Is Bosch off on its speed calculations?

[rudymsmith]

My Raleigh Motus with Bosch Purion display also reads the speed about 10% high but the odometer seems to be as accurate as my car, motorhome and GPS display. I have also doubled checked distance with the measurement tool on Google Earth on PC. Speed inaccuracy means absolutely nothing at all on a bicycle but the odometer does play a role in battery calculations. I just don't understand why, once you have calculated discrepancy, it should be a problem other than to anyone who has OCD tendencies and, having family members who are OCD 'sufferers', I can appreciate their frustration. They can't help it and nothing you can do will alter the fact that everything has to be perfect, as they see it. As for Bosch conspiracy theories, well, what do they gain. My ebike is UK spec which means assistance should cut out at 15.5mph, roughly 17mph indicated taking in the 10% discrepancy yet, it doesn't. Mine seems to be 'faulty', in that assistance seems to cease at about 22mph (indicated) 20mph actual. This seems to be in common with many other Motus owners so I fail to see, even with the high readings, what any form of conspiracy could be seen in this. The high assistance cut out isn't something I'm likely to complain about, same as others in the same situation. If the odometer is accurate, the motor assists to at least 15.5mph (actual), the speedometer is within the 10% high legal tolerance then where is the cause for complaint?

People tend to miss the point of this thread entirely - which is - If manufacturers ADVERTISE 28MPH before a class III motor stops assisting, then it should do just that - 28 MPH before dropping out. I've ridden bikes all my life, motorcycles, cars etc... ad nauseum, and what I want is simple: TRUTH IN ADVERTISING! If they were honest and said the motor cuts out at 26.5 mph, oh well, I'm good with that. Just don't lie to me to beef up your sales or make me think your battery or motor is better than your competitions or to be on the safe side of government oversight. The current overestimate of speed v. motor assist is software driven.. period, yes you can fiddle with tire size and magnet location etc... but as manufactured, and sold, the software used by my 3 Bosch E bikes, all class III's, is wrongly optimistic (for whatever reason) it is incorrect and the faster you go the more incorrect it becomes.

 

[Amazer98]

As for the speedometer accuracy I have found, on my particular machine, that adjusting the position of the magnet on the wheel DOES have an effect on the readout. I moved the magnet nearer to the rim and that actually did reduce the rather optimistic speed reading though I'm not sure exactly why it does....not that I really care.

Biffo1262,
Doesn’t the magnet on the wheel have to be aligned with the sensor mounted on the chain stay? If you move the magnet elsewhere on the spoke, it wouldn’t trip the sensor to display your speed. And no matter where on the spoke one fixes the magnet, even if the sensor were moved to align with it, the displayed speed would have to remain the same.

The sensor measures the rate of wheel revolutions in a segment of time... and the wheel makes X revolutions/minute, no matter where the magnet is located. If the displayed speed is inaccurate (as it also is on my Bosch system), it’s because the incorrect wheel or tire size was programmed by the dealer or manufacturer.

 

[FlatSix911]

Haha this thread is crazy... Bosch (maybe others) have it set in firmware so that if you have tire circumference, tire pressure etc all perfect the bike will go a tiny bit slower than legal limit but are still displaying 20mph etc.

They don't want you to be going faster than than the legal limit or else the law will come down on them. And with a zillion motors in the world, it's just too risky to bump it up a little higher, even if there is actually a tolerance permitted by EU law, etc.

So bike companies take on the risk by trying to introduce a bit of an offset to bring it as close to legal limit as possible. Maybe they don't get it perfect of course, or they don't even bother to mess with the motor supplier defaults. But the dealer is able to adjust the circumference by +/5% and get you the 1-2mph that is missing. It's really that simple. German companies are very law-abiding

Nonsense... I will not support Bosch products going forward.

VW, Robert Bosch agree to pay $1.55 billion to settle U.S. diesel claims

Volkswagen AG <VOWG_p.DE> has agreed to pay at least $1.22 billion to fix or buy back nearly 80,000 polluting U.S. 3.0 liter diesel-engine vehicles to settle claims it fitted illegal emissions-cheating software to the cars, court documents showed.

www.reuters.com

 

[rich c]

Nonsense... I will not support Bosch products going forward.

