Page 1 of 1
Tesla SDU HVIL
Posted: Tue Apr 06, 2021 6:44 pm
by EV_Builder
I was wondering. Is the HVIL working? Is it done by Software or external interface (like original?).
I have a V6 and i remember that it didn't seem to matter to solder it or not.
Re: Tesla SDU V7 wrong capacitor
Posted: Wed Apr 07, 2021 9:10 am
by johu
HVIL is connected straight to the NAND gate, so will HW-shutdown the PWM.
Re: Tesla SDU V7 wrong capacitor
Posted: Wed Apr 07, 2021 4:07 pm
by EV_Builder
EV_Builder wrote: ↑Tue Apr 06, 2021 6:44 pm
Is it done by Software or external interface (like original?).
Ok and who decides to shutdown then? And where?
Re: Tesla SDU V7 wrong capacitor
Posted: Wed Apr 07, 2021 4:52 pm
by johu
It is a dedicated pin of the timer that generates the PWM. As soon as the pin goes high, all 6 PWM outputs go low with no software interaction. Also an interrupt is generated The interrupt will send the software to stop mode. Depending on the "tripmode" setting the main contactor is then opened.
EDIT: wait a minute, I know what you mean. The CMOS NAND gate has so low input impedance that a floating pin that wasn't high before, will stay high for quite a while. This never occurred on my systematic tests since they use push-pull drivers to stimulate the inputs. But I did realize this last week because I had to test the boards in a more manual fashion because my Analog Discovery is broken.
That's ok for the gate drive fault signals, because they are actively pulled low on fault.
But HVIL is just interrupted and left floating. Pull-down needed!
Re: Tesla SDU V7 wrong capacitor
Posted: Wed Apr 21, 2021 8:00 pm
by EV_Builder
johu wrote: ↑Wed Apr 07, 2021 4:52 pm
It is a dedicated pin of the timer that generates the PWM. As soon as the pin goes high, all 6 PWM outputs go low with no software interaction. Also an interrupt is generated The interrupt will send the software to stop mode. Depending on the "tripmode" setting the main contactor is then opened.
EDIT: wait a minute, I know what you mean. The CMOS NAND gate has so low input impedance that a floating pin that wasn't high before, will stay high for quite a while. This never occurred on my systematic tests since they use push-pull drivers to stimulate the inputs. But I did realize this last week because I had to test the boards in a more manual fashion because my Analog Discovery is broken.
That's ok for the gate drive fault signals, because they are actively pulled low on fault.
But HVIL is just interrupted and left floating. Pull-down needed!
Let me know my needed mod since i wanted to use HVIL like it was designed for example connect it to the charger/hand brake system and the e-stop system of the car.
Am i right in the assumption that if one disables HVIL the motor will behave like in 'Neutral'? Since all GATES aren't active anymore?
Re: Tesla SDU V7 wrong capacitor
Posted: Wed Apr 21, 2021 8:28 pm
by johu
One pin of the HVIL connector is connected to 5V and the other is the detection input. Just pull down the detection input with 10k either externally or by modding the PCB.
Yes the motor will free wheel instantly when all IGBTs are off