Karen Nakamura: July 2008 Archives

Today I visited my local flea market and picked up a real find. A used Razor e300 (or maybe e200, not sure) electric scooter with charger for the unbelievable price of $25.

The hitch was that the scooter didn't work when I tested it out. The seller originally wanted $30 but I bargained down to $25 because of its condition. The LED power lights came on, but it didn't move. The battery read full but no motor motion. I decided that for $25, I could part it out and still get my money back and so I bought it.

I came back home and the batteries were indeed dead. A little massaging from my higher-end Soneil 2408 charger and the batteries came back to life. They don't seem as though they have their full capacity, I'm doing an amp-hour test on them with my Watts-Up tonight.

I put in the SLA batteries from my Pocket Rocket and took it for a zip around the neighborhood. Lots of fun -- until a car pulls out in front of you. Still, you can only go 10-15 mph so there's limited trouble that you can get into. And even with the newer SLAs from the Pocket Rocket, my range was only around 3 miles.

I'm going to see what I can do to hop it up, perhaps use some NiMH batteries instead of SLAs.

The ELF 100 ESC (electronic speed controller) arrived today and I installed it in my Piaggio Boxer. The install was a cinch and everything seemed to be going very smoothly.

I did some test runs with the rear wheel elevated using my 24V SLA batteries. Here is my data table:

DialMotorRear SpindleTire / SpeedAmp Draw
0-10 rpm0 rpm0 rpm0.03A
241116214 rpm / 1 mph -
3118033531 rpm / 2.5 mph -
418011495120 rpm / 10 mph6.6A
523901950156 rpm / 13 mph8.2A
628502330185 rpm / 15 mph9.4A
733272720217 rpm / 18 mph10.16

I ended the no-load test at dial position 7 on the servo tester as I didn't have a secure way of securing the moped on the center stand and if it came off the center stand during the test, the kinetic energy would rocket it off my back deck.

I tried some partial loads and the amps consumed quickly jumped to 30A as the ESC tried to maintain motor speed.

The motor sounded very good and heated up only a very small amount (to 38 centigrade). So everything looked good for a road test. I took it to my front yard and dialed the servo up to 4.

The moped took off and I dialed it to 5 and went down my street at about 15-20 mph. At the end of the street, there is a slight down incline and in order to come to a stop, I dialed the servo back to 0.

The rear wheel skidded to a halt and the ESC beeped menacingly at me then shut up. I tried to dial speed back up again and nothing happened.

I tried resetting the system and it didn't work. That's when I noticed the ESC was getting VERY HOT. I now have a blister on my index finger. I yanked my main system power and pushed the unit back home. I'm just glad it chose to blow near my house and not 5 miles out!

I let the ESC cool down and then plugged it back in again. It quickly started heating up again, drawing 10 A off the battery (240 watts) and not making any sounds (it normally will "sing" to you to let you know it's OK).

Although I didn't see where it went, I have the feeling that I let the magic smoke out of my $30 ebay special ESC. I've since yanked it off the bike and will likely go and buy a "real" ESC (like a Phoenix HV100).

Until then, I'll try to conduct an autopsy on my ESC. I've got the protective shrink wrap off and the chip that seems to be the one getting super hot is marked 78M05 and it has three leads coming off of it. I'm hoping that if I post some good autopsy pix, maybe Fechter on E-S can help me diagnose it.

In any case, the moral of the story is that the HXT outrunner will work, but it really really really needs a good ESC. And you should make sure your ESC settings are correct before taking it for a test drive. In the back of my mind, I'm wondering if the ESC was set to brake at throttle 0, which would have been disastrous.

p.s. A quick google reveals that a 78M05 is a voltage regulator chip. Makes sense that that would what blew. My guess is that when the ESC told the motor to brake, the motor said "sure, here's A GAZILLIO AMPS in back current to ya!" and that blew out a chip on the ESC.

I spent a few hours this afternoon installing the HXT outrunner motor. It was a bit larger than the MY1016 that I had previously had as a prototype. And as it is an outrunner, this necessitated a different type of engine mounting.

Here you can see the base end of the outrunner motor with the power leads on the left. The X shaped bracket came with the outrunner and is what I used to attach it firmly to my Boxer's left engine support bracket.

This is the view from the top. The main body of an outrunner motor rotates, so you can't attach the motor by the side. It has to be face mounted.

Here is the left side spindle with the V-belt pulley installed. The nice thing about retaining the CVT back end on the moped is that I don't have to adjust the pulley tension, it's done automatically.

