you can unplug everything in the engine compartment (labeled of course) and pull everything into the interior still hooked up to the fuse box and remove the entire harness. the hole in the firewall is like 30 or 40 mm
72 3.0 CS 2240344 Restoration
- Thread starter ScottAndrews
- Start date
-
- Tags
- restoration list
OOO!!. good to know!! That sounds easier than going the other way...The under side of the dash is truly a rats nest of wires at this point. Too many hands over too many years dinking around in there.
I am planning on remaking the entire harness while the body is at the body shop. Too many old crusty wires and terminals.. So I am documenting the locations of the branches and where they go. I plan to make a plywood skeleton to hold the old harness in the approximate spatial position, and then rebuild it from there with all new wire and connectors.
Is there a source for the various firewall grommets?
Last edited:
Regarding the wiring, I think you have the right idea. The blue book wiring diagrams are accurate but hard to read. I have used the Prospero diagrams available on eBay (search e9 wiring). I have heard others complain they are inaccurate, and the first one I ordered for a euro CSi had a misprinted reverse side, but the seller corrected the issue quickly and they did make further wiring much easier since they are in color. Alvaro also has a set of color wiring diagrams that I am sure are a lot better, they are a bit more expensive. Most of my disassembly comes with numerous photos to allow a rebuild, this component is at least something that is fairly easily rebuildabke with a diagram. You have the right idea recording WHERE each bundle runs - the connections are by the diagram
Sorry for double reply. Message Don about grommets, he will tell you where to source. w&N has a pedal rebuild kit which I got which looks nice and has all of the pedal parts id that’s something you need to refresh.
There is a list in the FAQ section on Tools. Just next to the FAQ section
I do have an (almost?) complete fully original (for the originality afficianados!) toolset; pm me if you're interested. Currently skiing with Family, so I can send pics early next week earliest.
Oh Darn! Does that mean we'll have to travel to the Netherlands to get them and avoid the tariff? Maybe drop in and visit my old friends in Portugal and finally get to Madrid and Bilbao...There is a list in the FAQ section on Tools. Just next to the FAQ section
I do have an (almost?) complete fully original (for the originality afficianados!) toolset; pm me if you're interested. Currently skiing with Family, so I can send pics early next week earliest.
Seriously, I may be interested. Let's talk when you get home. Stay safe!!
I am very close to having an empty shell to take to body and paint.
In this process I have been documenting labeling and removing the wiring. As a result, I have identified a variety of unexpected and missing wires, and also developed some thoughts on what I might change in a revised new harness.
Here is a rough sketch of the "geography" of the harness. The car is basically separated into five geographic areas: Trunk, Passenger Compartment, Cowl, Engine Bay and Grille. The separators are: Trunk Bulkhead, Firewall, Cowl, and Radiator Support. There are some sub areas, like the shift console, and doors as well.
Here is a general map of the car wiring. Note, this is not a wiring diagram, it is a crude map of the wire bundles and branches. I plan to refine this substantially going forward, but it's a start.
The blocks like "fuse box, relays, instrument panel" are complex, and will need much greater detail, but I wanted to get the basics down first.
Some observations and questions:
In this process I have been documenting labeling and removing the wiring. As a result, I have identified a variety of unexpected and missing wires, and also developed some thoughts on what I might change in a revised new harness.
Here is a rough sketch of the "geography" of the harness. The car is basically separated into five geographic areas: Trunk, Passenger Compartment, Cowl, Engine Bay and Grille. The separators are: Trunk Bulkhead, Firewall, Cowl, and Radiator Support. There are some sub areas, like the shift console, and doors as well.
Here is a general map of the car wiring. Note, this is not a wiring diagram, it is a crude map of the wire bundles and branches. I plan to refine this substantially going forward, but it's a start.
The blocks like "fuse box, relays, instrument panel" are complex, and will need much greater detail, but I wanted to get the basics down first.
Some observations and questions:
- I was able to find the trunk switch, but I do not see a trunk light, or a location where one would go. Where is this located?
- I found four wires on a branch at the top of the left rear wheel arch. These go to the center of rear seat bulkhead on the trunk side. I suppose two of these are supposed to go to the trunk light. Any idea what the other two are for? They have no terminations, but appear to be original in the bundle.
