Pagoda SL Group

W113 Pagoda SL Group => General Discussion => Topic started by: r.bartl on June 17, 2006, 17:13:10

Title: speedometer shows 1.6 times the real speed
Post by: r.bartl on June 17, 2006, 17:13:10
my conversion from miles to kilometer seems to be an endless story.

after I sent it to Palo Alto I got the unit with counter working but the speedo needle didn't move.

than I got it back, put it in again and when I drive 100 km/h it shows 160 km/h. I checked the real speed with my GPS.

So I send it back the second time to Palo Alto and no respond for weeks.
Now they tell me, that everything is perfect with my unit.
They cannot understand, why I get an inncorrect result with the speedo.
when they checked and measured, it exactly show the right speed.

they asked me to replace the speedo cable.

but this is what I already did at the beginning.
I bought the cable from mercedes benz.

is there a ratio, what they didn't check or could I have the wrong cable?
1.6 times is nearly exactly miles to km/h.

any idea?

230SL, 1967
ivory with chocolate brown top
Title: Re: speedometer shows 1.6 times the real speed
Post by: Cees Klumper on June 18, 2006, 11:37:06
I think you found the answer, in that they put a mileage 'wheel' inside a KM speedometer. It's too far off, and too coincidentally close to the KM/MPH conversion ratio (1.609) to be anything else?

Cees ("Case") Klumper in Amsterdam
'69 white 280 SL automatic
Title: Re: speedometer shows 1.6 times the real speed
Post by: Ricardo on June 18, 2006, 14:02:55
Wouldn't there also be a different speedo gear for each rear end ratio as well? With 4 different rear axle ratios and either Miles or Kilometer speedo's, there could be 8 different speedo gears, although technically the 3.27 rear end wasn't available to the 113's.....
Title: Re: speedometer shows 1.6 times the real speed
Post by: Chad on June 18, 2006, 20:54:31
That's an interesting situation. I think that makes perfect sense Cees.
It seems strange that this reputable company has had some trouble with this and perhaps not been so "speedy" to troubleshoot/fix the issue. Since it's a conversion it is somewhat understandable how it could occur.


1967 230SL (Manual)
Title: Re: speedometer shows 1.6 times the real speed
Post by: rwmastel on June 19, 2006, 07:38:22
quote:
Originally posted by cees klumper

I think you found the answer, in that they put a mileage 'wheel' inside a KM speedometer. It's too far off, and too coincidentally close to the KM/MPH conversion ratio (1.609) to be anything else?

Cees,

I've never been inside a speedometer before to know how it looks & works.  That said, I don't understand your theory of KmPH and MPH gears.  All you do is change the fact plate, right?  The needle would be at the same spot on the "dial" for 62 MPH as for 100 KmPH, right?  You're just changing the number under the needle.  If Roman's speedo is showing 160 KmPH when driving 100 KmPH, then the needle is way past the mid point (12:00, if you will).  I guess it is a gear problem, but I wouldn't think of it as MPH v. KmPH.  Either way, something is not right at all and Palo Alto need to fix it and pay for shipping this instrument back and forth extra times.

Rodd
Powell, Ohio, USA
1966 230SL, Euro, Auto, Leather, both  tops
1994 E420
Title: Re: speedometer shows 1.6 times the real speed
Post by: rwmastel on June 19, 2006, 07:39:39
quote:
Originally posted by Ricardo

Wouldn't there also be a different speedo gear for each rear end ratio as well?
Yes, and Roman should have visually confirmed his rear axle gear ratio and communicated that to Palo Alto.

Rodd
Powell, Ohio, USA
1966 230SL, Euro, Auto, Leather, both  tops
1994 E420
Title: Re: speedometer shows 1.6 times the real speed
Post by: Cees Klumper on June 19, 2006, 22:07:09
I think you are right Rodd - probably that is all there is to converting the speedometer. Then we're back to how there can be such a difference in the actual vs indicated reading. The greatest difference between differential ratios is 1.25 (4.08 vs 3.27 if I am not mistaken).