VW, Robert Bosch agree to pay $1.55 billion to settle U.S. diesel claims

Volkswagen AG <VOWG_p.DE> has agreed to pay at least $1.22 billion to fix or buy back nearly 80,000 polluting U.S. 3.0 liter diesel-engine vehicles to settle claims it fitted illegal emissions-cheating software to the cars, court documents showed.

www.reuters.com

That’s funny. Ask any engineer, if they give you an honest answer, they’ll tell you that EVERY manufacturer plays with software and certification. I know, I worked at Caterpillar for 30 years. Since the measurements are taken at certain parameters, engineers know what to do. It’s not measured in real world conditions. As far as not hitting the advertised speed to the exact digital decimal point. If we had an analog gauge, the width of the needle would be considered just fine.

 

[dblhelix]

As far as not hitting the advertised speed to the exact digital decimal point. If we had an analog gauge, the width of the needle would be considered just fine.

Yes, bc random fluctuations of needle half-width are expected but not a systematic bias of -1.5 mph.

I’m asked daily about where the assist cuts off. I tell them that 28 mph is what’s claimed vs an actual draw down to somewhere between 26 mph-27 mph. The response is always a frownie face.

The Specialized Turbo line advertised 28mph for S models vs 26 mph for X models if I recall the models correctly. I view the practice under discussion as Bosch-specific. Perhaps Yamaha as well? I’m not familiar with the Yamaha OEM bikes.

 

[AlanDB]

I'm in the same boat with my speedo being overly optimistic. I set my tire size as small as possible and it's still about 0.5mph fast. Did I do extensive testing? No. Do I care that much? Not really. My bike goes in for its complimentary 250mi tune up so I'll post back if they "fix" it. The shop is pretty confident that the Bosch will be more accurate than my satellite based GPS but said they would check it. My bike is a 2019 Raleigh Redux, class 3, Bosch speed motor, Purion display.

I can almost guarantee they won't "fix" it. I also disagree with your shop rep who said the Bosch display is more accurate than satellite GPS. I have observed this discrepancy with my Gazelle Arroyo using both the Intuvia and Nyon displays. It always reads 1 to 2 mph faster than I am going. I have just learned to live with it so it doesn't really bother me. I often use my Garmin Oregon side-by-side with the Nyon, so if I want to see an accurate speed, I look at the Oregon.

Interestingly, most car speedometers also register high by 2 to 3 MPH at highway speed.

 

[FlatSix911]

That’s funny. Ask any engineer, if they give you an honest answer, they’ll tell you that EVERY manufacturer plays with software and certification. I know, I worked at Caterpillar for 30 years. Since the measurements are taken at certain parameters, engineers know what to do. It’s not measured in real world conditions. As far as not hitting the advertised speed to the exact digital decimal point. If we had an analog gauge, the width of the needle would be considered just fine.

I am also an engineer and will tell you that Bosch went far beyond the point of reasonable error...

Please read the linked articles that describe the wholesale cheating that was programed by Bosch for VW.

Bosch fined $100 million over emissions-cheat software

Bosch has agreed to pay a $100 million fine for lapses in supervisory duties that enabled automakers to engage in emissions cheating, German prosecutors said.

europe.autonews.com

Prosecutors fine Bosch 90 million euros for emissions cheating role

Automotive supplier Bosch has agreed to pay a 90 million euros ($100.21 million) fine for lapses in supervisory duties which enabled carmakers to engage in emissions cheating, German prosecutors in the city of Stuttgart said on Friday.

www.reuters.com

 

[rich c]

I am also an engineer and will tell you that Bosch went far beyond the point of reasonable error...

Please read the linked articles that describe the wholesale cheating that was programed by Bosch for VW.

