If you are new to this blog, I suggest starting above with Project 1975 or Cafe Concept (on mobile devices, click the 'home' bar above for other pages).
Blog posts are in reverse order; to start from oldest to newest page, select May 2016 from the blog archive on the right. Or skip right to the completed project under 'Final Reveal' above.
Tuesday, February 28, 2017
Charging Troubleshoot
Feb 2017
Two miles from the end of a 72 mile ride, I briefly stalled the 850T at a light and could not get it restarted. The battery appeared flat as the starter just clicked and would not turn over. Quickly assessing traffic and noting a very slight downhill, I snicked it into second gear and duck-walked just enough to bump start it to life. Very relieved, those last two miles had me giving thanks that I had not stalled it earlier in the ride, and/or somewhere on a flat grade.
A check of the battery at home proved that there was a fault in the charging system; it was only 10.5V, having left before the ride with at least 12.5V. It had been charging properly at last check about 6 months earlier. Consulting several troubleshooting guides, I got to work to trace the usual suspects and identify the problem. The charging system is essentially made up of four parts. The rotor and stator take rotational energy from the crankshaft and turn it into AC power, upwards of 100V per phase depending on the type. The regulator and rectifier (r/r) cut off the AC peaks and restrict the DC output to a range that the battery can handle (roughly 12.5V to 14.5V).
First check, what's coming out of the r/r? At first I got a few erratic swings, one of which was up to 18V*. Later trying again the expected voltage of 14.5V was not reaching the battery when the engine was revved up to 4000 rpm (staying somewhat flat at 12.5V). But the solid state r/r was then ruled out when resistance and diode measurements using a multimeter figures checked out as expected. Then, resistance of the rotor windings measured between the copper slip rings also checked out, as did grounding of the rotor. And then, resistance between the phases and grounding of the stator was measured. This too checked out. A final test involved running the bike again and revving from idle while measuring AC voltage output from each of the yellow phase wires. Something was clearly up when only 3V was registering in each of the phases. These should be showing at least 50V AC each at higher rpm.
So this pointed to a bad rotor or a bad stator; stators are not known for breaking, but rotors are (averaging around 40K miles or so). But, both of these independently produced normal readings in testing.
This puzzled me for some time, prompting follow up with Gregory Bender who provided the wiring harness. His prime suspect was the rotor, but also ensured that I check r/r grounding. While re-doing several of the charging system diagnostic checks, I realized that one thing was overlooked. The rotor resistance had been taken directly via the copper slip rings, but the leads to the brushes which contact these rings were not routine items on the checklist(s). Sure enough, one was measuring contact and the other nothing. As these are hard to inspect while mounted, I pulled the alternator off the rotor. The second photo below shows the significance of the contact problem. The forward brush was worn to a nub; there was no more pressure that the spring could provide to push it down (inside the white plastic housing) enough to contact the rotor. As the charging system had worked many rides earlier, this slowly got to a point where contact was lost. Again, thankfully it came to light in a place and circumstance where I could resolve it without being stranded.
So new brushes arrived, and once installed the output of the stator is back to ~50V per phase. However, the r/r is putting out over 18V at 4000 rpm, so back to diagnostics again (has to be a dodgy r/r**)!
Pretty routine stuff on any bike, modern or classic, actually, but another example of the many potential issues and problems that are commonly associated with riding and maintaining a motorcycle.....
*(failing brush connection discovered later helps explain both erratic readings and eventual flat voltage)
**(r/r returned to manufacturer; testing revealed it was faulty, and they replaced it under warranty)
Two miles from the end of a 72 mile ride, I briefly stalled the 850T at a light and could not get it restarted. The battery appeared flat as the starter just clicked and would not turn over. Quickly assessing traffic and noting a very slight downhill, I snicked it into second gear and duck-walked just enough to bump start it to life. Very relieved, those last two miles had me giving thanks that I had not stalled it earlier in the ride, and/or somewhere on a flat grade.
