To my other GT6 pages.

April 18, 2020


Horns

My car, and most Triumphs of this era came with a pair of Lucas 9H Windtone horns.  These were pretty widely used at the time, even seeing some duty on motorcycles.  The horns look similar, but are different in that one honks a higher note than the other.

Considering where the horns live, right inside the front grille, they didn't look too bad--just dirty and rusty.  You can just make out the "H" and "L" markings on the mouths of the horns, indicating the high and low note units, respectively.




I tested the horns to a 12 volt power supply, and they both were trying, but the tones were weak and erratic.  I measured the resistance across the terminals.  It should be something less than one ohm.  The fact that the resistances weren't infinite was a good sign, but he high readings probably indicated dirty contacts inside.

I do know that generic horns for cars are a commodity item--they can be bought cheaply at any auto store.  There are even reproduction Windtone horns available for not much money.  Either of these options would be thoroughly sensible.

But that's not the way I roll.




So I had to break into the horns.  They are each held together with six rivets.  Drilling those out revealed that the horn was composed of two body halves with a thin metal diaphragm sandwiched between.  One body half was a pot metal casting of a rolled up horn shape.The other half contained the electrics: a coil of wire forming an electromagnet, and a set of contacts in series with the coil.  With the contacts normally closed, 12 volts applied to the external connections will energize the magnet.  It pulls in a steel slug attached to the center of the diaphragm.  After traveling some small distance, a collar on the slug contacts a finger on the contacts, opening them.  This stops the current, and the slug retreats to start the cycle again.  This of course happens hundreds of times per second.  The vibrating diaphragm gets air moving, and the horn shape of the channel leading to the outside reinforces the sound.




Everything cleaned up pretty well with a light blasting.  I powder coated the horn casting, but was a little apprehensive about subjecting the electrics to the oven heat, so that half was just primed and painted.  Not much hope of finding those gaskets, so they are home made.




Cleaned the contacts by pulling a strip of very fine abrasive paper through them, and put everything back together.  I did one unit at a time in case I had to refer to the virgin unit.




The second one went a lot faster.  There are two adjustments on each of these horns.  First, there is a big threaded stud and lock nut.  The stud  is a stop for the diaphragm stud, limiting how far it can travel.  Second, there is a small screw that bears on the fixed side of the contacts, allowing it to be moved slightly.  Oddly, that screw is left hand threaded.




I measured the protrusion of both of the adjustment screws before I removed them, so I could at least have a good starting point for tweaking them.  Neither of the screws changed the pitch of the horn--they are just mechanical adjustment of the circuit-breaking action of the contacts.  If they are too far from optimum, the horn just clicks or grunts.

One horn worked fine after assembly without much tweaking.  The other wouldn't do anything.  After fiddling a lot with the adjustments, I finally measured the resistance.  It was around 12 ohms, suggesting some crud still in the contacts.  So, apart it came for another contact cleaning, and that did the trick.

Factory frequency spec for the horns is 390-400Hz for the low tone horn, and 490-500 Hz for the high one.  My horns were in or very near spec, together approximating most of a harmonious G major chord.

I like quick and fairly easy projects like this.  Low stress, good chance of success, and little or no investment.  And it was good for a few pleasant afternoons in the shop.  Cost was essentially zero.

Comments to Ed  at elhollin1@yahoo.com

To my other GT6 pages.