Delfast Top3.0i / QS205 V3 hub motor with ND72450 controller.

[doingresearch]

hi all!
last year i got my new delfast top3.0i ebike. i use it in winter and few weeks in spring. then on a ordinary charging session the thing broke. smoke came out of motor controller and that was end of story. because i have full waranty i tried to get help from delfast support. did not work as expected. delfast stopped talking to me for no reason.
now i sit here and try to figure out what i need to repair my broken ebike. what i need in first place is a controller, that works like this one -> Brushless DC Controller for eBike







what also broke is the DC-buck converter which brings the voltage from battery (84V) down to 12V. the capacitor in this converter has blown up. by the way. i like to replace the electronics with high quality components only. i'm not intrerested to use 5$ buck-converter in a € 7000,- ebike.

i also checked the battery which is fine and has not been damaged in any way.

it would be very helpful if you can give me some tips where to look for such a controller or maybe i can upgrade to bigger motor/controller combination?

thank you very much
regards from austria!

 

[amberwolf]

I don't have a recommendation on DC-DC or controller, but I do recommend you check your charger, because if both the main items that would likely be connected to the charger during charging (other than the battery) were damaged in the way they were, it is likely that the charger is outputting too high a voltage.

It might not even do it all the time, but just during say, final stage of charging when the BMS disconnects the cells from it, so that the charger voltage could spike if it doesn't shut itself down fast enough and it has a problem with voltage regulation.

If it's a 72v pack, the charger should have a max voltage of 84v at any time.

I'd test it first without connecting it to anything except the wall AC cord.

 

[doingresearch]

finally i bought a Fardriver NS72450 and it works
i use v223 of FarDriverApp. got it from Alibaba where i bought the controller.

Electric Motorcycle 3000w Programmable Sinewave Foc Bldc Qs Motor E-bike Controller Fardriver Nd72360 - Buy 3000w Motor Controller,72v 400a Brushless Motor Controller,Qs Motor Controller Product on Alibaba.com

Electric Motorcycle 3000w Programmable Sinewave Foc Bldc Qs Motor E-bike Controller Fardriver Nd72360 - Buy 3000w Motor Controller,72v 400a Brushless Motor Controller,Qs Motor Controller Product on Alibaba.com

www.alibaba.com

now i'm looking for information on my hub motor (Rear Hub Motor for Delfast Top 3.0 and Top 3.0i)
i don't know how many pole pairs it has and i think i will need the correct this value within the Fardriver App. After AutoLearning procedure it came up with 4 pole pairs, but i don't think that this is correct.

there is only this number on the motor: 211221 1668 3.5T



have a nice day!

 

[doingresearch]

for the charger... i plan to use a zener diode in series with resistor to limit the voltage in case the charger goes crazy (this is called positive clamping circuit). this type of diode should work i think -> https://www.tme.eu/Document/1b6b0cbe4740a988d003793015c93864/nte5240a.pdf

It is crucial to note that resistor is required to limit the current through the zener diode and prevent it from burning out. The value of resistor can be calculated using Ohm's Law. R = (Vin - Vz) / Iz

have a nice day!

 

[amberwolf]

now i'm looking for information on my hub motor (Rear Hub Motor for Delfast Top 3.0 and Top 3.0i)
i don't know how many pole pairs it has and i think i will need the correct this value within the Fardriver App. After AutoLearning procedure it came up with 4 pole pairs, but i don't think that this is correct.

This may help:

How to find number of poles?

Hello, I have a BLDC motor that I am trying to configure the controller for. I need to know the number of poles in the motor but the motor itself has no useful markings on it. I understand one can find the number of poles by counting the number of cogs the motor does when rotated by hand...

endless-sphere.com

 

[amberwolf]

for the charger... i plan to use a zener diode in series with resistor to limit the voltage in case the charger goes crazy (this is called positive clamping circuit).
<snip>
It is crucial to note that resistor is required to limit the current through the zener diode and prevent it from burning out.

Note that if you put them in series with the charger, you are also limiting the current to the battery too, slowing your charging by however much it limits the current. If you don't mind the slower charging, then that's ok.

The resistor will need to be able to handle the full power of the circuit, meaning if you have say, a 2A charger and the resistor at that full 2A drops a few volts across itself (drop will decrease as current does toward end of charge), then the resistor will have to dissipate the heat from the several watts this will create. (V x A = W) So pick a resistor physically large enough to do this.

