Rear axle ratio identification and strange revs vs speed ratio (SOLVED!)

[FGN59]

Hi everyone,
Since purchasing my 280SL (late 1968 manufacture, 4-speed manual, originally US specs) last autumn, I have been puzzled by the high revs vs speed attained, even though I know that something like that is to be expected, at least with standard US axle ratios.
I was kept busy by some tuning issues until now, following an complete engine rebuild prior to my purchasing the car. I had to learn about the car, the engine, the FIP, etc, plus I am not an experienced mechanic, even though I usually find a (somewhat satisfactory) way out of most situations, after much huffing and puffing, more thinking (wrong at first, then less wrong, until light sometimes dawns), even more time, and a little doing once in a while (including a lot of wasted effort; but I do not deal with the 'heart' of the engine or transmission/suspension for lack of knowledge, experience and tools). The car now runs quite well, after I restored the cold start valve, dialled the correct program into the newly installed (but wrongly programmed) 123 distributor and adjusted the timing (also completely wrong for this engine). There are still some small shortcomings to deal with, but the overall performance of the engine is acceptable (plugs look all right, just brown, not white, not dark or oily, and driving is easy and fun in 98% of circumstances).
On to the revs vs speed issue.
To give everybody an idea of what I am dealing with, the car currently revs at 4.325 rpms in fourth gear to reach 100km/h or 62 mph, whereas even with the shortest axle gear ratio (4.08) listed in the table published by SLS on their website (which according to the info I read on the forum seems to be valid) the car should run at 3.475 rpms. Now, for some reason, the previous owner had tires installed that are slightly too small (circumference wise): they are 195/70R14, which is a little bit less than 5% smaller than the regular 185/80R14. So, logically, I would have to run the engine slightly more than 5% faster to reach the expected speed. 5% more than 3.475 rpms gives a theoretical result of 3.650 rpms to reach 100km/h. As detailed earlier, the engine is running at 4.325 rpms to reach such speed, which is nearly 25% faster than normal. 5% of those 25% are accounted for by the wrong tires (which I will change soon), but that still leaves me with a 20% gap which I cannot explain. Unless somebody tells me it's not a bug, it's a feature! In which case the car could be a contender for the fastest 100 feet dash (with exception maybe for some highly unorthodox, if still impressive, and quite frankly, quite desirable M127 V8 contenders).
Plunging under the car earlier today, I snapped a picture of what I hope is the rear axle ratio plate, although I must say I have a hard time making any sense of it.
So, after a very long introduction, my very simple question: can anybody help me make sense of this? Any ideas much appreciated, for I have none whatsoever!

[FGN59]

Sorry about the pictures, they are not the right side up. I tried correcting this, but failed (so much for working my way out of most situations...)

[Benz Dr.]

That isn't your rear axle ratio but it could be a casting number or date of manufacture.  Gear ratio is stamped into the bottom left side of the main housing. You will see some numbers, which are the serial number of your axle, followed by the gear ratio.

I tend to watch my tack more than the speedometer. Tack is generally more accurate.

[mbzse]

For position where ratio is stamped, see picture

[FGN59]

Alright, thanks. Sorry for the confusion and waste of time. Will look the right number up tomorrow.

Dan (if I may call you Dan, especially after you provided the tip that solved my cold start valve problem last winter): I agree with the tach reading being more accurate that the speedo. In this case I forgot to mention that the speed being quoted in my previous post was GPS derived (exactly for the reason you mention, and to take the tires out of the equation); I took many readings, all in straight line, always over a reasonable amount of time to get some stability (driving with a constant tach reading); altogether I'm fairly confident of the speed reading being accurate to within +/- 1 km/h. Still, even assuming an error in the speed calculation/reading, it cannot be 20%... I am not that deaf and blind (yet) 

[Benz Dr.]

Everyone can call me Dan or something else if they don't like me - that's their choice.

Something is odd with your findings, which you have researched, and didn't just state something with nothing to back up your findings. That's good work in my opinion. And, I have no logical answers to that question. It doesn't make sense, yet there it is.

[JamesL]

I just had to double check you are a manual not auto as I wondered whether your car was really shifting to 4th.
I'm guessing you double checked that 

[Pawel66]

Not coming in with knowledge, but an idea. Is your tacho reading ok? If you have a stroboscope lamp  with rev counter - it will display the revs. Or maybe you have the 123 with bluetooth.

Maybe tacho is playing with you.

You will know more when you check the gear ratio.

[FGN59]

Hi James,
I do confirm the car has a manual gear box (hard to confuse with auto), and I’m fairly confident I was driving in fourth gear when making these measurements  . All the more puzzling...

