The Bagger Project

[Radioguylogs]

Appreciate the input

TC: Thanks for taking the time.

Your advice and (Brant) is well taken for ACCT. Although I had read the spring that pushes most of the ratchet designs is stiffer than the OE spring, I can't say that would worry me to the point of rejecting the ACCT idea. I'm listening! Maybe I'll become a convert at the next valve shim.

Regarding the tach circuit, I think the fabled diode is in the headlight relay circuit, and the tach receives a straight shot from one of the stator windings- hence the negative and positive voltage peaks.

I think the regulator reaches maximum output voltage when RPM reaches 2500 RPM (yet can provide more current about 2500 RPM), but I think the signal to the tach is pre-regulator, so it did seem curious that the voltage seemed to saturate above 4k RPM. The frequency must increase, however.

Anyhow, I had to rationalize why I bought an oscilloscope for at a garage sale 5 years ago, so I finally did something with it!

Keep your kind advice coming.

[Radioguylogs]

Hmmmmmm

Hey Mike,

Not sure how much trouble it would be to swap tachs among your 2 bikes to compare? Also, another thing could be clutch slippage?? And same rear tire size/brand??

T.C.

I thought about swapping tacks to confirm if the twitching problem is inside the SG tach.

It didn't occur to me to swap the tachs to check the speed vs. RPM.

Clutch slippage is a very fascinating thought....I cleaned/sanded the metal clutch plates. I bought new OE friction plates. I put in new OE springs. I adjusted everything. Yet this is one idea that might explain the high RPM.

Tires are same size, albeit different brands. Can't believe that could cause 20% difference.

[Radioguylogs]

Staying busy

Replacement tach from Andreas is on the way. Maybe Tuesday?

So, I took apart the gas cap tonight and soaked the parts in Evaporust, and re-assembled it with a new rubber gasket from Mikes XS. Looks much better now. Made sure the vent is breathing well.

Now I'm going to start working on the cracked bags and the cracked cover.

[TopCatGr58]

Regarding the tach circuit, I think the fabled diode is in the headlight relay circuit, and the tach receives a straight shot from one of the stator windings- hence the negative and positive voltage peaks.

Hey Mike,

Yep, you're right, sorry about that. I remember NOW when I first put my mini race ALT unit on my bike, I tapped into 1 of the legs/phases of the windings and ran it directly to the OEM tach, and it worked just like the OEM alt did. I had calculated and set the pulley size on the crank to the same size as the pulley on the Mini Race Alt so that the RPM's would be essentially the same, and since both the OEM and the Mini Race Alt had the same # of poles/phases it provided the same number and frequency of pulses to allow me to still use the OEM tach. I have since put on an aftermarket gauge set, and run the electronic tach off of the ignition coil pulses now!!

I will tell you that when I did my rebuild in 2000 I also put in new frictions and OE(Vesrah) springs, worked okay for a few years, but then started slipping again. That's when I put in the extra Steel plate even though the springs were still within spec (before "WE" had learned about the more robust Barnett Springs), and that stopped the slippage.

With this NEW rebuild I'm doing, I'll finally be putting in Barnett Springs. Not sure if I'll take OUT the extra steel. There was a discussion a while back about how the pressure plates and the shafts "MAY" have stretched a little bit over the years, and so the amount of compression that is produced may be reduced since, and so there may still be a chance for slippage if I just use the OEM amount of frictions/steels even with the Barnetts. I get stressed every time I have to take that Star Plate off and put it back on. I broke one the first time I worked on it decades ago, and so I'm always a bit leary when I work on it, even though I now know what causes them to break, and how to prevent that!! So instead of having to take the whole clutch basket out to remove the steel, it'll be easier for me to just put the Barnett Springs in and button it back up.

But getting back to your bike, as the clutch wears, the throw out rod moves closer to the throwout bearing, and that can eventually lead to not allowing the clutch to fully release when you let the lever go, and that can also cause a little slippage. You may want to recheck the adjustment at the ENGINE lever to ensure that it's properly releasing completely.

T.C.

[Radioguylogs]

Hey Mike,

You may want to recheck the adjustment at the ENGINE lever to ensure that it's properly releasing completely.

T.C.

Good suggestion- I was thinking the same thing after you mentioned the clutch, but I forgot to do it when I was putzing this evening. Now you reminded me.

[madmax-im]

Oops, need to have this picture, the magic moment when it starts and runs:

What a sweet ride...hope to see it and you at XSSE...

[Radioguylogs]

Hi Ben, the Bagger I and look forward to seeing you in 5 weeks.

Tach

Good news! The tach arrived from Andreas yesterday and it fixed the twitchy tachometer problem. I only had a chance to take short test ride last night, but it did confirm the RPM at 70MPH is more reasonable than the 'old' tach was showing....but it still seemed to be around 5k RPM. Seems a little higher than the SF (maybe 4700 RPM). I'll hope to get a chance to ride it more tomorrow.

