First EFUN experiences in 2010

100% is whatever voltage is fed to the hall sensor. And yes, 5V is the standard for that. In my case while measuring my throttle grip 100% was just 4.5V from three 1.5V batteries :-)

Yes, the L setting flattens the voltage gradient so it is virtually useless, because the reduced max. voltage can easily be achieved in H setting by not turning the grip to full blast...

I accessed the throttle signal at the 3-pin connector at the end of the throttle grip pigtail - in my case that is accessible behind the front fairing. However, as my original throttle grip had been malfunctioning (the motor would sometimes creep even with the grip at idle...) I had gotten a replacement from my importer, so after drying the connector (it had been flooded with water...) and sealing the wire entries into the connector I measured it on my desktop :-)

In my experience the "L" setting is useful to have better control when low power is required (and also to limit power request to the controller).
"H" setting is difficult to controll, you go from zero to full throttle in too few degrees! (not to mention that any road axperity gets translated into a "power axperity")

Next week I will try to see if I can access the connector without too much hassle and will try to make some measurements/mods.

Edit: I was able to remove the two small screws on the throttle side, and while the side panel couldn't come away I could see the contacts of the L/H switch.
The switch is closed in "H" mode and open in "L" mode, so this suggests that the output of the throttle passes through a diode (or something similar) that adds a voltage drop and that closing the switch you bypass it sending the full output to the controller.
Too bad I realized this when I had already closed it back... because if my assumptions are right the switch could be the right place to probe the signal.

Edit2: on second thought it could also be that the switch is actually working on the supply voltage of the throttle...

Well, then I guess the efun 6-phase(?) controller simply is no good for controllability. Ah yes, NOW I remember: my first pick would have been a 4kW efun maxiscooter, but I found during a test ride that the throttle was no more than an on-/off-switch. I had to drop that plan because I never would have been able to ride it in winter on slippery roads with that kind of sudden, jerky full-blast takeoff. I KNOW what you are talking about, from mine own seat-of-the-pants :-)

The Kelly in my Thunder has almsot perfect controllabilty, completely linear from walking pace up to maximum speed. No need at all for the H/L switch, apart from the odd trial just to see if L still works :-) And yes, when I have to ride on icy roads, then it does come in handy for that extra little gentleness at takeoff.

I think that they didn't care too much about correctly matching the input stage of the controller to the typical hall throttle repsonse.
You can actually modulate the power, but this happens in too few degrees.

Anyway the response of the controller has improved since my first tests but there is still room for (lots of) improvement.

I think that they didn't care too much about correctly matching the input stage of the controller to the typical hall throttle repsonse.
You can actually modulate the power, but this happens in too few degrees.

Anyway the response of the controller has improved since my first tests but there is still room for (lots of) improvement.

After the winter rest I started using the scooter again but the range seems shortened... voltage drop during acceleration is 1.5V higher than before and overall voltage drops below 60V under acceleration (I have 20 * 40Ah cells) after about 30km (I remember that I was getting about the same voltage about 10km later)....

Could it be the winter "stop" that caused some imbalance? I checked the voltage during the stop time and it was ok, and gave a couple of charges even if the voltage was still ok)

I'm going to check all the individual voltages, but right now I don't feel confident about doing the complete resume to work without a intermediate charge.

Any ideas?

I tested the batteries after a charge and they were all between 3.34 and 3.41V a few hours after overnight charge. So small imbalance but nothing dramatic.

The real problem was the rear tire, almost flat after the winter. It was also affecting odometer reading.. 10.4km instead of the usual 9.6!
While the tire was getting inflated to the correct pressure the scooter got lifted by 2-3 cm!

Range now seems ok.

I made some further tests on the scooter and the throttle works differently from what was expected.
The H/L switch goes to the controller and limits the current (according to my current meter - maybe not that accurate) to around 50A, while in the High position maximum current is in the 80A ballpark.

In the meanwhile I had to fix a problem with the ignition key - a wire had got desoldered completely shutting down the scooter. So if you experience a complete shutdown check it (solderings were far from ideal...)

