Coospo Heart Rate Sensors

Which heart rate sensor should you use? Is an optical one enough, or do you need an ECG sensor – in other words, a classic chest strap? In this article, I’ll show you which technology suits which kind of training and what their limitations are. I’ll also compare them with an optical sensor on the wrist. On top of that, I’ll review the Coospo products, and I’ve prepared charts comparing the sensors in interval and endurance training. That’s where you’ll see where accuracy matters.

⚖️ Comparing the technologies

First, I’d say you only need to worry about all of this if you actually analyze your heart rate data or train by it. If it’s just a number you like looking at but never act on, it’s really a pointless purchase. Depending on what you expect from a sensor and how accurate you need the results to be, you can also choose a technology that’s good enough for you and cheaper than going straight for the most expensive one, with the most accurate data you don’t actually need.

The first and less accurate technology is the optical heart rate sensor. It can be worn either on the wrist, often as part of a smartwatch, or on the forearm or upper arm, where it can be more accurate. It works by shining light through the skin. To simplify a lot, the sensor tracks how light is reflected by the blood and uses the blood’s movement to detect your heart rate.

This technology naturally has several limitations. It all rests on an algorithm that tries to estimate your heart rate as accurately as possible, so the result depends on both the hardware (how many LEDs the sensor has) and the software (how good the algorithm is). The basic setup is two green LEDs. For better data, though, manufacturers are starting to use more LEDs – up to five – and different types of light to make the result as accurate as possible.

Smartwatches usually measure heart rate on the wrist, which is one of the most problematic places to do it. The wrist moves a lot, doesn’t have such good blood flow, and it’s not easy to tighten a sensor there just right. On top of that, the wrist is full of bones that the algorithm has to account for in its calculation. Accuracy suffers. That’s why optical sensors for the forearm or upper arm are appearing, and they can be much more accurate. The upper arm is closer to the core, so it has better blood flow, the cold doesn’t affect it as much, and above all it stays relatively stable, whereas the wrist is constantly moving. There’s simply far less interference.

But accuracy isn’t affected only by where the sensor sits. Skin tone or a tan can matter too – darker skin is harder to shine light through. Body hair causes the same problem. Cold also plays a role, especially on the wrist if it’s uncovered: the hand doesn’t get as much blood flow, and you can see it in the results. Limb movement has a big influence, too. A sensor behaves differently when you’re riding a road bike with your arm almost still than on a descent, while running with swinging arms, or whenever you’re using your arm actively in some other way.

An ECG sensor works on a completely different principle. It doesn’t try to estimate your heart rate from reflected light; it reads the actual electrical impulses from your heart. Of course, an algorithm still calculates the heart rate here too, but it’s based on a real signal, so the result is much more accurate and less prone to errors. It’s attached to your chest with a strap. The strap has a conductive layer on the back that carries the signals to the sensor. For more accurate data, it’s best to wet the conductive layer a little – it’s pretty cold at first – but once you’re working hard your body moistens it by itself. So if you want truly the most accurate data, a chest strap is the only option.

The downside is that it isn’t as comfortable to wear. You have to get the strap adjusted so that it doesn’t slip, but also doesn’t squeeze too tightly or get in the way of your breathing. It squeezes your chest a bit, and with more movement and sweating the strap rubs against your body, so chafing, skin irritation or even a rash is more likely. Basically, it’s not something I’d want to take to a race. That’s exactly why I was looking for an accurate alternative that doesn’t require wearing a strap around my chest.

💪 Coospo HW9 Armband

Coospo is a Chinese manufacturer offering a variety of electronics for sports. Besides heart rate sensors, its range includes bike computers, power meters and even electric pumps. Compared with established brands, it naturally stands out on price, but we have to see whether the quality is there too.

The Armband is their optical sensor for the forearm or upper arm. With the strap, it weighs only 20 grams, so I barely notice it when wearing it. I often forget I have it on at all – at the ultramarathon I forgot about it completely. It has a built-in rechargeable battery (so it’s not a replaceable coin cell), which means you have to charge it with a proprietary cable with a magnetic connector, with USB-A on the other end. The magnet holds well. I’d prefer USB-C, but I understand this solution helps with waterproofing. Battery life is 35 hours, so it handles a normal training week, for many people even a month, and it lasts through most ultras. You’ll only have to recharge it while bikepacking.

The optical sensor has two green LEDs. Newer sensors from competitors have as many as five, so they will be somewhat more accurate, but also more expensive. I’ll show how the Armband compares with an ECG sensor in terms of accuracy in the comparison section.