Cees ("Case") Klumper in Amsterdam
'69 white 280 SL automatic
Title: Re: speedometer shows 1.6 times the real speed
Post by: rwmastel on June 20, 2006, 10:22:55
There would be a gear change for the main and small odometers.  Does that effect the position of the needle?

Rodd
Powell, Ohio, USA
1966 230SL, Euro, Auto, Leather, both  tops
1994 E420
Title: Re: speedometer shows 1.6 times the real speed
Post by: Ricardo on June 20, 2006, 10:36:19
Actually Rodd, I think the odometer gear remains the same, it's just the nylon worm gear driving the speedo needle that is different...I have one from a MPH speedo driving my Kilometer unit and it reads close to what Roman is experiencing or approx. 120K at 6200 rpm with a 3.69 rear ratio...anyone know if these gears are available either from MB or places like Palo Alto?
Title: Re: speedometer shows 1.6 times the real speed
Post by: rwmastel on June 20, 2006, 10:57:24
Ricardo,

Wow, I'm really confused now.  I think I'll let the experts resolve this one!

Rodd
Powell, Ohio, USA
1966 230SL, Euro, Auto, Leather, both  tops
1994 E420
Title: Re: speedometer shows 1.6 times the real speed
Post by: mdsalemi on June 20, 2006, 15:55:43
All,

IMHO, Roman's troubles are serious business.  Previously I had good luck with Palo Alto, but the true test of one's mettle isn't how well you do when things work right, but how well you do when they don't.  In this case there's precious little excuse for this near 2-month fiasco when they can't seem to do a conversion!  Remember, here is what they say they are and what they do (and remember they get a fair piece of change for their work):

  • VDO Factory Authorized Repair station.
     
  • We are the premiere source for instrument repair, restoration, and conversion.
  • Repair of mechanical and electronic speedometers, tachometers, and electronic clusters.
     
  • All services guaranteed and work performed by factory trained personnel.
     
  • We also make cable and housings and custom ratio adapters.

The issue of rear axle ratio or even tire size isn't that important; it will for sure make a difference in accuracy but not by a factor of 1.6!  Besides, they are supposed to compensate for these issues and even provide a form for doing so so they can accurately calibrate the instruments:

http://www.paspeedo.com/ratio.htm

Sad to say they appear to be giving him the run around and now he has more invested in this conversion in time and trouble then simply finding a used Speedo in kM.  Don't know if I'd recommend them any longer.  Remember they are NOT the only guys that do this!

Michael Salemi
1969 280SL
Signal Red w/Black Leather
Restored
Title: Re: speedometer shows 1.6 times the real speed
Post by: enochbell on June 20, 2006, 17:57:17
I hope Palo Alto can sort this out, but I think RM has a point, it is illogical to conclude this is a simple MPH-KM/H problem.  Something more is going on here, and my money says it is a coincidence that the "off" is 1.6.  There is nothing different about a Km setup except the face of the speedo, yes?  The speedo needle sweeps through the same number of degrees (approximately) on either car, so I don't think this is a gearing problem.  Isn't 50mph at about the same position as 90km/h

Please keep us posted,

g

'64 230sl, fully sorted out...ooops, spoke too soon
Title: Re: speedometer shows 1.6 times the real speed
Post by: gugel on June 20, 2006, 22:03:03
Is the odometer accurate, or off by the same percentage as the speedometer?  If the two indicators do not show the same percentage error, then the problem must be internal to the speedometer.  If they are both off by the same ratio, then the problem is likely to be external to the speedometer -- e.g., in the speedometer drive gear.  I wouldn't condemn Palo Alto Speedometer at least until the problem is known to be inside the speedometer.

Chris Earnest
Title: Re: speedometer shows 1.6 times the real speed
Post by: mdsalemi on June 21, 2006, 06:04:15
quote:
Originally posted by gugel

Is the odometer accurate, or off by the same percentage as the speedometer?  If the two indicators do not show the same percentage error, then the problem must be internal to the speedometer.  If they are both off by the same ratio, then the problem is likely to be external to the speedometer -- e.g., in the speedometer drive gear.  I wouldn't condemn Palo Alto Speedometer at least until the problem is known to be inside the speedometer.

