Better than the Stark Varg? 76v 90Ah 35kw Dirtbike Build

In the last years Fardriver had a lot of problems delivering the 108V series.
Actually there is only the ND108530 and the ND1081000 available on Aliexpress.
The last one will be more than enough for a QS138 70h.
I had used a ND72680 with up to 300bA and 630pA in my 190kg heavy scooter and had no problems with it.
Very small an powerful. Now I have installed a ND721800 and I use only half of its battery power 400A and 1400A phase current to get a good acceleration from a standstill with the installed 10kW hub motor.
 
Blacksheep Engineering said:
Ok, whats the tuning like with the fardrivers? Do they also use Vesc?
Click to expand...
They use their own app, it's not VESC.
It's not incredibly great but it's more than enough to do whatever you need. It's not like you'll tweak the motor configuration everyday anyway, once it's properly set up people usually don't keep making changes, even though they think they will when they start their builds (I know I was one of them lol).
 
Ok, looking into those controllers now. The KW rating seems to always be really low compared to what it should be based on the amps and volts. (ND961200 is 96v 600 amp peak) it says 8-12kw? I'm guessing its based off of the suggested motor continuous rating? My qs138 technically has a 3kw continuous rating even though E&C has torture tested it endlessly at 35kw with no issues.

Also on that note, their testing has been at 76v 35kw. Would higher voltage and lower amps equal the same stress on the motor at 35kw? More or less?
 
Blacksheep Engineering said:
I'm guessing its based off of the suggested motor continuous rating?
Click to expand...
Yes, don't bother with this it's meaningless. The real power is determined by battery voltage x battery current. nothing else matters.
Blacksheep Engineering said:
Also on that note, their testing has been at 76v 35kw. Would higher voltage and lower amps equal the same stress on the motor at 35kw? More or less?
Click to expand...
Higher voltage is usually always better.
The stress on the motor isn't too much correlated as far as I know because the controller kinda acts like a buck converter so the voltage from the battery is vastly different from the voltage you get at the motor. At least that's valid at low rpm when the controller delivers current to start from standstill. But maybe that gets more important when the motor starts spinning faster, I'm not entirely sure.

Usually higher current means higher losses so going for higher voltage should induce less heat, so in theory the motor should be under less stress. But that's just theory as motors are complicated things in permanent movement and I'm not an expert.

But yeah, higher voltage is always the way to go, as long as you can get a controller that tolerates it. Most of the time we stick with 72V or 84V because it gets more difficult/expensive to get higher voltage rated controllers and BMS.
 
I do not know if the QS 138 70H V3 can really make use of voltages like the ND108 series can handle.
It is a motor designed for 72V, with 108V you can get more than twice the power out of the motor, but the motor needs to handle the additional generated heat. I would go for a ND96 series.
The stress on the motor is nearly the same when you use higher battery voltage.
The big advantage is when you need to apply field weakening because you do not have a high enough voltage to get higher rpm.
 
Last edited:
You need phase amps to make torque and voltage for rpm.
When you are picking up rpm the motor start producing voltage, back emf. That results in that you need higher voltage from the controller to push the same amount of phaseA.
When the voltage isnt enough the phaseA starts to drop, and so does the torque.

MPHeMGE.jpg

Motor Simulator - Tools

Here is an example of running two identical systems at different voltage without battery current limitation (set to 1000A, way over what it will reach)
As you can see on the blue lines they both have the same torque, because they both have the same phaseA.
But because of the higher voltage system b can hold the max phaseA and torque to a higher rpm, and taper off at a higher rpm resulting in more power than you can get at a lower voltage without changing the kv in the motor.
 
Ok, the BMS I have now is limited to 24s. If I get a new BMS that can do 26s then I can run the ND961000 or ND961200 at max voltage. Is the difference from 88.8v to 96v worth it?

The BMS I would get goes up to 30s, so I could run a 108v controller. Is that worth the extra $500 for that controller?

I guess the main factor would be my battery connections. That's the main reason I'm debating raising the voltage. At 76v I have a 460 amp peak draw, which Im having a very hard time finding a reasonable way to make connections strong enough for. 88.8v is 394 amps. 96v is 365 amps. 108v is 318 amps. That seems more doable.

@j bjork I checked out your 450 build, very nice, what company did you use for the copper connections? Those look really nice. What amps are you pulling from that setup?
 
Another controller option could be the 96V Votol EM550 but this would saturate that motor very quickly. I'm running the little brother EM260S at 28s in my build using the 90h motor. So that fits within your BMS rating voltage rating and squeezes another 2 cells above 26s if thats the limit of the fardriver ND96XXX. From the looks below, you're all good to run those at 28s.
Link below: High Quality Votol Controller EM550 Manufacturers Suppliers Factory
 
Last edited:
You can run the ND96 series with 28s and up to 118V.
 
Id like to avoid the Votul controllers if there is a good option with a better interface, but I will keep that in mind. Thank you.

