Kalkhoff Agattu Pedelec - Road tests continued
30.11.2007, Raining, temperature 15 degrees C, wind gusty 20 mph, (not relevant).
The fitted saddle might well be comfortable for someone of Wagnerian proportions, but it's breadth together with slightly excessive leg reaching due to the bike being the wrong size for me impeded my leg action, leading to thigh cramps which would limit my day's riding. However, I wanted to ride the bike in rain to prove the waterproofing, since although the Lafree's Panasonic unit is faultless in this respect, the handlebar switch wasn't, so my first action was to obtain and fit a lower unsprung seat post and fit it, readjusting the handlebars to match. That made the bike much more comfortable to ride.
The fitted saddle might well be comfortable for someone of Wagnerian proportions, but it's breadth together with slightly excessive leg reaching due to the bike being the wrong size for me impeded my leg action, leading to thigh cramps which would limit my day's riding. However, I wanted to ride the bike in rain to prove the waterproofing, since although the Lafree's Panasonic unit is faultless in this respect, the handlebar switch wasn't, so my first action was to obtain and fit a lower unsprung seat post and fit it, readjusting the handlebars to match. That made the bike much more comfortable to ride.
The route is a short one, just 1.7 miles ridden in two directions and passing through an adjacent estate which is built along the steep flank of a hill. Instead of running an access road level along the slope, in an early attempt to deter rat-running, the architect used short sections of road at various levels up the slope, joining them with steep up and downhill sections, creating the cyclist's nightmare you see on the right. Using the Sanoodi link will show how the roads snake up and down along the hill's flank.
Riding this abrupt sequence of mainly 10 to 14% ups and downs is a good test of both climb ability and braking, since some of the hill terminations are T junctions with poor visibility needing downhill speed to be almost completely lost each time. The bike coped perfectly and as expected, the well enclosed handlebar switch and cabling was unaffected by the heavy rain I waited for before going out into it. More on waterproofing the next day.
1.12.2007 AM, Clear sky, wet tarmac, temperature 7 degrees C, wind SW 10 to 17 mph.
The route chosen is a circuit including a 450 feet (137 metres) 4 miles long continuous climb, then followed within 10 seconds by another continuous 450 feet climb over 3 miles.
The route chosen is a circuit including a 450 feet (137 metres) 4 miles long continuous climb, then followed within 10 seconds by another continuous 450 feet climb over 3 miles.
This was intended to be a harsh test of the abilities of the Li-ion battery, using continuous high power mode for the best part of an hour. At the 9.2 mile point close to the end of the near continuous 7 mile climb, the first of the handlebar meter's 3 LEDs went out, together with the first of the 5 battery LEDs! That's impressive indeed. Panasonic have clearly done their homework on the battery management for the usage to be so parsimonious under high load.
These weren't steep climbs, but as those who've suffered Li-ion cut-out problems know, it's not the steepness, so much as how long the battery is kept on high current usage.
Near the top of those climbs a group of club riders returning from the nearby popular hill climb trials section passed me going downhill, the front rider calling out with a grin, "It's easier in this direction!", a neat illustration of just how unobtrusive the electrical side of this bike is, that cyclist not noticing the compact motor unit and minimally proportioned battery.
I'd been frequently pausing momentarily to check the battery LEDs, and just at the start of the long downhill of Slines Oak Road at 13.8 miles the second of the 5 LEDs had gone out, the handlebar still with two of the three there lit. Following the previous day's heavy rain, I knew of a point where there would be deep water to ford, and never having tested a Panasonic unit in that, but knowing how well it's sealed, I headed there next. On the graph above near the 18 mile point you can see a very sharp dip, this being a fold in the Downs with very high level fields either side of the lowest point of this dip in Farleigh Court Road which has no drainage system.