Chris Earnest


IMHO, it really doesn't matter WHERE the problem is.  Roman sent them the speedo twice, paid them a lot of money, and it is Palo Alto's responsibility to make it work.  If he bought drive gears himself to do his own work, that's different, but he didn't.  He paid for a conversion, it hasn't worked out, and the poor guy has had no speedo in his car for about 2 months, and has shipped the thing back twice and they tell him its fine?  In the oft-chance that there is something wrong with the engine gearing that drives the speedo cable or the cable itself, I would think that they could help him out a bit better; these guys have been around for many years and must have seen everything by now.  Roman's only issue prior to all this was that his speedo read in MPH (accurately) and he needed it to read in kM/h in preparation for eventual shipment back home to Austria.  Palo Alto needs to demonstrate their expertise, and show better customer service and sort out the issue no matter how it has come about or what the problem is.

Michael Salemi
1969 280SL
Signal Red w/Black Leather
Restored
Title: Re: speedometer shows 1.6 times the real speed
Post by: gugel on June 21, 2006, 16:02:25
quote:
Originally posted by mdsalemi

quote:
Originally posted by gugel

Is the odometer accurate, or off by the same percentage as the speedometer?  If the two indicators do not show the same percentage error, then the problem must be internal to the speedometer.  If they are both off by the same ratio, then the problem is likely to be external to the speedometer -- e.g., in the speedometer drive gear.  I wouldn't condemn Palo Alto Speedometer at least until the problem is known to be inside the speedometer.

Chris Earnest


IMHO, it really doesn't matter WHERE the problem is.  Roman sent them the speedo twice, paid them a lot of money, and it is Palo Alto's responsibility to make it work.  If he bought drive gears himself to do his own work, that's different, but he didn't.  He paid for a conversion, it hasn't worked out, and the poor guy has had no speedo in his car for about 2 months, and has shipped the thing back twice and they tell him its fine?  In the oft-chance that there is something wrong with the engine gearing that drives the speedo cable or the cable itself, I would think that they could help him out a bit better; these guys have been around for many years and must have seen everything by now.  Roman's only issue prior to all this was that his speedo read in MPH (accurately) and he needed it to read in kM/h in preparation for eventual shipment back home to Austria.  Palo Alto needs to demonstrate their expertise, and show better customer service and sort out the issue no matter how it has come about or what the problem is.

Michael Salemi
1969 280SL
Signal Red w/Black Leather
Restored



Michael,

Apparently you know more of the facts here than I do, or than Roman has posted.  For example, I didn't know that his speedometer was accurate in mph before he sent it in.  I will note, though, that he sent them the speedometer, not the car, so if the problem isn't in the speedo itself, it will be rather harder for Palo Alto to solve.  Disclaimer:  I have no connection with Palo Alto Speedometer, and haven't used them (although I wish I had for my clock overhaul).

But, whether or not Palo Alto could have done a better job in this case, my main point was to try to pin down the problem a bit more, so that Roman, or Palo Alto, or perhaps one of us, could come up with a solution.  In fact, another question which occurs to me is whether the  new face plate is incorrect (a photo of mine is attached for comparison, in case it might help).  The more information available, the higher the chances are for a solution to the problem.

Chris

Download Attachment: (http://images/icon_paperclip.gif) InstrumentsSmall.JPG (http://www.sl113.org/forums/uploaded/gugel/2006621183231_InstrumentsSmall.JPG)
55.64 KB
Title: Re: speedometer shows 1.6 times the real speed
Post by: r.bartl on June 21, 2006, 17:15:48
I spoke with Brian from Palo Alto.
What he explained, doesn't help us a lot, but he mentioned, that "he is not a technician and doesn't know the details, but it will all be good at the end. they figured it out and I will get back now an accurate instrument what will show the correct km/h and correct milage counter".
what I am frustrated about:
- with just an "excuse" for the inconvenience they get out of the whole thing
- my shipping costs (once overnight and twice UPS standard), I end up with much more $$$ than expected
- I was without a speedo in my car since winter
- putting it in now the 3rd time will be lot's of hours spent
- this car is not made to exchange the speedo very often. everything (glove compartment, clamps, clock, electrical connections) doesn't get better with each assembling and disambling work
- they didn't really give me an answer what makes sense to me about the issue, except "everything will be good this time"
- I like Palo Alto's website and their statement, but their service sucks. it took way too long to the current status
- it still has to be proven, that everything works, as the unit is still in california

I will keep you posted!