The box contained two straps – one spare, which I think is nice. Both were size M, and according to the e-shop there are no other sizes. The maximum circumference is about 32 cm. If you have really big arms, you can wear the sensor on your forearm instead of your upper arm. You can also buy spare straps, which is a great basic thing to have.

The sensor is IP67-rated, so it survives brief submersion in water. It isn’t meant for swimming. To connect it to a watch or bike computer, you can use ANT+ or Bluetooth. It switches on and off with a button, and then you simply pair it – I paired it with a bike computer, a watch and a phone.

I recommend pairing it with your phone and downloading the app. The sensor has an LED with five colors that changes depending on which zone you’re in. In the app, though, you have to set your maximum heart rate, from which the zones are calculated. Unfortunately, you can’t edit the zones manually; they’re always calculated automatically from your max. Personally, I haven’t found much use for the colored LED.

I also recommend fine-tuning the alert in the app. The sensor starts vibrating if you exceed a certain heart rate. You can either set your own upper limit or simply turn the alert off. At the Zapomenuté hory ultramarathon (“Forgotten Mountains”) I hadn’t set it up at first, and when I had to push hard on one section, the sensor vibrated for almost an hour because I was above the default limit. So in the end I turned the alert off. Set up properly, though, it can be a great feature that warns you during a race that you’re going over your ideal heart rate.

Overall, I find the device simple and well made. It sits nicely on my arm, and I had no problems pairing it with several watches and bike computers or with my phone. The app, even though it has a very low rating, didn’t seem problematic to me at all. In fact, it looks well designed and is easy to use – with some Chinese manufacturers the hardware is fine, but the software isn’t worth much.

My only complaint is that on my first runs I had dropouts in the measurement. At first I thought it was down to the watch, because it hadn’t happened on the bike. In the end I moved the sensor from near the shoulder to the middle of my upper arm (which Coospo itself recommends) and turned it from the outer side of the arm to the inner side, where my veins are more visible and my skin is lighter. That solved the problem.

I wore the sensor at the Zapomenuté hory ultramarathon, where it ran for about 15 hours. I had no dropout problems there at all; it worked great and never got in my way. In general, I don’t like wearing a chest strap while running, and this is a great alternative. I probably won’t take it to a bikepacking race, because I’d have to take care of charging, carry the proprietary cable, and the data from my wrist is enough for me there anyway.

Get in on Coospo.com

♥️ Coospo Realzone H9Z

The H9Z is a classic ECG sensor, a.k.a. chest strap. I don’t think accuracy differs much between manufacturers here, and there isn’t much to reinvent. Still, there are a few nice touches I noticed compared with the competition.

The chest strap itself is nothing special. There’s only one size, and I’ve never found a groundbreaking difference between manufacturers. Another advantage is, of course, that there’s a common standard and straps and sensors are generally compatible with each other, so finding a replacement isn’t a problem.

This sensor is rechargeable. In practice that means it lasts less – about 50 hours, versus maybe 200 for a coin cell – but you don’t have to replace or keep buying batteries, and you can be more confident that water won’t get inside. 50 hours is a decent battery life that’s more than enough for regular training, so you don’t need to recharge it very often.

What I like is that, thanks to the LED, I know right away that the sensor is on. There’s no button; the sensor switches on by itself when the electrodes pick up a signal from your skin. You don’t even have to put it on your chest – just touch the electrodes with your fingers to wake it up, and the logo lights up and the sensor beeps. With my Polar chest strap, which I used for eight or nine years, I often couldn’t tell whether it was on or simply hadn’t picked up a signal. I really missed this.

Like the arm sensor, this one lights up in different colors depending on your zone, and you can also set an upper limit for the alert in the app. I don’t see the colors on my chest as a particular advantage, though – I’m not going to look down at my chest; I usually have my heart rate displayed on my watch or bike computer.

With the strap, it weighs 43 grams and charges via a magnetic cradle with USB-C. Dust and water resistance is IP67 – it survives brief submersion. The whole thing is quite light and plastic, but they all are. I’m curious how long it will last – my Polar served me for over eight years.

Here, too, I’m happy with the quality after the first few uses. Pairing works well, the app is OK, and as for accuracy, it was exactly what I’d expect from intervals. I can’t test it against a certified device, though. I compared it mainly with the optical sensors, and by eye the data looked good. The manufacturer claims accuracy of roughly ±1%, and the values matched my sense of how hard I was working.

What I find silly is that the app lets you have only one sensor paired – either the arm one or the chest one. If you want to set up the other, you have to forget the first. It’s not really a big problem, since I won’t change the zones and alert that often, but it’s an unnecessary limitation. You can then pair both sensors with your watch or bike computer at the same time (they support both ANT+ and Bluetooth), so you can switch between them as needed.