Sure enough it was filled with a several yard wide pond, so keeping the pedals horizontal and pulsing them to give flicks of drive power, I ventured in. It was deeper than I thought, just touching the bottom of the hubdyno and completely submerging the motor unit, so I still got my feet very wet anyway as I forded across. As expected the motor unit didn't notice it, the sealing as perfect as the old model's always was, and measurement to the hubdyno lower edge later showed that the water had reached 12" / 30 cm deep. With motor to ground clearance of 7.5" / 19 cm and lower pedal clearance of over 5" / 13 cm, and a central motor/battery unit away from direct shocks, that waterproofing makes this bike a very good overall choice where rough tracks and standing water are to be ridden.
Almost back home, the second of the 3 handlebar meter LEDs and the third of the 5 battery LEDs went out at 20.9 miles, the ride ending at 21.3 miles. That would correspond to a range of 35 miles, a great result after that rather brutal continuous high current hill climb test earlier in the day, and consistent with the first day's results. Average speed 11.5 mph and maximum at any one point 32.1 mph.
Early in the day I'd ridden the Kent Gate Way cyclepath again, with still air through being well protected from the wind that was experienced high on the Downs later, some of this a minutely negative incline of between 0.9 and 1.2%, and found the bike rolling so freely that I stopped pedalling, the bike then rolling along at 17 mph for well over half a mile before the incline was lost. That's the best free rolling I've ever known over the years round here. The Continental tyres must take much of the credit and definitely have lower roll resistance than the Marathon Plus I usually use, but nonetheless, for any e-bike to roll in this manner and be better than good normal bikes is remarkable. I'm going to be interested in seeing if A to B did a roll test in their forthcoming report.
Near the top of those climbs a group of club riders returning from the nearby popular hill climb trials section passed me going downhill, the front rider calling out with a grin, "It's easier in this direction!", a neat illustration of just how unobtrusive the electrical side of this bike is, that cyclist not noticing the compact motor unit and minimally proportioned battery.
I'd been frequently pausing momentarily to check the battery LEDs, and just at the start of the long downhill of Slines Oak Road at 13.8 miles the second of the 5 LEDs had gone out, the handlebar still with two of the three there lit. Following the previous day's heavy rain, I knew of a point where there would be deep water to ford, and never having tested a Panasonic unit in that, but knowing how well it's sealed, I headed there next. On the graph above near the 18 mile point you can see a very sharp dip, this being a fold in the Downs with very high level fields either side of the lowest point of this dip in Farleigh Court Road which has no drainage system.
Sure enough it was filled with a several yard wide pond, so keeping the pedals horizontal and pulsing them to give flicks of drive power, I ventured in. It was deeper than I thought, just touching the bottom of the hubdyno and completely submerging the motor unit, so I still got my feet very wet anyway as I forded across. As expected the motor unit didn't notice it, the sealing as perfect as the old model's always was, and measurement to the hubdyno lower edge later showed that the water had reached 12" / 30 cm deep. With motor to ground clearance of 7.5" / 19 cm and lower pedal clearance of over 5" / 13 cm, and a central motor/battery unit away from direct shocks, that waterproofing makes this bike a very good overall choice where rough tracks and standing water are to be ridden.
Almost back home, the second of the 3 handlebar meter LEDs and the third of the 5 battery LEDs went out at 20.9 miles, the ride ending at 21.3 miles. That would correspond to a range of 35 miles, a great result after that rather brutal continuous high current hill climb test earlier in the day, and consistent with the first day's results. Average speed 11.5 mph and maximum at any one point 32.1 mph.
Early in the day I'd ridden the Kent Gate Way cyclepath again, with still air through being well protected from the wind that was experienced high on the Downs later, some of this a minutely negative incline of between 0.9 and 1.2%, and found the bike rolling so freely that I stopped pedalling, the bike then rolling along at 17 mph for well over half a mile before the incline was lost. That's the best free rolling I've ever known over the years round here. The Continental tyres must take much of the credit and definitely have lower roll resistance than the Marathon Plus I usually use, but nonetheless, for any e-bike to roll in this manner and be better than good normal bikes is remarkable. I'm going to be interested in seeing if A to B did a roll test in their forthcoming report.