Bosch fined $100 million over emissions-cheat software

Bosch has agreed to pay a $100 million fine for lapses in supervisory duties that enabled automakers to engage in emissions cheating, German prosecutors said.

europe.autonews.com

Prosecutors fine Bosch 90 million euros for emissions cheating role

Automotive supplier Bosch has agreed to pay a 90 million euros ($100.21 million) fine for lapses in supervisory duties which enabled carmakers to engage in emissions cheating, German prosecutors in the city of Stuttgart said on Friday.

www.reuters.com

It’s still my opinion that the deep pockets are where all the lawyers, government, and went after. Cummins recalled 500,000 trucks for the same kind of violations. Volvo, Renault, Jeep, Hyundai, Citroen, and Fiat are also on the list.

 

[rudym]

Short answer: Yes, the Bosch speedo is showing faster than actual speed.

There is a long thread over at the German pedelec forum of this same subject.
Many are disappointed that their speed pedelecs that are supposed to assist up to 45 kmph cuts out already at 43 kmph while the display is showing 45 kmph. This is consistent with my experience of my Bosch Speed motor.
Real speed measurement can be done by any GPS device.

Funny thing is that the European type approval certificate (Certificate of Conformity) clearly states that the motor assists up to 43 kmph.

The odometer and average speed seems to be accurate by all measurements I've made.

So the Bosch system clearly measures speed accurately but is programmed to show it incorrectly on the display.

View attachment 23769

It’s now 4 april 2022 am I’m riding a Riese Muller super delight with Nyon display and guess what - the same speed discrepancy is present with the Nyon displaying approx 1.6 mph faster than actual speed. It is what it is

 

[fooferdoggie]

It’s now 4 april 2022 am I’m riding a Riese Muller super delight with Nyon display and guess what - the same speed discrepancy is present with the Nyon displaying approx 1.6 mph faster than actual speed. It is what it is

you can fix it easily

 

[rudym]

you can fix it easily

Do tell!

 

[fooferdoggie]

Do tell!

go to my ebike and look for wheel diameter and make it smaller till it has the right speed. all my bosch bikes have been off.

 

[rudym]

go to my ebike and look for wheel diameter and make it smaller till it has the right speed. all my bosch bikes have been off.

interesting - this doesn't require a reset/reflash at the bike servicing level?

 

[fooferdoggie]

interesting - this doesn't require a reset/reflash at the bike servicing level?

no but it is limited how much you can change it. but I hve done it several tiems. my trek seems pretty close as is according to a roadside speed sensor.

 

[rudym]

no but it is limited how much you can change it. but I hve done it several tiems. my trek seems pretty close as is according to a roadside speed sensor.

what screen / onboard controller and motor interface does your trek have? wondering what screen i could find this in on my nyon if at all?

 

[fooferdoggie]

its the same for gen 2 and 4

 

[AlanDB]

I tried changing the circumference setting on my Gazelle. It seemed to me that it didn't fully correct the displayed speed (still showed 1 or more MPH faster than was showing on my Garmin Oregon), but it did change the odometer reading for trip milage compared to what the Oregon recorded. I put the circumference back to the setting that came from the factory which seems to track pretty consistently with the Oregon's odometer, and just decided to live with the inflated speed display.

I am going from memory of what I tried a couple of years ago and I didn't keep a good record of the results, so I can't prove my assertions. It was just what I observed non scientifically on my particular ebike.

 

[fooferdoggie]

a gps is no the best for speed adjustment mile markers would be better. or a radar on the road. gps's jump all over the place.

 

[rudym]

@Sparky731 i started this thread back in 2019 jump in if you want

 

[Sparky731]

@Sparky731 i started this thread back in 2019 jump in if you want

Per your invite to “jump in” I will concur that my Trek bikes with Purion displays are both off by a non-linear (1 mph @ 10mph, 2.5mph @ 20mph) EXCEPT for my wife’s Verve+ 2 which uses a magnet on the spoke and is adjusted to the far end of the induced receiver. Verified many times by street radars. By my calculations, I am getting 10% less assist than I paid for.