Super Commuter: 6T MAC 48V/120A Kelly in a 26" rim

[madin88]

So far, i have not experienced any battery current limit, even if running more than 100A for more than 30 seconds. So the maximum Battery current is equal to the maximum phase current at the point of maximum power. This is at 40kph for this bike. As a result, at all other speeds it is less.

thats strange. during last year i have setup some of same controllers with MAC 8T, and i have seen always set battery current as maximum on the CA (with external shunt).

i believe you know how the current needs to be calculated with kelly controllers, but just that we talking about the same:

rated current (130A on KBS48121X) x phase current percent = max phase current
THIS max phase current x battery current percent = max battery current

no problems with this, but i noticed other problems:

- no difference noticeable between speed and torque based throttle
- regen only worked with big Ah battery (or low resistance) no matter what HVC voltage
- very sluggish and stuttering acceleration from 0-10kmh with DD hubs (i believe thats nomral on kelly controllers because of self protection)

 

[crossbreak]

The way you do, it should be right. But it is different in my experience. So you never saw currents greater than
130A*70% = 91A ?
I didn't play a lot with the current setting. So far max battery current was always the same as max phase current. I always programmed it to 70% battery current. Maybe this differs if the battery current is set smaller than this.

The 130A are more like 150A in real life. For the KBS48121X it is more than for the KBS72121X, which somehow does only ~120A when connected to a 48V battery.

I did not try the speed throttle, but yes, the current throttle is still a bit like a speed throttle. That really sucks

Yep, I experience the same sluggish acceleration at 0..10kph. No stuttering but it definitely does not the phase current i program. Below 5kph i cant wheely. That really sucks

Another issue: If a hall sensor fails, it does not fall back to sensorless mode like the x8m06c controllers do

- regen only worked with big Ah battery (or low resistance) no matter what HVC voltage

Did you ask Fany about this?

 

[madin88]

- regen only worked with big Ah battery (or low resistance) no matter what HVC voltage

Did you ask Fany about this?

yes,
she answered i should try a battery with higher Ah - than it worked. before it only worked for a second than it cut out.
i cannot understand why battery Ah (thus Ri) should play a role because HVC should watch over this and it never reached (have seen maximum 1V rise during regen). maybe the controller does measure voltage rise and if it exceeds a specified value than it cuts, but thats only a guess..

I never saw current more tan i have set. maybe there is an issue with the program because of different firmware or program version, and 70% should mean 100% of set phase current. afaik some kelly controllers do have 100% here as maximum value for choose.

 

[telnet777]

Hi crossbreak,
Please tell combination connection Kelly's wires (phases and sensors) to MAC motor.

Kelly Controller > MAC Motor
sensors B>Y; G>G; Y>B
phases B>Y; G>G; Y>B
This is true?

Thanks !

 

[madin88]

blue and yellow must be swapped on both phase and hall wires, so yes it should be ok

 

[crossbreak]

yep, just try it...you can't damage anything

 

[crossbreak]

added sale thread for the torque arms http://endless-sphere.com/forums/viewtopic.php?uid=21540&f=31&t=68835&start=0

 

[crossbreak]

small update about oil cooling. the shaft seals are tight, the side covers are tight, but i did not manage to completely seal the phase wires. There are always a few drops getting out.

I'll try using very thin silicone for sealing the shaft next. The used oil is Total Isovoltine 2 insulating oil. Very thin. Still eats about 150W more at 60kph now at no load :/. But Temperatures are very low now. Amazing how much effect this has. 3kW continous input power are not a problem anymore. I never reached more than 90°C after this mod.

Oil volume was ~110ml

As efficiency is quite low when climbing 15%+ hills, i think about adding a BBS02 750W middrive to help the MAC a bit on steep hills.

 

[Andytheswede]

So why the 6t instead of the 8t?

 

[crossbreak]

the 6T does 50kph when the battery is almost empty (42V). The 8T would only do ~42kph with such little voltage of an empty battery. Too slow for town traffic.

I see no advantage in using the 8T. I had a 10T MAC at 74V before. Performance is the same. Both systems draw around 4000W.

But handling voltages >60V is not legal without special design and education in Europe. And I found no fan-less charger that is like the HLG240. The others i found are low quality and/or expensive. That's why i went for 48V (54V fully charged).

 

[crossbreak]

ok i found out that the new V4.5 software wont accept the 120A peak setting. Max is around 85A with the new one. According to Kelly it is dangerous to use 120A, due possible damage to the controller.

