Modern Vespa

Hi all,

I've been trying to get the fuel gauge to work on my P2, to no avail.

First, some specifics: the speedo is a PT Danmotor unit from an Exclusive 2, to which I have no wiring diagram (if anyone does have a diagram to share, I'd love to see it by the way). The bike is a spanish 200ds (P2 with no turn signals), and the electrical circuit was made by myself (everything else works, except the oil level warning system, but that's less urgent ).

The speedo was a brand new aftermarket unit when installed, so I'm assuming it was functional. It worked once with a battery (showing full, half full and empty) when tested while building the bike, but once built, and ever since, no more gauge!

The sender unit is new and has been tested, and I've checked the continuity in all the relevant wires (ground, gauge, and reserve light) all the way to the speedo. (the one in the picture is the old one, shown just to point to the wiring).

The issue seems to be within the speedo itself, but my limited self-taught understanding of electrics has me stuck, as to what or how to test to check if it works, and if not, how to break it down to the faulty component.

As on most other Vespa speedos, the gauge unit has 3 screws which attach to the circuit. So ground, 12V, and wire to the sending unit. There is a 4th screw with no wire attached.

I tried connecting a 12v battery to the (supposed) ground and 12V plots, which should make the needle go to full if the sender isn't connected (I tried with and without the sender, and the needle didn't budge). I'm identifying the plots solely based on their resemblance with other Vespa gauges, and an online translation of the chinese characters on the housing, so not 100% sure.

While testing continuity, I tried testing from gauge plot to gauge plot, just in case it might be relevant somehow:

between ground plot and 12v plot: 215
between ground plot and "to sender" plot: 128
between 12v plot and "to sender" plot: 86

Not sure what that means, if anything.

So did the fuel gauge come from the same scooter at the fuel sending unit? I dunno if the ohm resistance for every single vespa (or vespa clone) fuel sender and gauge are the same, but the resistance between the two should be compatible.

No, the sender is for Piaggio px. But nonetheless, putting 12v through it without the sender should indicate a full tank, whatever its resistance range (from what I've read).

I removed the little gauge unit from the speedo, and tried attaching 12v and ground to the plots using various combinations to see if/how any setup would move the needle to full. The original gauge moved under no circumstances, and when I tried the same with a gauge from a cosa, I briefly got a working system (needle moving with the float), but with wires attached to the wrong plots. I tried again later, and now the thing is unresponsive vis a vis the float. Don't know wtf is going on.

I went to buy a new resistor for the original gauge, and soldered it in place; but the multimeter gives me the same readings. I'll go test it later on, fingers crossed...

When we know the sender and gauge are good at my work the problem we experience is bad grounding between frame, battery, gauge and float. Are all your grounds good, especially to the frame? 🤷🏼

Is there any corrosion on the sending unit wires? If so, you can clean it up a bit with some 500 grit sandpaper.

Any way to check resistance of the guage with the soldered resistor taken off?

So I tested the original gauge with new resistor:

No good. The needle is all over the place.

I tried the cosa gauge again, which had briefly given me good results the other day then given up:

Of course, this only shows that it can work, not that it necessarily will.

To answer your questions: the sender unit is brand new, and has been tested with consistent results.

The grounds in the frame have been checked for continuity, as have been the other wires. I still need to doublecheck under the column, where wires of different colors meet up (but that has no effet on the gauge acting up ex situ).
With regards to the test here:
-battery -, sender and gauge are connected to the frame
-battery + goes to the gauge
-sender goes to the gauge.
I checked all wires for continuity to avoid false negatives, including the grounds.

Would checking the resistance of the gauge give some indication of what's wrong? I already soldered the new resistance, but at this point I can always remove it to do more measurements if useful.

Atm the plan is to adapt the cosa gauge to the speedo housing, reassemble the speedo, and see if it works again in the frame.

I posted to check resistance without the resistor to confirm the gauge is actually still good. Its strange that after both setups they died. Checking to see if the the gauge isn't open after the test confirms its not fried.

FYI my knowledge of electronics is comparable to yours (as you have described lol). I know on my stella, the fuel gauge uses regulated AC to function. I'm curious if that matters in this case. There is some explanation and test procedure in the stella manual. I have attached the relevant section.


If I understand right, the videos were made after the first post? Meaning the gauges actually do seem to intermittently function. It seems like the danmotors one wants to work. Could it be dirty? Can you try cleaning it with contact cleaner?

between ground plot and 12v plot: 215
between ground plot and "to sender" plot: 128
between 12v plot and "to sender" plot: 86

So I think these gauges are "balanced coil fuel gauges". Based on that, see the attached wiring diagram.

The resistances therefore look good. Between 12 V and Sender you have one coil, between Sender and Ground you have a different coil. Between 12V and Ground you have the the two coils in series. This matches your resistances. So I think the gauge itself is good.

