fizzit said:
yes, don't need any software upgrade for it since no display restriction on current field.
do you want it for your castle application ? we have RC firmware for rc application, can you tell me your voltage limit ? 55V ok ?
Speedict Ebike - Anybody seen/tried this?
- Thread starter newb
- Start date
yes, don't need any software upgrade for it since no display restriction on current field.
do you want it for your castle application ? we have RC firmware for rc application, can you tell me your voltage limit ? 55V ok ?
N!co
10 W
TheBeastie said:
Wh/km is possible.
Ahoog
100 mW
I bought a ZTE Blade III to use exclusively on my bike for Speedict.
However, this phone has Android 4.0.4 and the Speedict eBike app isn't compatible as it seems.
I was in contact with Danny regardig this and he had no solution. I searched the net for other ROMs with no luck.
Has anyone here had any luck with Android 4.0.4 and the app?
/Anders
However, this phone has Android 4.0.4 and the Speedict eBike app isn't compatible as it seems.
I was in contact with Danny regardig this and he had no solution. I searched the net for other ROMs with no luck.
Has anyone here had any luck with Android 4.0.4 and the app?
/Anders
N!co
10 W
I run it on 4.04 on my SGS2, no problem.
N!co said:
i don't mean that 4.0.4 is incompatible, the reason is 4.0.4 is custom ROM not official from Google, some phone company modify "too much" without fully tested ! i have friend with custom ROM phone without Google maps installed
Ahoog
100 mW
speedict said:N!co said:
i don't mean that 4.0.4 is incompatible, the reason is 4.0.4 is custom ROM not official from Google, some phone company modify "too much" without fully tested ! i have friend with custom ROM phone without Google maps installed
So it seems I was just unlucky when I choose my phone....
I should have picked the one next to it with 2.3 but too late now. Maybe I should go for the CA instead if it is an unfixable problem.
The app gave some kind of crash report and since I'm kinda new to this whole android thing I just pressed "send report" without knowing where that info went. To Speedict I hope
If you got it, please look into it when you have some resources to spare from the eBike V2 project.
Any suggestion on how to get a non tweaked OS onto the phone is greatly appreciated.
/Anders
fizzit
10 kW
speedict said:
Yup that is exactly what I would use it for. 55V would work well.
I think I will buy a speedict after X-mas, if santa doesn't bring me one.
fizzit
10 kW
speedict said:
Unfortunately I have not downloaded the data log in a long time, so I don't know how accurate it is. I just downloaded it and I don't see any measurements above 60A, which is surprising. I think the log might be from when I had a smaller motor.
If you told me the part number of the shunt that is used in Speedict, would it be possible for me to order another shunt and solder it on top of the one that is already in the speedict? That way, it could measure exactly twice as much.
Ken Taylor
100 W
speedict said:
I've been looking unsuccessfully for a field oriented controller for ebike motors to reduce noise from torque ripple and here I see one demonstrated which is nice. Can I buy it? How does this relate to the Speedict product? Is the Speedict going to be built in to motor controllers in future, will there be two products or is this a side project? Also the heading says "dynamic change via Android Device" and I'm wondering why you would want dynamic change other than for demo videos. Once you have FOC I can't see any reason to change back.
Ken Taylor said:
it is still not available, we stopped the development for a while because we are busy on Speedict Mercury, Mars, Venus and Jupiter production in recent month, we will resume the development later after those products launched, on the other hand this is just a speed controller without any surprise, internally we treat it as base module, on top of it is feature we developing on Mars, Mercury such devices e.g. Android / iPhone client, remote firmware upgrade, cloud data store, live view, data logging, throttle controls etc. ....
Ken Taylor said:
you are right, just for demo purpose, however having both type of commutation change dynamically can compare the efficiency
Ken Taylor
100 W
speedict said:
Wow that was a fast reply. I thought FOC controllers would be available in the RC world similar to this trapezoidal ESC from hobby King http://www.hobbyking.com/hobbyking/stor ... duct=10331 but as far as I can tell they aren't available anywhere. I can't understand why all BLDC motors that use hall sensors are not FOC as their doesn't seem to be any negatives except perhaps MCU cost. Is their a gap in the market, is it the extra MCU cost or is it something I'm missing?