1.12.2007 PM, Clouding over, signs of faint drizzle, 9 degrees C, wind SW 10 mph
With all the really tough testing done, time for something mundane, so on went a pair of large Dutch throwover panniers for a shopping trip to the supermarket! Here's the bike
With all the really tough testing done, time for something mundane, so on went a pair of large Dutch throwover panniers for a shopping trip to the supermarket! Here's the bike
parked outside Sainsburys, looking every inch the tough working bike that it is, huge businesslike panniers and down tube still splattered with mud picked up from the road straight after the morning's aquatic adventure. I must say the motor unit looks nicely washed though!
Note the rigid temporary seat post with saddle tilted back slightly to ease pressure on the back of the thighs while pedalling. It will suit many, especially heavy riders, but I'd change it for something rather slimmer.
Note the rigid temporary seat post with saddle tilted back slightly to ease pressure on the back of the thighs while pedalling. It will suit many, especially heavy riders, but I'd change it for something rather slimmer.
Here's the route on the graph with the complete return journey shown, and as you can see, we don't do flat in my area. Going somewhere means going up or going down, but more usually both. The slight differences in graph symmetry are due to small one way systems variations. Despite the high pannier position, the bike was completely stable and balanced with a full load of shopping on board. No measurements taken for this short run of course.
3.12.2007 PM, Clear sky, 9 degrees C, wind W, variable 0 to10 mph
The previous tests had not really been a fair assessment of range due to them being designed to be very difficult in various ways, and also including individual special tests. Therefore today's was to be on a route designed to be more average in nature and not including anything else but straightforward riding. Since climbs are unavoidable in my area, I still rode an elevation range of 520 feet (159 metres) to 850 feet (260 metres) above sea level, but instead of going directly over 6 miles, I used a zig zag route avoiding the very steepest hills and including some flat sections and totalling 11.6 miles to the high point. The included hills were up to 10%, and since they were moderate I haven't bothered with a Sanoodi graph since much of it would repeat the previous areas anyway.
With those more average route conditions, I never used high power mode and took advantage of the bike's low roll resistance by riding without power when it was easy. The longest such section without power was 3.6 miles in urban conditions, often slowing right down for junctions and roundabouts, and peaking at over 18 mph at times, the average over that stretch being 12.1 mph. In a rural area without so many intersections that would easily have been at least 13.5 mph average unpowered. For one brief period over half a mile I tried Eco mode, but otherwise the whole ride was on standard mode or switched off.
With my hands and feet usually freezing in 9 degrees after about 20 miles or so, I already knew that I was never going to ride for 40 to 50 miles, so I'd already determined that I would assess the full range by accurate calculation. Two days before, the first LED of the five had gone out at 9.2 miles, but since meters aren't linear, the full range wasn't five times that but 3.8 times that at 35 miles. Therefore, when the first handlebar meter LED and first of the five battery lights went off at 13.71 miles, the multiplication by 3.8 gave a range of about 52.1 miles.
In fact that was likely to be conservative, since the 3.8 times test basis was from the 9.2 miles which included the continuous 7 mile climb on high power, so was a worst case. However, with this almost unbelievable very long range indicated in moderate territory, the figure demanded checking, so I carried on until the second LED went out so that a cross check calculation could be made to prove or disprove the result.
Two days ago, the second LED had gone out at 13.8 miles, the multiple of which to give the 35 mile range then being 2.54. So with the second LED extinguished at 21 miles exactly, the 2.54 multiple indicated a full range of about 53.3 miles, very close to the 52.1 indicated before. I'm certain there wasn't any serious error in these measurements, but even after allowing up to 10% for an error, the range would still have been nearly 50 miles, so I'm confident that if I'd carried on repeating those circuits with the 520 feet climb within each, the distance to final cutout would have been a minimum of 50 miles (80 km).
I indicated on the first day that the manufacturer's claim of "Up to 80km / 50 miles in ideal conditions" was correct, but mine were hardly ideal conditions: Running a Li-ion battery in 9 degress C, a 71 year old rider, a hilly area, so how could I be beating that by so much? I think the manufacturer's figure is with leaving the power on and using the Eco mode extensively, but I found the Eco so ineffective that it wasn't worth using, just wasting the battery. I found it just as easy to simply ride with power off when the going was very easy, and with just one button press for power on or off, and two button presses to change to Eco, the former was more intuitive and became so automatic I wasn't aware of doing it.
