HD TPS vs MG TPS Replacements

The word on the HD TPS as it applies to Marelli 1.5M systems on newer Guzzi's:

On their 1.5M injected bikes, Guzzi uses a Magneti Marelli PF3C TPS ($212 retail)
MG part number is 01530500
On their 1995-2000 Electra Glide Models, Harley uses a Marelli PF4C TPS ($42 retail)
HD part number is 27271-95

   Both TPS' are identical in form, fit and electrical connections.  Both are 0-5v, and both are dual slope models.  That is, they are comprised of 2 linear curves that have a transition point somewhere in between. The transition point occurs at 3.338 volts, which is 30.095 degrees of throttle rotation as reported by the Motorbike Diagnostic Software Tool for a 1.5M ECU system. Go to:

to see a graph depicting voltage vs. degrees open.  No matter what TPS is in there, the software tool always shows the transition at this point, because it is an electrical/virtual transition.  The actual physical transition where the curves change slope vs. throttle grip position is slightly different for each model (within a few degrees).

   So, what is the difference you ask?  Well
MPH Cycles graciously sent me a Harley PF4C TPS to find out.  Here is what I did:

1. Fashioned a pointer and marked the throttle grip in 10 different locations
2. Noted the Guzzi TPS voltage output at each of these locations.  MY 2002 Stone idles at a TPS voltage of 541mv, which is 3.7 degrees of open throttle as reported by the ECU using the Motorbike Diagnostic Software Tool.  Spec is 3.4 +/- .2 degrees.  At 3.7 I am a little above spec, but that's ok.
3. Next, the Harley TPS was installed so that is read 541mv at the same throttle body linkage settings.  That is, I slapped in the Harley unit and set it at 541mv without touching anything else.
4. Voltage readings were taken off at each of the marked throttle grip settings. Note, all of this was done with the bike off!  There is no need to have the bike running.

   So what were the results?  Well, below the transition point, the Harley TPS tended to lead the MG TPS.  That is, it gave a higher voltage (max difference was 50mv) than the Guzzi TPS.  After the transition point, the Harley TPS tended to give a lower voltage (max dif of 38mv) than the Guzzi TPS.  The voltage difference between the 2 TPS units was not always constant.  That is because the slopes of the curves are different between the 2 units.  That is, the HD unit started at the same voltage as the Guzzi unit, but was reporting a higher voltage by the time it reached the transition point - thus if one plotted physical rotation of the TPS vs. voltage out, they would see that the slope of the HD curve would be higher.  That is, below the transition point, the HD TPS picked up "voltage per degree moved" faster than the Guzzi TPS. So, that means if you go cruising your bike down the road at the exact same throttle setting, a Harley TPS will be telling the ECU the throttle is open more than the Guzzi TPS, which results in a richer mixture.  Above the transition point the HD TPS reported a lower voltage for a given throttle opening.  This would mean your Guzzi would think it is running with less throttle opening when using the HD TPS at spots above the transition point, thus leaning out the mixture.  In a previous investigation, I found that cruising speeds usually fall below the transition point.  You are usually above the point if you are accelerating or cruising at an elevated RPM.
  See a graph of this

   Now, the next question, "how does this affect my bike?"  Well, considering the 1.5M ECU is an 8-bit digital system, it takes that 0-5 volt signal from your TPS and chops it up into 256 evenly spaced steps. Each step is about 19.5mv.  So, at each 19.5mv step, the ECU grabs a value off of the fuel table.  Now, with the HD TPS, below the transition point, you will be grabbing a fuel value 1 or 2 steps ahead on the fuel table from where you should be grabbing.  Above the transition, you will be grabbing a fuel value 1 to 2 steps back on the fuel table from where you should be grabbing. How does this affect drive ability?  Well, the folks out there roaming the country with the HD TPS claim no perceivable difference.  The question is, "is the HD TPS right for me?". That depends.  Are you one of those guys that obsess about fuel mileage to the point you have an 8 gallon tank, a bicycle speedo, and a GPS mounted so you can get speed and mileage down to the gnat's ass?  Well, you'll be the type to spring the big bucks on the MG unit.  Also, if you have had your ECU laboriously mapped by someone, you would probably buy the Guzzi TPS also.  As for all you stockers out there, you'll probably be buying the cheap HD unit.

