Is there a way to calibrate the Purion?

[ilanarama]

New e-bike (Priority Embark) with Bosch Active Line Plus and Purion. The speeds and distances shown on my computer don't match my GPS - yes, I know GPS has inaccuracies, but this is outside what's expected, plus has been confirmed by one of those speed limit sensing signs that told me I was going 22mph when my display claimed 24. Is there a way to calibrate this system?

 

[Deleted member 18083]

Is there a way to calibrate this system?

Based on my experience: NO.

The Purion cannot be calibrated by us, but check whether your Bosch-certified dealer can make the adjustment for you.

 

[tompat]

Most of the Bosch displays seems to show speed a bit on the happy side.
I guess this is due to Bosch wanting to stay on the safe side of the legal limits - if the bikes were to go faster than the lawful limits by ony a wee bit they could face dire economic consequences.

The bike itself registers the speed correctly, it just displays it with some added percent speed.

Easiest way to verify this is to pedal at an exact and steady speed, let's say 20 mph.
View the average speed on the display, then reset the statistics while keeping steady pace.
You'll notice your average speed after the reset will never reach the 20 mph you are keeping, but rather will be lower at 18-19 mph !!

100 % proof that the bikes registers both speed and distance correctly, but it does not display it correctly nor does the cutoff occur at exactly the legal limit but rather a few points below.

 

[ilanarama]

I can understand slightly under-calibrating for legal protection, but 10% is a huge difference! Grr.

 

[Mr. Coffee]

Most of the Bosch displays seems to show speed a bit on the happy side.
I guess this is due to Bosch wanting to stay on the safe side of the legal limits - if the bikes were to go faster than the lawful limits by ony a wee bit they could face dire economic consequences.

The bike itself registers the speed correctly, it just displays it with some added percent speed.

Easiest way to verify this is to pedal at an exact and steady speed, let's say 20 mph.
View the average speed on the display, then reset the statistics while keeping steady pace.
You'll notice your average speed after the reset will never reach the 20 mph you are keeping, but rather will be lower at 18-19 mph !!

100 % proof that the bikes registers both speed and distance correctly, but it does not display it correctly nor does the cutoff occur at exactly the legal limit but rather a few points below.

Just to put a caution out.

If you reset your average speed when stopped then started riding again your average speed would in fact be less than 20mph. And from a practical standpoint under any realistic riding conditions it is going to be challenging to maintain a precisely steady speed on any e-bike. And for a Class I Bosch-equipped bike maintaining a steady 20mph is going to be even more challenging because you will have effectively zero pedal assist.

One other thing that I suspect happens with the Bosch systems is that they are measuring speed with a very high sample rate, much higher than a radar gun or GPS will, and displaying the maximum speed they get in a given time period (the display seems to update at a lower sample rate, 3-5 times per second -- which is still higher than the sample rate you'd get from a GPS or a radar gun).

It is well understood that Bosch systems gradually reduce pedal assist as you approach the maximum assisted speed. It seems like this effect happens in the last 2mph (e.g. 18mph on a 20mph limited e-bike). For a lot of people they perceive that there is basically zero assist above that last 1mph, but I routinely can hear my motor running and see at least a tiny bit of assist at speeds as high as 19.6mph.

My point is not to defend Bosch but to emphasize that there are a lot of variables that go into what is on that display and unless you wished to do the hard and technically challenging work of systematically measuring the speed of a Bosch-equipped bike and reliably tracking what is displayed everything posted on this topic is wild speculation. My own personal suspicion is that the Bosch systems overstate speed by 3-5 percent. But again that is my own personal suspicion and I certainly do not know that for a fact.

 

[Deleted member 18083]

Bosch systems… are measuring speed with a very high sample rate, much higher than a radar gun or GPS.

Isn't moving speed measured once per wheel revolution as the spoke magnet passes the sensor? Put differently, the interval between 'pings' is timed.

 

[Mr. Coffee]

Isn't moving speed measured once per wheel revolution as the spoke magnet passes the sensor? Put differently, the interval between 'pings' is timed.

