#5 The Deep Dive Fix Files: Week Ending 26 June 2026

This is a weekly roundup of significant sports tech problems that the endurance community has encountered and, where possible, resolved. Each entry covers the issue, the affected hardware or software, and the current best fix or workaround known to the community.

Garmin Fenix 7, Forerunner 970 and Fenix 8 route calculation error missing routable roads when rerouting

On the Fenix 7, Forerunner 970 and Fenix 8, the watch returns Route calculation error. Missing routable roads in this area when it recalculates a route. It happens when you leave a planned route, when you ask for routing back to start, and when you reverse direction, so the recalculation fails on roads and paths that are present on the map. In some cases the map page then renders dark with no paths and the watch reboots. Owners on this thread report it across the Fenix 7, Fenix 8 and the Forerunner 970, on paths that show normally in TopoActive.

The fault sits in the TopoActive map data and the routing engine that reads it. The same paths route correctly one moment and fail the next, and the error appears on devices where TopoActive was never updated, which points at a routing algorithm change rather than a single corrupt map file. It reaches beyond Europe. Owners in the United States and Australia report the same failure.

The workaround owners confirm is to replace TopoActive with a free alternative map. Disable TopoActive in the map settings, install an OpenStreetMap base map and a contour map for your region, keep the Worldwide DEM basemap active, then reboot. Route calculation drops to seconds and recalculation works when you leave the route. The custom maps from alternativaslibres.org are the set most owners on the thread used.

There is a partial fix without changing maps. Stop completely, request the route, and stand still until the calculation finishes, which lets it complete more often. It is worth noting that spoken turn by turn guidance may not play on a route calculated on the watch, so set routing to Use Map if you rely on the map page rather than voice prompts. Did replacing TopoActive with a custom map restore reliable recalculation on your watch?

Garmin Venu 4 logging sleep before bedtime and missing the whole night

On the Garmin Venu 4, the watch can log sleep before you actually go to bed. If you lie down to relax in the evening, the small drop in movement and heart rate is read as the start of your main sleep. When you get up to go to bed, that movement is read as waking for the day, so the real night that follows is recorded as awake time and dropped from the analysis. Owners on this thread describe the same pattern with a correctly set sleep schedule and Sleep Mode off.

The result breaks the rest of the day. No sleep stages are calculated, the sleep score falls to a very low number, and Body Battery never recharges, so it sits near its minimum. The watch holds the raw heart rate and movement data, but the algorithm will not reprocess the night once it has locked.

The avoidable trigger is reclining before your real bedtime. Stay upright in the evening and go to bed when you intend to sleep, which stops the watch starting sleep early. The bug is the false early start, so removing that start is the reliable workaround until Garmin corrects the detection.

Editing the sleep times in Garmin Connect changes the duration only. It draws a flat block with no stages and does not rebuild the score or Body Battery from the stored data. It is worth noting that stable software 17.33 fixed several Venu 4 faults but did not address this one, so a corrected build is still pending. Did avoiding any lie down before bedtime keep your Venu 4 sleep record intact for the full night?

Suunto Race S and Race 2 pool swim distance dropping after you save the workout

On the Suunto Race S and the newer Race 2, the pool swim total is correct during the swim and then falls when you save. The watch counts every length and shows the right total on the wrist at the end of the session, for example 2150 yards, then records about 3.75 percent less once the activity is saved and synced, so 2150 becomes 2069. Owners on this thread report the same fixed percentage in both yards and metres, on firmware 2.50.26 and 2.53.52.

The reduction is applied after the count, not during it. The per length data is right, and the watch displays the correct accumulator before you end the session, then subtracts the percentage when it writes the summary. This is a regression. Owners who used older Suunto watches such as the Spartan Ultra did not see it.

The correct distance is still in the file. Open the activity in the Suunto app, go to the lap or interval data at the bottom, and add the individual lengths, which sum to the true total rather than the reduced summary figure. That recovers the real distance for your training log while the summary stays wrong.

Setting the swim styles correctly is worth checking, since mismatched stroke detection is a separate known fault, but owners report the cut even when the styles match. A Suunto moderator on the thread states the distance error will be fixed in a future update, so the manual total is the interim answer. Did summing the lap lengths in the app give you the full pool distance on your Race S?