230SL, 1967
ivory with chocolate brown top
Title: Re: speedometer shows 1.6 times the real speed
Post by: rwmastel on June 22, 2006, 07:34:48
quote:
Originally posted by r.bartl

I will keep you posted!
Roman,
Please do keep us posted.

I would recommend that you explain to Palo Alto in detail the cost incurred by you because of their errors:
 - shipping costs
 - time spent (MB dealer shop rate = $95/hour) by you to remove/replace
Once they understand that, ask them how they plan to compensate you for your expenses which were caused by their in accurate work.  Remind them of their statement, "All services guaranteed and work performed by factory trained personnel."  If there is a guarantee, then you should not be incurring additional costs.

Rodd
Powell, Ohio, USA
1966 230SL, Euro, Auto, Leather, both  tops
1994 E420
Title: Re: speedometer shows 1.6 times the real speed
Post by: r.bartl on July 06, 2006, 22:01:09
finally got my speedo back from Palo Alto.
it works correct.

Still - I cannot recommend them with the experience I had.
Customer Service: very poor, didn't concede, that they did wrong
Turn around time: very slow. this whole action took me more than 6 months
I expected more, but finally I can get on the road with a Speedo again.

230SL, 1967
ivory with chocolate brown top
Title: Re: speedometer shows 1.6 times the real speed
Post by: rwmastel on July 07, 2006, 07:04:38
quote:
Originally posted by r.bartl

Customer Service: very poor, didn't concede, that they did wrong
Roman,

So, they never said what they had to change to make it right?

Rodd
Powell, Ohio, USA
1966 230SL, Euro, Auto, Leather, both  tops
1994 E420
Title: Re: speedometer shows 1.6 times the real speed
Post by: Benz Dr. on July 07, 2006, 10:30:06
The speedo measures the drive shaft speed corrected to rear axel ratio. My car has the 4.08 axel which is standard in all 5 speed cars. You would need a speedo that is correct to the 4.08 axel or it wouldn't be acurate - the more common 3.75 axel would result in a higher speed showing on the quage. Just to add a nice twist my car is also metric which would add an extra level of trying to find a proper speedo should it ever fail.
It works OK even though I haven't even looked at the car yet this year. Maybe I should, eh?

Dan Caron's
 SL Barn
benzbarn@ebtech.net
 slbarn.mbz.org
  1 877 661 6061
Title: Re: speedometer shows 1.6 times the real speed
Post by: Longtooth on July 08, 2006, 03:48:29
I agree with Cees.... the 1.6 factor is (was) not coincidental with the ratio of km/mile (1.609).

It's really a simple problem to deduce.... the pick-up gear is taken off somewhere after the tranny... whether before or after the rear-end is immaterial..... 4000k engine rpm at 1:1 (4th gear) is about 72-74 mph... so engine and drive-line revolves ~54 - 55 revolves times per mph in 4th.  That's also 54 - 55 revs per minute per 1.609 kilometers per hour... or about 34 revs/min. per km per hr.

The pick-up has a gear on the cable's end... so the cable rotates at a fixed rate function of the engine's speed (in a given gear... using 4th since it's 1:1 to drive shaft). But that fixed rate function depends on the gear on the cable's pick-up end.

How many teeth on the gear on the cable's pick-up end?  That depends on what speedo cable PN was purchased at the dealer.

Assume for a moment that the PN was purchased for a car with a km/hr speedometer unit.  It has n teeth on the pickup gear.  The cable therefore rotates x times for every revolution of the drive shaft.

Now assume the cable PN purchased was for a car with a mph speedometer unit in the car.  It has m teeth on the pick-up gear. The cable therefore rotates y times for every revolution of the drive shaft.

