Help! My next project

[saneagle]

Thanks Saneagle. I'm not sure I fully understand but it seems then that the system is working as it should. If this is so then it is hugely disappointing as it makes it impractable for the many V shaped bottoms with steep climbs out that I encounter. I'm not looking for great speed here but 60-80 watts very slowly ramping up on a 1:8 or greater climb is, frankly, bloody useless.
I don't understand why you reference 13mph.
The Ezee was fully there straight away, this is what I'm used to. This, of course, was speed control and I'm wondering if changing this system to speed control would help this issue.
I only started this project as the Ezee was an unmarked big motor that didn't cut off at 15.5mph. I was just trying to be legal. Hurumph !

13 mph is the point where the controller can no longer give full current.

Here's a simulation for any 201 rpm motor at 36v with a 20A controller. You can see the ramping down from 22km/hr where the red and blue lines are together. The corner at the top is where the controller is no longer the limiting factor:

Motor Simulator - Tools

Our ebike motor simulator allows you to easily simulate the different performance characteristics of different ebike setups - with a wide selection of hub motors modeled, and the ability to add custom batteries and controllers and set a wide variety of vehicle parameters you'll be able to see...

ebikes.ca

 

[matthewslack]

Why does it take several (how many?) seconds to give full power when reengaging at 6mph or so? I'd want and expect to be at full power in a fraction of a second.

 

[saneagle]

Why does it take several (how many?) seconds to give full power when reengaging at 6mph or so? I'd want and expect to be at full power in a fraction of a second.

Mine is instant. the control system doesn't have any intelligence that could do that. You open the throttle and it goes. You pedal and it goes. It doesn't care what speed you're doing or have been doing other than the consideration of the max cut-off speed. Maybe his battery is weak and can't provide enough power, though you'd expect it to have recovered after freewheeling.

 

[Benjahmin]

Now we're getting to it. The whole nub of the problem is this infuriating intermittant delay.
I run two 14.5Ah batteries in parallel, both made by Jimmy. My charging diaries tell me one is on 256 charges the other on 167. None of those will have been full cycle charges. They have been running the Ezee system throughout those charges with no problems. They charge to 41.7v.
Most of the time the system does respond instantaneously to pas and/or throttle, on sudden ramps, steep inside corners, pulling away from junctions. But it's at those crucial V bottoms it's letting me down. The inconsistancy is infuriating and it's this I/we are trying to find the cause of.
I'll repeat what I found today, in case it helps.
Scenario is going down hill.
If I don't pedal at all going down hill (just let the bike roll) and don't pedal until speed comes down to 15.5mph, then power pick up is nigh on instantaneous.
If I do pedal (bikes gearing will allow me to pedal up to around 30mph) then the system will not kick in until speed is very low (8mph or less) and when it does the increase in watts displayed in painfully slow. There was no sag showing on battery indicator at this point.

Saneagle, I know you've said you've had trouble with the 12 magnet pas's that I have fitted, do you think it could be something to do with that?

I'm trying to be as concise and accurate with my language and observations as I can. Thanks for all the help and suggestions so far for this apparently irrational problem.

 

[Sturmey]

..........If I don't pedal at all going down hill (just let the bike roll) and don't pedal until speed comes down to 15.5mph, then power pick up is nigh on instantaneous.
If I do pedal (bikes gearing will allow me to pedal up to around 30mph) then the system will not kick in until speed is very low (8mph or less) and when it does the increase in watts displayed in painfully slow. There was no sag showing on battery indicator at this point........

I think anotherkiwi had that problem (2019) and came to the same conclusion in terms of not pedaling until the speed goes below 25 Km/hr. He comments on it below. I hope its some help.