Unless Pawel is on to something? (Reading his input on the forum, I got the impression he is a very astute guy  )

The engine rebuild has been thorough and very well executed, so I have no reason ‘a priori’ to doubt the rev gear/counter. Yet, the guy who put the car back together (he was a very reputable car mechanic , but not a MB specialist) was different from the person who rebuilt the engine. He made some small mistakes with the ignition and cold start valve (wrong program, wrong wiring). So let’s just say the tacho reading is wrong. How would I check that the revs are correct (or not)? My strobe doesn’t have a rev counter, and the distributor is the plain MERC-6-RV (not Bluetooth).

[bogeyman]

If your tach is correct you would max speed at 93mph.
I would find a lonesome road and wind it out and see if that is correct.

[450sl]

François, I suggest you use a GPS or an app on your mobile phone to verify your speedo readings are correct.

Mark

[wayne R]

Francois, im going to talk tires,as i had a similiar situation to you,on a 1974
Alfa Romeo GTV i owned CA, you say about 5% differents from the original  size195x70x14,
but you may not realize there is there can be a big differents to actual tire diameter ,and rolling radias,
and also the tire  brand, tread type and tire rating   example should be at least H rating for our Pagodas.
A lower rate cheaper tire has lower rolling  radias than a good quality high rate tire, so i would go down that  way,
with tires first then,look for other reasons like others have said,regards.

[FGN59]

Wayne: thanks for sharing your experience; it conforts my decision to change the tires soon, but I have a hard time believing just changing the tires would improve the speed by 20 or 25%. I intend to use Vredesteins, which according to my research both on this forum and outside, offer what appears to be an excellent overall performance.

Mark: you’re absolutely right, this is the easiest way to verify the speedometer in a car; I did this, and can report the speedo is off by about 3-4%, i.e. it shows speed that is about 3-4% more than actual GPS derived speed, which, taking into account the wrong tires on the car (their theoretical circumference is a bit less than 5% too small), is actually fairly accurate. Rephrasing this, in a theoretical sense (as those who read my usually lengthy posts can probably guess by now, I’m generally real fond of theory, a bit less proud of my real world experience/performance  ), had the car the right size -factory installed- tires, the speedo would display speed readings quite close to reality (to be confirmed in a later post; for those of you who have been following this thread with keen interest -maybe because you’ve run out of new stuff to watch on Netflix and are tired of reruns- don’t hold your breath, this will not happen until October at the earliest, as we’re about to leave for the summer, leaving Ms Pagoda behind).

Mr Bogeyman (as in ‘bogey coming in at 11 high, break left’?): now you’re throwing me in confusion again (but I appreciate your help and input; if I may cheaply paraphrase a well known US phrase, ‘e pluribus unum’: out of many, one, or in this -slightly off- interpretation: the more input, the better the outcome). The measurements I made of the car’s actual performance would lead me to believe that it would top out at about 145 km/h (90 mph) for max revs of 6.250 rpms with the current setup (gearbox, rear axle, smaller than usual tires). With normal tires, this would increase to about 94-95mph, which is right in your ballpark.
Now I have not taken the car to such an extreme yet (and may still not in the future, unless the consensus on this forum is that, provided everything else checks out fine, it is safe enough to try for a -very- short  time), as 1/ it’s tuning wasn’t good enough to make this reasonable, 2/ after the engine overhaul, and despite all the good things I have to say about the way it seems to have been done -and who did it- I want to build confidence by slowly taking it into higher and higher revs, longer runs, hot weather, etc. (not even mentioning brakes, steering, suspension etc. -which all appear to be in excellent shape but which I haven’t tested to some again reasonable limit yet) until I’m willing to really ‘let it out’. I did take it to about 130km/h (80mph or thereabouts), and was not impressed by the resulting conditions (very high revs -5600 to 5650 rpms- and noise). Normal kind of noise, but still a lot of it for a German GT even of the 1960s. Handling and breaking so far seem excellent (for a stock car).
These babies being entirely mechanical (no over-the-air software updates or some such follies), everything should be proportional (with the tires providing some small slippage, as Wayne mentioned, but not much): 5600rpms at 130 km/h should equate to about 145 km/h at the 6.250rpms manufacturer’s recommended limit.

So, and again at this stage in an entirely theoretical construction, my car would reach 93-94 mph at 6.250 rpms, as you’re suggesting is the normal speed. But this doesn’t jibe with the 4.325rpms for 100km/h (62mph) listed on SLS’s website

Is the SLS information correct after all? How come a car advertised to reach 200km/h can only reach 145km/h in actuality ? The German press would have eaten the MB engineers alive if it were so. The mystery goes on...