Cracks

I am happy to provide a glowing endorsement for the Plasti-mend product. I used it to fix the cracks in my bags, and the side cover repair is in progress. This stuff is SOOO much better than other repair techniques, that we should have a report on it in the Maintenance\Finish Forum.

There is almost no work to use the product. You rub the area with a Scotch Brite and the provided solvent, then you paint the product on, pressing a screen in the first layer. The screen is same fiberglass screen used in windows. The product is like black honey. You just paint it on and throw away the disposable brush (or cover it in plastic for reuse). After 3 or 4 coats, you're done. It's that easy. I did do a little sanding for a few seconds to take off the bumps between coats.

I think the Plasti-Mend product contains methylene chloride (paint stripper), maybe MEK, acetone, mixed with ABS plastic. It dissolves the base ABS plastic in the host, and when the solvents evaporate, the repair becomes homogenous with the host plastic. It is the same material. It has the same expansion/contraction coefficient. It has the benefit of the screen material to add strength.

The only drawback to this stuff is that it would be difficult to get a good appearance on an exterior decorative surface, because the stuff doesn't like to be smoothed as you brush it on. You want to brush it once or twice quickly and move on. It doesn't leave a big pregnant area, but the repaired area has some unevenness in the shape.

I haven't tested the strength, but it seems like my repairs are good-as-new. I wanted to provide pictures, but you can't see the repairs in a picture, just a little unevenness in the black surface.


My to-do list on this project is getting very small!

[3Phase]

Nice!

I put an oscilloscope on the white wire inside the tach can...with the tach still connected.



Not the sine wave we would all expect. The voltage does increase with RPM, and the frequency increases with RPM. However, it does seem to saturate above 4k RPM - seems like less voltage increase with further increase in RPM.

I wiggled with tach and the wiring. Nothing really seems to reliably cause the twitching. Meanwhile it still happened above 4k.

Put it all back together and went for a ride.

Seems like next step is to try another tach.

Nice work with the 'scope and for finding a Tachometer that doesn't twitch!


In your picture, remember that you're only measuring the one phase coil pair that's connected to that White wire and there is a time gap while the other two pairs are working.

Without reaching through the screen and tweaking the Time and Voltage divisions on your 'scope to check if that's what's really going on, here's a relatively brief 4:15 video that shows what you're probably seeing:-


matchudcu: Showing unrectified alternator output on a picoscope




It's not anything mysterious, the XS1100 and XJ1100 charging system is almost exactly like an old Chrysler system with its two-wire voltage regulator on the firewall but with the alternator voltage rectifier diode bridge brought out externally. The more powerful XJ1100 alternator even uses brushes for its field coil and, yes, an old Chrysler/Dodge/Plymouth mechanical or solid-state voltage regulator will work just fine for either bike but there's no voltage rectifier, it just regulates voltage so you'd have to build your own bridge.

.

[Radioguylogs]

You got me going, Scott

Interesting discussion (to me anyway).

Of course Three Phase has point out the other two windings are missing.

I did realize the other two windings would fill in the dead space between pulses, but what surprised me was that the short duration of the pulses.

Luckily, my picture does capture the voltage and time scale setting. I don't (yet) see how they would be readjusted to tell the story -> 8 VAC spikes at 200Hz.

So, what was curious to me was that the small area between the curve and the zero axis. If you integrated the area between zero and the curves, both positive and negative, and you multiplied it by 3 ......and divided by the time period, you get an average voltage that is rather small.

So, I guess that's why the regulator doesn't output doesn't break even until 2,500 RPM!

[Radioguylogs]

Correction

Actually it shows 8VAC @ ~800Hz....forgot to multiply by the number of divisions..

[3Phase]

You've pretty much started one of the projects I've been planning to check and document the different systems and electrical components on the XS11 like the alternator, regulator/rectifier, tachometer, pick-up coils, the TCI and the primary and secondary ignition circuits.
Again -- nice work!

The waveform you captured does look a little narrow but my eyes are mostly toast nowadays so I had to download the larger size image you posted. With a little Zoom in the image viewer I can see the white stripe on the knob for the Volts setting looks like 5V/Div but the knobs on your 'scope have more than one 'grip' sticking out of them so I can't see if the stripe on the Time switch is set for 2sec/Div or if there's a parallax error and it is not really set somewhere in between 2ms/Div and 5ms/Div.

The picoscope in the Youtube video I linked shows the individual alternator phase waveforms are around 8Vpeak to 9Vpeak. I assume that your picture was taken with the engine idling and the old tachometer and regulator/rectifier both connected and doing their job so you should see roughly the same peak voltages as shown in the video because you can only get the ~14Vrms you were looking for by measuring across two ~7Vrms phases, not one 7V phase to ground.