A check of the battery at home proved that there was a fault in the charging system; it was only 10.5V, having left before the ride with at least 12.5V. It had been charging properly at last check about 6 months earlier. Consulting several troubleshooting guides, I got to work to trace the usual suspects and identify the problem. The charging system is essentially made up of four parts. The rotor and stator take rotational energy from the crankshaft and turn it into AC power, upwards of 100V per phase depending on the type. The regulator and rectifier (r/r) cut off the AC peaks and restrict the DC output to a range that the battery can handle (roughly 12.5V to 14.5V).
First check, what's coming out of the r/r? At first I got a few erratic swings, one of which was up to 18V*. Later trying again the expected voltage of 14.5V was not reaching the battery when the engine was revved up to 4000 rpm (staying somewhat flat at 12.5V). But the solid state r/r was then ruled out when resistance and diode measurements using a multimeter figures checked out as expected. Then, resistance of the rotor windings measured between the copper slip rings also checked out, as did grounding of the rotor. And then, resistance between the phases and grounding of the stator was measured. This too checked out. A final test involved running the bike again and revving from idle while measuring AC voltage output from each of the yellow phase wires. Something was clearly up when only 3V was registering in each of the phases. These should be showing at least 50V AC each at higher rpm.
So this pointed to a bad rotor or a bad stator; stators are not known for breaking, but rotors are (averaging around 40K miles or so). But, both of these independently produced normal readings in testing.
This puzzled me for some time, prompting follow up with Gregory Bender who provided the wiring harness. His prime suspect was the rotor, but also ensured that I check r/r grounding. While re-doing several of the charging system diagnostic checks, I realized that one thing was overlooked. The rotor resistance had been taken directly via the copper slip rings, but the leads to the brushes which contact these rings were not routine items on the checklist(s). Sure enough, one was measuring contact and the other nothing. As these are hard to inspect while mounted, I pulled the alternator off the rotor. The second photo below shows the significance of the contact problem. The forward brush was worn to a nub; there was no more pressure that the spring could provide to push it down (inside the white plastic housing) enough to contact the rotor. As the charging system had worked many rides earlier, this slowly got to a point where contact was lost. Again, thankfully it came to light in a place and circumstance where I could resolve it without being stranded.
So new brushes arrived, and once installed the output of the stator is back to ~50V per phase. However, the r/r is putting out over 18V at 4000 rpm, so back to diagnostics again (has to be a dodgy r/r**)!
Pretty routine stuff on any bike, modern or classic, actually, but another example of the many potential issues and problems that are commonly associated with riding and maintaining a motorcycle.....
*(failing brush connection discovered later helps explain both erratic readings and eventual flat voltage)
**(r/r returned to manufacturer; testing revealed it was faulty, and they replaced it under warranty)
 |
| exposed stator/alternator on front of 850T; three yellow wires for three phases |
 |
| a 'spring loaded' brush for each copper slip ring on the rotor; note forward brush worn to a nub |
Sunday, January 29, 2017
Got Flakes?
Early 2017 Update
It was a matter of time before the cylinders came off for inspection, and a persistent minor leak on the right was the reason this moment finally came. I had re-torqued the head a few times already with no change, so after a little procrastination I got some new gaskets on order and got to work.
At first glance I thought all looked OK, until I noticed some
scoring on the back side of the piston. Then it was clear that one of the rings was
ground into the groove near the exhaust side and the piston was scored. A check of the
corresponding cylinder bore location revealed an unsettling mark. Crud, the chrome
had flaked off in an area about the size of a 'Tic Tac'. This cylinder is toast. And a little scary at how close it might have been to failure, had I not checked.
Options ranged from sourcing used jugs, to NOS cylinders, to new Gilardoni cylinders. And with or without pistons? Rings? Since an eBay search showed several reasonable used items, I decided to go the cheapest route (my usual!) and buy used. Including the pistons, which is good, as the scoring on my right piston has gone a little far for comfort. I hope the photos and descriptions live up to the condition required to use them. More when they arrive and the installation is done.
Follow-up: the used cylinder arrived, with chrome bores in very good condition. And, the piston with which it is matched is in good shape, and the rings are free from obvious wear and move freely in their grooves. Wrist pin also looks great; that was popped into the freezer for later installation.