We had a little bit of rain this morning (not expected by the weather specialists, so this tells us something about theoretical constructions , but the sun is shining again, the outside terrain is drying, and pretty soon I’ll be able to get back under the car and hopefully bring back information on the actual axle ratio (assuming the inside matches the outside stamp).

[450sl]

François, this one is from the net. have fun

[Peter van Es]

I'll add some fuel to the fire. My US derived 280SL has Vredestein Sprint Classics in the correct size (185/80R14). I have a calibrated trip master for rallies, and a very accurate app on my phone for speed via GPS.

My speedometer is spot on, exact with the GPS. I have driven my car at 140-150 km/h for hours on end. Of course it's loud, it revs hard, but it definitely is doable. And I'm not the only one, Jon B, John Dreu, Martijn Sjerps and Rick Bond drive at the exact same speed for hours in a convoy in Germany. That's a 230SL, a 250SL and a couple of 280's.

So if you feel your car can't do that, there definitely is something weird going on. Check the axle ratio...

Peter

[FGN59]

OK, thanks to Hans (mbsze), the stamp on the rear axle ratio was located, 3 inches in front of the false flag from yesterday.

Let's cut to the chase, it is the standard 4.08 ratio, no surprise there (unless it has been modified inside, a possible conspiracy theory I am not ready to explore yet).

So, the plot thickens, with very few (logical) outs:

Here we have (I have):

- a stock US specs 280SL, manufactured in December 1968, with matching engine block number (according to the data card) and standard cylinder head as evidenced by the number embossed on it (assuming standard camshaft too) - VIN on data card, identification plate and stamped into chassis agree
- stock G72 4-speed manual gearbox (until proven otherwise) - serial number unknown as of now (I haven't been far enough under the car yet) - data card says 001506
- stock 4.08 rear axle (in good shape, no leaks, no noise, but no apparent recent overhaul)
- slightly smaller tires than stock (190/75 instead of 185/80), assumed to be -roughly- 5% smaller (circumference) according to https://tiresize.com/calculator/

This is pretty much rock solid, only a blind person couldn't formally agree.

Now comes the more fragile part:

- I clocked the car, on at least 5 different occasions (but always using the same instrument, an iPhone running Waze), and came up with an engine running speed of 4.325 rpms (not 100% accurate, but close) on the dashboard tacho for a (GPS based) speed of 100 km/h (give or take 1 km/h)

(Sorry for the digression, but I'm trying to build some credibility with this community  ): this was done 1/ out of a willingness to 'know' the car as well as possible -among many other endeavours to that effect-, and 2/ because I was planning to take the car to an oldies-style rallye in Morocco last spring; the pandemic took care of that plan (good for the car, not so much for the rest of us), but until late February I was really trying to get ready. This type of rallye is not a race -not at all-, it is a leisurely tour on normal roads only (no off road), peppered with 3 to 4 short portions each day where one has to observe (as strictly as possible) a fixed speed (of one's choosing, but always pretty low, like 21 or 27 mph; there are some fun parts where you have to drive 2.3 km at 37 km/h, followed by 1.7km at 43 km/h, ending with a severe headache (and a big goofy smile when you've messed up totally) unless you have an onboard computer, which my wife and I refuse to have; we do it with pencil and paper, the car's stock instruments (odometer, speedo mostly replaced by tacho) and a mechanical stopwatch. Any over or under clocking compared with the theoretical timetable will give you points (1 point per second), and you are 'flashed' by surprise at various (unknown) points during the selected portion. Lots of fun and no damage to the cars. So as part of the preparation, I produced several charts to help us make the necessary calculations faster, of which one is attached for your benefit. This requires a pretty accurate setup, as the winner last year came up with a 60 seconds penalty at the end of a week, 1.400 miles of road, and 17 clocking exercises (so close to a three-second accuracy per clocked portion, with an on-board computer of course; we managed 360 seconds, so more like 21 seconds per; but our car is on the brochure for the 2020 edition  ).

OK, back to business:

SLS publishes a sort of chart (on their rear axle page for W113: https://www.sls-hh-shop.de/main/en/35-rear-suspension-35-b-info-axle-ratio-c-3_2346_120), which says the same car should run at 3.475 rpms for 100 km/h (if I can still read a chart).

So we have, for the same speed, on the one hand 4.325 rpms, on the other 3.475 rpms, the former being 24,5% higher than the latter. As I stated in my first post of this thread, the tires, being roughly 5% too small, should account for roughly 5% of the difference, leaving 20% (roughly) unexplained.