Bafang conversion kit : Recovery from over-speed is slow

Heyo, I have a 250 W Dillenger front hub kit on a 10-y-old hybrid commuter which has transformed my commute, but: When I sail down a hill which puts the road speed over the limit of 15.5 mph, the motor freewheels (great), but when I start up the next hill the controller waits until the road...

www.pedelecs.co.uk

 

[saneagle]

Now we're getting to it. The whole nub of the problem is this infuriating intermittant delay.
I run two 14.5Ah batteries in parallel, both made by Jimmy. My charging diaries tell me one is on 256 charges the other on 167. None of those will have been full cycle charges. They have been running the Ezee system throughout those charges with no problems. They charge to 41.7v.
Most of the time the system does respond instantaneously to pas and/or throttle, on sudden ramps, steep inside corners, pulling away from junctions. But it's at those crucial V bottoms it's letting me down. The inconsistancy is infuriating and it's this I/we are trying to find the cause of.
I'll repeat what I found today, in case it helps.
Scenario is going down hill.
If I don't pedal at all going down hill (just let the bike roll) and don't pedal until speed comes down to 15.5mph, then power pick up is nigh on instantaneous.
If I do pedal (bikes gearing will allow me to pedal up to around 30mph) then the system will not kick in until speed is very low (8mph or less) and when it does the increase in watts displayed in painfully slow. There was no sag showing on battery indicator at this point.

Saneagle, I know you've said you've had trouble with the 12 magnet pas's that I have fitted, do you think it could be something to do with that?

I'm trying to be as concise and accurate with my language and observations as I can. Thanks for all the help and suggestions so far for this apparently irrational problem.

I don't think it's anything to do with the pedal sensor. All it does is pulse. The pulse speed doesn't matter.

The only thing I can think of is something to do with the clutch, but I can't think what. In one case, the the motor is spinning with the clutch disengaged, and the other the motor isn't spinning with the clutch engaged. Maybe the centrifugal force holds the rollers out and prevents them from engaging. I remember other peopl talking about weird effects like that with other controllers and motors. The more I think about it, the more I believe that's what's happening. The rollers in AKM clutches are on quite a large diameter, so would get more centrifugal force than most

If my theory is correct, you'd see something around 30 to 50 watts on the meter when the speed is below 15 mph, you're demanding power, but not getting any. The moment the clutch engages, the motor will go from no load to loaded, so the power should jump to normal levels, depending on what you have the power set to.

The solution would be to stop pedalling for a couple of seconds when the speed gets down to say 17 mph. That would give the motor a chance to slow down to engagement speed before you reach 15 mph. Alternatively, keep pedalling, but feather the brake to use the brake switch to cut the motor. The moment you're down to 15mph, you can either pedal or let go of the brake, and the clutch will engage to give all the power you want.

 

[Sparksandbangs]

I think anotherkiwi had that problem (2019) and came to the same conclusion in terms of not pedaling until the speed goes below 25 Km/hr. He comments on it below. I hope its some help.

Bafang conversion kit : Recovery from over-speed is slow

Heyo, I have a 250 W Dillenger front hub kit on a 10-y-old hybrid commuter which has transformed my commute, but: When I sail down a hill which puts the road speed over the limit of 15.5 mph, the motor freewheels (great), but when I start up the next hill the controller waits until the road...

www.pedelecs.co.uk

Interesting. Tomorrows test is to pedal down the hill then stop pedaling at the bottom of the V. Let it coast up the hill until the speed drops below the set point and then start pedaling again.

 

[Benjahmin]

Looks like Anotherkiwi came to the same no pedal conclusion. The two bikes he lists both have KT controllers.

If my theory is correct, you'd see something around 30 to 50 watts on the meter when the speed is below 15 mph, you're demanding power, but not getting any. The moment the clutch engages, the motor will go from no load to loaded, so the power should jump to normal levels, depending on what you have the power set to.

If the bug (let's call it that) is happening, then there is zero watts displayed. When it does decide to kick in it is around 50-80watts very slowly ramping up to pas level.
One of my test runs is a downhill side road coming to a main road blind T junction in the village. So I have to stop to see it's clear to pull out onto flat road. If I've pedalled down the hill to be changing down through the gears ready to pull away, then I get zero power as I pull away, only kicking in some way along the flat.
Generally the gear changing would be with brake switch operated, but I will test this further.


