Low Heart Rate

I've got reply from Garmin Support Poland and they confess it's a software, not hardware issue. Next update is planned for February.

I have this problem too, as an earlier post of mine attests. But FWIW, Garmin watches as a rule have mediocre heart rate sensors. You can verifty this by checking Quantified Scientist on YT. Lost of people will dispute this but none of them have done the testing he's done.

Unfortunately for android users like myself, or most anyone who values long battery life, only Apple and a very few other watches have acceptable (above 90%) accuracy during activities.

Garmin's latest HR sensor somewhat improved sleep monitoring, as can be seen on Quantified Scientist videos, but appears to have done nothing for HR accuracy during workouts. 

I am not sure if the problems QS picked up are the same as people here are experiencing or just reflect Garmin's generally poor accuracy with the sensor during activities. Either way, the only options are to switch to a better brand or use a chest or arm band. I've reluctantly gone the latter with a Coospo armband.

You can get margingally better accuracy than the Coospo unit by paying two or three times as much for a Polar or other, but the Coospo outperforms even the Apple Watch Ultra, and that's good enuf for me, or at least cheaper than buying an iphone (which I don't like) and an Apple Watch. Or, if you're one of those with a watch that's been broken by a Garmin firmware update (16.22?), you can just accept all your fitness metrics are worthless.

7017726 said:

Lost of people will dispute this but none of them have done the testing he's done.

I did. I compare the HR from the internal HR sensor and from my HRM-Pro strap regularly, and most of the time I can barely find any difference at all, even under the worst conditions - freezing temperatures (making it especially hard for optical wrist sensors), and with interval training, or hill sprints, where the HR changes abruptly.

Although I have a different watch, it is built with practically identical hardware as Epix 2. If you are interested, I posted some details including graphs, from my run two days ago at https://forums.garmin.com/outdoor-recreation/outdoor-recreation/f/instinct-3/400680/anyone-bought-a-instinct-3-50mm-yet-can-share-experience/1883500#1883500 

I know it won't help you, and I understand your frustration, but it is just to let you know that not everyone has as bad experience as you, with the HR on Garmin watches. I owned already several models, and have similarly positive experience with all of them.

2b2bff  Please report back after a run or 3 on if any improvmenent.

Sadly, no change. Did a run this morning. After about 18 minutes of lower HR data a sudden jump to the HR data that I was expecting for the effort. Back to square one...

Currently I'm contemplating to leave the Garmin eco system altogether, I'm very disappointed about Garmins politics not only in regard to that ongoing HR issue. Had a play with a Suunto Race S lately and really liked it. Especially their maps experience and GPS accuracy puts a shame to that Epix Pro. And if I have to wear an additional strap for good HR readings anyway, why in combination with a Garmin?

Yeah, I'm aware the problem doesn't afflict all Garmin watches, or even all Epix watches. But it's damned frustrating for those of us who are affected to be told we're not wearing the watch correctly or it's just the usual limitations of wrist OHR sensors. FWIW, the Epix is my third Garmin watch and the first to see this problem, possibly beginning with firmware update 16.22.

7017726 said:

I'm aware the problem doesn't afflict all Garmin watches, or even all Epix watches

I do not think there is any difference between individual watches, rather differences among users. In the way they wear the watch, in the way they move, their physiology - epidermis thickness, tan, hemoglobin level, blood circulation problems, peripheral vasoconstriction, tattoos,...

So for example, when I do not tighten the strap of the watch sufficiently, I'll get big differences between the Internal and external sensors, and the watch will lock on the cadence easier, as the watch bounces more on the wrist. See below my two activities from today - the first one was done, when I forgot to tighten the watch strap, and the second one just half an hour after the first one, when I tightened it firmer.

I had a Vivoactive 4, a Fenix 6S Pro, a Fenix 7 Pro (for about a week) and now an Epix Pro for several months. So I'm confident that I know how to wear a watch. I cannot be sure the Vivoactive 4 or Fenix 6S Pro have been accurate, but they never had such a problem as the Epix Pro has since late October. 

The first run with the Fenix 7 Pro had the hiccup, the first run with the Epix Pro had the hiccup, but after that I had no problem for months until end of October.

The problem is that the Epix Pro uses incorrect data and interpolates something. And needs a very long time to be convinced the new data it receives is the actual good data. That is totally different from cadence lock, a problem the Vivoactive 4 had sometimes...
Even losen and tighten the watch strap or even pausing and continuing the activity does not make it change its mind...

2b2bff said:

I had no problem for months until end of October.

It could be due to the colder temperatures causing peripheral vasoconstriction, hence it becomes much more challenging for the optical sensor to detect the HR due to restricted blood flow.

2b2bff said:

The problem is that the Epix Pro uses incorrect data and interpolates something. And needs a very long time to be convinced the new data it receives is the actual good data.

Not really interpolating, rather trying to detect and learn the HR signal strongly disturbed by the noise in the weak signal.

I recommend going out and doing a ~10 min warm-up before you start. Record the warm-up as a separate leg of a multisport activity, or trim it later in GC, if you do not wish that it skews the totals. Or just use a HRM chest strap - you'll get not only reliable HR, but also HRV which helps greatly with the accuracy of the respiration rate, LTHR, VO₂max, and all other derived metrics.