IMO the best cooling designs are where the mosfets are cooled from the bottom of the PCB they are attached to, not the top of the epoxy package. The Unity and Stormcore (and the new Aluminum PCB 75100) got this right. 3 Shul also seems to be designing it with cooling from the bottom of the PCB. It’s one of the reasons the 4.12 was such a poor thermal performer even with a heatsink added. The sandwiched mosfets just fed heat into eachother with nowhere for it to dissipate since it transfers heat much faster into the copper PCB traces than through the epoxy package.
The DirectFET metal package used in the VESC 6 seems to try to improve cooling by allowing a more efficient transfer of heat through it’s metal case to a heatsink, but in my testing that ESC is only middle of the road in terms of thermal performance.
From what I’ve seen in tests, copper vs aluminum heat sinks don’t make a noticeable difference in performance.
I ordered the aluminum 75100 I told u. Fregin steal on sale. With it running firmware that allows I think 300 amps I’d like to get up there as high as possible. Is the only risk an over-temp shut-down? My battery will do 130amps but like to set motor amps to at least 140. What’s ur forecast and any advice on cooling it? I guess just leave it in the wind and not add small sinks to anything inside?
And will do 72v full charge and run strong regen brake with 83100 mid drive motor.
Hello, we are testing this ESC hobbyking turnigy-sk8-v2-80a-3-12s-single-motor-skateboard-esc
With maytech sensored-motors mto5065-170-ha-c?variant=29503884492894
We configure the 14pole and 6S. We power it from 24V power supply 8A.
The VESC tool recognizes the sensored motor and we can rotate it. The problem is that we can stop it with hand and the highest current we get is around 4-5A. In the program we set 10A in motor config and at the bottom of the VESC tool 20A, but the current does not look to increase on the power supply. Another thing is that if we control it from the bottom panel we enter 5000RPM it rotates at 700RPM. Has somebody faced such a problem? What can be the problem here?
This list of VESCs has always been grouped by shunt design, but now I’ve added that text under each section to make it clearer for those wondering about features like silent HFI that recommend phase shunts and phase filters.
Shunts are used to measure current flowing to the motor so the ESC can adjust the on/off time of the mosfets to deliver the correct amount of current requested by the remote.
Phase shunts are placed electrically between the high and low side mosfets, and the phase wires. These are typically found on VESC 6 and 75/300 based designs.
Low side shunts are placed between the low side of the mosfets the system ground. These are typically found on VESC 4, Unity, Stormcore, and 100/250 based designs.
Phase shunts provide a more accurate measuring of current flowing to the motor, but the design is limited by the available current sense amps on the market today. Generally these are used on designs under 70-80v.
Low side shunts can work on much higher voltage, but are less accurate in measuring current.
Phase filters can be added to any design. They allow you to measure the voltage while the VESC is driving the motor. Without filtering you would always measure either Vbus or 0V (since this is the output of the drive). The phase filters “average” this, which is actually the signal you want to measure. This helps more accurately measure motor resistance and the motor position especially at startup. To quote Vedder: “It is slow speed under load where phase filters make all the difference”
Vedder added a feature that will show the design in the HW spec after connection is established.
Shunt orientation and filters will be displayed. Next update will have it.
The issue is that some hardware designs don’t include hardware phase filters, but they are running stock firmware (75_300_R2 or 100_250 for example) which has them enabled. This causes the motor detection results to be incorrect and the motor won’t spin properly.
Usually if the Hw: name (75_100_V2 in the screenshot above) on the firmware tab is named after the controller then it’s a good bet that it’s configured correctly for that hardware.
Yes. This also confirms the ESC manufacturer likely didn’t skip steps and try to be cheap and fast, but were more likely to be concerned with creating the best product they could.