MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - Brown wire to lightswitch.. gets hot.

I have been having problems with my headlight system on my '71 B. For the last while, the headlights go on with the parking lights. Now, after some other electric surprises the brown wire to the headlight switch gets quite hot when the switch is turned on full. What could cause the brown wire to get this hot? Would a short not be allowing the circuit to finnish?

Thanks

Paul C

If the wire is only getting hot near the switch it is likely caused by a bad connection at the switch or internal to the switch itself. The connector on the end of the brown wire my be corroded where it is crimped on the wire or is loose on the switch terminal.
John H

Not uncommon on switches 30 years old, particularly if you have fitted uprated headlights taking more current than standard, these should always be powered via relays and individual fuses to each filament. If the lights are lighting then it probably isn't a short, but simply bad (high resistance) connections inside the switch or where the wires connect to it as John H says. This will be causing dim lights as well as heat in the wire and switch.

If the headlights come on with the parking lights that is incorrect, the first position should be parking lights and only the second for headlights as well. If the brown only gets hot when the switch is on full then that implies something else is connected to that position of the switch that is drawing too much current. The brown is the input, red (up to 69) or red/green (69 on) is for the instruments and parking lights, and blue for the headlights.
Paul Hunt 2

If you want to take a lot of the current out of your lighting circuit, you could add some relays to handle the current differently. A relay will basically allow all of your lighting (head lighting) switches ~ dash and column ~ to deal with much reduced current which reduces pitting and burning of the contacts, some of which can be relatively lightweight for the amperage they handle. With relays (usually one for each filament circuit to your headlamps ~ brights and standard) a low amperage current sufficient to energize the relay is all they face. The much higher amperage lamp current goes through much heavier duty contacts in the relay, and that current is supplied, usually, straight from the battery, often by adding a heavy duty cable of #10 or even # 8 ga. wire to the battery connection at the starter solenoid and running it to the relay directly. A very good discussion of this option can be found at Dan Stern's lighting website, http://www.danielsternlighting.com/tech/relays/relays.html

However, whatever you decide, cleaning switches or replacing them can do a lot to reducing the heat of your supply wire. Also, be aware that, as Paul says, adding halogen lamps will introduce a greater current draw than perhaps the switches were designed to handle easily. One very good reason so many cars made since the 60s have utilized relays right from the get go.
Bob Muenchausen

All of which is good advice - my two pennorth is to suggest you have a good look at the wiring looms from the headlamps back to the main loom - they are very exposed and prone to corrode and short - sometimes the source of multiple reaction to a single switch.

Detach them one at a time and see if it solves the prob.

Roger
RMW

Thanks all, for the responses. The seperate relay has been mentioned to me, but this is the first time that the use of Halogen headlights has been offered as a potential reason. I switched out both headlights for halogen ones last week so in the mean time I will switch back to the old ones. That doesn't address the problem with them going on with the parking lights, but I will start there.
I will read that link, Bob.

Paul
P COXON

Paul Hunt wrote: "...these should always be powered via relays and individual fuses to each filament."

If you rewire your headlights to use relays (and you should), please listen to Paul, and DO NOT follow the wiring diagrans posted by Daniel Stern. I have discussed this with Daniel, and he has agreed with me that there ought to be four fuses instead of only two and agreed to change his web site accordingly, but has yet to do so.

If you don't listen to me, please listen to Paul. Paul is the aknowledged expert on matters electrical.

Dan Masters

Paul, VB sells a complete kit with the relays and wiring harness for the headlights for around $35 US.
John H

Thanks Dan, for mentioning the fuses. Paul is right and my own conversations with Daniel have made their importance obvious. Too bad Daniel hasn't updated his diagrams but I assume your own do show the fuses both Paul and you are always aware of. Rick Astley's new book, MGB Electrical Systems show them as well as underscoring their importance.
Bob Muenchausen

BTW, Daniel Stern HAS added the fuses to his diagram (I had not been out there in awhile), so he must have taken Dan Master's comments to heart.
Bob Muenchausen

Dan Masters has headlight relay kit available. He also has instruction manual with a wiring diagram available on his site.

http://www.advanceautowire.com/headlight.pdf

http://www.advanceautowire.com/
Kimberly

Danuiel Stern's website shows two fuses - one for the low beams and one for the high beams.