This is the right side of the motor / moped. In order to have the pedals clear, I had to shorten the right hand (prop side) spindle. I left a small nub so that I later attach a flange bracket to lessen some of the side loading on the main bearings.

I had to cut away some of the sheet metal on the right hand motor bracket to fit the motor and its spindle in. This considerably weakened the right side bracket, which is a structural part of the suspension system.

This is the right side of the motor again, from the top. The green, black, and red wires visible are the mopeds low-voltage (6V) power leads to the headlight and taillight.

When I cut part of the right side motor bracket to fit the spindle in, this weakened the right side as it was a structural component. In order to make sure nothing Bad happened, I reinforced the bottom part of the mount with a piece of L aluminium bracket.

Once the ESC arrives, I can wire everything up and take her for a spin. I'm hoping that current pulley gearing will give me around a 30 mph top speed. I'll be running her off a 24V SLA system at first, then when my NiMHs arrive, at 28.8 volts.

Higher than 30V and I need a new ESC, darn I knew I should've just bit the bullet and gotten the Phoenix HV100A.

The servo tester that I ordered last week arrived by mail today. Still no sign of the ESC from Hong Kong so I can't wire everything up, but it's a start.

I go the E-Sky EK2-0907 servo tester from Hobby Lobby. With shipping, it was $13.69 although you can get it for a few less dollars if you shop around. I just wanted a basic model and it fit the bill. Most of the units that I see on ebay are identical to this.

You really can't get much simpler. It has inputs for battery power (5-6V) and two server outputs. There were no instructions enclosed but except for wiring it up wrongly, I can't see what the instructions would've said.

On an RC model, you would hook the servo motors to the servo tester to make sure that they were operating correctly (i.e., moving through their full range) in order to diagnose whether your servos were sticking or your RC receiver wasn't sending the right signals.

I'll be using it as my throttle control. I'll be plugging the motor ESC (electronic speed control) into the servo teste. The ESC looks like a servo to the unit and so if you dial in 0, the motor should be at a low RPM or stop; and 10 would be full speed.

I took the servo tester apart and it looks like the main knob is a standard 5K potentiometer. That means that once my testing is finished, I can replace it with a Magura 5K pot throttle that is standard for many e-motorcycles. For now, I'll leave it be.

The main controller chip is most likely something like a 555 timer chip. Basically a servo tester sends out 10-20msec pulses every 60 msec and the servo throw depends on the length of the pulse (10 msec for minimal throw; 20 for max). This is an efficient design and one that's very resistant to RF noise and signal loss. Which means I can make the controller wire pretty long (and unshielded) without too many problems.

The other thing that arrived was a Venon Smart Temp. I was about to buy a cheap BBQ or auto indoor/outdoor thermometer so that I could monitor my engine temperature when I noticed the Venom was just a few dollars more. The neat thing about it is that it will also monitor my BEC/servo voltage too. And if the motor temp exceeds a pre-programmed limit, it can limit the throttle to a pre-programmed limit. So I can set it, for example, that if the motor gets over 60C, then the max throttle is reduced to 60% to protect the motor.

Hopefully my ESC will come tomorrow or Wednesday. My house is filling up with parts that I can't use! I should also be working on the motor mount as well as get some of the servo wiring and BEC placement done in preparation. I could also use some more Dean's connectors....

Well, the postal carrier came and left without leaving me an ESC + servo tester, so I still can't install the HXT outrunner. But, she did leave me a little package from Hong Kong that contained the laser photo tachomoter that I ordered from ebay. I'm still amazed at just how cheap Chinese manufacturers are pumping things out. A laser tach would have cost me over a $100 just a few years ago, now I can buy one for $23 w/ shipping from Hong Kong. Incredible.

The quality of the unit is quite good. The laser beam is nice and coherent with minimal spread. The unit is sensitive enough that I could do my motor readings using some duct tape rather than expensive retroreflective tape (that they kindly include a small sample of with the unit).

This will be lots of fun.

Anyway, I went ahead and did some motor speed readings on my Boxer to check out my gear ratios. This is with the back wheel raised on the center stand (no load test):

Motor spindle speed / front hub speed: 3272 rpm
Rear spindle speed / flywheel speed: 2091 rpm
Wheel speed: 168.8 rpm (14 mph)

Motor amp draw was 5 amps, so close to no load. From this, I can see that my motor to wheel ratio with the 1.5" front sprocket is 19:1 rather than the 39:1 calculated by using the gear ratios (I think there is another gear inside the flywheel that I didn't consider). When the motor is at full speed, the wheel is going at 14 mph.