COROS Pace and Apex watch dead and will not power on after the May 2026 firmware update

Owners report COROS watches that will not power on after the firmware released in the middle of May 2026. The failure is not always immediate. Some units die within hours, others go dark days or weeks later, and several owners describe a sharp drop in battery life or the digital dial freezing in the hours before the watch stops responding. The fault spans current models across the Pace and Apex lines. Coverage of the reports traces the deaths to that single firmware release rather than to charging hardware or user error.

COROS acknowledged the problem after owners raised it publicly, and the company chief executive replied directly on Reddit. The first response offered affected owners a 10 to 20 percent discount on a replacement, which most rejected as inadequate for a watch that failed through a firmware fault. On 17 June 2026 COROS changed position and confirmed it will replace any device that died after the update.

If your watch still powers on, run a battery calibration before anything else. Drain it below 10 percent or until it shuts off, charge it without interruption to 100 percent, leave it on the charger for a further 30 to 60 minutes, then restart it. If behaviour stays wrong, a factory reset through System, then More Settings, then Reset All clears a corrupted state. It is worth noting that a watch which has already gone fully dark and will not charge is unlikely to recover from either step, and that is the case COROS has agreed to replace.

Before you open a ticket, record your firmware version, the date you installed it, and whether the dial stopped responding ahead of the failure, because that sequence is the pattern COROS is tracking. Quote your 6 digit watch ID in the ticket so support can match your unit to the affected firmware.

Did a full drain and recharge revive your watch, or did COROS replace the unit?

Garmin Forerunner 970 rebooting when starting an activity after software 17.33

On the Forerunner 970, and on the 570 that shares its software, stable version 17.33 makes the watch reboot the moment you start an activity. It happens across every sport profile, running, cycling and the rest, and it repeats on each attempt, so the watch cannot record a session. Some owners also see the screen switch off and on during normal use, and one reports the watch restarting whenever the Garmin Explore app is opened on an Android phone. Owners on this thread confirm it began with 17.33.

The crash centres on the GPS start, the same area behind the wider reboot problems Garmin worked through last year. It is not limited to one unit. Several owners report the same behaviour, and the symptoms appear on other models running 17.33, which points at the release rather than failing hardware on a single watch.

The workaround owners use is to start from a profile that does not call GPS. Selecting Indoor Track or Treadmill lets the watch begin and record an activity without the reboot, which keeps you training while a fix is prepared. Where the watch reboots in daily use, clearing the temporary cache memory through a restart returns it to a working state.

Garmin has acknowledged the reports on the forum and is emailing affected owners for diagnostic detail to add to a ticket. Owners have asked to roll back to the previous build, and that route is closed, because Garmin does not release older firmware for manual install. The practical options are the indoor profile workaround or waiting for a corrected version.

Did starting from the Treadmill profile let you record a run without the watch rebooting?

Garmin Edge 1040, 840 and 540 battery percentage wrong after firmware 31.29

On the Edge 1040, including the Solar, and on the 840 and 540 that share the firmware, software 31.29 reports the battery level wrongly. The percentage can climb during a ride rather than fall, with one owner watching it rise from 84 to 98 percent across a four hour ride, and on other units it sticks at 100 percent for a whole day. Swiping down to the status screen shows the same wrong figure. Owners on this thread report it was raised during the beta and remains in the stable release.

The fault is a safety concern for long off road events. On a bikepacking or audax route the battery reading is what tells you whether the device will last the stage and hold a single track log to the finish, and a meter that reads high removes that warning. The error is in how 31.29 tracks the percentage after charging, not in the cell itself, because the hardware discharges normally while the datafield disagrees.

The workaround owners confirm is a full power cycle. Turn the device off and back on before the ride, and the meter resets to the correct value, with one owner reporting an accurate 44 percent immediately after the reboot. For multi day use a second method holds the reading closer: ride down to about 20 percent, charge while riding back above 90 percent, then repeat, which keeps the meter calibrated across the event.