Using no fuses at all is OK, using four fuses, one for each filament, is OK; Using one fuse is NOT OK, using two fuses is NOT OK. In my opinion, and I believe Paul Hunt agrees, if you are going to use fuses, you should use four, one for each filament as Paul noted in his previous post.

I personally know of three people who have had a headlight fuse go out while driving at night on a winding country road. One of them plowed through a corn field before his eyes adjusted to the dark well enough to see where he was going, one crashed into a large Oak tree, and the other was lucky enough that the fuse blew while it was still light enough for him the see where he was going. Luckily, no one was seriously injured in any of these episodes, although some serious damage was done to the cars in the first two cases.
Dan Masters

I just found something interesting.
I had removed the DPO Radio Shack special fuse "Box" last week and took the four fuse box out of my 74.5 GT and put it in the '71. I bought a brand new fuse box this past week as well to replace the one I took out of the GT. After pulling my hair out more today, and looking at the schematics from both years, I turned the box upside down, the one from the GT and noticed that the top and the second fuse down are connected on the side that the Red and Green wire attach. On the new fuse box I bought, there is no connection. Oddly enough, the schematic for the '71 does not show the connection at the fuse box and the one for the 74.5 does. I am going to switch them tomorrow and see what I get. What is the purpose of this difference with the connection.

Paul
P COXON

Look at the wiring diagrams Dan Masters has provided on his website. They show the fuses for the parking lights bridged.

http://www.advanceautowire.com click on stock schematics

Print yourself an enlarged copy of the lighting circuit. Click on "current view" in the print dialog box to print an enlarged copy. I made an enlarged copy of the wiring diagram for myself that is about 20"x14". It's great for seeing details such as pin numbers on components.

Wires getting hot near the switch can be caused by hight resistance in the switch. Use a meter to test for resistance and continuity.
Kimberly

"On the new fuse box I bought, there is no connection."

The new fuse box you have is probably for a TR6. The TR6 fuse box is identical to the MGB box from the front, but doesn't have the bridge on the back.

This has caused many heads to be scratched trying to figure out why a new fuse box is causing so many problems.
Dan Masters

This is for the Dans - Masters and Stern.

I have always understood that in a relay system such as a headlight upgrade, the relay's main power input (pin #30) should be fused. This is what is shown in Daniel Stern's diagrams.

On the other hand, I agree with the concept of using 4 fuses in a relayed headlight system - one for each filament. This would include 2 ganged fuses off of pin #87, as is shown in Dan Masters' diagrams.

So, where does this leave us? It sounds like the "belt and suspenders" approach is to have an in-line fuse on each main power lead (pin 30), and the 4-fuse setup on the component side (pin 87). Two headlights, four circuits, six fuses.

IMHO.

Rick
Rick Stevens

Dan Masters is the Barney Gaylord (the MGA Guru) of British Electricals, so follow his advice. You can't possibly go wrong.
Bob Muenchausen