As Sherlock Holmes (Conan Doyle actually) used to say: when you have eliminated the impossible, whatever remains, however improbable, must be the truth. Which is exactly where we stand:

- either the data on the car is wrong (my possible bad, but really not good for my self esteem)
- or the SLS chart is wrong (unlikely)
- or I'm missing something; just as bad, as it means I'm becoming blind or senile myself (if so, please tell me quickly, as it is said that a quick death is better than a prolonged agony, all theoretical of curse/course).

Would any of you gentlemen care to comment (you don't have to be polite), or even perhaps take your car for a spin (assuming it has a 4-speed mechanical gear box and a 4.08 rear axle), and report what you observe?

  :-

[FGN59]

Hi Peter,

I know you must be good with numbers in general, otherwise you wouldn't be entrusted with our economies (I hope  ) as an audit committee member, so I won't dispute the accuracy of your Pagoda numbers  (in fact, I respect all you say and write, especially since we compared tractors a few weeks ago).

I'm not saying my car couldn't reach such speed, but it would take me right to the edge of the red zone on the tachometer (6250-6300 rpms), and while I trust German engineering, I am not ready to do just that until I've really proofed the car, and certainly not for hours on end.

May I ask you at what maximum revs the engine is working when you drive at 145-150 km/h ?

 

[FGN59]

450sl,

I will look into it  , and report to you, after a few drinks only I promise 

[Peter van Es]

because I was planning to take the car to an oldies-style rallye in Morocco last spring; the pandemic took care of that plan (good for the car, not so much for the rest of us), but until late February I was really trying to get ready. This type of rallye is not a race -not at all-, it is a leisurely tour on normal roads only (no off road), peppered with 3 to 4 short portions each day where one has to observe (as strictly as possible) a fixed speed (of one's choosing, but always pretty low, like 21 or 27 mph; there are some fun parts where you have to drive 2.3 km at 37 km/h, followed by 1.7km at 43 km/h, ending with a severe headache (and a big goofy smile when you've messed up totally) unless you have an onboard computer, which my wife and I refuse to have; we do it with pencil and paper, the car's stock instruments (odometer, speedo mostly replaced by tacho) and a mechanical stopwatch. Any over or under clocking compared with the theoretical timetable will give you points (1 point per second), and you are 'flashed' by surprise at various (unknown) points during the selected portion. Lots of fun and no damage to the cars. So as part of the preparation, I produced several charts to help us make the necessary calculations faster, of which one is attached for your benefit. This requires a pretty accurate setup, as the winner last year came up with a 60 seconds penalty at the end of a week, 1.400 miles of road, and 17 clocking exercises (so close to a three-second accuracy per clocked portion, with an on-board computer of course; we managed 360 seconds, so more like 21 seconds per; but our car is on the brochure for the 2020 edition  ).

Ahhh, regularity rallies. That's why I have a precise tripmaster in my car. Like one of these, classical, mechanical... and indeed, timing tables and a stopwatch. A device like that will help you get much better scores than just using your odometer, also because it is easier to reset, and you can adjust it during calibration to correct for tire size changes.

The one I have comes from these guys: http://gtirallytwin.nl but it looks they've stopped making them.

But sorry, can't help you with your top-speed. I'll check my revs at 100km/h next time I'm driving it...

Peter

[FGN59]

Thank you Peter, first for taking the time to read my lengthy posts (!), and second for your kind suggestions. I am aware of these types of contraptions, but we kind of like it as it is, if we come out of it still together it's like a proof that the marriage can last another year (maybe) 

Forgot to mention in my previous post (yes I know, it was long enough, but this will be short, and it is essential):

- there is also the Pawel hypothesis, as a last resort (to be taken seriously: possible wrong reading of the tachometer)

 

Cheers from sunny Provence

[Pawel66]

I think any car shop should have a strobo lamp with tachometer. You may just drop by to one and have it checked in no time...

Probably to eliminate this option.

Next may be: modified gear box or modified differential, as you wrote.

Or something completely silly that none of us can see...

[doitwright]

I may have missed it in this lengthy thread, but have you ruled out the clutch?

[Benz Dr.]

I may have missed it in this lengthy thread, but have you ruled out the clutch?

Are you thinking his clutch is slipping? It would really have to slip to produce that much over-rev.

[Garry]

When i purchased my very first Pagoda, i always though it had high revs.  I drove it for several months and 100kph appeared to give me a rev around 4000 in 4th.  I finally got my data card and working out what it all meant, i realised that my car was a five speed. I had never tried to get into 5th. I went back to the pervious owner who said he was not aware that it was either and had never had it in fifth gear but had only owned it for a short while, having imported it from New Zealand.


Just saying it happens.

[doitwright]

Are you thinking his clutch is slipping? It would really have to slip to produce that much over-rev.

It's been established that certain basics were discovered to be out of order. Considering all the options discussed, it is just another component to be checked.