I can give you the old software, just ask

 

[izeman]

Do you know the best settings for the kbs48101? I also must say that acceleration from stand still is very poor. I also found that maximim power seems to depend on the motor connected. With the 10t max power consumed is much lower that for the 8t. It takes a really steep hill to go up to 2400w with the 10t. The 8t performs much better. And it's not because the 10t reaches nominal speed already.
I wish the kelly would be more powerful at low speeds. At a stand still at a steep hill i can be that the kelly cuts power instead of lifting the front wheel or slipping the rear wheel. Really disappointing if you think that you programmed it to 2.5kw.

 

[crossbreak]

a bit disappointing but good for a commuter. i have 3kkm on the clock and the clutch still holds up...pulling 4kW peak.

I now this behavior at low voltage and the 10T. try 74V with the 10t it will be fine or a 6T at 48V

 

[izeman]

I can't change voltage. That's fixed at this bike.
Do you know the amp limits for the small kelly?

 

[DrKraut]

Very fine build ! 8)
I also live in Germany and asked myself if you are not scared to ride in cities with such a bike. You know its not legal without a license :lol: ?
What's your plan if the cops stop you?

 

[spinningmagnets]

The pictures at the beginning are very well done. Pics + comprehensive text is always better than just one or the other. Thanks for posting this!

 

[crossbreak]

of course i dont ride on public roads

What's your plan if the cops stop you?

thanks spinningmagnets

 

[crossbreak]

maddin told me that hilumin is not pure nickel, it's steel plated nickel. so my calcs about the conductivity/ resistance loss of the cell connectors were wrong. I measured resistance of a 1m x 25mm x .15mm hilumin stripe. it's 35mOhms per meter. cross section is 3.75mm².

conductivity is

which can be written as ρ = l / (R * A) = 1m / (35 mOhm * 3.75mm²) = 7.62·10^6 m/Ohm. Copper has 58·10^6 m/Ohm. So this has 13% of the conductivity of copper. Nickel has 13.9·10^6, almost double the value of this Hilumin shit i used :/

you can easily test if it is pure nickel: If it's not magnetic it wont be steel. all the stripes i used are in fact ferro magnetic Edit: Nickel is actually ferromagnetic, too

20.4mm² hilumin is a copper equivalent is only 2.65mm², or AWG13. Still, the distance between cells is only ~20mm. As there are 12 cell connectors, total length is 0.24m. Resistance of the connectors can be computed: .24m * 6.8 mOhm/meter = 1.6 mOhms. This sounds reasonable, as total battery resistance is about ~50 mOhms (one cell is ~25mOhms), this is only about 2% of total resistance. I can imagine a 1.6mOhm shunt with the large surface of all these connectors wont get even warm

 

[madin88]

nice gadget for measuring resistance. may i ask from where you have it?

 

[crossbreak]

bought it for this purpose: http://endless-sphere.com/forums/viewtopic.php?f=30&t=65757&start=175#p1072783

 

[izeman]

Very fine build ! 8)
I also live in Germany and asked myself if you are not scared to ride in cities with such a bike. You know its not legal without a license :lol: ?
What's your plan if the cops stop you?

Two words: GERMAN ANGST.
One idea: don't get caught!!
Enjoy life!

 

[crossbreak]

hey izeman, i always go to Austria if i want to use public roads

what i think is the issue with the kellys: They depend highly on the motor used. If you use a slow setup (like 8T MAC and 48V) it wont ever reach the 150A. I never saw the 150A on the 6T/48V. I was (after calibrating my CA) more like 100A. But i did measure the 150A on a 10T MAC on 74V. Somehow it must have something to do with the wind AND the battery voltage used.

I always try to use a controller that is overrated for the motor and throttle it down to a reasonable behavior. I know that this is not the way how we should do things as it is not too cost effective, but it has always worked for me. the difference in controller price was not a big deal in the end. Kelly KBS-X ( the <120A ones) are all the same size and size plays an important role IMO. I never reached controller overtemps before motor overtemp so this is another reason to further go this route

 

[izeman]

weird. for me it's quite the opposite. my 8t pulls more amps and the 10t doesn't seem to reach higher amps.

 

[crossbreak]

as said, try it with 74V, you will see 130A+ too with both motors. in theory the 74V/10T should pull the same amps as a 6T at 48V (which means of course less power), but in fact the 10T at 74V pulls more battery amps (74V and 150A mean about 10kW, insane to pump it into a MAC) . The 6T pulls the same 150A at 74V. this does not matter anymore at some point where your setup gets ultimately fast. but at slower setups and lower voltage this somehow makes some difference

i really tried the 74V/150A on a 10T Bafang BPM last summer, it even survived almost 1km!! then I had to rewind since the windings where toast and the grease was more like a mixture of soap and ash. Sadly i had no temp probe in it at that time

 

[izeman]

I can't raise the bike's voltage. Battery setup is fixed. I need to find tbe the best combination of motor and controller and the correct setup (percentage current setting)