The buttons being right next to the left grip and needing only a slight thumb movement makes change easy. When switching on, all 6 LEDs light in a momentary self test, then drop to Standard mode and showing whatever is in the battery. The Power mode levels can be changed at any time, but pedalling must be paused momentarily when switching on for the self test LEDs to go out before pedalling again.
It seems to me that a free rolling bike like this with it's very long range battery gives considerable flexibility. An owner in average territory who has settled on a easy to get range of, lets say, 35 miles (56 km), could at any time extend that by 10 or 15 miles (16 to 24 km) if the need arose since the additional cycling effort would be so small. It's always very difficult to predict ranges for others, but with power left on most of the time, and across the spectrum from rather heavy rider with hills, to very fit regular cyclist in easy territory, ranges should span from 30 to 50 miles. Using the handlebar Off button more often when the going is easy will add to those in proportion to the rider's fitness.
Journey over, I arrived home with the third of the battery's five LEDs still not extinguished at 26.3 miles done, roughly double that being likely, so that was a second confirmation of my above calculation of 52.1 miles. Paradoxically, the moderate but rather hilly nature of the territory probably helped, since it meant plenty of long, gentle downhill slopes where power could be switched off, which with hills no steeper than 10%, I wasn't offsetting by using the High Power mode. In a fairly flat area, I'd estimate the range would be more like 45 to 50 miles using either a mix of Standard mode and switched off for the easier sections, or possibly Eco all the time. My average speed for the run was 12.3 mph.
The previous tests had not really been a fair assessment of range due to them being designed to be very difficult in various ways, and also including individual special tests. Therefore today's was to be on a route designed to be more average in nature and not including anything else but straightforward riding. Since climbs are unavoidable in my area, I still rode an elevation range of 520 feet (159 metres) to 850 feet (260 metres) above sea level, but instead of going directly over 6 miles, I used a zig zag route avoiding the very steepest hills and including some flat sections and totalling 11.6 miles to the high point. The included hills were up to 10%, and since they were moderate I haven't bothered with a Sanoodi graph since much of it would repeat the previous areas anyway.
With those more average route conditions, I never used high power mode and took advantage of the bike's low roll resistance by riding without power when it was easy. The longest such section without power was 3.6 miles in urban conditions, often slowing right down for junctions and roundabouts, and peaking at over 18 mph at times, the average over that stretch being 12.1 mph. In a rural area without so many intersections that would easily have been at least 13.5 mph average unpowered. For one brief period over half a mile I tried Eco mode, but otherwise the whole ride was on standard mode or switched off.
With my hands and feet usually freezing in 9 degrees after about 20 miles or so, I already knew that I was never going to ride for 40 to 50 miles, so I'd already determined that I would assess the full range by accurate calculation. Two days before, the first LED of the five had gone out at 9.2 miles, but since meters aren't linear, the full range wasn't five times that but 3.8 times that at 35 miles. Therefore, when the first handlebar meter LED and first of the five battery lights went off at 13.71 miles, the multiplication by 3.8 gave a range of about 52.1 miles.
In fact that was likely to be conservative, since the 3.8 times test basis was from the 9.2 miles which included the continuous 7 mile climb on high power, so was a worst case. However, with this almost unbelievable very long range indicated in moderate territory, the figure demanded checking, so I carried on until the second LED went out so that a cross check calculation could be made to prove or disprove the result.
Two days ago, the second LED had gone out at 13.8 miles, the multiple of which to give the 35 mile range then being 2.54. So with the second LED extinguished at 21 miles exactly, the 2.54 multiple indicated a full range of about 53.3 miles, very close to the 52.1 indicated before. I'm certain there wasn't any serious error in these measurements, but even after allowing up to 10% for an error, the range would still have been nearly 50 miles, so I'm confident that if I'd carried on repeating those circuits with the 520 feet climb within each, the distance to final cutout would have been a minimum of 50 miles (80 km).