In fact, number of teeth on the pickup gear on the cable for the mph speedo would have to have rotate 1.609 times more per revolution of the drive shaft than the pick-up gear on the km per hr speedo unit.

Assume further that the speedometer's mechanism for rotating the needle is of the centrifugal type.... the faster the mechanism spins the further the needle rotates.  

The spinning mechanism is tied to the cable's rotation... and assumedly there's a set of gears within the speedometer unit between the cable's entrance and the spinning mechanism that drives the centrifugal pointer to rotate.  

Now, if the speedo unit isn't changed when the PN for the cable is changed, the speedo unit will either spin at a rate 1.609 times what the gearing in the speedo unit is designed to spin the mechanism that drives the centrifugal pointer... so the pointer will be driven to a location on the speedo dial (degrees rotation) that is always 1.609 times further along the speedo dial, OR, it will spin at a rate of 1/1.609 = 0.622 times what the gearing unit in the speedo was designed for... i.e. read 62% of the actual speed.

r.bartl's story was to convert a mph unit to a kilometer per hr unit.... so the original cable's pick-up was designed with m teeth on the pick-up end of the cable.... and the gearing in the speedo unit was designed to spin the centrifugal spinner at the rate required for the m teeth mph PN cable unit.

I don't know whether the cable PN's are unique to a mph cable or a km per hr cable unit, but if r.bartl purchased a km per hr cable unit (with x teeth) at the pick-up end to replace the one with m teeth (mph cable unit), then the speedo (without changing anything from the mph speedometer unit) would spin at a rate 0.622x that for which the gears that drive the centrifugal spinner was designed... and the pointer would point to a speed that was 62% of the actual speed.... i.e. not rotate as far along the dial as it should have.

To correct this at the speedometer, the gears inside the speedometer needed to be replaced by ones that spun the centrifugal mechanism 1.609 times faster.... bringing the pointer back to the correct degrees of rotation to indicate the correct speed in mph on the mph dial on the speedometer.

But, r.bartl's statement was that the speedo (on return from PA the 1st time) read 1.6 x the actual speed.  This means that the centrifugal device was spinning 1.6x faster than it should have been spinning. This in turn means that the gears inside the instrument had been changed to increase the speed of the centrigal unit by 1.602 times it's required speed. This in turn means that what PA actually did the first time the unit was returned was to modify the gear set inside to reflect a km to mph conversion... rather than the other way around.  This stands to reason since I'd guess that 90% of PA's speedo business in conversions is from km to mph... not the other way around. Some tech simply read the order that said "speedometer conversion (mph to km) and didn't pay attention to the direction of the change requested). To correct this, either the pick-up end gears on the cable needed to be changed to one with 1.609 times as many teeth as was on the cable he was using (making the cable spin slower), OR the gears inside the speedometer unit that drove the centrifugal spinner needed to be changed to reduce the centrifugal mechanism's speed of rotation by 0.622 times... i.e. make the centrifugal spinner spin at a rate that was 62% of the rate it was spinning when he got the speedo back from PA the 1st time.  

In any event, unless PA told r.bartl to provide the speedo cable along with the instrument to make the conversion from mph to km per hr on the speedo, or to tell them the number of teeth on the end of the cable pick-up end (or some other tell-tale for determining the cables rate of rotation for each revolution of the drive shaft (and or drive shaft + rear end), the entire conversion resides within the speedometer instrument's gears between the entrance of the cable and the centrifugal spinner.
 
There are probably small differences in the location of the km speed units on the dial compared to the respective corresponding mph units on the dial... the differences being due to using round figure ratio's for the gear drive mechanism's... like 1.6 instead of 1.609 or 0.6 instead of 0.622.

There is probably a small adjustment mechanism for the centrifugal spinner for fine tuning... but something on the order of +/- 5%... not +/- 60%!

