I started up the Mercedes SL (R129) and reversed it out of the drive and as I straightened up to drive forward both the ETS and ABS amber lamps illuminated on the dashboard. This was shortly after completing the front road spring swap posted earlier on this site.

Apart from indicating more work to do, I began to wonder if I had disturbed anything when I had replaced the front springs or even earlier, the suspension top mounts.  I went about my business on that day driving the car with the fault lights illuminated, they would pop straight on as soon as the key was turned and would not clear down on any number of engine re-starts.  I had noticed disappointingly, that I had also lost 5th gear (5 speed version of the 1996 R125 with electronic control overdrive gear) and also cruise control operation.  This underlined it was a speed signal fault but I would have to investigate further to narrow the field.

I connected the car to my Carsoft code reader kit.  This vehicle being the 1996 model SL it had the Mercedes proprietary diagnostic communication system, accessed from the 38 pin diagnostic socket inside the ECU box next to the bulkhead (firewall).  You can read more about early MB diagnostic systems in this post.  The first read of the ABS system revealed a recorded fault code with the rear left wheel speed sensor (C1106). Rear wheel sensor plausibility.

On removing the road wheel and checking the cable routing, all looked well. Often the older series of Bosch wheel speed sensors have a habit of collecting large amounts of iron debris on the sensing tip over time, causing impulse read issues at low speeds.  This was in my mind as being the fault, though it was unusual that on this occasion it had not reset the warning lamps on restart and re-triggered them only when the car started moving.

I removed the sensor from its mounting hole on the rear hub carrier and inspected the tip.  It was not too bad and looked in good order.  I had previously measured the sensors static resistance of 1300 ohms (1.3k) and this is in the range I would expect across the two sensor wires. 1.2k to 1.8k is the norm, anything outside this and I would be suspecting the sensor or wiring itself.  A quick check to chassis ground, from either of the two sensor wires just proved that there is no chafing of cables to the vehicle chassis.  It all looked electrically good, so investigations continued.

Just a point on removing the sensor from its pocket on the rear carrier – invariably it will be very tight or even bound into the hole, unlike the front sensors where they have a large clearance into the hole, the rears are a ‘push fit’ for about one inch and any corrosion causes them to bind woefully into the hub.  The technique here is to remove the single hex cap head fixing from the sensor and spray penetrant all around the sensor. Push a screwdriver into the clearance hole of the sensor tab and begin to work it, levering very slightly to and fro, very carefully.  Continue application of penetrant. Tap the tag gently from the direction of the drive shaft then insert a thin screwdriver through the tab hole into the casting tapped hole and lever the sensor back to its aligned position.  keep doing this, rotating the sensor many times while spraying more WD40 or similar.  Eventually the sensor should be able to worked out from its hole.  If you have not done so already, unclip the cable from the three snap clip mounting points behind the backplate to facilitate easy removal.

I shone a torch into the hole where the sensor is mounted to reveal the reluctor ring. This is a slotted (punched) continiuos steel band that is held around the drive shaft outer CV joint body as it passes through the hub bearing centre.  Its purpose in conjunction with the sensor, is to provide measured impulses as the wheel rotates, each slot passing the sensor tip as it turns.  This passing (gap, steel, gap, steel and so on) induces a small AC voltage in the sensor coil and this is passed in the form of electrical impulses, through its connecting cable to the ABS controller, allowing it to determine wheel speed.

I prodded around inside the sensor hole and discovered that the reluctor ring, sometimes called a ‘tone-wheel’ was damaged.  It was split (like a letter ‘C’)and no longer gripped the circumference of the CV joint/driveshaft.  Obviously it was just spinning and not providing the correct road speed signal in comparison to that measured at the controller from the other three wheels, flagging a fault code. My heart sank… Although tapping on a new tone wheel is quite simple, this is a large job to undertake, as to gain access it involves removing the hub and withdrawing the drive shaft from the bearing centre – this usually destroys the hub bearing in doing so and it then has to be replaced, more £££ and time I just do not have at the moment.  Although driving the car without ABS was no major concern on my normal commute, what did matter was that 5th gear was disabled and fuel costs as they are, this would have made my toes curl with the additional cost over any length of time.

So what was I to do?

I had noticed that the tone ring was in very good condition and the surface where it once was attached was clean and oil free. I could just see its inner edge as it passed the rear of the hub, so I selected Neutral and rotated the hub so the broken edges could be seen – they were clean and nothing was missing – just a straight, clean fracture. Taking a selection of pick tools and jewellers screwdrivers from the toolbox I proved it was possible to wedge the tone ring onto the drive shaft by forcing the thin tools into the gap between the ring and hub casting around the accessible 1/3 that was visible of the ring.  The gap in the split ring was forced closed, all but for less than half a millimetre gap.

