TMR Sticks on PS5: The 2026 Aftermarket & DIY Mod Guide

The stock DualSense uses a rotary potentiometer pair inside each thumbstick gimbal. Two carbon tracks sit under sprung wipers, and the wipers scrape across the tracks every time you move the stick. The resistance value at each angle becomes the analog reading the W3C Gamepad API exposes to your browser. The mechanism is cheap, well-understood, and was the standard from the original DualShock 1 through the DualShock 4 and into the current DualSense. It has one inescapable flaw: the wiper wears the track, and that wear shows up as drift in the idle reading and a noisy curve under motion.

A TMR stick module replaces the entire sensor stage with a non-contact magnetic sensor. The gimbal still moves mechanically, but instead of a wiper sliding across a track, a small magnet on the stick shaft passes over a tunneling magnetoresistance sensor IC soldered to the module's daughter PCB. Nothing touches anything else electrically. The IC reads the magnetic field angle and outputs an analog voltage close enough to the original potentiometer output that the DualSense main board treats it as a drop-in.

Why TMR and not Hall effect? Both are non-contact. Both fix wiper wear. TMR sensors from TDK and Allegro Microsystems draw less current and resolve smaller angle changes than a comparable Hall sensor, which matters on a battery-limited controller. Fresh TMR idle noise typically sits at plus or minus 0.002 to plus or minus 0.005; fresh Hall sits at plus or minus 0.005 to plus or minus 0.010. Both are far better than a worn potentiometer at plus or minus 0.03+, but the cleaner TMR option is what Sony chose for the DualSense Edge and what most serious aftermarket vendors stock.

A DualSense potentiometer is a 14-pin SMT part bolted into a metal gimbal frame. Aftermarket TMR modules ship with the same pin pitch and hole pattern so they drop into the existing gimbal without machining. Get the wrong pitch and you will fight clearance issues.

The 2026 sourcing landscape splits into four tiers.

AliExpress and TaoBao. Where 80 percent of DIYers buy. A pair of DualSense-compatible TMR modules runs €25-35 shipped, less on TaoBao if you read Mandarin. The risk is product description honesty. Some listings say "TMR" but ship Hall or rebadged potentiometer parts. Pay only after the listing names a sensor IC (a TDK TMR-prefixed part, an Allegro AAS25-prefixed part) and includes a compatibility chart that lists the DualSense revision, not just "PS5".

Specialist EU vendors. A handful of small EU shops (German, Dutch, Polish) import the same modules in bulk and add QC, English docs, and a return policy. Expect €35-50 per pair. The premium is for the QC, not for a different module. If a vendor will not name the sensor IC, walk away.

US specialist shops. Pricing sits at €40-60 per pair shipped. Same sensor-IC rule.

Premium kits. A few vendors sell €60-90 "drift-proof DualSense upgrade kits" that bundle a TMR module pair, a fresh gimbal frame, a thumbstick cap pair, and a screwdriver set. The kits make sense if your controller has both stick drift and a worn gimbal.

What you should not buy: any listing that markets itself as "TMR upgrade" but lists a price under €15 per pair, omits the sensor IC, and ships from a generic warehouse. If you are tracking pricing across regions, PriceRunner and Skinflint are useful aggregators.

Three tests sort the real listings from the marketing.

The sensor IC test. A real TMR module ships with a sensor IC whose part number is public. If the listing names a part from the TDK TMR family or the Allegro AAS25 family you can cross-check the datasheet and confirm it is genuinely a TMR (or in some cases a high-end Hall) part. If the listing says only "magnetic sensor" or "drift-free chip", treat it as unverified.

The JoyCheck idle test. Install the module, plug in, open JoyCheck, rest both sticks. Idle should sit at plus or minus 0.005 or tighter for fresh TMR. Plus or minus 0.010 to 0.020 fresh is a marketing-TMR or low-grade Hall part. Plus or minus 0.03+ fresh is a relabelled potentiometer and the vendor lied. Document the value the day you install.

The 12-month test. Real TMR shows almost no drift over 12 months of normal use. Hall-sensor "TMR" will hold at plus or minus 0.005 to 0.010. Relabelled potentiometers drift back to plus or minus 0.03+ within months. Keep your baseline, re-test quarterly, and you will know within a year which sensor technology you actually bought.

Vendor red flags: no sensor IC named, no compatibility chart by controller revision, vague "100,000 hour" claims with no source, no return policy, listings copy-pasted across generic shops. Vendors that reference real axis values like those in the MDN Gamepad API documentation generally know what they are selling.

The canonical procedure is the iFixit DualSense controller guide. Follow it. Do not improvise. The summary below is for orientation only.

Tools. T8 security Torx (the DualSense screws have a centre pin so a regular T8 will not seat), Phillips PH00, plastic pry tool, ESD-safe tweezers, soldering iron with a fine tip, 0.5mm leaded solder, flux pen, desoldering braid or hot-air rework station, 99 percent isopropyl alcohol, magnetic mat.

