Which side for PAS sensor

[WestOz]

I'm having problems with my PAS system similar to Wadimka http://goldenmotor.com/SMF/index.php?topic=5662.0 which didn't have a solution posted, so I don't know if one was found or not.

However, it occurred to me overnight (when you always come up with the best ideas  ) that the side on which the PAS sensor is mounted would change the polarity of the magnet passing and the direction that the magnet passed over the sensor, both of which I believe are significant in how the sensor operates. 

So is there a right side and a wrong side for the sensor to be mounted? 

I currently have it mounted on the crankset side, because that was the side with the lip on the bottom bracket to mount the PAS.  The magnet disc has the arrows in direction of rotation so that all looks correct and its evenly and closely spaced to the sensor.  But it doesn't work properly, so am I on the wrong side? I can't easily test my thought as the Bottom Bracket doesn't have the lip required for the PAS to be mounted on the left and due to the mix of left hand and right hand threads the bottom bracket can't be turned around.  So before I go out and buy a new bottom bracket, does anyone know if it will make a difference?

[Bikemad]

I have mine fitted on the left hand side and it works fine:



On page 24 of the GM User Manual it also shows the sensor on the left hand side.

Mine is firmly secured by the locking ring on the bottom bracket, but I noticed that in the User Manual they suggest glueing it to the frame with Superglue, but you would need to make sure that both surfaces were thoroughly cleaned before applying the glue.

Alan
 

[WestOz]

Alan,

The problem is that my bottom bracket won't secure the PAS on the LHS.  So I made myself busy today and tried the sensor in every configuration and magnet disc combination I could think of and I came to the conclusion that the sensor is magnetic polarity sensitive AND direction of the magnetic field sensitive.   Which the disc arrows and LHS mounting would support.  I could get PAS to work (with the bike upside down) if I had the magnet disc a certain way around AND I peddled backwards.  So working with that I reversed the sensor...thinking the disc could then turn the other way and I would be fine...nope...didn't seem to work.  Which I found very odd.  So then I tried the disc the other way around...nope didn't work.  Reversed the sensor...No...reversed the sensor and disc...no....tried every way I could think of including mounting the sensor on the other side of the metal support...nope.  Hours of effort and....

In the end I am back where I started from.  PAS "works" but it its REALLY jerky when peddling starts until it gets up to a certain speed and then it seems to work OK. This OK speed is different for lower PAS settings.  Setting 1 still does nothing (Watts remain on 0) but I have settings 2 to 5 apparently operational....with the PAS ratio set at 60 at present.  Full test on my 22km commute will be tomorrow.  60 maybe to harsh but at least I have a starting point.

Cheers,

[Bikemad]

It might be worth reducing the acceleration setting to see if it makes it less jerky. I have my acceleration set at 70 and have the PAS Ratio at either 40 or 50 (according to different screenshots).
The assistance obviously varies significantly dependant upon which gear you are in (and I don't mean Lycra or Spandex).
I tend to mainly use level 2 or 3 for my typical low speed dog walking rides, but level 3 is a bit too strong for wheelies so I drop to level 2 when I'm having fun.

If I select levels 4 or 5 in a low gear I can easily control the speed of the bike simply by pedalling faster or slower without applying any tension to the chain as the motor is doing 100% of the work, in low gears it works more like a pedal controlled throttle than pedal assist.

Alan

[WestOz]

Alan,

Per your advice, I set the acceleration to 70, but left PAS on 60 and gave it a real run this morning.  Sadly it was a no go.  On a real ride, as opposed to just testing in my short street, the true  response came out and I am disappointed to say I had the surge through the PAS selection range showing a surge from 0 to 500 watts on the BAC with a 5 second or so cycle time.  I could not get a constant power feed.

So it would appear that Left Hand mounting is the correct side, yet my locking cap for the bottom bracket, and every other square taper bottom bracket that I can find all have the locking lip cap on the crankset side.  As the locking caps are left hand thread one side and right hand thread the other I can't simply swap them over.  So it would seem that off drive side mounting is very difficult and I don't think super glue is a realistic option.

I need to be able to simulate LHS mounting on the RHS.  I tired to do this yesterday without success, so I will have another look.  I think rotating the sensor so its the same orientation as it would be on the LHS and then changing the disc to the other side, should work.  But I didn't get that yesterday, so maybe I missed something.  I am assuming that the disc magnets are showing a pole (either north or south) out of the disc surface (ie perpendicular to direction of motion) and they aren't mounted with the poles along the direction of rotation.  Which if they are would give understanding to the specified direction of rotation and why my trials yesterday didn't seem to work.  I'll test this with a compass.  If need be perhaps I can push the magnets out of the disc and rotate them to match LHS mounting.  This is getting really complicated.

Cheers,

[WestOz]

Hi Alan,

A bit more testing and my understanding is currently as follows;

With the magnetic disc correctly aligned with the arrows going in the direction of rotation, I have used a compass to determine that the North pole is on the Right side of the disc and the South pole on the Left.  What I have also concluded, though I am not certain of this as I cannot check on the sensor part, is that the magnet must pass over the sensor from the non-cable end for a North pole and from the cable end for a South pole my (my PAS sensor is the rectangular design from GM unlike your sensor style shown in the photo).  If this is correct, then provided the magnetic disc is aligned so that the arrows go in the direction of rotation then it doesn't matter which side of the bottom bracket the sensor is mounted on as it is always correctly aligned with the magnetic pole on the disc and the direction it must pass over the sensor.

Unfortunately my PAS still doesn't work correctly even with the above understanding.  Thus I either have a faulty sensor, or have misunderstood the sensor relationship with the magnetic pole or, and I hope this isn't the case, I have a faulty controller board. 

My only other thing to try is the sensor alignment with the disc.  Perhaps it is misaligned inside the rectangular package, so even though the external package is aligned the actual sensor is not.  I tried varying the distance of the magnetic disc from the sensor and from less than 1mm to over 4mm it made no difference to the PAS effect.  Given the sensor appears to be less sensitive to the magnetic field strength than I expected, it probably isn't an alignment issue.  Still I will give it a shot.

[Wadimka]

  Hi WestOz!

  I read your posts. Problem 1:1 as at me. The PAS sensor here not was once again convinced that and. It isn't its wrong installation. Your experiments convincingly proved it.
 Good luck to you in the solution of this problem! 

  p.s. And my patience in 2 months ended and I opened debate in PayPal on a refund yesterday.

[WestOz]

Hi Wadimka,

Yes it does seem we have the same problem.  Does that mean we are the only two using PAS, or the only two with problems?

I think I will ask the forum.

Good luck with Paypal.

Cheers,

WestOz