Anyway, I'm glad I got the tach since it's telling me the real motor:wheel ratio rather than the calculated one, which was always a bit iffy since I had a centrifugal clutch and could never test the entire system.

For $23, the laser photo tach is definitely a good buy and it'll be even more useful when I get the HXT mounted and have a variety of volts and amps combinations to play with with the new controller.


After zapping the moped with the tach a couple of times, I decided to take it out for another spin. Seems like the battery is coming up to speed after a couple of charge/discharge cycles as I was able to get 5.0 Ah out of it today.

Trip distance: 5.3 km (3.3 miles)
Moving time: 26 minutes
Moving average: 12.2 kph (7.5 mph)
Max speed: 18.3 kph (11.1 mph)

Total energy: 5.0 Ah
Total Energy: 0.11 KWh
Peak amps: 22.14 A
Min voltage: 21.05 V

km / kWh: 48 km / kWh (30 miles per kWh)

I have some NiMHs on order, I can't wait until they arrive too! Running on two 7 Ah SLAs stinks!

I'm still waiting for the ESC and servo to arrive (maybe early next week) for the outrunner motor, so I took the Boxer out for a test run with its temporary Razor motor and SLA batteries installed.

Piaggio Boxer
Motor: 250 watt MY1016 (from Razor Pocket Rocket)
Controller: 30A (from Razor)
Batteries: 2 x 12V 7Ah SLA (from Razor)
Front sprocket: 1.5" pulley (motor to wheel 39:1 ratio)

Batteries were fully charged and run from 26.5V down to 20.98V (under load). I used a Garmin GPS as a trip computer and a Watts Up as a battery monitor.

Trip distance: 5.8 km (3.6 miles)
Moving time: 27:26 minutes
Stopped time: 2:10 minutes
Total time: 29:36 minutes
Moving average: 12.6 kph (7.8 mph)
Overage Avg: 11.7 kph (7.2 mph)
Max speed: 17.4 kph (10.8 mph)

Total energy: 4.4 Ah
Total Energy: 0.10 KWh
Peak amps: 21.93 A
Min voltage: 20.98 V

km / kWh: 58 km / kWh (36 miles per kWh)

Final odometer reading: 1532.0 km


Given that my batteries were rated 7.0 Ah (for 20h), I was pleased that I was able to get 4.4 Ah from them in less than half an hour. I think I could have eeked enother half an Ah for 5.0 Ah. The range of 3.6 miles is pathetic, but then again, I'm using 7Ah tiny little scooter batteries.

This time around, the motor didn't see to get too hot. I think it's because everything is finally breaking in, the belt seems well adjusted and it was mostly running at 10 A which is its nominal rating and only pulling 15 - 20A when it was starting off or pulling me up a hill.

The top speed of 10.8 mph and average speed of 7.8 mph is pathetic. But it's above my expectations of 5 mph. Going 10 mph on a sidewalk is just fine, but it's much too slow for a road with moving traffic, where you would want at least 30 mph.

I imagine that if I got a real 500 watt motor in this at 48 V, it would be just fine for tooling around town. But I do need more battery life, only half an hour of run time (or 0.10 kWh) isn't nearly enough. I'd like an hour at faster speeds, which would be around 10 Ah @ 48 V (0.48 kWh) which is nearly 5x my current amount.

But hey, what other vehicle gets 36 miles per kWh? :-) According to this website (http://car.pege.org/2008-fuel/kwh-mpg.htm), 36 m/kWh is equivalent to 702 MPG Miles per US gallon gasoline in city traffic.

702 MPG! Try to beat that!

While I'm waiting for the ESC and servo tester (aka throttle) to arrive, I'm trying to figure out what gearing I'll need with my new motor.

My Piaggio Boxer wheels are 14" in radius, so if my high-school geometry is right, that's 2 x pi x r = 88" in circumference.

If go 5 mph, that's 316800 inches per hour, that's 3600 rotations per hour or 60 rotations per minute.

My mo-ped's rear hub has a built-in 12:1 gearing reduction. The current 1.5" front fixed pulley I'm using with the little 250 watt MY1016 motor gives me a 3.25:1 gearing reduction to the back wheel. That gives me a final 39:1 gearing reduction.

5 mph = 60 rpm * 39 = 2340 rpm.......
10 mph = 120 rpm * 39 = 4680 rpm......
20 mph = 240 rpm * 39 = 9360 rpm......

