Complete Wiring Harness Overhaul

[Foswick]

My boat is 50 years old. It has had work on the wiring over the years as components and circuits were added to the original harness. In the late 1990s, a previous owner hand-drew some really nice color-coded schematics that I believe accurately reflected the system at that time. I don't see much that is currently on the boat that is not accounted for in those diagrams. Still, that doesn't mean that (a) the diagrams were 100% correct when created and (b) they are still correct. But, I think they will be quite useful for understanding what I am looking at when I dive into the physical wiring.

I don't have any specific electrical problems yet, but I have four concerns that have prompted me to strongly consider renovating the entire wiring harness:

• The wiring is old. Although some spot inspections are encouraging, I have concerns about the overall age of the wiring and connectors. I'm not keen to replace something just because it is old, but almost every exposed metal surface I have inspected has some corrosion.
• I think some circuits have fuses/breakers that are undersized and/or the wiring is undersized and the circuit protection is appropriate. In other words, I have the occasional fuse blow and probably not from a short, but from an overload.
• I don't really understand exactly how things are currently wired. A renovation will help me understand and trust.
• Very little about the current harness is well organized, easily inspected, or accessible for neat expansion. And, as mentioned, I don't know about documentation - looks good, but who knows.

I am pretty good on electrical knowledge and can look up what I don't know. What I don't know and find a bit overwhelming is how to approach this project.

My personality would have me do this:

1. Document what I have, to the best of my ability.
2. Generate a brand new schematic and wiring diagram that is ideal for my current and planned needs.
3. Remove everything.
4. Replace everything, according to the new diagram.

In other words, slash-and-burn.

My Jimminy Crickett suggests this might not be the most prudent strategy unless I want to try to sell a boat with no wiring but a half-finished rewire project that has been stalled for 10 years.

So, maybe a better approach would be to do #1 and #2, but then slash-and-burn just one circuit at a time. Something like:

1. Document what I have
2. Document what I want
3. Replace the distribution
4. Work outward from distribution, one circuit at a time, labelling relentlessly

I like to think of projects like this as being less for me and more for the next owner. That helps keep me conscientious about making good decisions and documenting well. I want the next person to not have to go through this and to have complete trust in the documentation. Also, there is a good chance the "next person" will be me 10 years from now, having forgotten what I was thinking 10 years earlier.

Any advice on how to approach this project, overall? Not looking for specific advice on wire, connectors, etc., but just trying to develop a good strategy for the undertaking.

 

[Dave]

Start with the main DC+ Bus and work out. Problems closer to the bus will have the most wide ranging effect and the most serious consequences. Think, batteries, inverter/chargers, alternators, and solar. Next work to the distribution panel(s). Last work on the individual circuits, prioritizing the most important circuits, VHF, AIS, chart plotter, instruments, bilge pumps. Work down the priority list, refrigeration, water pumps, propane system, etc. Last would be the lighting, it will also be the hardest because the wires will be buried the deepest and hardest to trace.

To help with prioritizing, this is a list of the emergency circuits required on an USCG Inspected Vessel:

VHF1 Standby

VHF2 Standby

AIS

Instruments

Vessel Monitoring

Chart Plotter

Radar Standby

Navigation Lights

Masthead/Steaming

Anchor Ligh

 

[JoeWhite]

I have planned and wired telephone wiring for new construction hotels. You must allow for future changes by installing sub panels or screw terminals in strategic locations and run thick wires back to the main panel. This will provide flexibility in a future expansion and addition of equipment. Since these are conduits for running wires, it limits the number of wires that you can run. Plan on running a few large wires in there rather than many small wires. Keep an accurate wiring diagram. Make them into PDF files and store them in your server and a printed copy on the boat.

 

[Foswick]

I think you see it the same way I do, @JoeWhite. I have seen a lot of people almost religiously eschew connectors and busses, preferring to run more uninterrupted wires. While I get the reasoning, it makes maintenance and troubleshooting a real "joy". And, I don't think that practice really has the benefits people think it does. That is, it doesn't really reduce points of failure, nor does it really reduce resitance under normal circumstances.

One plan I do have is that I will wire in tiny Bluetooth beacons at key locations. So, if something isn't working, I can look on my phone and know exactly where to start looking. It also lets me do a very easy all-systems diagnostic test. The modular approach you recommend is perfect for this.

 

[dLj]

1. Document what I have, to the best of my ability.
2. Generate a brand new schematic and wiring diagram that is ideal for my current and planned needs.
3. Remove everything.
4. Replace everything, according to the new diagram.