I indicated on the first day that the manufacturer's claim of "Up to 80km / 50 miles in ideal conditions" was correct, but mine were hardly ideal conditions: Running a Li-ion battery in 9 degress C, a 71 year old rider, a hilly area, so how could I be beating that by so much? I think the manufacturer's figure is with leaving the power on and using the Eco mode extensively, but I found the Eco so ineffective that it wasn't worth using, just wasting the battery. I found it just as easy to simply ride with power off when the going was very easy, and with just one button press for power on or off, and two button presses to change to Eco, the former was more intuitive and became so automatic I wasn't aware of doing it.
The buttons being right next to the left grip and needing only a slight thumb movement makes change easy. When switching on, all 6 LEDs light in a momentary self test, then drop to Standard mode and showing whatever is in the battery. The Power mode levels can be changed at any time, but pedalling must be paused momentarily when switching on for the self test LEDs to go out before pedalling again.
It seems to me that a free rolling bike like this with it's very long range battery gives considerable flexibility. An owner in average territory who has settled on a easy to get range of, lets say, 35 miles (56 km), could at any time extend that by 10 or 15 miles (16 to 24 km) if the need arose since the additional cycling effort would be so small. It's always very difficult to predict ranges for others, but with power left on most of the time, and across the spectrum from rather heavy rider with hills, to very fit regular cyclist in easy territory, ranges should span from 30 to 50 miles. Using the handlebar Off button more often when the going is easy will add to those in proportion to the rider's fitness.
Journey over, I arrived home with the third of the battery's five LEDs still not extinguished at 26.3 miles done, roughly double that being likely, so that was a second confirmation of my above calculation of 52.1 miles. Paradoxically, the moderate but rather hilly nature of the territory probably helped, since it meant plenty of long, gentle downhill slopes where power could be switched off, which with hills no steeper than 10%, I wasn't offsetting by using the High Power mode. In a fairly flat area, I'd estimate the range would be more like 45 to 50 miles using either a mix of Standard mode and switched off for the easier sections, or possibly Eco all the time. My average speed for the run was 12.3 mph.
3.12.2007
.
.
During these rides I've concentrated on reporting on the more difficult aspects like hill climbing and handling in poor surface conditions, since normal riding on flat smooth tarmac roads doesn't present a problem with very little power needed to propel a bike at or around the legal 15.5 mph / 25 kph assistance limit. Accordingly this bike made light of that, and owners of bikes using one of these Panasonic motor units will know what I mean when I say, "same again". For those who are not familiar, this is totally unlike any motor vehicle, since instead of a continuous application of motor power, the system adds to your muscle strength on each thrust of the pedals, so is truly bionic in nature, leaving the character of this bicycle unchanged from that of any unpowered bicycle. The more pedal pressure you exert, the more help you get.
With power on, you can cycle with the strength of two people and more, that strength available from standstill with every pedal stroke up to around 12 mph in top gear. Then, to comply with the law, the assistance power begins to reduce very gradually until it finally ends at the legal limit, the transitions so smooth on these units that it's impossible to accurately detect where they start and end, so not in any way disturbing the cycling rhythm. The limiting applies in every gear, so spinning the pedals fast in a lower gear will take you out of power assistance in exactly the way 15.5 mph in top gear does, but at a lower speed according to the gear chosen.
Conclusion
Once in a while an e-bike appears that stands out as rather special, usually because of some particular attribute, but only very rarely does one appear with a special level of all round competence. To date those have all used the various versions of the Panasonic motor unit, but since the demise of the affordable Giant Lafree Twist series, they have tended to be too expensive and have sometimes fallen short of that bike on grounds of weight and performance. None of these deficiencies is true of the Kalkhoff Agattu here. With a weight the same as the Lafree Comfort ST model and a superior specification including the high power mode long range version of the latest Panasonic power unit, it's very reasonable price immediately elevates it to the top.