Mixing up some quality epoxy two part adhesive, using a small screwdriver I built up the edge of the tone wheel to the CV joint body, pushing epoxy into the punched holes of the tone ring, then cleaning off the excess. Allowing this to dry hard at the repaired split section, the hub was rotated and further epoxy was applied to the complete circumference of the ring, filling in the slots, pushing and forcing the glue into the ring as I went.  Making sure I cleaned any excess that could interfere with the clearance between ring and sensor. Finishing the job, allowing it to fully harden before testing the clearance by inserting a screwdriver into the sensor hole then slowly rotating the hub confirming the lack of high spots – there wasn’t any so the sensor was refitted and the road wheel replaced.

Later the Carsoft reader was connected and the ABS fault code cleared once more, a road test proved the repair to be sound and reliable.  The 5th gear worked, cruise control functioned as it should and importantly the ABS was once more available should I ever need it. (Even better no warning lamps!)

I do realise that this repair is only a temporary fix to keep me going until I revisit the rear hub to replace the tone wheel/reluctor ring, but it worked perfectly and for me is a perfect short term solution.  The repair to the ring is not stressed, it merely rotates in front of a sensor with an air gap between the two components – if it does fail again at some point before I get around to the proper fix, I will at least know exactly what the problem will be!  Two weeks on – no problems to date, but I will say they are expected to return again at some point!

This has to be one of the most simple and worthwhile jobs you can do.  Since I purchased the SL over Christmas I have run around with the hard top on.  About three weeks ago when we had that first little burst of nice weather I stored the hardtop in the back of one of the Storage Units and have run with the rag top ever since.

I kind of imagined that in the 20 years since the car was manufactured, the roof has never seen any attention for waterproofing.  This was based on the fact that water just soaked into the material and left it looking like a drenched wash leather!  Even though it remained water tight throughout, I decided it was about time to proof the mohair fabric.

I researched a few products and what came out top was Grangers Fabsil in a can.  Other stuff looked to be a little messy and in many cases was an aerosol spray (I didn’t fancy masking off and dealing with overspray on the paintwork) so I opted for a brush-on product.  Fabsil is a silicone based ‘clear liquid’ that is free flowing and dries very quickly without any residue.  In fact, even in instances where I was a little careless with the brush it wiped off the paint without issue and left no marking what so ever.  Even the brush you used will dry out overnight and not harden – this was the product for me!  eBay Link to the Fabsil product in 2.5L Can – Here.

There is a little in the way of preparation to do if you want to do a good job, but I suppose the proofer would work equally well if you just slapped it on – it really is that easy!

I dry brushed the whole roof down to remove any debris and rain ‘wash-down dirt’ that collects in the lip and edge rolls, for this I used a stiff nylon bristle nail brush and worked with the grain.  Once this was completed and all other marks removed and brushed out, vacuuming the complete roof area proved to be fantastic at removing and collecting up all the loosened debris.

Obviously pick a dry day and one that looks like being so for at least three to four hours. Blanket off the windscreen (don’t fancy silicone smears!!) and the boot lid area, tucking the protective cloth under the rear roof lip.

Use an old sandwich box to decant a decent amount of Fabsil and obtain a new or ‘ultra-clean’ 2 inch brush.  As far a quantities used goes, I gave my roof a generous coating, twice, and I used a little under half a gallon of Fabsil.  I have seen this sold in half gallon cans, but my shop only had the larger gallon can, that was in my case, far cheaper than buying its equivalent volume in one litre tins! – Plus once in stock, in the future I can proof anytime I care too !

I read several painful reviews of people brushing on the proofer once in one direction of weave then the second coat the other way.  I personally went for the ‘flood and soak’ approach, to the point where the material was soddened with Fabsil – making any weave brushing just a total waste of time!  It worked perfectly.  There advantage of this method is that by keeping a wet edge to your work with the brush, you don’t get any brush marks whatsoever, and the agent ‘wicks’ into the most hard to get places.  Painting around the windows is a cinch!  Just get to within a couple of millimetres from the seam edge with a loaded brush and let capillary action do the rest. Easy…

After the first coat you will notice how the colour of your your roof darkens to a beautiful ‘as-new’ black.  As Mohair ages and gets an ingrained surface dirt coating from the elements of our poor weather, it tends to go a dark charcoal grey and it really does look fantastic once the proofer has dried and it returns to its ‘as-new’ black finish.  The second coat goes on much the same as the first, taking about four hours to dry totally.

Some of the reviews I have read, beef about the smell involved in using Fabsil. I have found this really, to be honest, unfounded and if you ventilate the car well and let it dry properly, I must say I can’t understand why it is worthy of comment.  It smells slightly like ‘turpentine’, which I suppose is the products base solvent.

Great Job and well worth the effort (not that much effort required) – Its child’s play!

I shall report back once I get chance to take a few photographs in rain, so far we have a couple of ‘dewey’ nights and the water droplets sit on top of the fabric – we have not had that before!

Added pictures of waterproofing after first downpour and hard frost a couple of weeks later…