The point of fuses is to protect the wires, and by extension, the car, from short circuits; as such, they need to be as close to the power source as possible. The point of the relays is to ensure maximum power to the load; as such, they should be as close as possible to the load, with as few connections as possible in between. The relay setup also lets you have the fewest large power conductors running long distances, desirable since they are the ones that melt harnesses and start fires. Adding excess fuses is counterproductive. I normally run two fused power leads, one each for high and low beams. This is a practical compromise that protects everything, and gives the driver the opportunity and responsibility for switching to the other beam if the lights go out. As near as I can determine, ONLY old Britcars have unfused light circuits. A survey of books and memory here shows Mazda, Subaru, Dodge, Ford, International, Chevrolet all having a single fuse, breaker, or fusible link for headlamp circuit protection. VW and late Jaguar have multiple fuses. If you are worried about the sky falling, by all means put a lot of extra fuses in the circuit, but it's mostly a waste of time and added inefficiency and trouble points. After the relays is worthless if the relays are where they should be. I have had the admittedly exciting experience of losing illumination maybe a dozen times, on vehicles with or without circuit protection, but only the cars without also caught fire or suffered severe damage. A competent driver will have his braking distance memorized, so that catastrophic results would only occur if you are actually entering or in a tight corner.
FRM
FR Millmore

"The point of the relays is to ensure maximum power to the load; as such, they should be as close as possible to the load,..."

"The relay setup also lets you have the fewest large power conductors running long distances, desirable since they are the ones that melt harnesses and start fires."

Main power for the headlights goes from the battery to the relay to the headlights. If you put the relays close to the headlights, then you have a longer run from the battery to the relay. If you put the relays close to the battery, then you have a longer run from the relay to the headlights. The total distance current has to travel is the same either way. Where you place the relays will have no bearing on the voltage drop over the wires, as the total length of current carrying wire is the same either way.

"...with as few connections as possible in between."

That's one of the reasons I prefer no fuses at all.

"...gives the driver the opportunity and responsibility for switching to the other beam if the lights go out."

Having your lights go out at night, leaving you temporarily blind, can be a panic inducing experience. How many drivers will have the presence of mind to remember to switch the beams should he find himself driving along and the lights suddenly go out? For most of us, by the time our mind clears enough to realize what's going on, it will be too late.

"...A competent driver will have his braking distance memorized,..."

Your braking distance will be of little concern if you can't see the road ahead of you. While you're the "panic braking" mode you might find you are braking on a path off the road.

"...so that catastrophic results would only occur if you are actually entering or in a tight corner."

Entering or in a tight corner is precisely the most likely time to have a fuse go out. Panic braking will be of no help at all in such a situation.

"I have had the admittedly exciting experience of losing illumination maybe a dozen times,..."

That's about 12 times more than I would like. You survived those experiences, but you could just as easily been killed. Maybe you value your cars more than I do, but I've never seen, let alone owned, a car I thought that much of. I would be much more willing to risk the loss of a car than the loss of a life. Every one is entitled to their own opinion, but I'm of the opinion that you either use four fuses or none at all.
Dan Masters

Dan. You do not list your website on your address header. Might I suggest that you do so? Paul Hunt and Chris Betson list their's. I have found their websites to be of great use over the years. I have also found yours to be of great value. When my computer "crashed", and my wife was not able to recover all of the book marks, having Paul's and Chris's websites available through their posts, was of significant value to me. A major problem I had was, due to Chris posting his website and my, then, being able to find his tech forum, solved rather quickly.

Your website is of value to me and, I believe, to others. It seems to be referenced on a regular basis. Thus, might I ask that you modify your existing header to include the website? Such would be appreciated by many of us.

By the way, I, very much, like the simplicty of the original two fuse set up. Bugger when there is a problem, but, a very simple system which those of us who are not "electronic wizards" seem to be able to figure out.

Thus, it would be a GOOD THING should you and Paul Hunt be able to put together an "Idiot's Guide to MGB Electrical Systems". Bob Muenchausen has some good ideas and should be included in such a venture. Such a thing could be published on the internet, cost a few dollars to download, and provide the rest of us with a great deal of expanded knowledge. Please consider this and mention it to the others. All three of you have helped me over the years and I greatly appreciate it.

If you do not have the wiring diagrams from the Bentley manuals, please let me know. I have found a friend who has a scanner, is an MG owner (midget and MGBGT), a computer geek, and is willing to scan the applicable portions of the manual and transmit them to you.

Please let me know.

Les
Les Bengtson

Les-

Dan has included the diagrams from the Bentley manual on his site.