This sort of explains the 8mph max speed that I'm getting with the 250 watt Razor motor. By the way, I bought a Doc Wattson which tells me that this motor draws 20 amps at "stall" speed (since I have a centrifugal clutch, it never actually stalls):

When I'm running just down the street, it draws about 10 amps.


So when I put on the HXT outrunner, it's obvious that I'll have to go to a bigger front pulley. If I use a 2.5" front pulley, the v-belt gearing shifts to something closer to 1:1. Doing the same math but with a 12:1 reduction gives me:

5 mph = 60 rpm * 12 = 720 rpm.......
10 mph = 120 rpm * 12 = 1440 rpm......
20 mph = 240 rpm * 12 = 2880 rpm......
30 mph = 360 rpm * 12 = 4320 rpm......

Since I bought the 130 Kv motor and was planning on running it at 24 volts or 3120 rpm, it looks like my max speed would have been limited to 21 mph.

If I instead run the motor at 36 volts, this would give me a max rpm of 36 * 130 = 4680 rpm which would give me a max speed of 32 mph.


In addition to stepping up to 36v, I could tighten the v-belt so that the rear gearing shifts to something like 1:2, which would give me a 6:1 reduction and effectively double all of my speeds @ 36v:

5 mph = 60 rpm * 12 = 360 rpm.......
10 mph = 120 rpm * 12 = 720 rpm......
20 mph = 240 rpm * 12 = 1440 rpm......
30 mph = 360 rpm * 12 = 2160 rpm......

For a max speed of 65 mph at 4680 rpm............. which is a bit too fast.

What I might do is a 6:1 reduction @ 24 volts which would give me a max speed of 43 mph which is just perfect.

This is all theoretical, I'll have to do some playing around when the rest of the parts arrive. But for now, it looks like the pulleys and v-belts I currently have will be just fine.

p.s. If I had access to a machine shop + was actually skilled, I could machine the HXT spindle so that it would fit on the old CVT from the moped. It would take some filing to make it D-shaped, and then some tapping so that the nut could screw on. Then I could take full advantage of shiftless gearing from 0mph to 65 mph.

(above, the spindle on the old ICE with the CVT front pulley attached)

Well, the motor from United Hobbies arrived! Very fast shipping, I ordered on Friday and it got to me on Wednesday from Hong Kong. The packaging was secure and everything seemed in place:

The motor is much heavier than it looks and it rotates with a definite clunk-clunk-clunk. I got the 130Kv version which means that at the 24 volts that I plan on running it, my max RPM should be 3120 rpm, which isn't that different from the current brushed motor I'm using so I won't have to fiddle with the gear ratios much.

The motor base unit is 4" long. Each spindle is 1.5" long so the entire unit is 7" long.

I do like the odds and ends they include, including the very handy motor mount. Since this is an outrunner motor, the outer shell actually spins around so you can't mount it like a normal motor, you have to mount the one stationary end. It has two spindles, I'll be using the one closer to the base to run my V-belt and affix a small propeller blade on the one on the far end to help cool the motor.

Unfortunately, I ordered the ESC and servo tester from separate places and they still haven't arrived, so there's nothing to do right now but twiddle my thumbs.

At the current moment, I have the ICE engine out and the Razor motor in and wired up. Using a small hub on the 250 watt motor, I'm getting about 5-8mph and not nearly enough power to make it up hills. Fortunately, this was designed as a moped and I can pedal assist whenever I need to. Also, the centrifigual clutch means I can't stall out the motor.

The problem is that I don't have enough power, the motor is running very hot (70 C after 10 minutes runtime), and the rinky dink batteries I have now won't give me enough distance.

As mentioned above, I've ordered an HXT outrunner motor from Hong Kong (per Scott @ Endless-Sphere's experience) and will wire it in once it arrives. I'm hoping to be able to go 30 mph on straightaways if I can with larger SLA batteries on it.

This is my very high-tech throttle control. The Rocket controller is not speed adjustable, you just have to short two wires. I short one pair and turn on the bike using the mini-headphone jack on the left (aka: "the key") and then turn on the motor using the switch on the right. I avoided doing anything nicer as the HXT outrunner + ESC that I'm using in Phase 2 will be an entirely different control system.

I finally got around to downloading the photos of the first half of my scooter's EV conversion from my digital camera. Here we go:

This is the Piaggio Boxer mo-ped that I'm converting:

This is the donor Razor for the motor, batteries, and controller:

The right side is the bicycle pedal and chain section, don't want to mess with that side:

The left side has the gasoline ICE and you can see the CVT drive belt:

Front of the CVT spindle up close:

Front of the CVT spindle dismantled:

I managed to shim the v-belt pulley and jerry-rig a motor mount, so I mounted up the little motor onto my Piaggio Boxer scooter, put the v-belt on, strapped the batteries on with duct tape, and shorted the throttle wires, and it ran!