Try and see if there is a diode somewhere in there by swapping leads for each measurement. If there is a difference in measurement, then power and ground attachments are possibly important. If the fuel sender is confirmed good, then the gauge is dirty or there is and issue with resistance matching possibly? Or dirty contacts in your setup?

If you can't tell im spitballing now, word vomiting whatever comes to mind. Hope it helps

At this point, there is no doubt I have a family of gremlins hiding in my garage, and messing with my tools and anything electric when I'm not around. My tools, I always find after a while (usually when I no longer need them); but the electronic issues tend to turn into long-lasting mysteries.
As mentioned, I've been doing the same tests over and over, with varying results despite similar testing techniques.

Thanks for the documents; I'll check the Danmotor gauge again with that test, see if it can help me make some sense of it.

If not, hopefully the cosa gauge will work out.

I finished installing the cosa gauge in the speedo with the Danmotor faceplate; turns out sealing compound is pretty good for gluing stuff in vibration-prone areas when screws no longer fit. The speedo plate didn't quite accommodate the gauge needle at first, but that was resolved with some filing.

I pondered changing the speedo needle for an orange one from the cosa to match the gauge, but finally decided against potentially giving myself something else to repair.

Aiming to fully troubleshoot before installing the handlebar cover, I once again checked for continuity through the whole line from sender to connector, and through the speedo itself.

The sender to gauge wiring proved correct; however, when I tried simply lighting up the speedo idiot lights by connecting them to battery via hot wire and ground, no dice. Not one lit up. (Neither fuel reserve, neutral, hi beam, oil reserve, nor general speedo lighting). Getting anxious to see if anything would work on the scoot's own power, I plugged the speedo connector to the headset, and kickstarted the bike.

The gauge went to full, although the tank is half empty, and still no idiot lights (neutral should be on by default) Looks like the next things to investigate are the ground wires in the speedo loom itself.

Well, I guess this whole operation results in a half-victory: the gauge moves, but acts as a reverse tacho instead of a fuel gauge Start the engine (low revs), the needle goes to full; twist the throttle, it goes to empty. Looks like the fluctuations of the stator output affect the gauge, but why??

I went out to get some gas, and on the way home, the needle seemed to stay on full. If it keeps moving with the revs but goes farther towards empty as fuel runs out, that would at least be workable. I checked that the reserve lights turns on when the buoy is pushed down, and it does. That's the one thing I really need to work, whatever the gauge says...

Is there a chance the gauge is DC??

It's always possible, but I have a hard time figuring it out from the wiring diagram. I doubt there were cosas without battery, so I assume this diagram is correct for my gauge, but I can't guarantee its provenance.

Is the cosa sender unique? It has three connections. Or is one leg for the reserve light?

Why on your original picture does the sender post have two wires connected to it?

So apparently the stella has a diode in the fuel gauge, which would make it half wave rectified dc. That would lead me to believe that these things need DC to function. Your original pictures say AC is going in as the 12V source. Can you give it the 12V DC source instead and see what happens?

The sender seems to work on a similar basis as the PX: one wire to ground, one to reserve light, one to gauge.

I'm not sure which original pic you're referring to, but the tests were done usually one function at a time, so usually just the ground plus the gauge or reserve light.


What differences are there between DC and AC units? I'll have to check one of my T5 speedos; they were on bikes with batteries and I've used them on AC bikes as well. Come to think of it, they would turn on when you turned the key, so they worked with DC as well....

The picture in the very first post says "12V AC". The Stella uses half wave rectified DC in the actual coils, and the other wiring diagram (one that says Figure 1) uses pure DC since it is connected to a battery.

Since yours is operating as a tach, it would seem like its getting some kind of unregulated voltage to it. So either a bad ground like you said or its connected to the wrong electricity type.

The cosa diagram stinks. Unfortunately its a black hole when it comes to the instrument cluster.

Oh, ok. The captions on the first pic were my understanding of where the wires go to, and I added AC because it comes from the regulator. So if my gauge is disturbed by the variations in current coming from the regulator, what's the fix? Installing a capacitor before the gauge?

I checked the wiring diagram for a PX200 without battery, to see how a gauge is fed on an AC loom. Looks like the gauge is supplied with 12V AC.

So the P200 diagram shows after the regulator it goes to a rectifier. Is it actually being rectified (turned to DC)? Thats not super clear.

Attached is the off bike test circuit for the fuel gauge and sender I would use to see if it needs DC

PX speedos use a rectifier diode for the fuel gauge, that's what keeps the AC feed going one way. If I recall it's a 1N4007…

HA I think we are on to something! To answer Frank N Stein's question, at minimum you need a diode to rectify the AC into a form of DC

Thanks for the pic, Safis. I checked the reference you gave, and it certainly looks like it. If I get this right, the diode simply needs to be on the 12V line. I'll go get a few tomorrow, and find a way to bolt the diode perhaps inside the speedo housing to keep it safe.