Ken Taylor said:
it is difficult to develop Sinusoidal commutation controller on rc world because its lack of hall sensor, also rc just need high rpm instead !
Ken Taylor said:
Sinusoidal commutation is complicated on both hardware and software that's what you referred to "cost" e.g. hall timing and sequence is critical on sinusoidal, Trapezoidal communication need to control 1 x pwm signal but Sinusoidal need 3 ...
you can see most of the china made speed controller still use transistor to drive its mosfet because of COST, cost is everything to them, i have conversation with "some of them" as per their mindset if they can save $1 component cost each, when sold 10,000 units they can save $10,000
they don't even think about if they put $2 component cost they can sold 100,000 units
Ken Taylor
100 W
speedict said:
I understand most of the noise in RC motors is from torque ripple which Sinusoidal commutation would remove and that would be very nice.
I only just realised that field oriented control and sine control is not the same thing from this post by Burtie at http://www.endless-sphere.com/forums/viewtopic.php?f=30&t=44910#p654534 but the R8C/25 application note you posted says on pg3, "As many
brushless motors have sinusoidal BEMF (called Permanent Magnet Synchronous Motors PMSM), it is possible to
match these motors with a sinusoidal driving voltage". Therefore for these motors FOC and sinusoidal commutation produces the same controller output and the sinusoidal commutation algorithm is simpler. Burtie's Microchip dsPIC33 kit has algorithms for sensorless position detection that would be suitable for sinusoidal commutation but he finds it complicated to get them to work as described at http://www.endless-sphere.com/forums/viewtopic.php?f=30&t=44910#p654203
speedict said:
The hardware cost could make it uneconomic but software cost, is once off and if sinusoidal is better it should be able to be spread across every motor controller sold. It needs good engineers but Speedict seems to have them.
speedict said:you can see most of the china made speed controller still use transistor to drive its mosfet because of COST, cost is everything to them, i have conversation with "some of them" as per their mindset if they can save $1 component cost each, when sold 10,000 units they can save $10,000
I'm interpreting your statement "they don't even think about if they put $2 component cost they can sold 100,000 units" as a belief that there is a gap in the market.
Ken Taylor said:speedict said:
I understand most of the noise in RC motors is from torque ripple which Sinusoidal commutation would remove and that would be very nice.
I only just realised that field oriented control and sine control is not the same thing from this post by Burtie at http://www.endless-sphere.com/forums/viewtopic.php?f=30&t=44910#p654534 but the R8C/25 application note you posted says on pg3, "As many
brushless motors have sinusoidal BEMF (called Permanent Magnet Synchronous Motors PMSM), it is possible to
match these motors with a sinusoidal driving voltage". Therefore for these motors FOC and sinusoidal commutation produces the same controller output and the sinusoidal commutation algorithm is simpler. Burtie's Microchip dsPIC33 kit has algorithms for sensorless position detection that would be suitable for sinusoidal commutation but he finds it complicated to get them to work as described at http://www.endless-sphere.com/forums/viewtopic.php?f=30&t=44910#p654203
speedict said:
The hardware cost could make it uneconomic but software cost, is once off and if sinusoidal is better it should be able to be spread across every motor controller sold. It needs good engineers but Speedict seems to have them.
speedict said:you can see most of the china made speed controller still use transistor to drive its mosfet because of COST, cost is everything to them, i have conversation with "some of them" as per their mindset if they can save $1 component cost each, when sold 10,000 units they can save $10,000
I'm interpreting your statement "they don't even think about if they put $2 component cost they can sold 100,000 units" as a belief that there is a gap in the market.
FOC sensor-less is hard for e-bike market, for washing machine it is ok ...
Ken Taylor
100 W
speedict said:
Yes, that's what Burtie found.
Similar threads
