What controller for 72v, 100+ amps?

PRW

100 kW
Joined
Nov 10, 2011
Messages
1,044
Location
Melbourne, Australia
The Phasor is about to receive a new battery - a 21700-40T pack at 20s6p. I have been advised max amps should be around 105A.

Therefore, it also needs a new controller - the existing one is 70A max. The motor is a Mxus 3k. Motor will get fins and ferrofluid

The other constraint is the space to place it in - maximum 240mm x 120mm x 100mm. Otherwise, it will need to go on the outside of the frame, which I would prefer not to do. Ideally, it is relatively plug and play, with a CAV3.

I have identified the following:
Votol EM-100 100A battery, 330A phase - but there are pages and pages on ES where people have struggled to get it to work
https://www.qsmotor.com/product/votol-controller-em-100/
Kelly KLS-S - the 7230S looks like it would just fit, and price looks acceptable
https://kellycontroller.com/shop/kls-s/
Powervelocity the 15kw controller that looks just too big http://powervelocity.com/home/55-24f-15kw-sinewavesilent-controller-android-ios-bluetooth-programmable.html, or 24F Nextgen that looks great, but overkill, and costs $599 http://powervelocity.com/home/60-18f-ebike-sinewavesilent-controller-android-ios-bluetooth-programmable-8kw.html

Comments please, and is there anything else out there? thanks
 
Thanks, I looked at the Sabvoton after I posted. The 72150 would work, but is slightly too big. I ended up booking a Nucular 24F - wait is supposedly 5 months. I didn't get an answer back from Powervelocity if they would supply a 24f Nextgen outside of the USA
 
Does your current controller get hot when running at its existing current limit?

If not, then just beef up the shunt(s) (cover 1/3 of it with solder), put new thermal compound behind the MOSFETs, and go for it.
If it uses current sensors, route wire (2 gauges thinner than the existing wire) in parallel with the existing wire. I did the shunt mod to one of my controllers (40 amp -> 70 amp) and it has run cool and fine for over a year.

If it does, then it's time for a new controller.
 
thorlancaster328 said:
Does your current controller get hot when running at its existing current limit?

If not, then just beef up the shunt(s) (cover 1/3 of it with solder), put new thermal compound behind the MOSFETs, and go for it.
If it uses current sensors, route wire (2 gauges thinner than the existing wire) in parallel with the existing wire. I did the shunt mod to one of my controllers (40 amp -> 70 amp) and it has run cool and fine for over a year.

If it does, then it's time for a new controller.
Interesting, thanks - didn't think of that. Existing controller (ZL 48-72v) has been very reliable, has never overheated. I'll give it a go!
 
dont forget he could also replace his mosfets to depending on how much amps they rated for

and his caps

that would ensure his controller handles it fine
 
thorlancaster328 said:
If it uses current sensors, route wire (2 gauges thinner than the existing wire) in parallel with the existing wire. I did the shunt mod to one of my controllers (40 amp -> 70 amp) and it has run cool and fine for over a year.

If it does, then it's time for a new controller.

Could you explain the mechanism here? By "in parallel" do you mean just through the current sensor, right along side the original (current measured line)?

Could you wrap (loop) the single wire around one time and accomplish the same thing? Or would that half (signal of )the output current? How exactly does this work? Is it just like the metal shunt bars, fooling the controller into thinking it is not putting out as many amps like the solder on the bar mod, by decreasing the resistance of what travels through the current sensor (with the paralleled wire)...

I did not know you could fool Hall sensor shunts like that. Huh. Cool.
 
The additional wire does not go through the current sensor loop. When current flows, some of it flows through the original wire and some now flows through the new wire you just soldered on. Therefore the sensor reads less current than is actually flowing, and the controller increases phase current to compensate and produces MORE POWER.

I use this same method in my prototype controller and it works very well. It allows me to use cheaper and more full-featured current sensors to sense the ~500 amp peak phase current even though the sensors are only rated for 50.
 

Oh! Ok Thank you I got a little confused there. Cool. I will have to try this someday. Ty.
 
serious_sam said:
PRW said:
I ended up booking a Nucular 24F - wait is supposedly 5 months.
Nucular is well worth the wait.

FWIW, 12F would be fine for your setup. 84V (20S). 150Ab. 250Ap. MXUS3k.
Ah - good to know. I was told the battery will produce 350 phase amps, so believed I needed to base it on that!
 
PRW said:
Ah - good to know. I was told the battery will produce 350 phase amps, so believed I needed to base it on that!
I would not put 350Ap through a MXUS3k.

Depending on the wind, that could be more torque than the axle (or frame) could handle. The 4T is approx 1Nm/A. The 5T is higher. Maybe the 3T would be ok at 350A, but it's going to get hot so quickly that it would be impractical. Even with ferrofluid and hubsinks.
 
Back
Top