The speed was horrendous, maybe 6 mph on a slight downhill and 3-4 mph on a level surface. But that's fine for the purpose of my test run.

Here are the current specs:

Base model: Piaggio Boxer Mo-ped
Motor: 24V 240 watt MY-1016 from a Razor Pocket Rocket mini-bike
Batteries: 12 V SLA 7ah x 2 (from Razor)
Motor Controller: 24VDC 30A controller (from Razor)

After running it for 10 minutes up and down the street, the motor was extremely hot, hot enough for the label to peel off. The controller was also hot (enough to wrinkle the duct tape holding it on). The batteries seemed fine.


1. I definitely need to push this up to 48V so I can get more wattage at the same amperage. There's enough space on the mo-ped runners to mount 2 sets of 2 SLAs on each side.
2. The current motor mount blocks airflow to the motor. I need to mount it better so that it is in the direct air current.
3. I need to find a more powerful motor too, while I'm at it.
4. I need to figure out how to get it to be even a higher gear ratio that what I currently have. Maybe find a smaller front pulley somewhere or figure out how to mount a large sprocket on the rear and have it chain-driven.

Anyways, I'm very pleased with myself that it at least works. Smiling Now to just make it work better!


p.s. Photos to follow soon.
(Originally posted: July 6, 2008)

Well, I went to Tractor Supply Center (TSC) today and all they had were imperial sized pulleys (1/2" and 3/4" bores). The motor that I have must be a metric size one because the 1/2" hub pulleys were too large. I bought some 1/2" hub pulleys anyway since I wanted to see if it would work as a proof of concept (and just around how much more power I'll need).

Now I just have to make a mount for the motor. I think I'll just use a piece of 2x6 wood that I have laying around and make a simple mount.

I took the motor, motor controller, batteries, and everything else out of the Razor. I tried to play with the motor controller. It turned out the "throttle" on the bike was only a switch and not a potentiometer. I tried using a 5K pot but the motor speed (unloaded) didn't seem to change any. I'm not sure if this controller is variable or not, but it wouldn't make any sense not to be, right? Maybe I need more resistance.

If I can get the pulley and V-belt on next weekend and cut the mount and drill it out, I'll be able to get the electronics in and the mo-ped on the road to test it out.


To answer some of the comments on the previous post:

1. Definitely 240 watts is too little, if that is the actual rating of the motor. Once I get this to work and can show proof of concept to my partner, my budget for getting a bigger motor should be approved.

2. I thought about a hub motor and might go down that route if belts/pulleys become too much of a bugaboo. It'd certainly solve a lot of problems except the financial ones -- most ones I've seen are $300-500 right? That's about $200-400 over my current budget (until I can prove that this will work, and that I haven't ruined my partner's mo-ped in the name of science).


Original forum posting and comments: <a href="http://visforvoltage.org/forum/3973-modifying-my-piaggio-moped-part-2">VisforVoltage.org</a>

My partner had bought an old Piaggio Boxer mo-ped a couple of years ago but sitting in our garage had rusted out the engine. I decided to make it an EV conversion.

So far, I've removed the old ICE cleanly. It looks like the ICE was connected to the rear wheel through a belt-driven CVT / clutch.

My replacement motor is a small 24V / 250 watt one taken from a RAZOR electric (toy) motorbike. It's about the only thing I can find that would fit in the small area of the former engine compartment (about 4-5" wide). If the 24V works and doesn't have enough power, I'll have to see if I can find a more powerful pancake unit or mount the motor in a separate place.

I'm driving of to my nearest (? 1 hour+) tractor supply center today to see if I can find a v-belt pulley that will attach to the motor. The old front CVT pulley from the ICE won't fit on the DC motor because the spindle isn't long enough.

IF I can find a new pulley today and find some kind of engine mount, I think I can finish up the conversion this week. But I'm just guessing at the pulley gear ratio and I don't think that the 250 watt motor will be powerful enough.

The controller is rated for 24V / 30A so I could put in a bigger motor if I could find it..... Or I could hop up the voltage to 36V by adding another L-A battery.

About this Archive

This page is a archive of recent entries written by Karen Nakamura in July 2008.

Karen Nakamura: August 2008 is the next archive.

Find recent content on the main index or look in the archives to find all content.