My perspective is step 1 is a total waste of time.

Piece meal is the best way to take a really long time getting it done - if you ever actually get it done. My approach would be:

1) Decide what I want and where.
2) Create the schematic for how it needs to be done.
3) Remove everything. (I'd try to pull wires and leave a line to pull new wires back through in places where that seemed best.)
4) Create the new system.

It's amazing how many wires you'll likely find that go nowhere and are simply left from some some previous wiring work...

dj

 

[Project_Mayhem]

One plan I do have is that I will wire in tiny Bluetooth beacons at key locations

Please do tell more!

 

[BAD ORCA]

Slash and Burn. I statrted to try and document where everything was and tried to trace every wire. In the end it was just easier to rip it all out, every single wire except the engine wires/harness and redo it all. The documenting wasnt all that helpful because I knew what was on board already and had a decent idea of future upgrades. That being said, my wiring was a horrible mess, not a reasonable done install that was maintained over the years.

I started with new distribution panels to handle what i had, what i wanted and a little more room for expansion. Bought new batteries that would feed the system. Bought my chargers and inverter. Rebuilt the battery boxes and got batteries installed. Installed the new distribution panel (AC & DC) then installed by busbars, shunts etc...etc. Once the main electrical backbone was installed I ran all my branch circuits. Like dlj mentions, there were so many random wires that went nowhere, so much junk and sketchy add ons, not to mention multiple ground loops and a few other surprises like a main ground wire to engine block that was full of water. It was worth the effort, expensive as hell but worth it.

 

[dlochner]

The remains of the second pass at cleaning up the electrical system and removing the old NMEA 0183 system.

Save the wire, copper wiring of any size has good value at the metal recyclers.

 

[Timm R Oday25]

I did this perhaps a year after buying our boat . Quite a few wires were run over the cabin and under the deck .
This made removing some wires impossible .
I wanted to know where each wire went and what it did .
Now 20 years later ,I am ready to do it again .
I plan to draw out a schematic and follow it.
This time I will also use pvc conduit to make pulling wires easier and cleaner .

 

[Foswick]

Please do tell more!

Nothing much to it, really.

The long story is that I use a service dog whose job it is to get assistance, when needed. I designed, built, and carry this device:

I wear it on my belt. When required, he pulls the tab which activates the device, which uses Bluetooth to connect to my phone. I developed a simple app that runs on my phone and, when the device connects, it initiates an emergency response sequence using my GPS location and some other data.

This works great, but it has always bugged me that the device must actually connect to my phone. That connection takes time and, as most people have experienced, Bluetooth connections can be a little fussy and unreliable. Also, the way I designed this, the device is always on and not rechargeable. So, I need to replace the battery every month.

So, I recently decided to improve the device. My new one (nearly done) has wireless recharging but is also ordinarily powered off. It only turns on when the tab is pulled, using a NC reed switch and magnet. And, it doesn't need to connect to my phone. It just broadcasts its ID. That's it's only job - turn on and broadcast a unique ID. My phone just detects that a device with that ID just became visible - even just for a split second. This is much more reliable and crazy fast.

For this device, I designed a custom PCB to fit everything into a compact form, but that isn't necessary for a harness monitor for the boat.

Then, the code that does the main job is just about a dozen or so lines. To wire this into a wiring harness, I would use an ESP32-C3 (about $4-$5), a 12v-3.7V mini buck converter (about $2) and 3D print a case in PETG (pennies). So, at maybe $7-$8 per unit, I can put one of these pretty much anywhere I want in any circuit.

Since each one has a unique ID, I can immediately tell which are powered and which are not by looking on my phone. I'll probably write a simple app that shows active vs inactive beacons in a list - maybe with a diagram to help me visualize, if I get really motivated.

If there's any interest, I'll post instructions for making these, but it is really super simple - just solder the buck converter to the BAT +/- pads on the XIAO and program the microcontroller with Arduino IDE. The code is really just:

#include <bluefruit.h>

void setup() {
Bluefruit.begin();
Bluefruit.setName("Your device name here.");
Bluefruit.setTxPower(4); // Max power for range

// Configure advertising packet
Bluefruit.Advertising.addFlags(BLE_GAP_ADV_FLAGS_LE_ONLY_GENERAL_DISC_MODE);
Bluefruit.Advertising.addTxPower();
Bluefruit.Advertising.addName();

// Broadcast forever
Bluefruit.Advertising.setInterval(32, 244);
Bluefruit.Advertising.setFastTimeout(30);
Bluefruit.Advertising.start(0);
}

 

[Project_Mayhem]

Nothing much to it, really.