Believe me... this isn't rocket science... it's simple ratio's of spin rates which rely on the number of teeth of one gear with to another.... and a simple centrigal spinner or other some-such analogue rotating device that rotates a fixed number of degrees of rotation per unit of input to it.... it could be a simple magnetic solenoid meter for example... the spin rate of the spinner driving the solenoid a given distance determined by the gears between the cable's input to the instrument.... but that actual analogue drive unit is probably independant of what's on the dial or whether everything is based on a km or mph unit.... or what rear-end is in which car.

Point being that PA goofed and used the wrong gear set (used a gear set that was normally used to change a km/hr unit speedo to an mph unit speedo) in their original conversion on the speedo instrument.  And from r.bartl's brief statements (indicated speed 1.6x actual speed, and reading km/hr on the dial), the only thing PA did right the 1st time is change the dial from an mph dial to a km/hr dial.  
 
So it's no wonder that PA didn't divulge their mistake.... who'ed want to admit to mis-reading the work-order.... or picking up the wrong gear set from the bin (absent mindedly using the normal gear set bin they're used to using).  If r.bartl lived in the US he could file for redress in small claims court.... and win!!!! without the defendand (PA) even trying to offer a defense....since they already know they have none.  

In fact, if I were r.bartl, I'd write a very nice letter (honestly... a nice letter... gentleman to gentleman) to the president of PA, explaining the amount of time having to be used to re-remove, re-reinstall the speedometer unit, the inconvenience experienced  and the local MB Dealer's unit rate of remove & replace speedo unit, requesting compensation for PA's error.  The only thing that's a little sketchy to me and which needs to be determined in fact, is what the end gears at the pick-up end of the speedo cable's are for a mph unit automobile and a Euro automobile, since if these are different, then part of the problem r.bartl experienced may have been due to the difference in the gear's at the pick-up end not being supplied at the time the conversion was originally requested.  Other than that little detail, it's clearly a doo-doo head dufus mistake at PA's end of the business... and they'll know this as well as I do... so a gentleman will compensate you for their error.

Longtooth
67 250SL US #113-043-10-002163
95 SL500
Title: Re: speedometer shows 1.6 times the real speed
Post by: Benz Dr. on July 08, 2006, 11:40:01
The drive shaft turns 3,750 times and the axel turns 1,000 times at about 70 mile per hour. The gear set at the back of the auto trans shows only one set in my parts book. The standard trans uses a different type of gear set that must also be a standard part. My 230SL parts book shows only one gear set and no options for gear ratios. The 230SL used 3.75 gearing in almost every car built except 5 speed which used 4.08
The speedo cable has nothing to do with how the speedo works except to drive it. There are two cables, one long and one short for standard or auto trans cars.

The speedo needle moves by spinning a magnet inside of a drum - a sort of hall effect. Small adjustments can be made to make the needle climb faster.

Dan Caron's
 SL Barn
benzbarn@ebtech.net
 slbarn.mbz.org
  1 877 661 6061
Title: Re: speedometer shows 1.6 times the real speed
Post by: Longtooth on July 08, 2006, 22:23:23
Given Benz Dr's response which states that the speedo cable's the same regardless of Euro or US (km/hr or mph), then the gears in the speedo between the input of the cable and the mechanism that rotates the needle (Hall Effect magnetic drive rather than a centrigual drive) is the only determinate of how far the needle moves per revolution of the drive shaft.

r.bartl's problem with PA's speedo instrument conversion was therefore clearly and unequivocably their incorrect selection of the gear set inside the speedo instrument.  Accordingly, they inserted their normal gear set for conversions of km/hr to mph which is most certainly their dominant conversion.  

It also occurs to me that a simple interchange (reverse order) of the gears in whatever set's in the speedo will produce the inverse gear ratio and hence drive the Hall Effect pointer to rotate by either 1.609x the actual degrees rotation per unit spin rate input or 0.622x.... depending on which gear set (km/hr or mph) was in the speedo unit in the first place.  Thus, it was also possible that PA's error/mistake in asm'ing the gears was that they inverted the order of a correctly selected gear set when they installed them in the unit.

r.bartl has a dyed in blue case to request PA re-imburse him the expenses / time he incurred in re-removing and re-reinstalling the speedo unit in his car.  They (PA) either
 1. Installed the incorrect gear set (their normal conversion set for converting km/hr to mph), or
 2. Installed the correct gear set, but inverted the order of the gears.