It is worth noting that a route or course file already loaded continues to navigate normally, so the bug affects the battery display rather than the ride recording or the map.

Did a power cycle before the ride keep the battery percentage accurate to the end?

Garmin Edge 540, 840 and 1040 Strava Live Segments not syncing after firmware 31.30

After updating an Edge to stable firmware 31.30, starred Strava Live Segments stop syncing and no longer appear on the device. The Training menu shows a generic Segments entry where Strava Live Segments used to sit, and re-starring or reconnecting accounts through Garmin Express does not bring them back. Owners on this thread report it on the Edge 540, 840 and 1040, and Strava support has told users the cause is Garmin firmware 31.30.

Strava released a backend fix on 19 June 2026. Most devices recover on their own after a few more syncs with Garmin Connect over the following days, so the first step is to sync again and check the Training menu before changing anything.

If segments are still missing, un-star every affected segment in the Strava app, sync Garmin Connect, then re-star them and sync again. One owner restored both the starred segments and the Strava Live Segments entry under Training this way after nothing else worked.

When the re-star cycle fails, remove the device from Garmin Connect and add it back. On the mobile app this is a single remove and re-add. Through Garmin Express you also need to remove the device from Garmin Connect Web in between. Afterwards confirm that Training and Planning, then Segments, is set to Strava Segments rather than Garmin Segments. Did un-starring and re-starring your segments bring Strava Live Segments back on your Edge?

Garmin Fenix 8 and Fenix 8 Pro battery drain after software 22.35

After updating to stable software 22.35, owners of the Fenix 8, Fenix 8 Pro and Fenix 8 Solar report a sharp rise in battery consumption, with the battery manager attributing almost all of it to System/Other. A restart helps for a short time, then the drain returns. Some owners measure around 7 percent an hour while the watch sits idle. Owners on this thread confirm the change tracked 22.35 rather than any change in use.

Garmin acknowledged that 22.35 carries a battery drain bug. The 22.38 update began rolling out on 23 June 2026, but its changelog lists a battery fix only for the Tactix 8 and does not name the Fenix 8. Several owners on 22.38 still report severe drain, including after a full factory reset.

The workarounds owners describe are limited. A restart clears the drain temporarily. A full factory reset, rather than a restart and re-pair, held normal consumption for about four hours for one owner before it crept back. One owner reduced the loss by connecting the watch to the phone for only a few minutes each morning and then disconnecting. The earlier 22.35 notes also addressed drain after using the phone assistant or accepting a call on the watch, so avoid those actions while you remain on that build.

If your watch holds charge acceptably on 22.35, there is a case for staying on it until a Fenix specific fix ships, since 22.38 targets the Tactix 8. It is worth noting that Garmin asked owners still seeing the drain on 22.38 to report back so it can collect data from affected units. Did 22.38 and a factory reset return your Fenix 8 idle drain to normal, or is System/Other still consuming the battery?

Garmin Edge 1050 showing the wrong Varia RearVue 820 battery level with a second light connected

The Edge LIGHTS battery field is meant to show the lowest battery among the connected lights. With a Varia RearVue 820 connected on its own, the Edge reports its level correctly. Add a second light such as the UT800 and the field switches to the headlight instead of the lower radar, so an RV820 sitting near 30 percent can read close to 80 percent on the Edge. Owners on this thread describe it on the Edge 1050 with the RV820 paired alongside a headlight.

The RV820 connects to the Edge over BLE as a radar, but it uses ANT for the light controls and battery reporting. The battery roll up across the light network stops returning the radar’s lower value once a second light joins, so the figure on screen tracks the headlight rather than the unit that will die first.

There are two ways to read the true level. Connect only the RV820 before a ride, where it reports correctly, or keep the Varia phone app open during the ride, which shows the radar battery and gives the audible low alert that the Edge speaker volume can miss. The third party Connect IQ battery fields written for older Varia units do not yet support the RV820.

Until the reporting is corrected, charge the RV820 every two to three rides rather than trusting the Edge figure. It is worth noting that the wrong reading only appears when more than one light is connected, so a radar only setup shows the correct level. Does the Edge battery field show your RV820 level correctly when it is the only light connected?

