Which cycle computer

[kevinfox]

What are the pro's and con's of cycle computers when used with an electric bike?
I'm guessing something that shows distance travelled is a good idea to enable battery range to be monitored, but is something more elaborate that monitors current usage worth having where hills are involved?

 

[Barnowl]

I don't think there's much, if any difference, between using basic cycle computers on an electric or non electric bike.
Most computers seem to give the same data. Speed, Max trip speed, temp, distance, average trip speed, total miles and time. Some are wire connected and some wireless.

There seems to be a great deal of difference in quality though. The quality of the display and how long it lasts. As far as wireless goes and interference from the motor I've never had any and mines a front hub bike. There does seem to be a preference for wired cycle computers on electric though so interference is a problem for some. I do find they get erratic when the temperature drops but this is down to the cycle computer battery and must be common to most computers.

As for battery monitoring range and current usage I've no knowledge. Same goes for health stuff like heart rate etc etc and GPS. I think I'd find it all a bit of a distraction.

Try doing a search on the forum and you'll find quite a lot of stuff on this topic.

 

[flecc]

I value a cycle computer on an e-bike for judging range and also riding at optimum speeds for various purposes.

However, if you want to monitor current usage directly, it means a much higher cost. The best device for this purpose is the Cycle Analyst which can feed back current usage and various other data on your battery state and usage, but it costs around £100 when I last looked:

Cycle Analyst

There's also the cheaper WattsUp at around half that price, not as capable but possibly giving you everything you need:

WattsUp
.

 

[wurly]

For current monitoring i have designed a small circuit, not too elaborate and very much like the CA. It takes the volt drop from a shunt resistor and uses the signal to override the throttle. You 'dial in' the amount of power usage taken from the battery.
Far from being a distraction, i find myself no longer viewing the leds on the throttle trying to find that 'sweetpoint' of just the right amount of assist and keeping the yellow light on.
It works very well. I have one on my bike and the wife's. I am building one for Frank and JohnP at the moment.
(still waiting for parts Frank, won't be much longer)
http://i219.photobucket.com/albums/cc224/gearhead188/ebike/IMG_7673.jpg
http://i219.photobucket.com/albums/cc224/gearhead188/ebike/IMG_7674.jpg
Mel

 

[Phil [OnBike]]

I personally much prefer wired cycle computers - wireless ones can be a problem, which I have experienced myself.
It also means two lots of batteries are needed and it's hardly far from the fork to handlebar to warrant wireless - just route the wire neatly.

On the simple side:
As long as the computer shows distance (which can be reset after each battery charge, so you know how far you've been on each charge) and an odo (so you know how far the bike and battery has done in total) you should be fine with pretty much anything, although do steer away from something costing £1 at a local market.
Extra features are just a nicety.

On the complete side:
Cycle Analysts are the bees knees and I have one on one of my electric bikes and that just tells you everything you need to know, including the all important Amp Hour consumption - these can be a little over the top though, depending on how technical you are, and they are a more difficult to install having to solder in the sensor etc.

Hope this helps

 

[Alex728]

I use a cateye enduro 8 on my Salisbury, works well and is robust...

 

[Mike63]

I too find the cateye enduro 8 to work fine and good value at about £18.

....Mike

 

[RobinC]

One slightly wacky idea I had would be to drive a cheap cycle computer from a circuit which measures the battery current.
A typical cycle computer uses a reed switch to sense when the magnet is near the sensor.

Basic idea:

Low value resistor (ammeter shunt) in series with battery + or -
-> voltage amplifier
-> voltage to frequency IC, perhaps driving frequency divider
-> pulses which turn on mosfet switch which shorts the sensor wire.

N.b. the circuit should probably be designed to short the sensor wires for short pulses to avoid flattening the cycle computer battery.

If it all works out correctly the displayed speed would correspond to battery current and trip distance would in effect be amp-hours used.

I don't expect it would be as accurate as a proper ammeter but might be an option for a stealth e-bike, plus if the display part was pinched it would be less hassle/expensive to replace.

Robin

 

[flecc]

.

If it all works out correctly the displayed speed would correspond to battery current and trip distance would in effect be amp-hours used.

Robin

I think the first problem could be the human component. It could only be reasonably accurate if using motor only, as the human contribution increased and the motor contribution reduced the reading would drift out.

The second and more severe problem would be road inclination, as the road steepened the speed would reduce as the current increased, bringing a serious mismatch of current and speed. Downhill would be the reverse. You'd need an inclinometer coupled to a linear potentiometer to correct for that.

The third problem would be headwinds and following winds, acting like up and down hills in seriously distorting the readings. You'd need an anemometer to measure wind speed to correct for that.

All in all it could get very complicated and appear Christmas tree like!
.

 

[Wooky]

Interesting thing about my NoName wireless computer is that it seems to get affected by security systems or cameras. I can sit outside a local Tesco and a Macdonalds and be doing 61.9mph whilst eating a sandwich. It plays havoc with my range calculations when I can't remember how far I had come! Move about 20 foot away and it's back to normal.

Terry

 

[RobinC]

I think the first problem could be the human component. It could only be reasonably accurate if using motor only, as the human contribution increased and the motor contribution reduced the reading would drift out.

.

I think it should be OK irrespective of user input, hills etc.

Trip distance is calculated from total number of pulses from the sensor. In the modified circuit the pulse frequency represents amps or charge/second so the number of pulses should be a measure of total charge.

One user input needed is to remember to reset the total distance when the battery is recharged.

If I ever try out the idea I will give some feedback here.

Robin

p.s.
Of course without one of the fancier ebike computers I won't be able to check the accuracy of the total charge measurement.

 

[flecc]

I think it should be OK irrespective of user input, hills etc.

Robin

I can't see how it can be if as you say, the speed is derived from the current being used. The current usage varies enormously at any particular speed, depending on the imposed load due to hills, headwind etc. Of course the bike type will affect things, a throttle controlled one likely to produce very large errors as loading variations are compensated by different throttle openings.

The voltage on lithium batteries also varies greatly with load, the range of variation can be almost 9 volts on a nominal 37 volt battery, so that could badly affect the voltage to frequency divider output.
.

 

[RobinC]

I can't see how it can be if as you say, the speed is derived from the current being used. The current usage varies enormously at any particular speed, depending on the imposed load due to hills, headwind etc. Of course the bike type will affect things, a throttle controlled one likely to produce very large errors as loading variations are compensated by different throttle openings.

The voltage on lithium batteries also varies greatly with load, the range of variation can be almost 9 volts on a nominal 37 volt battery, so that could badly affect the voltage to frequency divider output.
.

Perhaps I didn't explain properly.

The voltage to frequency IC will be controlled from an amplified copy of the voltage across the low value resistor in series with the wire from the battery. This will be measuring the battery current not the voltage.

All the computer will then be measuring is battery current so bike speed won't matter. E.g. when it displays 10miles/hour it really means 10Amps.
After one hour it will say distance travelled is 10miles but this really means 10Ah consumed.

Robin

 

[flecc]

Thanks for the further explanation Robin, I had misunderstood. All clear now!
.

 

[kevinfox]

I went out and purchased a Cycle Analyst which arrived today. Does anyone know what the calibration is for a Quando 2?
Thanks
Kevin

 

[daniel.weck]

I went out and purchased a Cycle Analyst which arrived today.

Hi Kevin ! Did you purchase your CA from the USA eBay shops ? How long did you have to wait for delivery, and how much shipping tax did you pay ? Thank you very much, Kind regards, Dan