Longtooth
67 250SL US #113-043-10-002163
95 SL500
Title: Re: speedometer shows 1.6 times the real speed
Post by: Benz Dr. on July 09, 2006, 00:18:47
No. The gears inside of the speedo only drive the mileage counter. The cable is inserted directly into the shaft with the magnet on the other end. The gears may be different inside the gauge for the mileage counter - that's something I wouldn't know about.
I think the main difference is really the face of the speedo since the drive gear at the trans is the same Euro or US spec. The needle sits in the same place at 100 KPH or 62 MPH.
I thought they simply changed the speedo to conform to DOT regs back when these cars were new or when they were converted at point of import. We never had to change Euro cars coming into Canada so most of them still have Euro headlights and Euro specs. Oddly, the Red Rocket still had Euro head lights when I bought it back in 1993 somewhere in NC.

Dan Caron's
 SL Barn
benzbarn@ebtech.net
 slbarn.mbz.org
  1 877 661 6061
Title: Re: speedometer shows 1.6 times the real speed
Post by: Longtooth on July 09, 2006, 04:20:18
Benz Dr... you fail to account for the observations made by r.bartl ... the needle was indeed rotating by a factor km/mile more than it should have.  This requires a coupling mechanism of some kind or an adjustment sensitivity of the hall output device to the needle that can be adjusted by 60% or more... not a 'small' adjustment.

The coupling mechanism is either a set of gears or levers (lever lengths) with ratio's that are fixed at km/mile factors of one another.  Whether that coupling mechanism is between the shaft and the magnet or the magnet's hall output and the needle's is immaterial in the net of things.

In the ideal design case the only thing that would need to be converted would be the dial face and the gears that drive the odometer.  If that were the case, however, then it would be impossible to get the needle to rotate by a factor of km/mile further around than dial than it should... and r.bartl's statement is clear.... it did rotate by km/mile further along the dial than it should.

So there's obviously a coupling (gears or levers) that allow the dial to move at different rates relative to the rotation of the cable's shaft.  Such a coupling is either between the shaft and magnet or magnet and output of the hall device to the needle.

Since the odometer's coupling from the shaft has to have a gear on the shaft that drives it, and since that coupling has to change by a factor of 1.6 going from mile to km on the Odometer, it's also likely that there's another gear or lever somewhere that is changed in making the conversion... whether that coupling is between shaft and magnet or Hall device output to needle... PA changed it in the wrong direction to get r.bartl's observation to be true.

Longtooth
67 250SL US #113-043-10-002163
95 SL500
Title: Re: speedometer shows 1.6 times the real speed
Post by: enochbell on July 10, 2006, 04:47:21
I still don't buy it.  The speedo needle doesn't care if it is in a metric car or not.  It still sweeps through the same (approximately) degrees of arc regardless of whether the dial face reads mph or kph.  It responds to shaft speed in a linear fashion.  I don't believe the issue here is a simple "wrong gear" problem.

The fact that the error was 1.6 is, I believe, coincidental.  Now the odo readout is an entirely different matter.

But I am just using logic here, and would sure like to know the real explanation.  

Can anybody help?

Best,
g

'64 230sl, fully sorted out...ooops, spoke too soon
Title: Re: speedometer shows 1.6 times the real speed
Post by: Ricardo on July 10, 2006, 08:09:34
Here is a photo of the worm gear that drives the magnet for the hall effect needle...note that there are three different gears here, though not all from a 113 speedo and I assume there are others.
The metal one was in my car when I got it and the speedometer was erratic. I found the nylon gear was badly chewed and I managed to cut and slide it on it's shaft to a different part of the worm and the speedo then was steady enough to get me through the saftey check and licenced, but it read too high, I've found these other gears and more at scrap yards and I now have a speedo that is nice and steady and approx. 20% too high, so there has to be an assortment of these depending on rear end ratio....
my opinion, though I'm not an engineer

Download Attachment: (http://images/icon_paperclip.gif) speedometer gear close up.JPG (http://www.sl113.org/forums/uploaded/ricardo/200671010431_speedometer%20gear%20close%20up.JPG)
41.26 KB
Title: Re: speedometer shows 1.6 times the real speed
Post by: enochbell on July 10, 2006, 13:33:23
Ricardo,

Thanks for the pic of the speedo guts.  Is the worm gear driving or is it driven by the shaft?  I believe it's purpose is to drive the odometer gears, not the speedo needle/spindle?