Both needed a lot of cleaning, however; mud wasp nests and corrosion from the cylinder and carbon from the piston head. My soda blaster was put to use as well, helping to strip decades of build-up; a water bath then neutralized the soda and all was dried. Following the thorough scrub and cleaning (and aluminum paint on cylinder fins), it was time to install the piston, gaskets and put the head on. I'd never held the 'beating heart' of a motor in my hands before, so was a little nervous about putting this all back together. I further did not expect the wrist pin (a press fit) to be so snug. A tool had to be made to press the old one out and the new one in (and installing it fresh out of the freezer overnight helped as well). Below photos show the tool in action pressing the old one out, and then the parts of the tool. Next time I will use a more robust bit of PVC tubing, but this worked well and I got the new piston installed without too much effort.
The head bolts had to be torqued properly before the valve clearance adjustments could be made. While cold, the head was bolted down in a star pattern until very snug, but just under full torque. At roughly 24 hours, the head was torqued to 34 pd/ft. Then again, roughly 24 hours later the head was re-torqued to the same 34 pound feet. Valve clearances were then both set at 0.006" (slightly tighter than Guzzi spec, I know). Other note, both were at first measuring tight from prior setting, as evidence that the new gaskets are (obviously) not quite as compressed as the old ones were. A day later I took the 850T out for a ride; it started quickly as usual and settled into an idle. It rides at least as well/normally as before; revving easily no smoking or oil leaks. Hooray!!
It was a matter of time before the cylinders came off for inspection, and a persistent minor leak on the right was the reason this moment finally came. I had re-torqued the head a few times already with no change, so after a little procrastination I got some new gaskets on order and got to work.
Options ranged from sourcing used jugs, to NOS cylinders, to new Gilardoni cylinders. And with or without pistons? Rings? Since an eBay search showed several reasonable used items, I decided to go the cheapest route (my usual!) and buy used. Including the pistons, which is good, as the scoring on my right piston has gone a little far for comfort. I hope the photos and descriptions live up to the condition required to use them. More when they arrive and the installation is done.
Follow-up: the used cylinder arrived, with chrome bores in very good condition. And, the piston with which it is matched is in good shape, and the rings are free from obvious wear and move freely in their grooves. Wrist pin also looks great; that was popped into the freezer for later installation.
Both needed a lot of cleaning, however; mud wasp nests and corrosion from the cylinder and carbon from the piston head. My soda blaster was put to use as well, helping to strip decades of build-up; a water bath then neutralized the soda and all was dried. Following the thorough scrub and cleaning (and aluminum paint on cylinder fins), it was time to install the piston, gaskets and put the head on. I'd never held the 'beating heart' of a motor in my hands before, so was a little nervous about putting this all back together. I further did not expect the wrist pin (a press fit) to be so snug. A tool had to be made to press the old one out and the new one in (and installing it fresh out of the freezer overnight helped as well). Below photos show the tool in action pressing the old one out, and then the parts of the tool. Next time I will use a more robust bit of PVC tubing, but this worked well and I got the new piston installed without too much effort.
The head bolts had to be torqued properly before the valve clearance adjustments could be made. While cold, the head was bolted down in a star pattern until very snug, but just under full torque. At roughly 24 hours, the head was torqued to 34 pd/ft. Then again, roughly 24 hours later the head was re-torqued to the same 34 pound feet. Valve clearances were then both set at 0.006" (slightly tighter than Guzzi spec, I know). Other note, both were at first measuring tight from prior setting, as evidence that the new gaskets are (obviously) not quite as compressed as the old ones were. A day later I took the 850T out for a ride; it started quickly as usual and settled into an idle. It rides at least as well/normally as before; revving easily no smoking or oil leaks. Hooray!!
 |
| wrist pin pressed through the rod, and cylinder about to be dropped down onto base gasket |
Take a Seat, Cover
Early 2017
Ugh. This seat issue has dragged out for far too long; besides the added effort and cost, it is interfering with riding time too! To recap, the seat and pan as purchased on bike were not stock; being made of aluminum, the pan also was losing shape and not a good platform for my project. I purchased a cafe seat which looks great, but was not designed to fit with the 4" rise of the stock rear fender. Much modification later I got it to fit, but it was too low (not a lot of leg room) and did not meet the tank in an appealing way. So the next plan was to add two inches of foam to raise the seat height and cover the seat with a stock(ish) cover. But the leather (vinyl) of the V7 style cover I bought does not stretch enough to retain the foam shape I'm looking for (cafe style similar to my after market cafe seat).
