Did you check the place its attached for more "grease" ?
I wonder if the grease softens the carbon so it scratches off easier?
Gulikit is a company that makes hall effect sensors for controllers. The king Kong pro 2 controller is a good one. They have hall effect stick replacement for steam deck and are making some for the switch joycons that I'm really excited about.
For analog control there's also photo resistors. Put a tiny LED in there and open and close a window to the light based on stick position.
@@YoureUsingWordsIncorrectly rotary encoder is digital. Photo resisters are analog with no interfacing parts.
Check out the “tremolo” circuit in old fender amps.
@@justovision If it needs to be analog for whatever reason why not just use a Resolver?
Sega Saturn and Dreamcast controllers had Hall Effect sensors from the start because the engineers had predicted that pots would wear too much.
@@YoureUsingWordsIncorrectly I'm saying an LDR is far cheaper than an "encoder" or hall effect sensor, like the trim pots on these sticks only it doesn't involve two surfaces grinding against each other. I don't design controllers but as long as you can tune your ADC to match the range it could be a direct replacement with the addition on one or two (also cheap) LEDs. They're also readily available. It's probably been tried in some design lab somewhere...
How is the carbon deposited? Perhaps they use a binding agent that is not consistent between pots.
@Ben, I think that's yellow zinc plating, similar to what you find on fasteners
You were moving the sludge and I was like here comes the taste test as a joke then you actually tasted it lmao.
Is there any voice monitoring device which does not need charging
You should look at the Guilikit King Kong 2 wireless controller. It's the Xbox controller form factor but uses hall effect sticks.
Maybe worth looking into trying to mod in the Magnetic Sensors they are putting out for controller, again. Like they had on the Dreamcast.
i think its the bonding agent in the carbon that was attacked by the greases PH. or other chemical terminology.. etc etc. it does seem easier to remove in that one instance. but at were the carbon was worn was the resting position of the controller where the grease was left to fester. just my 2 cents.
All carbon resistors have a published lifetime in their datasheet. Those look closer to trim pots that don't last long vs. the large pots in guitar or AV equipment. I assume the center is carbon coated for lubrication and that's not a great idea in this case.
hi ben, why don't you try to make a photo actuated analog? you could use one led and a photo transistor/diode, it would be an interesting option over the hall sensor (it would probably be more expensive but I believe it would need less changes to the analog circuit)
Would it be possible to add carbon back to the potentiometer case? I'm guessing no and that it probably requires very expensive industrial equipment, but you never know until you pose the question.
The leaking grease combined with the flaking carbon would be more abrasive than the carbon alone, right? With the grease it's basically scotch brighting the inside of the carbon layer.
I dunno if this is crazy or not. I have carpal tunnel. I take out the rumble motors by desoldering and just replace the place with some lead sinkers for fishing that I hot glue in place. Same nice weighty feeling no numb hands after a while of playing games.
why the inner ring isn't a solid metal? it is like it was designed to fail..
I wish these big companies would move away from potentiometers. Hall Effect joysticks are a much better option
Edit: look at that, you brought it up at the end lol.
Was that a Last Action Hero reference I heard? I'll have to re-watch it.
the two tines create a voltage divider between the two
as one gets lower in resistance, the other gets higher, but the total of the two is ALWAYS equal to the reference of direct center
unless, as you saw, it was damaged
electricity always takes the path of least resistance.... but that doesnt mean that it never takes the path of most resistance, it merely means that it will put less current through the path of most resistance, and both end up being a lower resistance than one
Could a thinner carbon layer be an energy savings strategy?
Perhaps also a rougher surface/sharper edge on the contacts.
How hard would it be to retrofit a hall effect sensor in place of the pot? It's still hard to believe, after decades of pots wearing out, that we still depend on them for these controllers.
They stopped using pots for a while, the PS3 controllers used Hall Effect sensors. So did the Dreamcast.
@@Phos9 I think its barking at the wrong issue Pots are Tried and true same with hall effect and Optical. something else is efffing up. Dont have stick drift on anything except new controllers.
I had a POS offbrand Xbox360 controller... and it had a slight stick drift. It was due to the return spring not returning to true, and had nothing to do with the pots at all.
Btw, any variable resistance on the "follower" in middle of the pot will be common on readout and not matter. The voltage divider is outer section and it must be what is causing the issue.
Center tap that is read is feeding the adc at high impedance. a little more or less will not be noticed.
The carbon layer on the inner ring was 80ohms. If it drops to near 0 on that spot rubbed bare it would read differently.
Bud is such a himbo baby
I think it's always the carbon film pot. I have a Switch and all the controllers develop drift after 6 months and become unusable after a couple of years.
I have PS2 and PS4 controllers and an original XBOX, never had any drift with quality old school pots. I did have a cheap controller which didn't recentre well but I think that was spring imbalance.
I haven't pulled apart my Switch controllers yet, I have at least 3 joycons with significant drift now and several more with minor drift. Previously I just replaced them under warranty.
I really think it's designed obselescence and they should go back to the old style.
Edit: Can you pull apart a PS1 pot, I'd love to see the difference because I heard only the latest Gen controllers used the new graphite and the last gen XBOX pro controller.
I do have a PSOne somewhere too, no issues with it's controllers except the plug rusted on one.
Edit 2: Loving your tests. If you could get to the bottom of this mystery I'm sure all the people that use controllers would be mighty impressed, the manufacturers might not though :P
Edit 3: Awesome vid, I'd love to see more on this. Could use a little more singing though :P
PS1 uses graphite.
@@vyor8837 it's strange the new ones are so weak.
@@Commander_ZiN the carbon is likely applied differently or less/more thickly.
@@vyor8837 they need to go back to the old way.
They should make the tines carbon as well, it should mitigate wear having 2 materials of the same hardness rubbing together.