Ok, flipping my thinking on the controllers here since you point out something I hadn't thought about @dominik h . For the BMS, the series connections are the important part since it has that many wires, but for controllers, the output voltage would be 96v max in this case but I could have a bigger than 96v battery feeding it as long as the max charge of the battery doesnt exceed their peak voltage rating, the ND961000 I'm looking at is 115v. In theory, a 28s battery would have a 117.6v max charge so if I could get one that has the 118v max that you're talking about, that would work. That would also prolong the percentage of the battery life that could deliver the full 96v right?
 
My ND961800 has a max voltage of 121,5v, but it reads a little high so in reality it is 120,5v before it shows overvoltage.
I would expect the whole 96v line to have the same cut out voltage, but I cant say for sure.

The battery bus bars comes from aliexpress, I think the store name is wellgo or something similar.
 
My ND96850 has it's overvoltage limit at 121.5V and release at 119,5V but it reads 1.3V to high.
It tested it up to 123V to see what happens, it only says OVP and does not turn the motor, if you get under 118V it works normal again.

28x4,15V should be no problem when the motor is running. And 118V without any load is also no problem.
You also can set the upper voltage limit for the battery , so the controller does not charge the battery to high with regeneration braking.

If you really want to buy a Fardriver, buy it from Siaecosys or direct from Fardriver or maybe from DUN. My ND721800 is labelled DKYS and their Support technican could not solve any of my problems.
When I bought my ND96850 from Siaecosys there was a wrong Firmware on it. It had set it's max voltage to 88V. They first sent me a video how to upload the firmware, but it did not work with my phone. Then I had two times 2 hours an online Session with a technican from siaecosys and they really tried to help me. They could not a that time because the Android app had a Bug, so I could not upload the new firmware they sent me. They showed me live how they did it with a controller and the chineese app version. A few days later I managed to upload it with an IPhone. And with the next Android App Version I could also do it with the Android phone.
 
Last edited:
Fardriver lists all those as 115v max. I know the numbers are often incorrect, is that what your controllers said too? Just making sure they didnt make an update and dont allow more than 115v anymore.

So which ND96 model should I get then?
-Im guessing 850 isnt enough phase amps.
-The 1000 would be great I think but it seems to have been discontinued, but still available from a seller on ali express.
-The 1200 is widely available from the places mentioned but it weighs 5kg! The 1000 weighs 3kg.
-If Im going to have all that extra weight, the 1800 is only .4kg heavier than the 1200. I dont think it would have any benefit over the 1200 in my case though.
 
Blacksheep Engineering said:
Fardriver lists all those as 115v max. I know the numbers are often incorrect, is that what your controllers said too? Just making sure they didnt make an update and dont allow more than 115v anymore.
Click to expand...
The software allows you to set it higher, my nd96680, which was purchased ~3 weeks ago and listed 115 as the max voltage let me go up to 120v.
 
Exactly what I wanted to hear, thank you.

I started on a model of a 28s 14p battery. This would give me 7.25kwh using the cells I have!
Heres pictures of one half. The other half would be the same thing mirrored.
1731115243710.png1731115268123.png
 
My bike has the qs138 90h water cooled, a 20s10p P42a pack, and the fardriver ND721800. I have the Line amps at 450. I don't get much heat from my battery. Granted it only sees the full 450 under hard acceleration. I get about 8v of sag when I pull the full 450. I used the copper nickel sandwich method with .2mm copper sheet cut with scissors and .15mm nickel plated steel for the sandwich. My pack is insulated with sheets of heat press foam rubber and silicone inside of a steel housing.The bike is capable of 80+mph speeds and it will easily lift the front wheel if I take off too hard. I can go 30 miles if I stay under 40mph and avoid too much hard acceleration. If I keep the RPM down I can go 60+ miles but that's on the extreme end going slow and purposely trying to save battery. My bike is relatively lightweight compared to what you're building so take that into account as well. It's under 200lbs. I would agree that a 96v battery is the way to go. You will get a wider usable RPM range as stated by everyone else. Although I think there's still more potential on my setup with a larger 72v pack, it will probably come down to gearing the bike to use the extra torque on the low end when I do build a bigger 72v pack. I can feel it lose a ton of torque when it hits the back EMF wall. Field weakening will get it to a little over 5000rpm but it loses a lot of torque above 4000rpm. My controller stays relatively cool. I can't overheat it with this setup and the motor barely gets warm. It's generally operating at 29kw when I ride it hard. That's about as much as 10p will get me with the p42a before the voltage sag gets to be unreasonable. It looks to me like you're on the right track. There's a lot of very knowledgeable people here. This forum has been indispensable to me for sure. Just figured I would throw my experiences and my 2¢ in. I'll be following along with your build for sure. Sounds like it's going to be awesome!
 
What phase amp settings do you have with the water cooled QS138?From which speed do you get 29kW?
 
Thanks for sharing, that does make me less scared of the amps, from the research I had done, to get 460 amps out of my battery I would have needed 3 times the ability of a copper nickel sandwich. So now especially with 1mm copper busbars I shouldn't have any problem even if I was still going to have 460 amps of draw. Still feel extra cozy with 365. That sounds like a sweet bike.

