Device Used: PC with i9-12900K, RTX 3090, and 64 GB RAM
Bike: Wahoo KICKR CORE with Zwift Ride Smart Frame
Connectivity and Sensors: I’ve got a Garmin Instinct 2 for heart rate monitoring. That and the trainer are connected to my phone (which is running the Zwift Companion app) via Bluetooth, which is connected to my PC via a USB-C cable.
Network Environment: My PC is connected via Ethernet cable.
Region/Country: United States
There might not be a real problem here, it might just be bad form on my part, but I’ve noticed my power is extremely spiky and inconsistent despite my cadence not changing that much. I didn’t think much of it until I saw some pictures of other people’s power charts and I was shocked by how smooth they are.
I’m not being very consistent since I’m usually doing something else while biking, but should my power really be that erratic when my cadence is only varying by 10-15 RPM or so? I’m wondering if something is wrong with my equipment. I’ve attached a picture of the graph from a recent ride. I don’t think there’s any problems with connectivity, since I tried uploading logs to Zwiftalizer and it didn’t report any problems there.
If your phone is close enough to the PC to be connected with a cable, why aren’t you connecting directly to Zwift on the PC? And why are you connecting the phone to the PC with a cable at all?
My PC doesn’t have a decent Bluetooth adapter, but my phone does, so in order to run Zwift on my PC I’m using the phone as an intermediary. As far as I’m aware there’s no way for me to directly connect the trainer to my PC via a cable.
Disable the Bluetooth adapter in the PC by going into Device Manager, right-click on it, and choose Disable. Do that after you unpair any other stuff like keyboards, mice, headphones. Then stick in one of these: UB500 | Bluetooth 5.4 Nano USB Adapter | TP-Link
I’m not 100% sure this will help but it is a reliable and well-traveled adapter. You could also upload a Zwift log file to trainerdx.com and see if it identifies any problems like Bluetooth dropouts.
I did try uploading a log to Zwiftalizer already and it didn’t find any connectivity issues. Is that other site more reliable? If so I can give that one a shot and see if it notices anything.
Another thing to try would be shut off Zwift, run the Wahoo app, do a ride in that and see if it shows similar problems, then talk to Wahoo about anything unexpected that you see there.
Hard to tell, what’s the scale? Whats the min max of the peaks and valleys? What route, looks like a big climb? What cadence is associated with the power value? Lots of questions…
Most of the ride was between 70 and 85 cadence (average 77), with an average wattage overall of 112 and a peak wattage of 296. The lowest of the lows there is around 40, excluding the ones where I stopped temporarily. It was about half of Big Loop, so mostly fairly steep uphill, though I have the trainer difficulty moved down pretty low so the grade wasn’t affecting it as much as it normally would.
I’ve noticed that it’s extremely difficult for me to maintain a consistent wattage even if I am paying attention. If I keep my cadence stable my wattage tends to abruptly start dropping after a few seconds even if I’m keeping the same cadence.
I’m not in great shape, but I’ve been putting in pretty decent time on the bike for the last six months. Apparently I’ve put in 1,536.1 miles and a total of 5d, 1h, 56m in rides since getting my bike back in June.
I’ve been surprised (and pretty disheartened) by how little progress I’ve made according to my Zwift stats since I’ve definitely felt like things have gotten a lot easier, but I figured I must have just been doing something wrong with my workout intensity.
If someone’s power graph is very smooth, then it will have been artificially smoother by the trainer. Some trainers have a setting to do this, and the resulting smooth graph is much more like their workout power target rather than their actual power output.
Even pros don’t have smooth power output in reality.
I’m not saying that there can’t be another issue here, but ignore those super-smooth power graphs.
I would take Pauls advice and get a bluetooth adapter.
I was using my internal bluetooth and was a fairly new PC so should of been no issues but it did randomly start having connecting issues on first pairing after about 18 months. Made no sense but given the adapters are so cheap i bought one and its been solid since.
ps stick the devices me mentions i have tried another in theory newer/better and it was not as good.
I don’t think it unusual for power to spike like that. A change in 10-15 rpm is quiet a big difference on a flat course. Power just normally fluctuates during a ride, it is just how it works. On my bike computer IRL I use a 5 second average which smooths out the power figure to stop it fluctuating up and down so that the figure is usable. As a check go to somewhere flat, like Tempus Fugit, keep an even cadence and same gear and see what happens to the power chart. The ups and downs should be less than the figure in the original post although it still won’t be entirely smooth (ERG mode would do that though!).
Part of me thinks the power graph looks erratic because of the low range of power in that example with an average of 112W, coupled with getting little resistance change from gradients with a low trainer difficulty.