Rick Astley has written a book (recently published) specializing in MGB wiring (190+ pages). Unfortunately the delivery date is in December from Amazon. There maybe other sources for it due to some of the comments I have read about his book.
http://www.amazon.com/gp/product/1845840577/sr=8-15/qid=1153011250/ref=sr_1_15/103-0710746-6624669?ie=UTF8
Here are some comments about Rick's book
http://www.mgexperience.net/phorum/read.php?1,439739,439934#msg-439934
http://www.mgexperience.net/phorum/read.php?1,438554,438554#msg-438554
http://www.mgexperience.net/phorum/read.php?1,429403,429681#msg-429681

-Kimberly
Kimberly

I do agree it should be one fuse per filament, or none at all like from the factory. With none you have a chance that the short will not be so solid that it shunts *all* the current away from the filaments, leaving some glow albeit greatly reduced, along with smoking wiring! The fifth fuse feeding the relays is optional. Without, and depending on where you pick up a 12v supply and mount the relays and fuses, there could be just a matter of a few inches of wiring with the potential (pun not intended) to short. If you do add a fifth supply fuse then it should be rated at least four times higher than an individual filament fuse.

As regards loss of lights remember that if your headlights fail the 'other' beam may still work if you flip the dip-switch. If both fail then you have another chance by pulling on the headlamp flasher, is that is fed from a separate fused circuit. But like steering when the ABS kicks in under panic braking, how many of us would have the presence of mind to do it?

On the linked fuses, from 1970 MG changed from a 2-fuse box to a 4-fuse box, the top two fuses being for the parking lights - one fuse per side. They are fed from the same circuit of course, which is only one wire, so rather than a wiring harness link between the two fuses it is cheaper to provide a link on the box itself. Great fun when first fitting it upside down, as you can't turn the ignition off, as well as no parking lights on two corners. For two years before 1970 there were two in-line fuses feeding the parking lights - one for the front and one for the back, before that no fusing.
Paul Hunt 2

Had some time to spend with it yesterday so I separated the two blue wires at the switch and tested, first.. the larger of the two that go directly to the headlights. The headlights went on fine when given power from the brown wire. I then tried the smaller blue wire which goes to the front and rear marker lights. The wire immediately got hot. Then, I removed all the bulbs from the marker lights and made a connection again, with the blue and brown wire. All was good without the bulbs in. One by one I put the four bulbs back in. The forth bulb was at the rear passenger side. as soon as I put that bulb in, I knew that was were the problem was. After searching for a cause, I removed the black fuel vapour separator from the side of the trunk and checked the wires in behind it. AAAHA.. The blue wire going to that light was missing some of it's casing and the bare wire was touching up against the inner fender. I taped that up and tried again and all the lights work now and no heating up of the wires. Too simple.
I took the switch out of my GT and hooked it up. I know it worked properly in the GT so it should here, right.. Well, when I put the switch to the middle position for just the parking lights, nothing happens. No lights go on. I put it to full headlights and everything goes on. I tried my old switch as well and got the same thing. So I guess it can't be a switch problem now.
I will put in the relay system for sure next spring, but I can't figure out how the current is not flowing to just allow the parking lights to go on in the middle position without it being a rheostat problem. I am using Diagram 10 - 70/71 MGB From Bentley E3211W as my guide.
The origional switch is a four prong, the GT and the new switch I bought were three prong. Mine has two blue wires and one red / green wire. The GT has two green / red and one Blue.
P COXON

You were lucky removing the bulb made the difference, it must have joggled the wire, a short from the blue to the ground would still be there even with the bulb removed.

The red/green must be on the wrong connection of the switch, either that or a PO has cut the red/green somewhere else and connected the end going to the lights to a blue wire for some reason. If connecting a jumper lead between a 12v supply and the red/green wire lights the parking lights you know it is the former, if not, but connecting 12v to the blue wire lights *all* the lamps, then the latter - good hunting.
Paul Hunt 2

How about 6 fuses!