It's starting to look like it's down to a change of riding habit. Not sure yet how that's going to work out. Often, as one approaches the bottom of a hill, one has to be pedalling in order to change gear ready for the climb out.

As AK quotes the same bug also with KT controllers I can't understand how others have not noticed it. AK also lived in a very hilly area so maybe it's only apparent to us hillbillies.

You live near some pretty big hills Saneagle, does yours not do this?

 

[matthewslack]

If I don't pedal at all going down hill (just let the bike roll) and don't pedal until speed comes down to 15.5mph, then power pick up is nigh on instantaneous.
If I do pedal (bikes gearing will allow me to pedal up to around 30mph) then the system will not kick in until speed is very low (8mph or less) and when it does the increase in watts displayed in painfully slow. There was no sag showing on battery indicator at this point.

Just to be completely sure I have the right end of the stick, is the above behaviour consistent in every V bottom, I.e. 100% correlation with pedal/ no pedal?

Now we're getting to it. The whole nub of the problem is this infuriating intermittant delay...

 

[RogerA]

I have an AKM100 motor and 36V 15A KT controller. I have a few similar situations near me where I'll hit 30+mph coasting downhill and then go straight back up the other side. I've never noticed this issue, but I do start pedaling uphill in high gears before the motor would kick in, and then it kicks in normally when I drop to the cut off speed. If I've had to stop at the bottom of a hill, as I do pulling out onto a main road near me, then the motor always kicks in OK.

 

[Benjahmin]

Just to be completely sure I have the right end of the stick, is the above behaviour consistent in every V bottom, I.e. 100% correlation with pedal/ no pedal?

Yet to be further tested Matthew.
Thanks Roger, this is part of what I'm trying to establish.
I am contemplating getting in touch with the supplier but the thought of trying to explain this in simple translatable English is daunting.

 

[saneagle]

Yet to be further tested Matthew.
Thanks Roger, this is part of what I'm trying to establish.
I am contemplating getting in touch with the supplier but the thought of trying to explain this in simple translatable English is daunting.

Suppliers don't have a clue how their stuff works. You'd be wasting your time. there is nothing wrong with what you have. You just have to figure out how to use it. Try what I suggested in post #206. I'm pretty sure that it'll solve your issue.

 

[AntonyC]

The Ezee was fully there straight away, this is what I'm used to. This, of course, was speed control and I'm wondering if changing this system to speed control would help this issue.

That's a worthy straw to clutch at as "current" control's not as simple as speed control. With either, at the limit speed the power should fade in and out gradually to prevent annoying 'bouncing' off the limit, like when you're pedalling downhill and the bike slows on the up. It's OK for power to respond quickly to user input, like brakes or onset of pedalling.

But it's still a straw because the geometric PAS levels suggest KT's non-speed mode might work similarly to speed control.

 

[Benjahmin]

Weather and work committments just allowed me a 20 mile ride.
I think I'm starting to get the hang of it, though I will do some more testing before I draw conclusions.
I have tried four test parameters.
1 Pedal dowhill overspeed (18/19mph+), no brake feathering.
Seems to produce the 'bug'. Power will not kick in until speed is down to around 12mph. I'm not sure there is consistency in this. Will test further.
2 Pedal downhill overspeed with brake feathered.
Also seems to produce the 'bug'.
3 Coast downhill overspeed but no pedalling. Careful not to start pedalling until 16mph or below.
Power does come in straight away but at about 60-80w, then ramps up at moderate pace.
4 Down hill overspeed with pas set at zero. Both pedalling and not. Pas switched back on at around 16mph.
Power comes back in timely manner, wattage ramps up smoothly and quickly.

None of this is yet conclusive but does seem to confirm what Saneagle has said about motor rotation (even though I couldn't fully follow it, not being conversant with the internal of the motor).