For many the Lafree series were the best electric bikes ever made, a title which could only realistically be held by those using the Panasonic system to date, since no other motor systems have yet been able to retain the full cycling attribute of the power coming directly from the pedalling, all having some degree of the motor vehicle about the way they operate. Taking every factor into account, design, features, build quality, strength, performance, long range, and the UK price, in my opinion the Kalkhoff Agattu is now the best electric bike ever, and the Lafree has lost that title to a worthy successor.
So should you all rush to buy one? That depends entirely on how you view e-bikes. This is a proper bicycle and if you enjoy cycling, like to share in the effort of cycling, and revel in the resulting fitness, this is ideal. If you view an e-bike as being a form of moped and like to sit back and let the bike do most or all of the work, this is not for you.
For Lafree riders, this is the obvious replacement, since in particular it completely answers the few problems it had. As well as the frame flexure previously mentioned, there were an occasionally weak headstock and weak drop outs seen here, and a front mudguard that lashed around like a caged wild animal over every ripple in the road. The closely hugging front mudguard on the Kalkhoff doesn't move a fraction, the frame is very stiff, and the headstock and rear dropouts are epic in scale. I think the Kalkhoff designer was probably very familiar with the Lafree and knew exactly what was needed.
With power on, you can cycle with the strength of two people and more, that strength available from standstill with every pedal stroke up to around 12 mph in top gear. Then, to comply with the law, the assistance power begins to reduce very gradually until it finally ends at the legal limit, the transitions so smooth on these units that it's impossible to accurately detect where they start and end, so not in any way disturbing the cycling rhythm. The limiting applies in every gear, so spinning the pedals fast in a lower gear will take you out of power assistance in exactly the way 15.5 mph in top gear does, but at a lower speed according to the gear chosen.
Conclusion
Once in a while an e-bike appears that stands out as rather special, usually because of some particular attribute, but only very rarely does one appear with a special level of all round competence. To date those have all used the various versions of the Panasonic motor unit, but since the demise of the affordable Giant Lafree Twist series, they have tended to be too expensive and have sometimes fallen short of that bike on grounds of weight and performance. None of these deficiencies is true of the Kalkhoff Agattu here. With a weight the same as the Lafree Comfort ST model and a superior specification including the high power mode long range version of the latest Panasonic power unit, it's very reasonable price immediately elevates it to the top.
For many the Lafree series were the best electric bikes ever made, a title which could only realistically be held by those using the Panasonic system to date, since no other motor systems have yet been able to retain the full cycling attribute of the power coming directly from the pedalling, all having some degree of the motor vehicle about the way they operate. Taking every factor into account, design, features, build quality, strength, performance, long range, and the UK price, in my opinion the Kalkhoff Agattu is now the best electric bike ever, and the Lafree has lost that title to a worthy successor.
So should you all rush to buy one? That depends entirely on how you view e-bikes. This is a proper bicycle and if you enjoy cycling, like to share in the effort of cycling, and revel in the resulting fitness, this is ideal. If you view an e-bike as being a form of moped and like to sit back and let the bike do most or all of the work, this is not for you.
For Lafree riders, this is the obvious replacement, since in particular it completely answers the few problems it had. As well as the frame flexure previously mentioned, there were an occasionally weak headstock and weak drop outs seen here, and a front mudguard that lashed around like a caged wild animal over every ripple in the road. The closely hugging front mudguard on the Kalkhoff doesn't move a fraction, the frame is very stiff, and the headstock and rear dropouts are epic in scale. I think the Kalkhoff designer was probably very familiar with the Lafree and knew exactly what was needed.
The surprisingly small and very light 10 Ampere hour battery seen here is one of the secrets behind the performance, the very clever management being the other. Since the power supplied is a ratio of the rider's input, it follows that only the strongest rider can draw the maximum possible current. But the strongest rider will have the least need of assistance and use of the high power mode, so will be the least likely to draw very high currents. That's clever design, using natural human attributes to make management of the system more effective. Being a lithium battery, once it's run in initially, it can be charged at any time, the more often the better, and Panasonic have had the longest experience of all in using these on e-bikes. Running in and further charging advice is in the Extra section after this review.