Yes, but the net effect at realistic speeds is that the sample rate is still far higher. Most radar guns take around 3-5 seconds to generate a sample, and most handheld GPS don't seem to update the speed calculation much more frequently than that (there are actually very good technical reasons to not have a very high sample rate with a GPS).

I still see the Bosch display update at a fairly high sample rate and I don't fully understand why. This is most noticeable at higher speeds.

 

[AlanDB]

It is not just the Purion. My Bosch Intuvia and Nyon are both calibrated correctly (tire circumference based on rolling distance), and they both show a similar speed discrepancy to what you report. I haven't tried it, but I doubt if I could change the calibration enough (decreasing the circumference) to correct this. Also, the odometer distance seems to match the GPS trip meter distance pretty closely, and I think changing the calibration could have a negative effect on recording accurate distance.

Another interesting thing I have noted about the sampling rate is by watching the speed display when you come to a quick stop. It takes a bit (fraction of a second) before it drops to 0 MPH.

 

[ilanarama]

It is not just the Purion. My Bosch Intuvia and Nyon are both calibrated correctly (tire circumference based on rolling distance), and they both show a similar speed discrepancy to what you report. I haven't tried it, but I doubt if I could change the calibration enough (decreasing the circumference) to correct this. Also, the odometer distance seems to match the GPS trip meter distance pretty closely, and I think changing the calibration could have a negative effect on recording accurate distance.

Another interesting thing I have noted about the sampling rate is by watching the speed display when you come to a quick stop. It takes a bit (fraction of a second) before it drops to 0 MPH.

Ooh, you know, I've checked the speed against my GPS but not the distance. I should do that. I assumed that the 10% overestimation meant that my "Bosch-mileage" is 10% shorter, but it would be interesting to see if this is actually the case. Though since the Purion doesn't report fractions of a mile this would be pretty tough to check other than on a long ride, and even then it there would be a lot of slop.

 

[Mr. Coffee]

Ooh, you know, I've checked the speed against my GPS but not the distance. I should do that. I assumed that the 10% overestimation meant that my "Bosch-mileage" is 10% shorter, but it would be interesting to see if this is actually the case. Though since the Purion doesn't report fractions of a mile this would be pretty tough to check other than on a long ride, and even then it there would be a lot of slop.

My own experience with long rides (40+ miles) over varied terrain is that a Garmin Etrex or Garmin Edge overstates distance 2-3 percent over what Google Maps displays for the same route, and the Bosch Intuvia display overstates an additional 2-3 percent.

Interestingly, the Garmins seem to wildly overstate elevation change by ten percent or more.

 

[ilanarama]

Interestingly, the Garmins seem to wildly overstate elevation change by ten percent or more.

The Garmins which calculate elevation based on GPS are terrible at it, which has to do with the way GPS works to calculate distances: differences in the plane perpendicular to the satellites (that is, horizontal) are far more accurate to derive than differences in the plane parallel to them (vertical). However my current Garmin is a Fenix 3 (wrist mount - I'm primarily a runner) which has a barometric altimeter - which is still not perfect but is much more accurate. The program I use for activity tracking includes an elevation correction which maps the (horizontal) route to a topographic database and produces a fairly good elevation track. (However if there are any weirdities in the GPS track, or if I'm running or biking in a canyon or on a mountain trail where the resolution of the topo map and the GPS slop can combine to produce artificial terrain jumps, errors can be introduced during the correction!)

 

[King me]

Posting after a 7+ month pause in this thread, hoping for a reply. I have read all the posts here, but I'm not trying to address all the questions or issues raised. I have noted the same inaccuracies on my Bosch purion display. I have checked it against other riders in my group, their's being both GPS based and wheel circumference. No offense to technology when I say that the old school bike computers using a pulse received from a magnet triggered sensor are the most reliable, given an accurate rolling circumference of the tires. So speed is one issue, but I also note excess mileage in my trip total, which tells me that calculation is based on Bosch's measured and displayed speed. I have already taken the bike to my dealer, had them put it on the Bosch diagnostic app, and adjusted the the rolling diameter to my carefully measured length. Again based on real time riding comparisons with other riders, the adjustment seems to have shaved nearly 1mph off the discrepancy. There remains about a 1mph difference. Quit while I'm ahead? Also, I assume the posts referring to elevation change/gain are only in regards to Garmin devices? Otherwise I've failed to discover that feature in my Purion! Lastly, that would be my main complaint about the Bosch systems: overly simple, under-featured displays.