Also, a zener clamping circuit doesn't put the zener in series with the current flow, it is in parallel with it (see diagram below), with the cathode at the point the resistor (that *is* in series with the current flow on the positive wire) connects to both zener and battery side of the positive wire. The zener anode connects to the ground or negative wire from the charger that also goes to the battery.

You may be able to use the zener as a series diode drop instead, to limit maximum voltage from your charger, but it may work better to just adjust your charger down to the voltage you want it to end at (or get a charger that is adjustable or programmable, like the Cycle Satiator from ebikes.ca).

 

[doingresearch]

today i managed to get regenerative braking to work on my QS205 hub motor. it needs further improvements but for now i'm quite happy with the results.

here are the actual parameters in farDriver App 2.2.4
















for some reason 'EABS when brake release' did not work for me. i got no regeneration with this setting.
'EABS when release throttle' works.

regards!

 

[doingresearch]

i'm on this regenerative-braking-thing... i'm wondering how much breaking is possible with the QS205 bldc hub. when i drive down the hill and reach 50km/h the regerative breaking puts out -470W at the most. i've set the parameters on different rpm speeds to 100%. playing around with maximum back current seems to have a little impact but not much. it is set to 20A actually.

 

[amberwolf]

i'm on this regenerative-braking-thing... i'm wondering how much breaking is possible with the QS205 bldc hub.

That depends on the specific way your controller does it's breaking, the speed you're going vs the winding of the motor vs the votlage of your battery, and the specifcs of the controller's settings that relate to it's braking operation

If your controller and system supports enough braking curent (either negative or positive, depending on braking method), it could lock up the wheel.

FOC controllers can change the current flow in the motor to what would make it spin in reverse, while it'smoving forward, to actively brake the wheel. This doesn't genrate power back to the battery, so it isn't really regen. Whether a specific controller actually is programmed to do this you'd have to check, and on what current limits it has, to know if it's suitable for the braking forces you need. Doing things this way can heat up the motor, cables, and controller quite a lot, so if you have to brake repeatedly like this you could overheat things if there's no automation in the system to prevent that (which would be able to take away your braking unexpectedly, so not usually a good thing to have--a warning would be better).

Regulard cotnrollers use at least two ways of regen braking--one simply accepts any voltage higher than what the battery is and feeds it back to the battery to generate current in the system that loads the motor down, dragging to slow it down. This is only useful at higher speeds.

The other repeatedly shorts t he windings and uses the voltage spikes to get that current flwo that drags on the motor, so it can get braking down to a much lower speed...but not zero, and the higher the battery voltage the hihger the lowest speed it can brake effectively at.

 

[dominik h]

stop back current is the limitation for the controller , the max back current has to be set 50% higher than stop back current according to the manual, but it has no visual impact.

 

[doingresearch]

i'm still learning how to setup the parameters correctly. thank's for all the information!
in the meantime i looked for a GPS app that can run on my android-device (cat S62 pro).
i tried this one -> GitHub - woheller69/gpscockpit: Android GPS dashboard and head-up display



i think i can reach 80 km/h with this motor/controller combination.
regards!

 

[doingresearch]

found out the exact motor type now. it's a QS 205 V3. temperature sensor should be KTY83/122. motor should have 16 pole pairs.
found some useful information here -> http://www.siaecosys.com/upfile/202203/2022032841777345.pdf

what i noticed while testing the regenerative braking is, that sometimes i have to release throttle 2 times to turn regerative braking on. it's ok if you know it...
regards!

 

[E-HP]

i'm on this regenerative-braking-thing... i'm wondering how much breaking is possible with the QS205 bldc hub. when i drive down the hill and reach 50km/h the regerative breaking puts out -470W at the most. i've set the parameters on different rpm speeds to 100%. playing around with maximum back current seems to have a little impact but not much. it is set to 20A actually.

Make sure you monitor your temps on long descents. Regen downhill can heat up your motor as fast as climbing the hill, and it's a drag to have to stop on the way down to let the motor cool.

