Suunto SHRB2 Firmware 3.2.6.8 Update

Inge Nallsson

@VoiGAS said:

Tried it today with wet belt and Merino Shirt. No artifacts

While my latest running session got totally destroyed by HRV faults! OK, I used three layers of polyester on the torso (autumn is starting to bite), but also a 70 mm length of rubber tubing as ESD shielding - the length still doesn’t interfere with the strap electrodes.

I had done a factory reset of the SHRM2 to see if the Flight Time impossibility numbers would correct themselves. The did not on today’s trail run (Avg. 2 ms FT over 6 km…), though I’ll make a separate post about a couple of experiments yesterday.

Even if some ESD circumvented my rubber tube, the disturbance should not have disrupted the HRV readings in such a severe manner as below. My guess is that the HR monitor suffers from major bugs in the software.

Anomalies 22.5%
Artifacts 37.6%

Runalyze R-R graph:

Suunto App. Most of the HR also got corrupted:

Garmin Forerunner 955 control watch. Correct HR:

VoiGAS

@Inge-Nallsson Wow, thats even worse than everything I produced with the first Gen Smart Sensor. Really seems to be the Shirts material, but even then I don’t think Gen2 works better than Gen1. More likely to be the other way round which raises the question what’s behind the claim of having a better signal quality with the new belt

atoponce

@Inge-Nallsson As much as the data was pointing to the module itself, I’ve started wondering if the strap is picking up ESD. Despite every effort I’ve thrown at the problem to eliminate ESD at the module, it still shows up in the R-R interval data.

Regardless, if the ESD is picked up at the module or at the strap, the marketing promised that this would be mitigated in the firmware, but I’m not seeing it. I specifically purchased this HRM because I was battling this with my Polar H10, and the feature that artifacts would be removed was promising! In practice though, where the rubber meets the road, it’s not doing anything any other HRM on the market isn’t doing.

As it sits, ZoneSense isn’t trustworthy for me. When the first 12-15 minutes look like this, how do I know ZoneSense has a good baseline for the rest of the run? This was a 10×3:00 threshold interval run this morning.

plot (4).png

stromdiddily

@atoponce where do you find this graphic?

2b2bff

@stromdiddily Runalyze…

Inge Nallsson

@atoponce said:

As it sits, ZoneSense isn’t trustworthy for me.

Exactly. Also, never knowing if base HR figures are correct or not (without using a control watch) takes the pain too far. Today I packed the SHRM2 up, this my second unit to try, and shipped to the seller asking for a refund. I lack the stamina necessary to continue experimenting and wondering.

stromdiddily

@2b2bff said:

@stromdiddily Runalyze…

haha yeah, got that…there are 500 different tabs and analytics links in Runalyze, which one shows this? The closest thing I can find is aerobic estimation but it doesn’t look like this graph.

snow*

@stromdiddily If you want to take a quick and easy look at R-R data, I recommend

https://smlanal.szmigiel.design/

Just upload the activity’s .json file, and you’re all set

atoponce

@stromdiddily When clicking on an activity, scrolling to the bottom there is a “Heart Rate Variability” box. This is in the new site redesign.

stromdiddily

@atoponce thanks!! I somehow had “hide hrv plot” checked so wasn’t seeing that

this…doesn’t look great for a completely steady treadmill effort

atoponce

@stromdiddily Yup. I have no idea how to fix this. Chest strap HRM’s should be “plug-n-play”, but that’s definitely not the case with this one.

I use electrode gel with the strap, spritz the front of my shirt, and have a anti-static bag taped over the module, and I still get data that looks like this:

Dimitrios Kanellopoulos

If you want HRV analysis also quantified-self.io offers this for free FYI

just upload a normal fit file
Screenshot 2026-10-23 at 18.18.46
Screenshot 2026-10-23 at 18.19.16
Screenshot 2026-10-23 at 18.20.34

matram

My contribution from a very short test run today.

Running top is 97% Polyester.

VoiGAS

@matram Now thats a strange pattern

matram

And a run from later today. This run is with an IceBreaker 200, Merino Wool, top. It does not seem to improve things.

The watch is a Race 2, and Stryd Duo pods are synced as foot pods.

Inge Nallsson

@matram Very strange. From what I’ve seen of your previous posts regarding the SHRM2, you usually have good data from the unit (judging by hearsay; think this is the first screenshot you post).

My first thought went to the old Suunto pod (Smart Heart Rate Belt ‘version 1’) where I discovered “fuzzy ‘ripple’ or ‘echo’ areas” when the module was snapped on upside down:

Suunto Smart HR Belt - Beware module orientation - HRV and ZoneSense
https://forum.suunto.com/topic/13372/suunto-smart-hr-belt-beware-module-orientation-hrv-and-zonesense/9

But, I’ve tested the reverse orientation with the SHRM2 as well, on a short 1 km run, and got perfectly valid HRV data…

Very strange.

matram

@Inge-Nallsson
I have been focused on running efficiency and running dynamics with SHRB2.

I think HR traces looks OK compared to my Polar H10.

I do not really use R-R values or ZoneSense. So any issue that affects only R-R and derived metrics is not a big concern for me.

With regard to running dynamics. I think SHRB2 is a big improvement over the watch but Stryd is generally more accurate and responsive and is what I now use for running form improvements.

Of course if the underlying issue is one of ESD, that is something Suunto needs to address urgently.

VoiGAS

@matram This doesn’t look sooo bad, because you have a visible line. Real bad grapghs have desolved curves, more like clouds of dots.

snow*

This is an outdoor rock-climbing session, during which the SHRM2 was subjected to a lot of stress. I’d say the result is satisfactory.

zoom in the 6:20–8:00 interval, a few hitches

ditto for 13:20-16:00

same setup: lcebreaker Tech Lite shirt + chest strap wet

matram

@VoiGAS
I asked Claude to analyse some recent runs. Claude’s key points …

The pattern is an early detection of a Heart Beat followed by a longer R-R for the following beat where the sum of two beats is accurate, so the problem is not likely to affect HR.

Early incorrect detection of a Heart Beat could be caused by ESD or poor belt contact.

For today’s longer run, the problem appears to go away after 12 min when running with Merino Wool. Which could indicate a problem with electrode wetness, maybe in combination with weather conditions.

For today’s run with SHRB2 there were 4.6% faulty beats. This is likely on the boundary when run-based DFA and HRV computations will no longer be accurate.

We did find one recent run with the Polar H10 which had 2.8% faulty beats. And many earlier runs with <0.5% faulty beats.