- The splice for rear side markers seems wonky. These appear to come off the harness in the middle of the rear branch pretty much where the license plate light wire branches off the main bundle. The left one then doubles back to the side marker, and the right one parallels the tail light branch to the right side marker. Seems odd, since this approach adds two extra independent branches instead of being bundled with the tail lamp wires. Is this a retrofit for US cars because they added the side markers?
- The AC compressor wires do not come in through the main harness bundle into the engine bay, but instead pass through the firewall with the AC hoses and follow the right inner fender up to the compressor and then to the AC fan. Again, seems added on.
- I am likely to delete the seat belt wires that light up the dash indicator, since that indicator has been terminated, with prejudice! Interestingly, I do not recall seeing any seat belt contact wiring.
- I may also re-wire the harness for the relay based window control. This will require reducing the size of the switch wires and adding a power wire (probably a dedicated ground too). I am still kicking around the idea of just adapting a modern LIN based single wire control for the windows, since that would vastly simplify the wiring in the passenger compartment (which, outside the instrument panel, is mostly comprised of window wiring).
Last edited:
1. Trunk light mounts in center of trunk bulkhead on the thin cardboard panel
2. The four wires are for trunk light and speakers
3. Since side markers are on with license plate light they are grouped together and installed at the factory
4. A/C components came in a box in the trunk and were installed by the dealer. The wire from the temp gauge follows the a/c hoses and goes to the relay and also branches off to the compressor. Wires for the relay to power the radiator fan come from battery and radiator support.
5. Wires for seat belt were under the seat next to the tunnel and can be removed along with the warning module on the dash top.
You are certainly in deep!
2. The four wires are for trunk light and speakers
3. Since side markers are on with license plate light they are grouped together and installed at the factory
4. A/C components came in a box in the trunk and were installed by the dealer. The wire from the temp gauge follows the a/c hoses and goes to the relay and also branches off to the compressor. Wires for the relay to power the radiator fan come from battery and radiator support.
5. Wires for seat belt were under the seat next to the tunnel and can be removed along with the warning module on the dash top.
You are certainly in deep!
when you have your harness out you can make some changes - like headlights (low beam relay) and windows (relays + direct power).
Ahh. Makes sense. There is no cardboard panel in my trunk, so that's why I have those loose wires. The unterminated wires would make sense for a speaker.
Seems like the license plate light and side markers would be powered on when the tail lamps were on, so that wiring seems wonky. I noticed it when I had to deal with he extra branch when pulling the harness through the small opening between the trunk and the passenger compartment at the C-Pillar. I may revise that wiring to simplify things
Were the side markers a US only thing? If so, then that wiring makes sense, one added wire at the license plate light splice, and then separate branches right and left tot he markers.
I'll check my seat belts, but I do not recall undoing any wires when I removed them.
Yeah, in DEEP!! I'll be reassembling from an empty shell.
my harness was sucked into the engine compartment - and it wasn't labeled (done by the paint shop). i figured out what was what - not that hard. figuring out where everything goes on the inside of the firewall ... that's a little more challenging.
Pondering the wiring harness and the huge bundle of large wires driving the windows...
One way to drastically simplify this while also adding some nice features is to go electronic. Below is a diagram of what I am thinking. The PIC devices are single chip super basic microcontrollers (no operating system, hard coded), usually 18-20 pin packages. They cost about $2 each. The LIN bus is a single wire networking protocol. In this setup there is one PIC that is connected to the window switches. All it does is sense when any of the switches has closed to ground. The window switches are single pole double throw, so the common pin goes to ground (changed from B+ in the existing setup), and the other two pins go to to the PIC inputs.
This approach has a single power wire to each window module, and a single LIN wire from the con=sole to each window module. So, a net reduction from sixteen wires to eight (two at each window), and a huge reduction in the wiring to and from the console (sixteen to four, and smaller gauge wire as well)
In operation the PIC sits there spinning in a loop, doing nothing. When any of the window switch contacts closes to ground the PIC throws an interrupt that branches off to handle the event.
The PIC first determines which pin changed, and then issues the appropriate LIN command. There are four small gauge wires that lead from the console PIC to each window module. Each module has a LIN address, so the LIN command contains the address for the window module corresponding to the switch that was moved.