Sorry to the board if I am getting boring with this, but it has me puzzled and I generally don't like to stay puzzled.

Best,

g

'64 230sl, fully sorted out...ooops, spoke too soon
Title: Re: speedometer shows 1.6 times the real speed
Post by: Benz Dr. on July 12, 2006, 23:06:16
Mr. tooth;
no one seems to understand how these things work least of all anyone who has never ben inside of a speedo gauge. The neede moves in direct relationship to input - the wrong input will give a higher or lower speed than what the car is actually traveling. The speedo is set to the rear axel ratio and there is usually a number stamped on the back housing of the guage indicating what the ratio is. Miss match the speedo to the rear axel and it can be way off. The hall effect on the needle is not that great and minor adjustments can be made to bring it into line with real speed.
The input shaft of the speedo runs the set of gears that run the odometer and trip counter so this is a hard mechanical drive. The needle on the other hand, is run by a magnetic flywheel and perhaps the magnet is different for higher gear ratios. All I know is if you put a 3.27 axel in a 280SL that had 3.92 gearing it will show a 20% difference on the gauge. When the speedo shows 60 MPH you will be really going about 75 MPH. In this case the speedo is too slow because the speedo measures drive shaft speed not rear axel speed.
The amount of adjustment inside the gauge is really for small fine tuning so if you want to use the same gauge ( there is no 3.27 for a 280SL ) the gears and the magnet or spring tension on the guage has to be changed. The cable and gears at the trans really have no bearing on this from what I can see.
This is why the gauge is set to the rear axel ratio. What everyone seems to want to know, is not how they work, but rather how one gauge is different from the other and how those differences relate to function.
 I was never good at math but a KM is .62 miles. I multiply the number of KM by .62 to come up with miles which I guess is the same as dividing miles by 1.6 to come up with KM. However, a 60% difference in guage function would mean that instead of showing 80 KPH the gauge is showing 128 KPH? Is this what Roman saw? If this is the case, this IS the difference between KM and miles and it would seem they got it backwards somehow.
If the change would have been made using a standard metric guage none of this would of happened since the change is as simple as instaling it. The gauge is set up and makes the conversin for you as long as the axel ratios are the same as the gauge.



Dan Caron's
 SL Barn
benzbarn@ebtech.net
 slbarn.mbz.org
  1 877 661 6061
Title: Re: speedometer shows 1.6 times the real speed
Post by: Ricardo on July 14, 2006, 08:45:09
Hi Greg, Dan
I must admit I have made an error here and upon reflecting on this and opening up another parts speedo, I have come to a different puzzling conclusion. Greg you are right, the pictures I showed were of the gear that drives the odo and not the speedo, and I was forgetting that when I pulled mine apart to get a steady working speedo, that the repairs I made then, weren't about the cross shaft with the nylon gears. What I discovered then was that the spinning magnet, which is directly connected to the speedo cable, plugging into the end of it, was able to move in and out from wear and this caused the speedo to fluctuate. I later repaired and replaced the nylon gear to get my odo working...memory isn't what it used to be....
I'm thinking now that the difference in speedo's for different rear ratio's may be changed by the length of the shaft to which the magnet is attached, causing the magnet to be in a slightly different position within the cup that connects to the needle itself. This cup has a sloped shoulder internally and perhaps the position of the magnet relative to the slope may influence the actual reading....??
Last night I drove the car at a steady 3000 rpm and my reading was 116 kph, (should be 105 or so with a 3.69 rear ratio and 205/70 tires) so it can't be simply k's or mph, there must be something internal to the speedo which gives the different readings depending on rear ratio and since the cable is directly connected to the spinning magnet...where else can the difference be made up???