User Tools

Site Tools


Sidebar

astra_conv:conversion:wiring:wiring

This is an old revision of the document!


wiring - general

The descriptions in this chapter may provide too much detail for easy reading. Actually however, I want this documentation to be sufficient for understanding the circuitry. In case that one fine day I get abducted by aliens and someone else has to maintain my contraption, she or he will be thankful.

This chapter covers the additional wiring and circuitry that I installed in the course of the conversion process.

The existing original wiring and electronics was mostly maintained, only the engine cable harness and the engine controller were removed.


wiring diagram

To have an overview and to keep track of all the wire connections that I made, I started an artful hand drawing on a sheet of paper:

After having sketched the most important components such as e.g. radio and subwoofer and having drawn a few connections, I realized that the sketch might get overloaded eventually and it would be better to start a proper computer drawing. Note that despite of the little information in above sketch, it already contains a fatal error: With the negative pole of the traction battery connected to the chassis, the car would not comply with the European regulation ECE R-100. I recognized and corrected that error later (with quite some effort).

Below a “thumbnail” of the wiring diagram in its current state. It still resembles the original structure, but with some more detail. That is, it has grown out a bit too, over time. I will discuss the drawing in detail further down this page. To view in more detail, I recommend to right-click on the link below the picture and view it with a picture viewer that allows to zoom and scroll. Most below descriptions refer to this here drawing (I have provided additional detail extracts of the diagram to go with the explanations, but these are difficult to keep up to date).

link to wiring drawing (jpg), right-click and save as file (or directly open with a picture viewer)

The drawing also includes in detail smaller electronic circuits, such as drivers or little annoying “warning” gadgets. These circuits are also described in subchapters on this page. Two devices, that is the heater and the “current-to-RPM converter that drives the rpm gauge, are only shown as “boxes”. They have their dedicated subchapters with dedicated wiring diagrams within the “conversion” chapter.


general remarks

regulation

As already said, in Europe, wiring of electic vehicles must be in accordance to ECE-R 100. Regulation foresees that the traction circuit is electrically isolated from i.e. floating against the 12V grid of the car. This is why you will find a couple of optocouplers in the diagrams.

Cables or cable ducts that contain wires belonging to the traction circuit must have an orange coating, according to ECE R-100.

All conductive parts of the traction circuits must be protected against accidental touching. This will be tested using a “test finger” and, for the interior of the car, even a “test wire” of 100mm length and 1mm diameter. While making the installations test finger proof seems an understandable precaution, the “test wire” test (that any 230V plug socket in your appartment would fail) in my humble opinion rather causes unnecessary efforts.


wiring practices

I luckily found a retail shop that sells two-colored stranded single conductor wires, available in any length and in many different color combinations. For most connections that don't carry higher currents, I used these, with 0,5mm2 gauge (0,75mm2 gauge for ground connections).

The wires are usually bundled in ribbed tubes.

Wire joints were usually made using simple block terminals and wire-end sleeves. Such connections can be done and also disassembled quickly, don't need further isolation (when within a wiring box) and also offer good mechanical robustness. After opening such a connection, it must however be taken care that the free wire ends don't make any unwanted contact. The wire ends were labelled with sticker tape and felt-tip pen.

In contrast to block terminals that reliably provide ample contact pressure, crimped connections made by a clumsy amateur with an €7.90 crimping tool can probably not be trusted. It may have catastrophic consequences, if a wire end slips out of a cable lug and touches another conductive part. Besides, bad crimped connections may have high and varying electrical resistance creating overheating and circuit malfunction, and may fail early due to corrosion.

Soldering over a crimped connection is not admissible, since the solder will be soaked into the stranded wire, make it rigid and create a potential place of fracture where it ends. This hazard can however be overcome by protecting the wire end with heat shrink tube, and by avoiding bending and vibration of the wire by properly fixing it. So, with investing a bit more time, you get a connection that is strain proof, has outstanding electrical properties, is durable and 100% illegal.

For the electronic circuits, I used breadboards (single sided or with groundplane), mainly to save time and to be more flexible. The wire connections were made with circuit board type block terminals and again, wire end sleeves.

For developing the circuits and for first tests, my experimental board came in very handy - actually it is a christmas present I got thirty years ago and still appreciate a lot … Below photo e.g. shows the first “draft” of the coolant heater's control circuitry. The coolant heater will be discussed in a later chapter.