Each widow module also has a PIC controller, and each one receives the LIN command. Each PIC examines the address in the command, and if it matches their address it energizes the appropriate signal line to turn on the window motor driver (a MOSFET bridge circuit). If the address doesn't match the PIC does nothing.
There are two drivers, one for the UP motor winding, and one for the DOWN motor winding. Depending on the command (UP vs DOWN) the PIC activates the appropriate driver. There is a small resistor in the motor driver line that senses the current drawn by the motor. When the motor reaches its limit, or if it encounters a barrier, like an arm or a finger, the current rises, and the PIC stops the motor.
There are two different ways to operate this concept.
One is that you can just react to the change in the switch state window switch. That change starts the motor which continues until one of two criteria are true: 1) The switch is pressed the other way OR 2) The current sense reaches a threshold (either because the window is all the way up or all the way down, or there is something in the way of the window). So this is a one-touch-up, one-touch-down setup.
The other is that the motor runs as long as two criteria are true: 1) Switch is pressed, AND 2) Current sense is below a threshold value. This is the more traditional approach where the window moves only while the switch is pressed, but it if gets to the top or the bottom it stops, even if the switch is still pressed.
Not fully sure I will take this sacrilegious leap, but it is interesting to consider!
One way to drastically simplify this while also adding some nice features is to go electronic. Below is a diagram of what I am thinking. The PIC devices are single chip super basic microcontrollers (no operating system, hard coded), usually 18-20 pin packages. They cost about $2 each. The LIN bus is a single wire networking protocol. In this setup there is one PIC that is connected to the window switches. All it does is sense when any of the switches has closed to ground. The window switches are single pole double throw, so the common pin goes to ground (changed from B+ in the existing setup), and the other two pins go to to the PIC inputs.
This approach has a single power wire to each window module, and a single LIN wire from the con=sole to each window module. So, a net reduction from sixteen wires to eight (two at each window), and a huge reduction in the wiring to and from the console (sixteen to four, and smaller gauge wire as well)
In operation the PIC sits there spinning in a loop, doing nothing. When any of the window switch contacts closes to ground the PIC throws an interrupt that branches off to handle the event.
The PIC first determines which pin changed, and then issues the appropriate LIN command. There are four small gauge wires that lead from the console PIC to each window module. Each module has a LIN address, so the LIN command contains the address for the window module corresponding to the switch that was moved.
Each widow module also has a PIC controller, and each one receives the LIN command. Each PIC examines the address in the command, and if it matches their address it energizes the appropriate signal line to turn on the window motor driver (a MOSFET bridge circuit). If the address doesn't match the PIC does nothing.
There are two drivers, one for the UP motor winding, and one for the DOWN motor winding. Depending on the command (UP vs DOWN) the PIC activates the appropriate driver. There is a small resistor in the motor driver line that senses the current drawn by the motor. When the motor reaches its limit, or if it encounters a barrier, like an arm or a finger, the current rises, and the PIC stops the motor.
There are two different ways to operate this concept.
One is that you can just react to the change in the switch state window switch. That change starts the motor which continues until one of two criteria are true: 1) The switch is pressed the other way OR 2) The current sense reaches a threshold (either because the window is all the way up or all the way down, or there is something in the way of the window). So this is a one-touch-up, one-touch-down setup.
The other is that the motor runs as long as two criteria are true: 1) Switch is pressed, AND 2) Current sense is below a threshold value. This is the more traditional approach where the window moves only while the switch is pressed, but it if gets to the top or the bottom it stops, even if the switch is still pressed.
Not fully sure I will take this sacrilegious leap, but it is interesting to consider!