@ Luigi: the 12v wire only connects to the rectifier and the blinker relay for practical reasons; the current isn't modified by being connected there. It would have to go through it (like a 3 pole regulator, say) to be affected.

Your test diagram is basically the same as my test setup, except that instead of the grounds bring connected to the frame, they're interconnected. What result am I looking for? Needle on full?

Otherwise, I went on a 150km trip to the countryside today, and the gauge pretty much acted as yesterday: riding in low 4th gear, needle on full; riding in high 4th: needle on empty to half.

Yes, you can tell I was hesitant to trust the diagram because something seemed fishy. Safis clarified by showing us the diode. That makes clear the AC from the rectifier is later rectified by the diode.

In theory, using the schematic I provided the needle should show correct sweeping direction with actuation of the sender. It may not be perfect because the resistance matching issue is still not settled.

Ok cool👍. I'll go get some diodes tomorrow, and do the test while the headset cover is off.

Thanks all for the input guys!

Damn, that bike looks great.

Thanks, I'll tell him!

I got a pack of 1N4007s today (1€ for 5 units, not bad!), and I've been reading up on its properties. There's a direction for these (anode for incoming, cathode for outgoing), so that means I'll have to place the stripe closest to the gauge.

One question I have, is why 1N4007 instead of 1N4001? The former is good for up to 1000V; the latter up to 100V. Since the current will be rectified to 12V (coming from the regulator), wouldn't the lowest value be more appropriate? Probably a dumb question; I'm reading this in hopes to better understand how this stuff works...

Probably for margin of error. Or if the regulator craps itself, to cover the speedo? When my regulator died, I saw at least 50V AC then shut the bike off and that was just a quick rev.

Back to testing the gauge. It occurred to me that I might have messed up the resistor on the original gauge by putting it in backwards. Upon checking, resistors, unlike diodes, are bidirectional, so the issue is elsewhere

On the topic of resistors, here's how to read those with colored stripes, for future reference: (grouped side first, or silver/gold side last)

If you have the old one still, double check you read the resistor right using your multimeter. It wont be spot on but within some tolerance, probably 5%

Right. I'll retrieve the old one and check it against the new one.

Speaking of which, I checked out the cosa gauge, and compared it with the Danmotor. The cosa has a little black diode that in fact IS a 1N4007, which the original one didn't have, confirming what you guys were saying about gauges made for DC (or not).

I checked that the current went the right way through the cosa gauge diode and not the other, and it did! (Reading of 560 for what it's worth). Not sure what to do with the cosa gauge, since it doesn't work right despite being functional. Would a bad resistor explain the weird needle movements?

I checked the same for the new diodes bought (for fun), and the continuity readings were consistent with the polarities.

On the other hand, the Danmotor gauge lacking a diode offers some hope: I'll try soldering a diode in line with the resistor, and see if that works.

I also tried the prescribed tests on a different gauge on a T5 speedo; the needle didn't budge in any circumstance. This one I'll have to open at a later date to check the components: it might simply be fried.

Interesting you have 3 sets of bad fuel gauges. Are you 100% sure that sender is good? That's the common denominator here. Also good battery voltage?

You can do your resistance test on this new guage too. You should get three resistances with the the two smallest summing to the largest. The diode means the resistance test will only work with the correct lead orientation with your multimeter (as you have learned the diode is unidirectional).

You need 12V AC, a battery won't do…

Do you mean, for the test on the T5 gauge?

Luigi: 3 bad gauges: yes
But the test on the T5 gauge was done independently, not on the bike. The sender is new and has been checked extensively: it should be good.

So I checked the resistor on the cosa gauge, and it should be 178 ohm, but the multimeter says 87-90. I have a new one which is 180 ohm (the guy said 178 wasn't available any more, and that would only slightly affect the reading). So the plan is to solder that one in place of the defective one.

Checking the diode now, numbers flash, a battery sign appears on the screen, and it goes to 1. Not sure if that's good news and there's no resistance, or not. Reversing the polarities, nothing at all: it stays on 1. On continuity, it says 526.

Following the connections on the circuit board, power comes through the diode, connects to the windings, then goes through the resistor, then leaves for the float unit. The other side of the winding goes to ground.

Regarding the original gauge, I checked the original resistor again using the colored lines and find 5.4 ohm (which thankfully happens to also be the measured resistance). The new resistor installed says the same, although it tends to go all over the place before settling on 5.3-5.4 ohm.


So basically I'll be adding a diode to the old gauge, and changing the resistor on the cosa gauge; hopefully between the 2 I'll have one that works!! And I'll have learned a lot along the way

Old gauge: diode soldered to the resistor; resistor value and continuity of the pair checked. On paper, this one should work.

Cosa gauge: resistor soldered; resistance checked @ 88 ohm. Old resistor checked @ 173 (within specs).

It seems like the values are altered when the resistor is soldered to the circuit (I thought measuring between the poles would prevent that).