   I have found by marking the grip at 10, 20, 30, 40%, that most of your cruising takes place at 20%-30% throttle (40mph 2nd gear, 60mph, 3rd gear).  By looking at the voltages, you can see that the HD TPS would richen your mixtures somewhere near your 60% column and lean your mixtures above that point.

Armed with actual numbers now, I make the following observations:

-The MG TPS must be rotated 2 degrees from its closed position to give you 150mv and a 0 degree reading on the Motorbike Diagnostic Software Tool (MDST). The output of the MDST is always 2 degrees less than the actual physical rotation of the TPS unit.

-The HD TPS must be rotated 6 degrees from its closed position to give you 150mv and a 0 degree reading on the MDST.

-The slopes of the two TPS units ARE different. Also, the HD unit does some funky things on the top end not previously
realized. Also, both units do some funky stuff on the bottom end that was not previously realized.

-If you set up each TPS with the throttle screws backed off and the TPS reading 150mv, for a given set throttle grip position, the MDST will actually think the throttle plates are open LESS on the HD unit (vs. a MG unit) until you get to 10% of throttle, then the MDST thinks the plates are open more with the HD unit than with the MG unit for a given fixed throttle grip position. Once you get above 10% throttle or so, the HD equipped machineís ECU will think the throttle plates are open more than they actually are, thus giving you MORE fuel. Below 6.7% of span (335mv on a 5v system) on the HD TPS, they really do not specify the slope of the curve.  I interpolated it below that point, but am probably close at best.  Since they really do not specify anything below that point, I would believe that they do not intend it to be used at settings below that point, thus setting it at 150mv with closed throttle is a hit or miss proposition. Those that have been seeting up the HD TPS at 150mv with backed off screws and throttle plates have probably been getting lucky on their set-ups.  It would be more appropriate to set the TPS at it's proper position (500-525mv) at the correct idle speed (1100 rpm) when using the HD TPS with a 1.5M ECU.  Still, the bike will run slightly richer with the HD TPS. Affects on ignition timing when using the HD TPS are unknown at this time. For those interested, on the Guzzi TPS, they specify a slope all the way down to 3% of span which is... 150mv for a 5v system. See where this is headed?

-Now, the weird shit happens on the top end. Since the HD unit is rotated more (6 degrees vs 2 degrees), you actually reach a point in the top end where the HD TPS says thatís it, we're done, while the MG says there is still room to open the plates more. This happens a shade over 80 degrees of throttle opening as seen on the MDST. So, once you reach 80 or so, the ECU will put out the same fuel for 85 degrees as it will for 80 as the TPS output flatlines. Meanwhile, the MG equipped machine actually sees a TPS difference between the 85 and 80 degrees. I told you something funny happened up top. Now the question is, how does this effect wide open throttle performance? Well, once you get above 80.5 degrees on the MDST, it pope up a little flag saying you are at wide open throttle. Now, what happens to the fuel map above this point? Sorry, I donít know the answer there.

Anyway, there you have it. Also, for those folks that think they have a bad TPS, hook up a multimeter or the MDST and very, very, very slowly roll on the throttle and look for a sharp dip or jump in the TPS output.. If you get a sharp dip or rise, tour TPS is bad at that point. If your TPS opens (over volt) or shorts (0 volts) , it will store that as an error on the 1.5M computers that can be retrieved via the MDST. Note, most of your cruising is done in the 1.4 to 4.0 volt range (The slope on the TPS changes at 3.333 volts - I have found that most of my cruising is near this transition point....), so pay particular attention to values there if your bike does funny things at cruise at you suspect the TPS.

Last note: Those soft T17 TPS torx bolts should be replaced with M4x12 socket head cap screws.

Jeffery H. Brannen


MPH Cycles.com