If the carbon is so easy to wear off, why do they keep using it...
Oh yeah, because money. It's cheap to make, and if it breaks you spend $50 on a new one.
glad hall effect controllers are coming out now so we don't gotta worry about this stuff
seems like over a long period the grease breaks down whatever adhesive is used to bond the carbon powder together and to the surface. Bring back hall effect sensors.
Maybe the grease acted as a solvent to help the metal scrape away the carbon more easily?
the greese may have either caused an insolating barrier or if it is conducting maybe bridging the center and outter though im not and expert at this stuff
your tv latency seems high
did you tried setting it to "game mode" or something?
15:07 - Oh, I saw Melora Hardin in a movie this week. Reckless Kelly, with Yahoo Serious! Remember Yahoo Serious? He did like three movies and that was it.
There's a very young Hugo Weaving in it too. He's Agent Smithing hard in that movie.
It's just flappin' in the breeze!
You know, it's interesting that the outer track has TWO wipers (the bump), but the inner is single.
I wonder if each inner side had two small, individual, wipers (feet? heh), if it would be less susceptible since it wouldn't have as much spring tension...?
Hey Ben, are there any fixes for unreliable A button. Every Xbox controller I have had since the XBOneS has had this issue including the series S and X and the Elite series 2 controller. It’s like it only registers like 9 out of 10 presses. Seems to be mostly the A and Y buttons.
Check for dirt on the conductive part in the rubber dome that corresponds to these buttons.
@@guycxz they are like this brand new out of the box. I'm starting to think it's not actually missing. You can see this behavior when entering text using the onscreen keyboard and it will double enter letters when you press the button once. Some games may see this as too fast of an input and ignore it?
13:40 in case you are still wandering it is called anodized metal. I think the color is caused by zinc plating but it could also be aluminium!
There are not any hall effect based replacements?
There are for the steam deck
I've lost a ps5 controller because the usbc connector died in it and now the firmware is too old. I have another ps5 controller with the same hardware fault with newer firmware so its a matter of time. If you want them to play around with you can have them. The pads are tore on the port because the solder flood doesn't make a good mechanical connection for the report.
New years day 2018 will never be the same
Ben has been holding out the truth about analog sticks potentiometers for this long!
I love how mischievous Bud is
Based on evidence presented, it looks like the carbon film is getting melted, causing the grease seen. And then further friction easily erodes it in its softened state.
When Ben scraped the carbon film off manually, it produced a lot of clingy debris.
However: There was none of this debris when first opening up the module. Just the mysterious oil.
When Ben overcharged the stick, the oil became present, and some of the Carbon looked as though it were misshapen.
If I wanted to confirm it, I'd ask Ben, when he was overcharging them did he match positive/negative as it would be in an unmodified controller? Because it's strange that it appeared on the outer ring. But I suppose, past of least resistance.
I have heard of a hal effect joysticks. That would solve the issue with pots wearing out. And I'm sure the pots are being made cheaper. My oh Xbox controllers still work flawlessly but my 360 controllers all get stick drift in about a couple years of play. I'm sure the joycon sticks are cheapest, it's like they get drift out of the box
At the least they should be giving us adjustment pots, if not going full hall-effect for the prices they charge these days. They've known they've been selling us bad sticks for 15+ years. great vid as always, but this topic always makes me a bit angry. 🤣
I dont understand why they don't use coils same principle as slot car controllers etc
Ben, we met at Maker Faire. I'm a longtime friend of Jeri's.
Anyway, I have invented a way to make driftless joysticks, NOT using Hall effect [which is the current best way].
My invention is something you could scratch build in a couple of days.
It has a lower BOM.
50/50 everything when we talk about it.
PS: I named Darlington.
There are basically 3 technologies for potentiometers: Carbon film, wirewound and conductive plastic film. Wirewound would be real hard to do in this size and expensive, but I'm surprised they don't use plastic film, as it used in high reliability applications as it has much more wear resistance. I guess carbon film is "good enough" and cheaper than the plastic film pots.
Carbon film == "planned obsolescence" == more money on replacement gamepads
If the only difrence between the pots on the controller and the one in your parts bin is max resistance but they only use such a limited airea in the center of the pot couldn't you just add a serface mount resistor to both the outer leads to coppy the larger total resistance.
Or calibrate it.
Were you successful in sourcing the hall effect modules from gulikit? Would love to see how those turn out
I got a couple of the steam deck modules to poke at. The magnet holders have a similar shape to a joystick potentiometer, except the casing has no traces and the bit that has the swiper on it just hold a magnet at the bottom. The actual sensor is soldered to the board directly under the magnet holder casing, and isn't very visible. It outputs between 1.3-2.3v and rests and 1.6-1.7(on my units).
Ifixit have made a video on the modules if you want any more details.
Do you think there would be a way to integrate touch-controller (capacitive) sensor into a joystick? In other words, the joystick would just move a conductive probe over a capactive surface, thus eliminating electrical contact? The Steam Controller tried a capacitive stick, but it wasn't popular - maybe an actual 'stick' that just moves a small conductive 'finger' over an internal surface would do the trick? Love the videos and Bud! Cheers!!
Oh sure, totally doable. You can also use Hall effect sensors, or optical techniques.
@@absalomdraconis Luckily hall effect sensor joysticks in controllers are starting to become a thing
so if they have been using the same type since the ps1, how is my 14 yo ps3 controller having no problems while still being used to this day
The winters in Wisconsin didn't always suck.
On your custom one handed ones, why not upgrade them to Hall effect joy sticks to prevent RMA's?
Because it would consume MCU's which we can't get with these endless parts shotages.
using pots for joysticks in 2022 is planned obsolescence. ps bud is super duper cute
So many Ben Heck videos lately, I feel spoiled