Relays and Wiring Lucas 14W Wiper Motor
Caution: These early diagrams did not work for me - See below
At some point, I purchased a Lucas 14W windscreen wiper motor from eBay and installed it, described here. This is a 2-speed motor with a self parking feature. It requires a special switch that I understand is simply not available anymore. Besides, I really like the switches I have and would not change them anyway. No, they are not original and probably not even British. I think I got them from a French car at some salvage yard many years ago. But, I digress.
The problem with not having the original switch is dealing with the auto-park system. The motor has an internal park switch, one side of which is always hot with the ignition on. So when the wiper switch is set to off, the motor continues to run until the wipers reach their park position, then the internal switch opens and the motor stops. I don't know how this motor is wired internally but I have read, and been told, that a direct short occurs if either low speed or high speed is actuated while park is enabled. I have also read, on the internet, that Lucas put a resistance in series with the low speed field winding for high speed. I am not an electrical engineer but that somehow doesn't make sense to me. I do find a slightly higher resistance through the high speed windings than the low but it seems that would be due to more windings to increase the magnetic field rather than some resistor. Anyway, either way it is not important for this discussion.
Based on that information, I constructed the single relay system on the left here that I
found on that same internet and gave it a try. The park function is isolated by the relay
and both slow and fast engage the relay to disconnect park. Well, it did not work and I
don't remember exactly why it failed, if I ever knew. I tried again with this two relay
system, also found on the internet. Well, it failed too. Looking at it now, I'm not sure
it ever made much difference from the single. So I gave up and eliminated the relays
entirely. This meant a separate somewhat hidden switch to manually engage the park system.
It worked all right so I left it that way for a few years. Remember, the car was not really
used during that period so I can't make any real claims about how well. That diagram is
below.
So now, a few years later, restoration is again progressing and I ran across some of these
old notes. It seems that I should be able to make the system work right and automatically
park, without that manual hidden button. Looking at the diagrams, two thoughts came to mind.
Maybe the relays were slow enough to engage that the direct feed to high speed occurred too
soon and caused a short. This seemed possible but unlikely. Maybe the miniature relays from
Radio Shack were not able to carry the load although they are rated at 10 amps. That really
didn't seem likely either. Anyway, I got some Bosch automotive relays and built a prototype
system using this new two relay diagram of my own design (not from internet this time). It
uses one relay to enable/disable the park function and the second to select between low and
high speed. The relays are rated for 30 amps and neither slow nor fast can engage before park
is disabled. It definitely should work. In all the diagrams, except the manual one, the feeds
are: 1, black, earth; 2, green, hot; 3, black/green, high speed; 4, white/green, slow speed.
In the manual one, the slow speed white/green wire goes directly to pin 5 along with the out
feed from the park switches. Also, the actual internal windings of the motor is likely not
as I have indicated but it works for my design purposes.
I removed the wiper motor from the car to the work bench and began testing. Things did not work as I expected. I began disconnecting various wires, one-at-a-time until I discovered that the internal park switch never stops the motor. So park is running all the time. Looking at my manual park diagram, it appeared that all park really does is feed the low speed winding. If true then park and low speed are really the same. Checking my actual wiring, seemed to agree. There should be no conflict if both are hot at the same time. I can't say that about high speed yet though. Now it was time to open up the motor again and discover why park never shuts off. The answer was the internal park switch had apparently failed and was always closed even when the cam tried to open it. Thus park, or slow speed, was always on. My manual switch worked because it only applied power to slow speed when pushed. I did an internet search hoping to find at least a good used switch. I found one on eBay but it was eBay.uk. More search found new ones at Holden Vintage and Classic, Ltd., also in England and I ordered it.
There was a secondary benefit of having to remove the internal park switch. It is completely isolated electrically from the motor. Since it just feeds the slow speed winding, I don't believe there is any way it can cause a short, at least with direct low speed. I can't say that for certain about high speed but I am beginning to suspect it would not. Probably the only way it could is if low and high speed are two distinct windings and I find that unlikely.
That's where I am now, waiting for delivery from England. When it arrives, I will put the motor back together with the new switch and try again. If it works, and I fully expect it will, I will try the same circuit with those same miniature relays and see if it works. I would rather use the minis as that will fit in a smaller project box that easily fits in the same compartment with the motor.
Well, I finally got the switch from Holden and there is an interesting back story. I ordered the switch on May 20 and Holden delivered it to the Royal Mail on May 21, pretty good service. The Royal Mail delivered it to the US Postal System in Chicago, IL on May, 23, more pretty good service. According to USPS tracking, the package was "delayed delivery but in transit" until finally delivered to the local Post Office in Oklahoma City on June 30, not very good service at all, who delivered it to me on July 1. Thirty-eight days to make less than 800 miles is less than 20 miles per day. A younger person than I could have walked it that distance in that time.
So I reassembled the motor and tried it the next day. It ran fine on low speed and high speed. Low and high are relative terms for a Lucas wiper motor, but I digress again. It failed to park, however. It would hesitate at the park position but immediately take off again. It did this many times. I tried adding manual resistance to the cable thinking that the motor's momentum was simply carrying it past the park cam which again turned the power back on. I thought the additional friction of the working system might be enough to overcome that momentum. This had no effect. Next, I thought maybe the button was possibly not raising the contacts soon enough or far enough. I removed the switch from the motor and attempted to operate it manually. This seemed to work but apparently I held it too long, probably why it worked anyway, and the smoke escaped, melting the thinner of the wires on my test system and slightly burning my finger.