Get in on Coospo.com

📈 Accuracy comparison – which sensor to use when

For testing, I compared the following devices:

• Garmin Fenix 6 – optical sensor on the wrist; it should be said that newer models have a more accurate optical sensor
• Coospo HW9 Armband – optical sensor on the upper arm
• Coospo Realzone H9Z – ECG on the chest (my reference, since ECG is always more accurate)

I tested the three of them in three different scenarios:

• Cycling intervals
• Cycling endurance
• Running endurance

Cycling intervals

Intervals are where it really counts. Heart rate jumps up and down quickly, and that’s where it’s hardest to capture the changes. I did five one-minute intervals at the highest power I could manage.

The Fenix on the wrist simply couldn’t handle it. It often didn’t register the intervals at all, or registered them with a delay. On top of that, it was fairly cold and I was wearing a short-sleeved jersey, which is partly why the Fenix was so far off during the intervals. From experience I know that sometimes the watch captures heart rate well and the graph looks quite decent, but at other times it deviates from reality significantly.

The Coospo arm sensor and the ECG were much more in line, and apart from a few spikes the curves look almost identical. Still, you can see that the arm sensor “drifted” in some places. They aren’t differences that would fundamentally affect how you evaluate the workout, but they’re there. The most prominent spike came at my maximum heart rate: the optical sensor on my arm showed roughly 193–194 bpm, while the chest strap showed around 180. That’s a difference of about 13 beats, which is quite a lot when you’re determining your max heart rate. If I wanted to set a new max heart rate, for example, I would always measure it with an ECG sensor only. The optical one can have anomalies, which is why I definitely wouldn’t change my zones based on it.

So if you want to do intervals, an optical sensor on the wrist may not be enough, especially in worse conditions. An optical sensor on the upper arm is good enough for them, though.

Cycling endurance

On the endurance ride, in the endurance zone, it was warm, and even so the Fenix on the wrist was off in places. In some spots it overestimated my heart rate a bit, in others it was somewhat lower than the other sensors. You can see it best at the moment I stopped for a short while: the arm sensor and the chest strap reacted to the drop in heart rate very quickly, but the optical sensor on the wrist lagged behind and responded slowly to the change. A similar delay shows up in other places where my heart rate was dropping. The algorithm, which first has to calculate the value, may be to blame.

But if I look at the average heart rate for the whole ride, the difference isn’t that big – the wrist just reacts a bit more slowly. So if you ride endurance rides where your heart rate doesn’t jump around much, the accuracy on the wrist is, in my opinion, good enough and you don’t even need an extra sensor.

Running endurance

Unfortunately, I didn’t have much energy for running, because I was still wrecked from Zapomenuté hory. So it’s only a short test, about 12.5 minutes. Even so, there’s an interesting difference to see.

At one point the optical sensor on the wrist reported around 150 bpm, which I think was complete nonsense. I was running really easy, with no special pace, so I definitely wasn’t at that level. The arm sensor showed about 13 beats less, and the chest strap another 15 or so less. So the difference between the wrist and the chest strap was about one heart rate zone, maybe two. But the longer I ran and the more I settled into my usual pace, the more the values converged and the more similar they became.

So in general, it’s often not that the less reliable sensors show wildly different data. They just take a bit longer to get to the value that should be there.

👨‍⚖️ Conclusion

Overall, I was pleasantly surprised by Coospo’s products. The build quality seems good, and there are several small touches I make use of. Pairing works flawlessly, and the app is functional too. The dropouts with the arm sensor surprised me, but once I fine-tuned the placement I had no more problems, and during a 15-hour run it behaved properly.

The arm sensor is a great alternative for me, especially for running and longer events. Its accuracy is at a very good level, with minimal differences compared with ECG. I’ll definitely use it for all my runs. I often run in the cold wearing short sleeves, when the wrist doesn’t get as much blood flow and measuring there isn’t great, and I don’t like wearing a chest strap while running anyway. That’s exactly where I notice the difference, even in endurance activities.

On the bike, I’ll probably stay with the chest strap. It doesn’t bother me much on the bike, and it’s slightly more accurate. Especially for intervals, where heart rate goes up and down. Plus I get a real value for my max heart rate from it – I wouldn’t trust an optical sensor for that because of the spikes. For endurance rides, though, I’ll happily use the arm sensor when I can’t be bothered to put on the chest strap.

If I could only have one sensor, I’d probably choose the one for the arm. It’s more versatile – I can use it for running, cycling and intervals – and the difference in accuracy isn’t that big. A chest strap is mainly for when you want the highest possible accuracy, because ECG is always more accurate. For most people, though, an optical sensor on the upper arm is perfectly sufficient.