In the meanwhile I had to fix a problem with the ignition key - a wire had got desoldered completely shutting down the scooter. So if you experience a complete shutdown check it (solderings were far from ideal...)

that's a common fault

it will also start to fail internally

basically it's a 12v ignition switch, switching full battery voltage to a capacitive load (the dc-dc usually)

the work around is to put an inductor in series, to reduced the current spike at turn on

Do you have some practical values to suggest?
Also... on the photo I can see multiple contacts... I think that wiring them in parallel should make them last longer...

I used two of these wired in series on a thunder model (72v battery):

http://www.jaycar.com.au/productView.asp?ID=LF1278

that worked

before wiring contacts in parallel - I would first check they are isolated

even in parallel, they will still wear - the first contact to close will get the current spike

So about 1mH should be enough - thanks for the suggestion.
Next time I open the scooter I will give it a go. I will also take the occasion to remove the "speed sensor" on the front wheel (actually it's a flexible transmission for a analog "tachometer" that is installed but not used (as confirmed by Amoyee Chen)

Well, it seems that the ignition switch didn't like being criticized.

After some turn on and off (I was making some tests) the scooter refused to turn off.
Tripping the circuit breaker under the seat didn't help (but I can't understand why)

I was forced to dismantle the plastics to have access to the connector that goes to the ignition switch to be able to turn the scooter off.
I admit that I was a bit scared. What if I was away?

Anyway, the switch smells of burnt plastic and gives a 20 Ohm reading no matter the position of the switch...

Did anyone trace che actual wiring of the circuit breaker?

the circuit breaker is between the battery and the main power wires for the motor controller

the ignition switch is between the battery and the dc-dc

turning off the circuit breaker only disconnects the motor controller from the battery, but not the ignition switch or dc-dc

looks like you need to get a new ignition switch

Yep, I need a new ignition switch. Waiting for a reply from seller and efun.
I'd rather get just the electrical switch, to avoid having to change all the keys.

In the meanwhile I'm using a 13A switch. The contact must have already been pretty bad, because I notice that now my Scooter Analyzer shows the correct voltage (compared to the bike voltmeter) while before sometimes was ok and other times was giving a lower voltage (I take my reading after the switch)

When I get the new one I'll surely use some inductors to limit inrush current.

Ordered a new switch from efun plus a new dc-dc converter with external enable so that the ignition switches only few milliamps.
Hopefully it's going to be a definitive solution

Installed the new dc-dc converter, it required only minimal rewiring.
Since the scooter was open I also applied fetcher's modification for the throttle, and after a week of riding I can say that it is very good. Now throttle is more progressive and does not have any longer the "hole" it used to have in the first degrees of rotation.

Btw the old dc-dc outputted 11.7V , this one has much more heatsink and gives a steady 12.8V (front lights are better now)

I recently encountered a problem with my scooter: when turned on after a long pause (few hours) the scooter seems stuck in low power mode (max 35-40A); after some running it reverts to full power (80A).

Today, while the scooter was showing this behaviour, I flipped the H/L switch to " L" and the power went suddenly to zero.

My guess is that since with the 6-phase technique we basically have 2 controllers working on 3 phases each one of the controllers for some reason has difficulties starting. The H/L switch works on the enable of one while the other is always enabled.

In my case the controller having problems is the one that is always enabled.

Now some questions:
1-What can cause a controller not to start?
2-What do you suggest I should check?

After some checks and fiddling with the motor connectors it turned out that the culprit is most likely the "fetcher" mod I did to the throttle.
This mod inserts a small value resistor on the gnd line of the throttle thus raising the first part of the throttle curve.

Probably with the first colds the initial voltage is perceived as "too high" from the controllers that fail to start until they see a low wnough voltage.
Turned the trimmer backwards a little bit and now the scooter works fine again!

A quick update on the scooter, running fine and approaching 7000km.
Upgraded the analyzer with new TFT display and relay output to be able to use Fetcher's mod vith low temperatures (relay has the added R on the NC contact and opens after a delay)