It did occur to me again today that all this may be that I'm just not used to the subtly of the KT system. There is no way you coild call the Ezze subtle, it's either on or off.

 

[guerney]

1 Pedal dowhill overspeed (18/19mph+), no brake feathering.
Seems to produce the 'bug'. Power will not kick in until speed is down to around 12mph. I'm not sure there is consistency in this. Will test further.

Not knowing what the heck is happening, my brain has reacted to the statement above by blasting out yet another noxious fart, wondering if one wheel magnet is insufficient for your KT controller's firmware to calculate high speed fast enough to resume pedal assist when speed returns to <15.5mph... if the controller is using one wheel magnet for speed calculation, which it may not be, because I have been bigy bamboozled by the labyrinthine WTF PLC settings on this thread, and haven't the faintest idea how your controller is calculating speed using what sensor or sensors. Would adding wheel magnets and making the appropriate change(s) to WTF WTF setting(s) required help? I don't recall my brain eating fish recently but WARNING! This brain fart may be a red herring clutching a straw between it's teeth.

 

[Benjahmin]

If my theory is correct, you'd see something around 30 to 50 watts on the meter when the speed is below 15 mph, you're demanding power, but not getting any. The moment the clutch engages, the motor will go from no load to loaded, so the power should jump to normal levels, depending on what you have the power set to.

The way this occurs, I could see that a spinning mass that has inertia might be playing a part - somehow.
As per my post 214, pedalling overspeed downhill produces the 'bug'/fault condition. As the wattage display is at zero whilst doing so, it is hard to see how/why it is getting a rotation signal. And why does it not produce the bug when not pedalling? Seems to point to pas being involved somewhere, though I thought it's signal would be damped/negated when overspeed.
However if the motor is spinning as the bike comes down to assist speed I can see that back emf would produce the very slow ramp up of wattage consumed . I think !

Guerney, I feel the same when reading the Bafang programming posts. And lets not mention drive train and cadence issues
Like all I would like my kit to work out of the box. However it is nice to be able to tune them to personal requirements, so I guess this is the price we pay for that and generic parts replacement possibility.

 

[matthewslack]

A frequency meter attached to one of the hall sensor outputs would shed a bit more light.

 

[Benjahmin]

Just been to my favourite seaside cafe, Kerallan fish curry - I thankyou.
So it's a climb out of this valley, down into the next, climb out then down to the coast.
I can now confirm that if I pedal at over 19/20mph (with or without brake) there will be a delayed and slow re-application of power.
If there is no pedalling above cutoff speed power application will be smooth and timely when pedalling starts again.
I now know how to control and use the system and it is, as has been said, a beast.
The climb out of the seaside village is either a brutal wooded climb or a long undulating climb out. Today I chose the former. As I got to the top (just as it flattened out) I came across unannounced resurfacing works blocking the road. I had no choice but to turn round meaning a second climb out of the beach. This gave me some concerns for range because, in colder weather, I have very nearly run out of battery on this run.
Gotta say this system seems more efficient than the last and I arrived home with plenty of battery to spare.

I have C14 set at 3 which is giving pas1-4 higher wattages. It's working well for me. It's giving around 120w in pas 1 which, on the flat, gives me around 14mph with my effort. Perfect.

So, in conclusion, a 328rpm 48v motor run at 36v in a 700c wheel was just plain the wrong choice. This 36v 210 rpm running at 36v is just the ticket. Always with the option of going 48v as old gitness advances.
My son in law needs a replacement folder for commuting, I shall look at using the 328 for him.

Thanks for all the help, tech ponderings and brain farts.

 

[AntonyC]

If ever there was a dark art it's hub motors in steep terrain. The actual change in assistance may be small but if the remaining gradient is near your limit it's a big deal.

No wait, councils are a dark... something. I explained the above and still we got a new 'road table' on our 12% hill, "only adds 3%" they said.

 

[Sparksandbangs]

After three runs with a different controller and with all other components the same there has been no replication of the fault. I'm calling this as a software issue with the KT controllers.