Here is what the battery is looking like with some more completed CAD.
I added very small cutouts where I want to bend the copper so that it will bend nicely. My plan is to bend them in my vise and then fill in the holes with solder before spot welding them to cells.
The long grey connectors will fold towards each other and have braided copper stringers to attach them.
The smaller blue ones will be my positive and negative connections, also folded towards each other.
I remember seeing someone on here made their cell holders with little walls that slide up between the bus bars to keep them from touching. Thanks for the idea.
Im going to cut out a new reference model to fit into the bike and as long as it fits like Im thinking I'll go ahead and order the busbars, new BMS, and controller.
1731178597935.png1731178640028.png1731178946076.png
 
How comes nobody get overheating from their controller/motor? I have a Qs180 and cl1000 and the controller overheat everytime I ride the bike and the motor get close too. Maybe the 180 is a bad motor in efficiency compared to the 138?
Also for the battery Ken bao from wellgo is very nice always respond fast to email! Your battery design looks very good
 
dominik h said:
What phase amp settings do you have with the water cooled QS138?From which speed do you get 29kW?
Click to expand...

I have the phase amp setting at 1800. I can't pin the throttle from a dead stop because it has too much torque. It will just wheelie. I feather the throttle so that the front wheel is just about to lift up off the ground and it will pull like that to about 40mph or so before the torque starts to drop off. During this time is when I see the 29kw. It pulls almost exactly 400 amps and the voltage drops from 83.9 or so to around 76.
 
Last edited:
rivvs said:
How comes nobody get overheating from their controller/motor? I have a Qs180 and cl1000 and the controller overheat everytime I ride the bike and the motor get close too. Maybe the 180 is a bad motor in efficiency compared to the 138?
Also for the battery Ken bao from wellgo is very nice always respond fast to email! Your battery design looks very good
Click to expand...
Don't get me wrong my controller does heat up. It's rated for 800 amps though. I'm only using half that and for short bursts. The ND721800 is a giant controller. It has a large aluminum block on the bottom which is in thermal contact with the MOSFETs. It gets hot but not to the point of overheating.

Blacksheep Engineering said:
Thanks for sharing, that does make me less scared of the amps, from the research I had done, to get 460 amps out of my battery I would have needed 3 times the ability of a copper nickel sandwich...
Click to expand...
I know where you're coming from. I had the same fear before I built mine. It's my first battery pack. I saw people building these huge ridiculous packs with copper sheet that, by the math and theory, should be impossible. I think that a lot of the calculations people use are drawn from ideal circumstances. By that, I mean the goal of the calculations is to find the ideal amount of copper for a system that will never heat up. In practice this isn't necessary because very seldom will you actually fully load the buss bars for extended periods. On top of that, the distance through the buss bar isn't necessarily taken into consideration. In these battery packs we build, the distance is very short compared to something like a 480v buss bar in a factory for example. A 480v buss bar in a factory is something that will be loaded down 24/7 in certain cases so it's important that the heat output never exceeds what can be dissipated. As long as you have enough thermal mass available you can reach a balance point. Even if you're slowly building up heat over time, it's ok because you'll be able to deplete the battery before it catches up to you. With that said its definitely better to be safe than sorry. Go as big as you feasibly can.

Your model looks great to me. I like the copper braid idea. It will be easy to assemble that way. I'm not sure how you're planning to attach the main battery wires to the blue tabs but If you're planning to use lugs and bolt them to the tabs make sure to consider where the bolt shanks will end up in proximity to the sides of those cells on the end. I've seen a few builds around the net where they bolt lugs on that way and the bolts are scary close to the cans of the cells. Ideally, you would put this into a sturdy enclosure where nothing is going to move around anyway but things can happen. I folded the end tabs the same way in my pack and I soldered the cables to the tabs as well as glueing pieces of silicone sheet between the folded portion and the sides of those end cells. I stripped about 6 inches of insulation off the 4 gauge cable and soldered that whole length to the end tab.

This is a cool project! I'm definitely following along with this one. 👍
 
Ok, Ken from Wellgo is suggesting .3mm copper for the busbars? I saw @j bjork got 1mm. I was thinking .5 should be good. Is that right? Wouldnt hurt to save some weight but .3mm seems too small to me.
 
You must log in or register to reply here.

Similar threads

Ananas
    • Like
Battery Enclosure Design Considerations
Replies
9
Views
1,634
Etienne
E
G
Need some help/advise on a build.
Replies
0
Views
712
Gemini8026
G
R
    • Like
    • Brainy
Really good article comparing Bafang vs. Tongsheng vs. CYC
Replies
14
Views
27,253
Chalo
Chalo
T
*USED FOR SALE* Grin Technologies 52V eBike DIY Build
Replies
2
Views
2,222
tomceg123
T
L
FOSSHW Category L7e Hybrid EV (full build: not a conversion)
2
Replies
46
Views
7,628
classicalgas
C
Back
Top