Enlightenment: Finally I Understand the Whole Story
On opening the switch, I discovered that when the button is pressed, it does break contact with pin 4 but also creates a dead short between pin 2 and pin 1, which is ground. As you can see in that last diagram above, I had the constant 12 volt supply connected to pin 2 this time which, being sent directly to ground, caused a dead short. No wonder the smoke escaped. In doing so, it also did minor heat damage to the switch internally rendering it no longer able to function. The button was locked in the pressed position. After a fair amount of "cleaning up", I think I have it now functioning again and will try again when I know a little more about it. If only I had known about that hidden ground but hindsight is 20/20 and foresight is not. I did it that way just because it made the pin numbers match on the last 2 connectors.
Additional internet search found one article that claimed Lucas sent the motor a short blast of
reverse voltage to act as a brake causing it to stop more abruptly. This seemed plausible, when
considering the way it coasted through and only paused for me at park. The question was, where
did Lucas get that reverse voltage? I finally found this rather good description of the manual
switch intended for this motor at Jag Lovers, I think. Study of that and the resulting diagram
of the park switch with its internal short, finally began to make sense. When the park switch
encounters the park cam, it first breaks contact between power at pin 2 and pin 4 where it had
been returning through the manual switch to the low speed winding. This removes voltage from
the armature windings causing the electrical field to collapse. This collapse, in turn induces
a reverse voltage which is sent back through pin 2 which is now shorted to pin 1 which is ground.
This allows a reverse current flow, attempting to run the motor backward, effectively working as
an electrical brake. Hopefully that will be enough to make it park rather than hesitate. Tomorrow
will tell, maybe.
The diagram shows that in the "Off" position, the switch connects its pin 2 to pin 6, which then connects pin 2 to pin 5 at the motor. With the internal limit switch not in the park position, this directs power from pin 4 to pin 2 then back through the manual switch to pin 5 at the motor, engaging slow speed. When the cam reaches the park switch, it disconnects pin 2 from pin 4 and the power to run the motor. Then it connects pin 2 to pin 1 to supply a path to ground for that reverse voltage from the collapsing electrical field. In either run positions of the manual switch, pin 2 is not connected to anything, effectively isolating the park switch from the rest of the circuit. In low speed position, it connects its pin 4, which is always hot with the ignition on, to its pin 6 which connects directly to pin 5 at the motor for slow running. In the high speed position, it connects its pin 4 to pin 8 which then connects directly to pin 3 at the motor for fast running. Again I point out that slow and fast are highly relative terms in Lucas speak.
Our goal then is to duplicate the function of that Lucas switch #35927 with whatever we have to
use. In my case, it is a simple SPDT switch with center off or Low-Off-High. My final design,
the left diagram here, is a 2 relay system that should work with any similar, simple switch to
exactly match the Lucas switch performance. Note the corrected configuration of the park switch
and its wiring. The first relay performs the function of isolating the park switch when either
low or high speed is selected. When energized, it disconnects pin 2 at the motor from the circuit
and connects the ignition voltage instead, like pin 4 on the Lucas switch. Note that pin 2 will
still be hot when the park switch is not on the cam but it is not connected to anything. When the
switch reaches the cam, pin 2 will still be grounded but then it is no longer hot. The second
relay routes the power from the first relay to either pin 3, when energized by the high speed
selection, or pin 5, when not energized, on the motor for high or low speed
operation. The park function uses low speed as well. The diode at the first relay is to prevent
the low speed selection from also energizing the second relay which would select high speed
instead. The inputs for both diagrams are 1) black, earth; 2) green, 12v; 3) black/green, high
speed; 4) white/green, low speed. Both are shown in the off and parked condition. Note that in
both diagrams I have rearranged the wiring to the motor to better reflect how I now think it
probably works.
The right diagram I believe will also duplicate the function of the Lucas switch although from a slightly different approach. Instead of using a second relay to route the single power source to the correct destination, that power comes directly from the signal leads of the manual switch. Note that the "always hot" input lead at pin 2 does not feed through the relay normally open pin as it does for the 2 relay system so it is not used in either run mode. This may be an advantage in using the miniature relays that are rated for lower current in that only the park load for less than a full sweep is carried by the relay. In my case, all the wires are adequate to carry the full running load. In some applications, the B/G and W/G wires may not be and the running load must be borne by the G, always hot wire. This system is enough simpler that I will try it first and use it if it works as expected. The second diode is needed here since both signal leads perform dual functions. As I said, tomorrow, maybe.
So "tomorrow" came a couple of days later and I was able to test both the single and dual relay circuits with complete success. In trying to repair and reassemble the new switch that lost its smoke, I managed to lose a tiny spring. I thought I could just retrieve one from the original switch. On finally getting it opened, there are no springs, just springy brass contacts. I found the main contact arm was bent enough to prevent operation. I straightened it and it all seemed to work after reassembly so I used that for the test. There is a weak point in that arm where it bent before and I do not trust it to not bend again. Before reinstalling the motor, I will probably order yet another switch. Hopefully this one will not be lost in the mail for six weeks.
The next step was to duplicate the circuit with the miniature relays. Another spec I found about the motor is that it draws 1.5 amps on low speed or park, 2 amps on high speed. The relays are rated for 10 amps so should be more than adequate for the purpose. The only current they will carry is the park 1.5 amps and then only for less than one full sweep. I hoped to rewire the old board I made originally but it turned out impractical. So, I ordered some more project boards and started over.