Some quick math:
8Vpeak / 1.414 = 5.7Vrms
5.7Vrms * 2 phases = 11.4Vrms and is discharging the battery

9Vpeak / 1.414 = 6.4Vrms
6.4Vrms *2 phases = 12.8Vrms and is barely maintaining the status quo on a conventional battery but losing ground for an AGM battery

10Vpeak / 1.414 = 7.07Vrms
7.07Vrms * 2 phase = 14.14Vrms and is charging the battery

To get a useful RMS 12V to 14V charging voltage from a single phase to ground the alternator would have to generate something like 18Vpeak to 20Vpeak and that would give you 12.7Vrms to 14.14Vrms. You'd have to use more and larger diodes in the rectifier because running the positive phase and negative phase through a single diode pair in the rectifier bridge would give you a little over 28V and that would let all of the magic purple smoke out of the electrical system.



Anyway!

Did you try checking the waveform voltage from White wire to White wire = phase to phase?

Does the waveform look different now that you have installed the new tach'?

Is the single phase waveform to ground the same on the other two White wires?

What does the waveform on the Red wire at the Rectifier output look like?

What does the waveform on the Brown wire at the Regulator Vsense input look like?

I had to go out of town for a week but now that I'm back I'm in the middle of researching and ordering a couple of pocket oscilloscopes for my projects. For about $250 I can get 6 channels to work with: 4 analog, 2 digital and built-in function generators. The digital channels on the Quad Nano aren't really useful to me but it's a better 'scope so I might just buy two of them to get 8 channels and skip the DSO 202 Note II.



SainSmart DSO 202 NOTE II




SainSmart DSO 203 NANO V2 Quad MINI



.

[Radioguylogs]

More Thinking about Charging

Scott:

I was pondering your thorough evaluation of the operation.

At first, I couldn't intuitive understand why I should think about the peak-to-peak voltages, because the waveforms of the three winding would surely be staggered in time so that no positive pulse would line-up with the negative pulse from another winding. I think I might be missing something.

Then, I later started to ponder if the RMS calculation is valid for waveform that is not a sine wave. think the "divide by 1.414" might be for a sine wave?

Still, your thinking has me understanding more than I did before.

BTW, scope was at 5msec/div.

(1) Did you try checking the waveform voltage from White wire to White wire = phase to phase? Nope, not yet

(2) Does the waveform look different now that you have installed the new tach'? Didn't try it, but I imagine it looks identical, because I imagine the input impedance of the Tach is negligible, but I don't know.

(3) Is the single phase waveform to ground the same on the other two White wires? Didn't try it (yet).

What does the waveform on the Red wire at the Rectifier output look like? Didn't try it (yet).

What does the waveform on the Brown wire at the Regulator Vsense input look like? Didn't try it (yet).

Sorry, I dragged the scope back to the basement when I got the tach to work. This week I'm going to concentrate on getting the bags done, so I can get some shake down miles on it before XSSE.

So, it will be interesting to understand how this charging system works. First I gotta take care of some other priorities. Chime in with your thoughts and findings!

[3Phase]

Leave the 'scope in the basement, stick to your plan by finishing your bike and don't mess with the charging system! You just got a tachometer that doesn't twitch so don't mess it up!

Yes, the alternator puts out sine waves, three sine waves but since they're all converted to DC ~ AC RMS you don't have to worry about any fancy polyphase polynomials. Unless you bang your head on the frame late one night after removing and replacing the saddlebags for the seven-ump-teenth rear tire change/final drive spline input lube opportunity and suddenly decide that you want to make your own alternator from scratch and that it will also be used to run a tachometer: let it be.


I have no idea how the voltage regulator actually works down at the component level -- yet! It's not being used to run an induction motor or any sensitive piece of AC equipment that requires a sine wave so the odd shape of the waveform you captured could be a low-cost voltage regulator/rectifier design changing the shape of the wave as it does its job more like a PWM duty cycle and a faux square wave; could be a component or several somewhere else in the circuit affecting the waveform; oscilloscope or the probe could be out of kilter, or something could just be baroque. En.


The tachometer on Columbo, my '80G, started to jump and bounce after the engine spit a spark plug up in Utah and had to be carried home in a U-Haul. So far all of the individual parts test good on the bench and by Mark I eyeball so after I get the engine rebuilt and the wiring harness put back together and running I have some troubleshooting of my own to do. My time has a lot more value to me now after my heart attack so it's time to get some new tools instead of trying to do things the hard way.

All together now: "Just one more thing!"

.

[CaptonZap]

Ya gotta love those DSO Nano's.

CZ

[3Phase]

Ya gotta love those DSO Nano's.

CZ

You know it!

Mike, sorry about temporarily hijacking your thread but I've paid more for dinner with no movie afterward and the Sainsmart site just put all three DSOs on sale. I went whole hog and got the 201, 202, and 203 for ~$300 with free express shipping. 9 ports, crikey!


.