June 2016

Traction inverters installation.

Finally, time and materials are in place to install front and rear inverters. I ended up using two RMS PM100DZ inverters - 100kW each. As discussed before while initially I was planning to have 200kW inverters per axle the battery will not support this level of power output and I would not be able to take advantage of them. 200 kW drive system is not as spirited as 400 kW one, but even that is still much better than stock A6 Audi with its 2.8 L engine, and that is no slug either.

So the task was to find good accessible location for them as close to the respective motors as possible. This will minimize their inductance and radiated EMI. I managed to place each inverter right against terminal box of the motor it is connected to - in fact so close that no shielding will be required. Since motor current usually exceeds battery current and the battery cable is AWG 4, I decided to use AWG 2 size cables for 2 phase connections, although this is likely overkill. However this relatively thin cable is easy to work with, bend and crimp, so the choice made sense. It is unlikely motor current will ever exceed 200 A rms and at that level even continuous current would leave AWG2 cable cold. The liquid cooling lines from both inverters will join together and coolant run through a common radiator. Depending on the space up front the radiator may be large enough to cool off both motors as well, I'll have to come up with this decision later depending how much room I have.

An HV distribution box was constructed and bolted to the top of the drive shaft tunnel. This allows to route HV wiring toward DC/DC converter. A/C and anything else potentially running directly off of 750 VDC traction voltage. Auxiliary fuses are mounted next to power splitting blocks made from solid aluminum and color marked for right polarity.

Here are few photos representing major steps of the installation, in no particular order.

Front inverter placed on the stack of wood blocks. Once in desired position, metal brackets were fabricated to hold it in place.
An AWG 2 cable is too thin for the feed through glands provided with inverters - they don't quite grip it. Two shrink tubes were installed on the cable. Because it has to be water tight, the tube has a layer of meltable glue visible on this photo.
A thin ferrule is placed on the end of cable before it is inserted into the inverter's clamp. Noalox is applied to the bare strands to prevent cable oxidation. Because the ferrule is thin, there is no need to crimp it - it will conform to the shape of the clamp.
Tightening the cable to each phase output. 20 Nm of torque applied.
Front inverter is in place, battery cables are attached.
The motor side cables are ready to be cut to actual length - turns out they are only 15 to 20 cm. long, so no need for shielded version.
I've tested couple of crimping dies to pick right size - you obviously don't want loose crimp, but you don't want to over-squeeze the lug's barrel so that individual the copper strands break inside of it. We'll skip whole science of crimping technique here, important point to be made is that the compression has to be just high enough to form what's called cold weld where the copper at the end becomes one solid rod rather than individual strands, but not more. On this photo I'm cutting test-crimped section in the middle to inspect the crossection whether this is the case.
It is hard to see on this photo, but the copper is solid. The crimper pressure and the die size are about ideal for these lugs and cable.
Cable lugs re ready to be crimped on.
This hydraulic crimping hand tool is very handy, but the problem is it leaves two flat extrusions of copper on both sides of the barrel I have to grind off later. The handy crimper I've made out of a bolt cutter does not have this issue and leaves very smooth barrel. The only shortcoming is that it does not have insertable dies and its jaws shaped for one particular lug size. I may consider making a few inserts to accommodate different lugs, however the cutter was large, so it is good if crimping is done outside the car and not in tight spaces.
The lugs are installed onto studs inside motor junction box and tightened down with M6 nuts The feed-through glands on this photo are not tightened yet.
Completed connection, each cable is protected by corrugated split loom, but again, the cables are so short that no other protection is needed.
Power to each inverter is routed using pair of twisted AWG 4 cables. You know why twisting is needed here, don't you?


The final connection to each motor is low voltage interface - shaft encoder and winding temp sensors. Since PM-100Z inverters have different connectors for encoder and everything else, inverter's end of cable is also split into two groups - encoder wiring and sensor wiring (23 pin Ampseal and 35 pin Ampseal respectively)

This really concludes inverters installation and connections. Granted main interface cable to 35 pin plug still has to be made, bit it is only a matter of plugging it into the inverter and I can leave inverters alone - this part of the project is completed.
Now, to route high voltage power to the front inverter and split off the power for the DC/DC converter and an air conditioning inverter, a small distribution box with color coded aluminum blocks and few HV fuses has been constructed.


Next - vacuum pump

Home of the project