I think that your window concept is intriguing, as is the possibility to reduce the wiring load. My only concern is any future revision. As you’ve discovered (and have many of us), previous owners have often intervened in ways we can’t determine. As I’ve added wiring, I have tried to add a label indicating what the wire is so the next owner knows what’s what. The relay adds are fairly well documented here and seem to be standard issue - once you understand how a relay works it’s immediately obvious what’s going on. Your proposed addition, however, would be somewhat of a one-off and may therefore be harder for anyone later to decipher. In addition the relay adds involves adding wiring (power to relay from battery, e.g.) but is relatively reversible. If your option involves removing a bunch of wiring, that’s nice for current cleanup but may also make future revision more complicated. Just my 0.02. As one example, I added fog lights to my car and since I don’t need a cigarette lighter, I’m wiring a second light pull switch in place of the cig lighter. That process involves running an additional wire to the switch - I’ve labelled both ends, and I will likely actually wrap the cigarette lighter in some light foam wrap and tape it to the underside of console so that it’s right there if anyone wants to reverse what I did. I spend my days dealing with revisions in my surgical work, and I try to assume that anything I do will have to be revisited by me or someone else later, and plan accordingly.
Yeah, this is my primary reservation. The change I am considering would involve not only wiring changes, but also adds dependencies on electronic hardware that may be obsolete in the future, and on software that once it is copied and embedded is pretty much impossible to understand.
I may play with this while the car is in paint, but if I were to implement it, I'd probably do it more in the relay style for all the reasons you suggest.
That said, I DO think I'll make some other modifications to simplify the re-installation (specifically either re-doing the rear side market wiring, or possibly just eliminating the side markers altogether...
interesting, yes ... do i understand it, nope. above my pay grade.
what i am doing is running a new red wire from the battery (in the harness) to a 4 fuse block (to be mounted in the fusebox area) and use 2 cables to feed 4 front / 4 back relays (mounted near the motors on each side - front + back). most of the wiring will remain the same in the car - the up (green) wire will trigger the up relay and the down (black) wire will trigger the 85 pole on the down relay. so the green + black wires that go to the terminal connection will go to the relay instead and a new wire will go from the 86 pole to the terminal connection (to power the motor) ... on the rear, since there are 2 feeds to the terminal connection, i will remove the green + black wires from the terminal and attach them to the 86 pole on the relays and add a new wire from the terminal connection to the 85 poles of each relay. will add new ground wires to each relay.
so there is mainly new red wires going to the 30 pole of each relay and short green / black / brown wires.
what i am doing is running a new red wire from the battery (in the harness) to a 4 fuse block (to be mounted in the fusebox area) and use 2 cables to feed 4 front / 4 back relays (mounted near the motors on each side - front + back). most of the wiring will remain the same in the car - the up (green) wire will trigger the up relay and the down (black) wire will trigger the 85 pole on the down relay. so the green + black wires that go to the terminal connection will go to the relay instead and a new wire will go from the 86 pole to the terminal connection (to power the motor) ... on the rear, since there are 2 feeds to the terminal connection, i will remove the green + black wires from the terminal and attach them to the 86 pole on the relays and add a new wire from the terminal connection to the 85 poles of each relay. will add new ground wires to each relay.
so there is mainly new red wires going to the 30 pole of each relay and short green / black / brown wires.
So after considering this a bit, and having the same desire to maintain some semblance of original wiring, I think I may just go the relay route, but with a twist. The relay approach basically just re-purposes the existing wires, and that's not necessarily a bad thing, since they are already in the harness, and the real issue is running power through the window switches and out to the windows. Not really overly concerned about eliminating wires to, for example, save weight.
What I may do is to implement the automatic up/down function using a PIC in the console. That would then run the relay control wires as if they were connected to the switches, but now with the features I described above. I'll have to look art the wiring and see if it is possible to implement the current sensing as part of the relay circuit, because I would not want to do an auto-up feature without a safety.
in my scenario - power that runs thru the switches only powers the relay to energize - main power to the motor is run thru new red wires.
Side markers were on all US, North/Central/South America deliveries for sure. They are on with running lights/license lights, harness makes sense to me. That is why all coupes have a four digit model code that allowed the assembly line workers to know if was for Europe/ROW, LHD/RHD, manual/auto, 2800/3.0, pre 74/74 and on, etc.
I did the wiring like @rsporsche with the new red wire going from battery to motor with relay trigger. My primary variation is that rather than having the additional fuse block in the fuse box area, which already seems tight to me, I added it and a distribution block in the battery tray and moved battery to trunk. One large cable goes from trunk battery up to the battery tray area and then splits to feed original fuse box and all of the old wiring as well as new fuse box and distribution block. Not an original look but I figured that way any additional wiring is easy access in the engine bay without needing to crawl under the dash