What I have done is run what I look upon as a ring main round the front of the car.

It is a loop of fairly heavy purple wire. I can't remember the exact spec. Each end is fused where it leaves the brown. The wire is well above the weight required for 4 headlamps and the 2 fuses are matched to the wire's spec. Purple so that it follows the Lucas/MG convention.

Close to each headlamp I have a relay hanging off the heavy purple. So as to avoid a joint I use flag style female spades. A continuous wire can run through one of these and I just strip enough insulation to solder the flag.

Each of my relays has a built in fuse holder. These fuses protect the wires to the headlamps and are well below the rating of the ring main fuses. The old wires from the switch now only carry a switching load.

The spec of the ring main and its fuses is more than sufficient to power 2 headlamps and the radiator fan with only one ring main fuse in operation.

Perhaps it is over kill!
David Witham

The problem with using fused relays is if *that* fuse blows you lose power to both filaments at the same time.

If you also have filament fuses, and the relay fuse is four times (say) each filament fuse, then whilst relay fuses will protect the wiring from the relays to the filament fuses, that still leaves the wiring from the 12v source to the relay unprotected.

If you want to protect the relays you should have a fifth in-line fuse where you pickup the 12v supply. That way it protects the wiring *to* the relays, and the relays themselves, as well as the wiring from the relays to the filament fuses.

If you have filament fuses, relay fuses, *and* in-line fuses where the 12v supply is picked up from the brown that is a minimum of seven fuses. If you have a loop of purple with a fuse at each end (which isn't how ring mains are wired as they only have a single fuse to the supply) as well as the above then you have eight fuses! Now that really is overkill.

I can't actually see how you have six fuses. You only seem to describe four, and would lose both filaments on either dip or main in the event of a problem at one headlight.
Paul Hunt 2

I should have said 4 relays. 1 for each filament. So 2 fuses on the purple and one in each of 4 relays results in 6 fuses.

The term ring main was used to help visualisation. I had not considered that it may not satisfy the IEE Wiring Regulations definition of a ring main.
David Witham

The only light failure I have experienced was over 20 years ago on a Triumph TR 250. The dip switch developed a short. I was lucky, it happened as I was driving into the garage. The circuit was not fused and I wasn't lucky with the wiring, there was lots of damage in the underdash loom. My MG has relays and two fuses. That doesn't seem to be adequate. That said I checked the wiring diagrams of my 2004 Tacoma and it has a fuse for each filament. My 2002 Mazda Protege 5 has two fuses, one for the left lights and one for the right.

Looking at some other wiring suggestions for relays and fusing I found these.
http://www.humanspeakers.com/audi/headlight-simple.htm
Uses two fuses.
http://www.ibmwr.org/ktech/headlight-relay.shtml
Uses one fuse, I think?
http://www.earlycj5.com/tech/electrical/headlight_relay/index.shtml
Uses one fuse.
http://zhome.com/ZCMnL/HeadlightRelays/JudkinsRelay.htm
Uses two fuses.
http://www.73-87.com/7387garage/exterior/headlight_relay.htm
Uses two fuses.
http://www.classictruckshop.com/clubs/earlyburbs/projects/head/lights.htm
Not fused.

FWIW, Clifton

Clifton Gordon

To minimize the number of fuses used, thus minimize the number of connections and still have light if a fues fails, why not one fuse per light? If a fuse blows, there would still be one head light on to provide light enough to get home. Just a thought. Cheers - Dave
David DuBois

Please help me, I very fat and do not know that to me to do with my body.
Diets do not help, nothing helps.
Who нибудь can advise me as me to grow thin?

LarryG1978

Larry,
Run as many relays as necessary to lose the weight desired and not blow a fuse.
mephisto

This thread was discussed between 22/09/2006 and 09/10/2006

MG MGB Technical index

This thread is from the archive. The Live MG MGB Technical BBS is active now.