After all my researching, shop time, and calibrating, it may be unavoidable to resign myself to a more crowded cockpit and mount a trusty Garmin on the bar.

 

[AlanDB]

With no GPS chip, no topo map and no altimeter, the Purion would have no way to measure elevation changes. My Bosch Nyon tracks elevation, but it has a GPS chip and I am guessing it may also have a barometric altimeter since it allows you to calibrate the correct elevation at the beginning of your ride. I have asked the question about how the Nyon tracks elevation but have never received a definitive answer. Commercial grade GPS chips are not all that accurate on the vertical measure, so if it is using that, considerable error would be introduced. Barometric altimeters are more accurate if calibrated correctly, but changing weather conditions can affect their accuracy over the duration of a ride.

As far as the inflated speed/distance I have just decided to accept it for what it is. In the big picture, a few tenths of a mile on a 25 mile ride, or a MPH or two discrepancy in the real time speedometer reading doesn't really matter all that much. I do sometimes use my old Garmin Oregon 550 mounted next to my Nyon … just because it is interesting to see what the discrepancies are.

 

[King me]

Thanks Alan for the reply. I forgot to mention another untried option than a Garmin computer. I have a Cannondale and the front wheel has a sensor on the spokes that is bluetooth connectable via their smartphone app. The smartphone would then have GPS tracking capability. Haven't tried it yet so I can't say yet what its features are. Strava would be another ride stat option. What is the Nyon? Is it the default display/controller with a certain Bosch motor or aftermarket option?

 

[AlanDB]

The Nyon is a Bosch display that has GPS, on board maps and other advanced features. It uses the same mount as the Bosch Intuvia display. It is possible to upgrade a bike with the Purion to the Nyon, but it requires more components as you have to replace the wiring down to the motor as well as the display, mount and thumb controller.

 

[fooferdoggie]

The Nyon is a Bosch display that has GPS, on board maps and other advanced features. It uses the same mount as the Bosch Intuvia display. It is possible to upgrade a bike with the Purion to the Nyon, but it requires more components as you have to replace the wiring down to the motor as well as the display, mount and thumb controller.

the nyon is not supported in the us. I guess the maps don't work. I wanted one but it could be a problem in the us.

 

[AlanDB]

Bosch may not officially support the Nyon in the US, but many of us "US" riders are using it. The nyon, app and maps all work fine in the US. Bosch has even removed the download/update restrictions on the app, so you no longer have to "jump through hoops" to get everything working. Upgrading from the Intuvia to the Nyon on my Gazelle Arroyo is one of the best decisions I have made.

 

[fooferdoggie]

been debating on the nyon and the cobi. since I have the purion I have to have the shop do the work. not sure how the two compare though.

 

[King me]

Quick update: I did a short ride today with the Cannondale app running. Pretty much confirming the sampling rate of the purion/Bosch setup. The tenths of a mph change are way quicker on the purion vs. cannondale app. Disappointing to find out the mph differential between the two displays are around 1 - 1.5 mph higher on the purion. That lands me smack dab in the middle of what this thread's original conclusions were. Oh well, choose my "speed reality" I guess. It's like you were saying Alan:

As far as the inflated speed/distance I have just decided to accept it for what it is. In the big picture, a few tenths of a mile on a 25 mile ride, or a MPH or two discrepancy in the real time speedometer reading doesn't really matter all that much.

 

[Bert Gijsen]

I have been told it's due to German law.
The bikes have to reduce power ...far before... achieving the "real" 25 or 45 km/u.

To make this less obvious the displayed speed is exaggerated.