Wahoo Kickr v6 disconnecting and dropping power in Zwift after firmware 5.6.13

After the 5.6.13 firmware reached the Kickr v6, the trainer began dropping its connection at random during Zwift sessions. The drops do not depend on how the trainer is connected. Owners report them over Bluetooth, over WiFi, and over Direct Connect, and across Windows, Mac, Apple TV, and iOS. Each drop lasts ten to twenty seconds, which is long enough to lose contact with a race. The update was pushed as a forced over the air install with no option to decline or roll back, and several owners on this thread had never seen a single drop across tens of thousands of kilometres before it landed. A few Core 2 owners describe the same behaviour.

Wahoo described 5.6.13 as a stability release with no new features. There is a second pattern worth separating from the in ride drops. The trainer disconnects silently while it sits idle in the pairing screen, so a setup completed five to ten minutes before the start can be dead by the time you begin to pedal, with nothing on screen to flag it.

The recovery owners rely on is to re-pair immediately before the start. Keep the session active rather than leaving it idle, confirm in the pairing screen that the trainer is still connected just before you enter the pen, and re-pair if it has dropped. This does not stop the random in ride drops, but it removes the idle disconnect that catches most people at the gun.

There is no supported rollback. Owners who asked Wahoo support to revert the firmware were refused, so the practical options are the re-pair routine and a support ticket. It is worth noting that attaching Zwift log files to that ticket is what affected owners have done to push Wahoo toward a fix. Did re-pairing the trainer immediately before the start hold the connection through a full race on your v6?

Suunto app 6.10 route library not loading and not syncing routes during an activity

On Suunto app version 6.10, including the 6.10.6 and 6.10.7 beta builds on Android, the Routes tab in the Library stops loading. The list hangs, and in several cases it jumps back to the map view instead of opening the saved route. Some planned routes take a long time to appear at all. Owners on this thread report it across the Vertical, Race, and Race S after updating the app.

The same release removed a function that worked before. You can no longer create a route in the app and sync it to the watch while an activity is recording. During a session the app now reports no connection to the watch, so a route planned mid run or mid ride does not reach the device, even when the route already existed and simply needed syncing.

For the Routes tab, disconnecting the app from the watch and reconnecting restores the list. The fix is temporary and the fault returns on a later session, so treat it as a way to load a route you need now rather than a permanent repair.

For the in activity sync, the only route owners have found is to stop the activity, sync the route, then start a new activity, which splits the recording into two files. It is worth noting that the Suunto community manager asked owners to send any route that crashes the app through the Sports Tracker web login, with the same account, so the specific files can be traced. Did disconnecting and reconnecting the app keep your Routes tab loading for more than one session?

Garmin Edge 1050 tone volume resetting to zero after power off

On the Edge 1050, the tones volume drops to zero every time the unit is powered off and back on. The device then starts the next ride with the turn prompts, alert tones, and lap beeps all silent, until the slider is raised by hand. Owners on this thread describe resetting the volume from System, then Sounds, before every ride.

Several owners trace the reset to the paired phone. On a Pixel 10, turning off Bluetooth on the phone or on the Edge stops the volume resetting on later power cycles, although it can still fall to zero on the first cycle after disconnecting before it holds. An older Samsung S22 Plus kept the volume at maximum across power cycles. That points at the phone, or a sound setting on the phone, rather than the Edge on its own.

The owner who started the thread reports the volume dropping to zero even with the phone powered off, which puts a firmware fault on the unit alongside the phone trigger.

Until a corrected build arrives, set the volume in System, then Sounds, confirm the tone categories are enabled, then power the unit off and on once and check the slider before you rely on it. Where the reset tracks the phone, turn Bluetooth off, re-pair, and confirm the volume holds after one full power cycle. It is worth noting that the affected owner has a support ticket open, which is the step to take if neither check holds the volume. Did disconnecting and re-pairing your phone stop the tone volume dropping to zero after a power cycle?

More from last week’s Fix Files roundup.

Browse every issue in the archive at the Deep Dive Fix Files hub.

Last Updated on 27 June 2026 by the5krunner