Procedure outline. Disconnect the battery first. Remove four T8 screws from the underside. Pry the back shell along the centre seam. Disconnect the battery, trigger, touchpad, and speaker ribbons. Remove the main board screws. Lift the main board out, taking care not to bend the L1 and R1 bumpers. The stick modules are on the back of the main board, soldered through 14 pads each.

The desolder step. The hardest moment. The pads are surrounded by ground plane copper, which sinks heat. With a soldering iron you will struggle without a wide tip and added flux. With a hot-air rework station at 350-380 degrees Celsius the module lifts in 30-60 seconds. If you have never desoldered an SMT part with through-hole pins, practice on a junk board or pay a phone repair shop €15-25 to do this step.

The solder step. Drop the new module into the same holes. Check orientation against the silkscreen. Tack one corner pin first, verify the module sits flat, then solder the remaining 13 pins. Add flux. Inspect every joint under magnification. A cold joint or a bridge will produce phantom drift that looks identical to a bad sensor.

Reassembly. Reconnect every ribbon in reverse order, battery last. Do not over-tighten the back shell screws or you will warp the plastic. Total time once experienced is 30-45 minutes; the first attempt is 75-90 minutes.

Plug the controller in over USB-C, not Bluetooth, so you remove any Bluetooth SIG wireless polling variability. Open JoyCheck in Chrome, Edge, or any Chromium-based browser. Press a button to wake the API.

Idle test. Rest both sticks centred. Wait three seconds. Read X and Y on both sticks. Fresh TMR should show plus or minus 0.005 or tighter on every axis. If one axis sits at plus or minus 0.010 or worse, you have a cold solder joint, a bent gimbal frame, or a low-grade module. If both axes on one stick show a stuck offset, the module may be rotated; pull and re-check the silkscreen.

Range test. Push each stick to all eight cardinal and diagonal extremes. The JoyCheck trace should reach plus or minus 1.000 cleanly with no clipping or notching. A short reach usually means a bent gimbal frame.

Button and trigger cycle. Cycle every button and sweep both triggers. L1/R1 should click cleanly and L2/R2 should reach 1.000 at full press with smooth intermediate values, the behaviour described in PlayStation accessories support.

Baseline. Note the date, the idle reading per axis, and the vendor and IC of the module pair. Re-test quarterly. Real TMR will hold within plus or minus 0.001 of baseline for years; drift away tells you either a sensor failure (rare) or mechanical gimbal wear.

A note on the lifespan column. TMR sensors are good for tens of thousands of operating hours, but the gimbal frame, the thumbstick cap, the springs, and the click switch under L3 and R3 all wear independently. Replacing the sensor extends the sensor's life by roughly 10x, but the total stick assembly lifespan is sensor-limited only if the mechanical components are also in spec. If your gimbal frame is bent or the springs are tired, swap those at the same time.

The DIY TMR mod is the right move for a specific player and wrong for several common ones.

Buy the DualSense Edge if you have never soldered. The Edge ships with TMR sticks from the factory and replaceable stick module assemblies that drop in without soldering, carrying a full Sony warranty. €239.99 is a lot, but the Edge is the only TMR path that does not require either a tool you lack or a vendor whose claims you cannot verify. The strategic comparison sits in our best TMR controller for PS5 guide.

Buy the Edge if you want warranty cover. Aftermarket modules void the Sony warranty the moment you open the shell. For a base DualSense worth €74.99, the warranty may be worth more than the mod.

Buy the Edge if downtime matters. A botched solder job costs the controller and a day of debugging. If you own one PS5 controller and need it tonight, do not start a 45-minute mod that might take 4 hours the first time.

Do the DIY mod if you have a fleet, basic SMT skill, and want to stretch the gear you own. €25-35 per controller buys two or three more years of clean play if your modules are real TMR. The decision tree across all paths sits in our joystick TMR PS5 strategy guide.

A few questions that come up repeatedly in DM and comments after sibling articles in this cluster.

Will the PS5 firmware reject an aftermarket TMR module? No. The DualSense main board treats the stick output as a standard analog voltage. The firmware has no idea what sensor produced it. Any module that hits the same voltage range across the same mechanical sweep is invisible to the controller's MCU and to the PS5 console. The W3C Gamepad API reading in JoyCheck after the mod will look identical in format to the reading before; only the cleanliness changes.

Can I mix one TMR stick and one stock stick? Yes. The two sticks are electrically independent. You can mod just the left stick if only the left one drifted. Some DIYers mod the left first as a test, verify the result in JoyCheck, then do the right a week later.

Will TMR fix L3/R3 click failure? No. L3 and R3 are tactile dome switches under the stick. They are separate from the analog sensor. If your click failed you need a new switch, not a new sensor. Replace it during the same teardown if you have one on hand.

1. W3C, "Gamepad API specification": www.w3.org/TR/gamepad
2. Mozilla Developer Network, "Gamepad API reference": developer.mozilla.org/en-US/docs/Web/API/Gamepad_API
3. Sony, "DualSense Edge Wireless Controller": www.playstation.com/en-us/accessories/dualsense-edge-wireless-controller
4. iFixit, "Game controller repair guides": www.ifixit.com/Device/Game_Console