The long story is that I use a service dog whose job it is to get assistance, when needed. I designed, built, and carry this device:

View attachment 236951

I wear it on my belt. When required, he pulls the tab which activates the device, which uses Bluetooth to connect to my phone. I developed a simple app that runs on my phone and, when the device connects, it initiates an emergency response sequence using my GPS location and some other data.

This works great, but it has always bugged me that the device must actually connect to my phone. That connection takes time and, as most people have experienced, Bluetooth connections can be a little fussy and unreliable. Also, the way I designed this, the device is always on and not rechargeable. So, I need to replace the battery every month.

So, I recently decided to improve the device. My new one (nearly done) has wireless recharging but is also ordinarily powered off. It only turns on when the tab is pulled, using a NC reed switch and magnet. And, it doesn't need to connect to my phone. It just broadcasts its ID. That's it's only job - turn on and broadcast a unique ID. My phone just detects that a device with that ID just became visible - even just for a split second. This is much more reliable and crazy fast.

For this device, I designed a custom PCB to fit everything into a compact form, but that isn't necessary for a harness monitor for the boat.

View attachment 236952

View attachment 236953

Then, the code that does the main job is just about a dozen or so lines. To wire this into a wiring harness, I would use an ESP32-C3 (about $4-$5), a 12v-3.7V mini buck converter (about $2) and 3D print a case in PETG (pennies). So, at maybe $7-$8 per unit, I can put one of these pretty much anywhere I want in any circuit.

Since each one has a unique ID, I can immediately tell which are powered and which are not by looking on my phone. I'll probably write a simple app that shows active vs inactive beacons in a list - maybe with a diagram to help me visualize, if I get really motivated.

If there's any interest, I'll post instructions for making these, but it is really super simple - just solder the buck converter to the BAT +/- pads on the XIAO and program the microcontroller with Arduino IDE. The code is really just:

#include <bluefruit.h>

void setup() {
Bluefruit.begin();
Bluefruit.setName("Your device name here.");
Bluefruit.setTxPower(4); // Max power for range

// Configure advertising packet
Bluefruit.Advertising.addFlags(BLE_GAP_ADV_FLAGS_LE_ONLY_GENERAL_DISC_MODE);
Bluefruit.Advertising.addTxPower();
Bluefruit.Advertising.addName();

// Broadcast forever
Bluefruit.Advertising.setInterval(32, 244);
Bluefruit.Advertising.setFastTimeout(30);
Bluefruit.Advertising.start(0);
}

Impressive but I love the simplicity in the execution!

 

[Indysailor]

I've done a piecemeal approach over the six years I've owned my boat. Some stuff I didn't see a need to document because I was either removing extraneous/unnecessary wiring or just improving poor wiring by a PO. When I rewired the engine and replaced the gauges, I did take the step of documenting what was there, then drawing out a new diagram with improved fusing. To me, passing along that diagram is an important part of selling the boat and like Foswick, it wasn't 100% necessary, but was 45 years old and showed evidence of someone else's tinkering that I didn't trust. Anyway, I think a sort of hybrid approach makes sense.

One question for others (sorry for the hijack): I could use advice on how people are labeling wires. I'd like to avoid a numbering system in favor of naming them, e.g., "depth", "bilge", etc.

 

[dlochner]

One labeling method used by professionals contains the source and panel position of the circuit breaker. A label would look like: LP1_1, translated it is the first circuit breaker on Load Panel 1. LP3_6 is the sixth circuit on Load Panel 3. The label goes on each end of the wire and at any junction. This is probably not the only way, however, it is simple and easy to trace.

 

[dLj]

One labeling method used by professionals contains the source and panel position of the circuit breaker. A label would look like: LP1_1, translated it is the first circuit breaker on Load Panel 1. LP3_6 is the sixth circuit on Load Panel 3. The label goes on each end of the wire and at any junction. This is probably not the only way, however, it is simple and easy to trace.

So to continue the hijack - any suggestions of how to make the labels?

dj

 

[dmax]

I use a labelmaker with waterproof labels.

 

[dlochner]

So to continue the hijack - any suggestions of how to make the labels?

dj

I use a Brother P-Tech label maker (Amazon Link) and then cover them with clear heat shrink. The P-Tech labels are not as sensitive to heat as the cheaper labels which turn can black when heat shrink is applied. It is clear on my work, at what stage I changed label makers.