On the wiring diagram, you will notice that I very generously populated the circuits with protection diodes and smoothing capacitors. Their purpose is to cancel surges that are caused by inductances along a certain wire or its load itself, that sneak in via the supply grid or that are induced by strong and varying magnetic fields from neighboring wires.


overview of the wiring

topology

Fuses, wire joints and electronic circuits are mainly concentrated in four wiring distribution boxes:

The engine bay has one dedicated box for the 100V fuses and wiring, and one for the 12V stuff.

The lower glove department in the dashboard also had to give way to wires and terminals. With the limited heat available, one will probably leave the gloves on anyhow.

Finally, there is the trunk distribution box, including the 230VAC distribution besides some more fuses and circuits.

In my drawing, rectangles with grey outline (no fill, i.e. white) mark the location: “Trunk” to the left, “dashboard” in the middle and “engine bay” to the right. The distribution boxes mentioned above are each represented by light-grey rectangles within these locations.

color and labeling conventions

To label the wires in the drawing, you will find little tags near both ends of each wire. The tags tell the color (e.g. “bu-rd” stands for “blue coat with red line”) and usually also the wire gauge in [mm²].

The color of the tags itself indicate the cable (or wire bundle) that the respective wire belongs to. That is,

  • the cable that goes from the trunk distribution box to the engine bay has light-orange labels
  • the cable that goes from the dashboard to the engine bay has sea-green labels
  • the cable that goes from the trunk distribution box to the dashboard has kind-of-lilac labels
  • the short cable between the central dashboard console and the dashboard distribution box has white labels (except wires that directly cross-connect to other cables in the dashboard distribution box, these have inherited their cables' code color).

The wire bundles or cables themselves are indicated as bold blue+black (12V grid) or blue+orange (traction grid) lines that go across all lines that are bundled together. At the end of this line, you find a label that names the cable, its fill color matching the fill color of the wire labels.

In circuits where the “grounded” 12V grid and the “floating” traction grid meet, a dashed orange outline demarks the “isolation” border between these grids.

Meaning of the different line colors associated to wires (note that the color of a wire's coat however is indicated on the label text, not by the line color!):

  • The supply lines of the 12V grid are drawn red (and also have a red coat)
  • Ignition lines are drawn with blue line color (and also have a blue coat)
  • lines belonging to the traction circuit are drawn orange or black (different coat colors)
  • the GSM remote control output and connected lines are drawn green (different coat colors)
  • the “230V active” indication contact and connected lines are purple (different coat colors)
  • lines dedicated to the radio and amps are violet (different coat colors)
  • spare wires are grey (different coat colors)

some more colors were used for certain lines to make them easier to trace in the drawing, e.g. for the input and output lines of the current-to-RPM converter circuit.

why semiconductors instead of relays?

It is quite difficult to find relays that can switch (and disconnect!) larger currents at 100VDC. So you will find some semiconductor based drivers to do this task.

A further reason is that these are much more reliable than electromechanical components. With relays, you are likely to over time encounter issues with contacts that are corroded, dirty or “stick”.

The driver circuits are built up with discrete components, as I could not find matching “solid-state relays” that would fit my price expectations.

Only later, I found out that I should have searched for “motor drivers”. These inexpensive integrated components have no isolation between control input and load output (as solid-state relays have), but this is anyhow not required for my application.


logics

Another thing that I would have tackled differently á posteriori is the implementation of all the logics that connects the different state and alarm signals and control inputs. While at the beginning of the project, their number and also their interdependence seemed easily manageable, the complexity increased over time.

So, my approach of using e.g. some combiner diodes to form a logical “OR”, to feed signals to both ends of a relay coil for an “A AND NOT B” or to use old-fashioned timer circuits eventually ended up in quite a messy harness of wires and circuit boards.

I did use a few microcontrollers (heater control, control of the cooling system, DC/DC converter control), but not a central processor - which would for sure have made the setup much tidier and easier to manage.


alarms

Following table shows the numerous alarms that have to be handled somehow, and their connection to control lines or logics circuits. Each row covers one alarm. “Trigger condition” means the condition upon which the alarm is raised. In column “goes to” we see to which destination the alarm is connected. For example, the “HV idle voltage too low” alarm (first row) connects to the “HV alarm unit” on the “BMS peripherals” circuit board. This circuit will raise the “combined HV alarm”, with “HV idle voltage too low” as one possible trigger (see column B).

As another example, the “HV isolation check failed” alarm generated by the HV isolation check board actually provides one “ground contact” and one “active 12V” output. The ground contact output goes to the immobilizer latch (which avoids that the immobilizer is triggered during driving), while the “active 12V” output again is fed to the HV alarm unit of the BMS peripherals board.

*** back to top of this page ***

astra_conv/conversion/wiring/wiring.1394211981.txt · Last modified: 2014/03/07 17:06 by richard