Update: Found High Beam column Flasher problem!
Problem's in the contact mechanism itself ... i.e. the switch contacts. I had to take the terminal board unit off the switch mechanical piece to find it though. Not a problem as long as you don't lose the little spring that forces the blinker Cu slide to the terminal board contacts, or lose the little Cu slide itself. It installs with the convex edge to the outside edge of the terminal board in case it falls off before you get a chance to see where it came from and how it was installed... the spring pushes it up and off when you pull the terminal board off (4 corner screws hold the terminal board to the switch mechanical section).
The entire switch & contacts for both the flasher and washer are inside / on the terminal board itself. The only part the mechanical section plays is that when you push down (washer) or pull up (flasher) on the stalk, the mechanical section pushes one of two little ceramic cylinder's out -- which pusher cylinder moves depends on whether you push or pull on the stalk ... The activated cylinder then pushes against the cantilever Cu spring (one for the washer switch, the other for the flasher switch) that serve as the contact mechanism in the terminal block.
I was able to trace and verify that both the flasher contacts are wired to the male plug end terminals (terminals 3 & 10... counting from the wire entry end with male pins facing you, with terminal #1 on the left, counting down to terminal #6 on the same side (left), then starting back at the wire entry end on the right side with terminal # 7 and down to #12 on the same side... right)... and the corresponding female plug too... but I haven't traced from there yet to insure the wires aren't terminated in air somewhere else. I'll trace that after I fix the actual contact mechanism in the switch.
Meanwhile, the mechanism is simple.... a copper cantilever spring is mounted on the terminal board. The spring end (not the mounted end)of the cantilever spring sits under another Cu contact leaf with some clearance between them. When you pull on the stalk, the stalk simply pushes a ceramic cylinder plunger thru a slot in the terminal board which plunger pushes against cantilever spring to push it against the fixed leaf and voila'... circuit is made.
The mechanism's design problem is that the cantilever spring's fixed end is not an integral member of the fixed terminal contact on that end, rather it's joined by what appears to be a tiny spud on the underside of the fixed terminal which presses down on the cantilever spring's fixed end.... which end is held firmly in place by simple spacial interference.... i.e. the tiny spud would have to be thicker (longer) than the space between the terminal and the bottom of a groove in the terminal board in which the cantilever spring sits.
What happened for whatever reason is that
1) either the tiny spud came loose and fell out or
2) the fixed end termination piece was initally installed incorrectly (i.e. mfg'ing tolerances) so that the interference between it and the spring's fixed end is no longer an interference, but rather a slight space.... so the spring has to be pushed on both ends simultaneously to make contact with the fixed end and the other contact. This won't ever occur if the space on either end of the cantilever spring to it's contacts is too large .... hence no flasher switch. If the space on either end is small enough, then pulling very hard on the stalk may allow the contact to be made on both ends... it depends on the space (distance) between the cantilever spring ends and their respective terminals whether pulling harder on the stalk will work or not. In my case it doesn't....the spaces are too large.
Either way, the mechanism isn't worn, the plungers both move freely by more than enough distance to push the cantilever spring to it's other contact... over pushing won't damage the spring or contacts... there are no wear patterns in the contacting pieces... so it's a mfg'ing defective condition. The likelyhood is that MB's vendor for this part used a high level of quality control (i.e. circa 5200 250SL's mfg'ed in all of '67), so the contact probably worked when it left the vendor's factory... and probably when it left the MB factory installed in the car. So the likelyhood is that the tiny spud was tenuously attached to the fixed terminal initially, and came loose, falling out shortly after the car was shipped and the flasher unit actuated a few times. This is what we (in my business) refer to as early life failures due to mfg'ing quality escapes, as opposed to fundamental reliability failures... i.e. like wear-out, fatigue cracks, etc.
Since this type mechanism is used for both the washer and flasher switch, there were 2 such mechanism's per shipped vehicle in '67.... i.e. circa 11K mechanisms in ~5500 cars. It's possible I suppose that I got a 1/10K defective unit -- my bad luck. On the other hand, what's more likely is that the vendor's process for attaching the spud to the fixed contact went awry for awhile, producing poor spud to terminal joints... causing several to fail outright at both the vendor's quality / test before shipping the part to MB and at MB before the car's were shipped.... the normal action under these circumstances is to stop shipment of the defective lot or lots, send them back to the vendor to rework or scrap (which depends on whether it's cheaper to rework or build a new one). BUT, since the mechanism of failure is a poor joint, some joints were good enough to get thru without failing outright... but failed a short time later.
Anyway, that's my best engineering guess on why the unit I have failed. BTW, I'll bet a dollar to doughnuts that rwmastel's 'pull real hard' method of making the flasher work is directly related to the same fundamental mechanism failure --- an interesting thought which might shed light on MB's design reliability and/or the vendor's process control.
To fix it I've got to re-establish a fixed contact of the spring's fixed end with the fixed terminal. There are a couple of possible methods depending on whether the tiny spud is still in-between the fixed terminal and the spring or not.... and unless I undo the tiny rivet joint which mounts the terminal to the terminal block I can't peer into the groove.
1. If the spud's still there I may be able to plastically deform the fixed terminal (bending it downwards in direction of the depth of the groove) to re-establish the interference fit --- so that the spud is wedged tightly enough between the fixed terminal and the spring to insure constant contact. This is a tenuous fix option... I'll have to bend the fixed terminal into a very slight "S" shape to enable it to press down with force on the spring (thru the spud)... though if it works it's the most reliable since there's no forces that can subsequently be applied to plastically deform the terminal back.
2. If the spud's not still there, or I can't make option 1) work reliably enough for a good fixed contact, I'll have to clean the Cu mating surfaces and stick solder into the groove under the fixed contact to wet (reflow solder) from the fixed terminal to the fixed end of the spring. The difficulty with this approach is that I have to keep the solder from wetting out the length of the cantilever spring... which is very short, btw, (or it will solder the cantilever spring on both ends... constant high beam), and still insure the solder wets to the mating Cu surfaces I'm trying to join. From a reliability point of view, solder joints of this type are not easily wetted to both surfaces, so I could end up with the same problem after a few actuators or even 1 actuation. I also don't like the potential for corrosion to set later due to the acid's used in the solder... short-term it's not a problem, but since I can't remove the entire asm to dip clean the acid flux (water soluble) & scrub (with a dental type tool) from unit I don't like the idea of leaving the acid flux around the solder joint.
Actually there's a 3rd fix option... to wedge a small piece of Cu between the fixed contact and the spring's fixed end (recreate the interference)... and to maintain the wedged condition... which is the trick. If I pre-tin the small piece of Cu wedge material 1st, clean it of flux thoroughly, then wedge it in position to re-create the interfence, then reflow the pre-tinned Cu wedge I might be able to wet the wedge to either the fixed terminal or the fixed end of the Cu cantilever spring. This is probably a better option than 2) since it precludes the possibility of too much solder wetting the entire cantilever spring and making contact with the terminals on both ends (i.e. permanent high beam condition). This still leaves some room for potential corrosion depending on whether I leave any flux in the pre-tinned wedge or not, so Option 1) is still the preferred approach.
I love these kinds of problems and hate 'em at the same time.
Sorry for the length of this, but it's just in case somebody of a like mind has a similar problem.
Longtooth
67 250SL US #113-043-10-002163
95 SL500