 

[Project_Mayhem]

I have a brother label maker that does wire tags. It was around $100. Sometimes the labels will come off. Adding clear tape helps a lot. THere's models out there that print on heatshrink but for a one off project, it's more economical to have them custom printed. I've started 3D printing labels. Typically white text on a black background. If you have the space, it's a great method

 

[Foswick]

I have a label maker that will do heat shrink labels, which I use all the time. One thing I'm resolved to do with this project is to label each end of every wire, plus anywhere in the middle of a run that I might be tracing the wire and wonder which wire I am looking at (like at routing bottlenecks). My plan is: name, amps, AWG. So, like: bilge (5A) - AWG 18 - 20A

This makes the system more or less self-documenting. I can look at a bus, for example, and see the load the wire supplying it supports and the load the the wires leaving it add up to. Tells me if I can safely add a wire or upgrade a device on that bus.

My original plan was to only use the load supported by the wire and not include the load of the device except on the device itself or on the wire right next to the device. But, I quickly realized that I might have a bus that is capable of 30A with two wires - each capable of 20A, but with only a few amps actual load.

So, having both numbers lets me see everything I need to know:

• What does this wire go to?
• What is its size?
• For the length of run, how many amps could it support?
• What is the actual load?

I've waffled back and forth on how/where to document the actual load. That is the number that might change on a given wire. So, I might not want to include it on the heat-shrink label. I was going to just label it at the device. So, if I wanted to know the actual load on a wire, I would have to go to each device it feeds. This has the benefit of making it easier to keep labelling accurate but the disadvantage of removing my ability to see actual load at a glance.

A hybrid strategy would be to just label the wire with its name, gauge, and capacity, then keep the actual load values on a laminated 3x5 card nearby, or something.

I'm undecided on the details but do favor this self-documenting approach.

 

[Stu Jackson]

My approach was like "How do you eat an Elephant?" One bite at a time. When I first got my boat, I had a very good manual which I reviewed and compared to what I saw in front of me. If I found anything different I documented it. i also had a tremendous c34 website, which included, among other things, an important article about the infamous "wiring harness" between the engine and the cockpit panel, and also included the color coding. I used this when I upgraded my alternator and installed a Balmar MC612 external regulator. For those areas that were undocumented by any other source, I investigated and drew my own wiring diagram. An example is my nav station where dealer installed equipment was built in with little documentation as to how it was connected.

An "all-or-nothing" approach usually leads to failure. Just take it a step at a time. Plan ahead, take notes and document what you do. I have an entire three ring binder boat Electrical System.

 

[dlochner]

I have a label maker that will do heat shrink labels, which I use all the time. One thing I'm resolved to do with this project is to label each end of every wire, plus anywhere in the middle of a run that I might be tracing the wire and wonder which wire I am looking at (like at routing bottlenecks). My plan is: name, amps, AWG. So, like: bilge (5A) - AWG 18 - 20A

This makes the system more or less self-documenting. I can look at a bus, for example, and see the load the wire supplying it supports and the load the the wires leaving it add up to. Tells me if I can safely add a wire or upgrade a device on that bus.

My original plan was to only use the load supported by the wire and not include the load of the device except on the device itself or on the wire right next to the device. But, I quickly realized that I might have a bus that is capable of 30A with two wires - each capable of 20A, but with only a few amps actual load.

So, having both numbers lets me see everything I need to know:

• What does this wire go to?
• What is its size?
• For the length of run, how many amps could it support?
• What is the actual load?

I've waffled back and forth on how/where to document the actual load. That is the number that might change on a given wire. So, I might not want to include it on the heat-shrink label. I was going to just label it at the device. So, if I wanted to know the actual load on a wire, I would have to go to each device it feeds. This has the benefit of making it easier to keep labelling accurate but the disadvantage of removing my ability to see actual load at a glance.

A hybrid strategy would be to just label the wire with its name, gauge, and capacity, then keep the actual load values on a laminated 3x5 card nearby, or something.

I'm undecided on the details but do favor this self-documenting approach.

You're making it too complicated. The image below is a portion of a wiring plan I am working on it has all the information that is needed for any circuit. It was easy to set up in Excel. The wires are labeled by panel and breaker/circuit. Columns could be added to indicate wire gauge and color. (This is a draft, the items in red were edited from an earlier version, highlighted so I could easily find the changes that needed to be made on other drawings and tables.) Wires are labeled by the Id Number. Note on LP1_8 there are 3 devices controlled by the one breaker. All three wires will have the same label, depending on how the wire is actually installed, we may add a second descriptor, eg. LP1_8 (Port Light). This particular table is for 1 of 6 load panels on the boat.