 

[doingresearch]

guys you are 100% correct on every aspect so far. today i checked temperature while regenerative-breaking and the degrees went up pretty fast. regen-breaking (in my case -470W) results in more heat, than driving constantly with 1500W of power. we have 34°C today, so maybe this has an impact on the electronic devices too.

i still have a problem with the temperature i think. far-driver app shows 90°C on hub-motor but i can touch it with my hands without feeling much heat. maybe i should try another sensor...

regards!

 

[amberwolf]

How long after you see 90C are you testing with your hands? It takes time to get the heat out of teh stator where the sensor is (usuallly right under the windings, which are the hottest part in operation), thru the air gap, and to the side covers. By the time it reaches those covers, it's usually cooled some, and so it can take a while to get the heat out of the motor (which is why it's easy to overheat them by producing more heat than it can shed in the same time period).

Next, what sensor does your motor use? And what sensor is your controller compatible with?

 

[E-HP]

guys you are 100% correct on every aspect so far. today i checked temperature while regenerative-breaking and the degrees went up pretty fast. regen-breaking (in my case -470W) results in more heat, than driving constantly with 1500W of power. we have 34°C today, so maybe this has an impact on the electronic devices too.

i still have a problem with the temperature i think. far-driver app shows 90°C on hub-motor but i can touch it with my hands without feeling much heat. maybe i should try another sensor...

regards!

Heating from regen is in way more dangerous. When the heat comes from running the motor hard, the controller can automatically scale down the power to keep the motor from overheating. But when heating from regen, the only thing keeping it from overheating is you watching the temps and stopping to let it cool.

 

[doingresearch]

did some tests to get a better feeling for the sensor and the motor temperature. after 20 minutes of riding constantly uphill/downhill i reach 55° on the far-driver app. flir-image shows 36° on surface of hub motor. as you mentioned, it took some time for temperature to reach surface.
what i noticed is, that the current never went above 40A when i monitor with battery-app. i wonder where i can tell far-driver how much current he can take from battery.
regards!

what i found out about my motor: QS 205 V3 (16 pole pairs), sensor should be KTY83-122
since i got no information on this whole topic from Delfast support i can only assume, that this is the right motor.
it works with Far-Driver ND72450. still trying to optimize parameter settings.

 

[doingresearch]

today i got a notification within fardriver app (i use 2.2.4) that i need the new app. downloaded from fardriver and it is still 2.2.4 but works now. did the self-learning procedure and for some reason i have more power in the lower regions now. did not know what happened to the App but i like how the motor performs now.
regards!

 

[nalu808]

today i got a notification within fardriver app (i use 2.2.4) that i need the new app. downloaded from fardriver and it is still 2.2.4 but works now. did the self-learning procedure and for some reason i have more power in the lower regions now. did not know what happened to the App but i like how the motor performs now.
regards!

Have u changed in settings in the app that u have posted? I have the same motor and was thinking about getting that same controller.

 

[doingresearch]

a few days ago i updated my fardriver app. had to go through the registration process again and found some sliders that i had never seen before. did some modifications on the 'GearHigh-, GearMiddle- and GearLow Power Output' (60%, 100%, 100%)
after saving i got a 'new' bike. hell yeah!!!
absolutely incredible! i can take over cars on a 15% uphill climb with no problem. the power is amazing now. i have no idea what the new app version changed in background, but this did a lot for my controller and the motor. i can send 110A to the motor now which was not possible in the earlier version of the app. btw i run the app on android11 (cat s62 pro). fardriver-app-version is 2.4.9 (App info details)
regards!

 

[natus]

a few days ago i updated my fardriver app. had to go through the registration process again and found some sliders that i had never seen before. did some modifications on the 'GearHigh-, GearMiddle- and GearLow Power Output' (60%, 100%, 100%)
after saving i got a 'new' bike. hell yeah!!!
absolutely incredible! i can take over cars on a 15% uphill climb with no problem. the power is amazing now. i have no idea what the new app version changed in background, but this did a lot for my controller and the motor. i can send 110A to the motor now which was not possible in the earlier version of the app. btw i run the app on android11 (cat s62 pro). fardriver-app-version is 2.4.9 (App info details)
regards!

please post the new settings FarDriver + qs205 very interesting

 

[doingresearch]

i attached the screenshots from controller. have a nice day!

---
by the way...
when i try to take more than 110A longer, the Battery switches off the power to controller. after few seconds the controller starts automaticaly.