I finally realized that I'd have to cut and sew my own seat cover. Although a reasonably capable tailor, not looking forward to this.
Starting with the fabric, I sourced a black vinyl product which has 4-way stretch; I would not characterize it as 'stretchy', merely that it has some give in it. Not a lot, however, meaning that my cover had better come very close to matching the foam shaped underneath.
Next was to finish the foam reconstruction of the seat pan to raise the height 2". This further required an extension to the front of the pan, by which I cut two strips of steel and riveted each to either side at the front. The foam also needed to be built up around the front to 'cup' the rear of the tank (dark grey section you can see in the photos).
After a few internet searches and YouTube videos, I then tackled the hardest part of the seat cover - a quilting effect on the top. I glued 1/2" foam to the back of a piece of liner fabric and then to the piece of vinyl I'd cut out in the shape and length I needed (here, I also needed to note that I'd extended the foam top forward about 1", and measure lengths accordingly). A few practice runs using heavy duty thread had the tension off and the thread binding in the needle. I had to resort to regular black cotton thread (meaning that however this seat turns out, the stitching will not last forever....). Sounds like a good excuse to get that industrial sewing machine I'd been thinking about. So my results are as pictured below.
I was noticeably off on one row, but when stretched a little and mounted I don't think it will be a problem. Was pretty happy with how it turned out in relation to the cafe seat I'd originally purchased; except the ridges are 'higher' than I'd wanted since the foam I used was likely a little thicker - should have measured 2" between them instead of 1 3/4". That piping I had to make from scratch as well.
Next is to measure the rear and sides; but before sewing these on, I'd always really liked the Moto Guzzi logo in white across the back of the V7 cover (and other models as stock). I hand cut a stencil using clear window decal paper and purchased a can of white vinyl spray paint. The first test (yes, testing first is always a good idea!) had the paint bleeding under the stencil due to the embossed nature of the leather-looking fabric. A second test showed that dusting the paint on in a single pass (or two) with 5-10 min in between passes resulted in very little bleed. Once I started painting the cut rear fabric, it would be all over for better or worse. Level, square and no bleeding were the goals.
I'll let you know shortly how it all turned out............
Ok, it's about 2 weeks later, and about that time! The logo painting worked very well, with just a little micro-bleeding under the the stencil (not visible except very close; below). And after MUCH puzzling and planning and measurement I finally decided how the sides and front would be sewn, and committed to doing so. The only section I was prepared to mess up was the front, as any errors would be covered by the tank. Indeed, I had to pull stitching out of the front two times due to flawed estimations. The rear turned out to be tougher than the front, reverse of what I'd been expecting. I realize now that any change in surface angle or direction requires a separate section cut out for it; sewn interfaces are much easier to conform to turns and corners, etc, than relying on the stretch of the fabric alone. But I made it work. 'First' Final product below (not fully tacked down):
I was happy but not entirely so. The seat on its own looked fine, but was a little bulky looking on the bike itself. And the 90 degree turns in the back led to bunched up fabric that did not look as professional as I'd hoped for. So I did some more thinking and looking on the internet for examples of seat shapes. I realized that a gentle curve up near the tail would both reduce the bulk and avoid sharp bends and thus reduce bunched up fabric. Even more puzzling and thinking led to a plan. I had purchased several tack strips but had not used them; one of these would be needed. I realized that if I made two cuts on each side of the seat pan it could be bent upward gradually towards the top of the fender. A length of tack strip could then be riveted into place across the gap, and over top of the fender, to continue the smooth arc from one side to the other. Again, my trusty 4" cut-off wheel did quick work of it and soon the idea was tested (below).
Once the cover was stretched over the seat pan, the latch mechanism was the last element to construct. I pulled the one off the seat that came with the bike and, after several measurements, drilled the holes for rivets. These holes had very little margin for error, as they directly dictated how the seat lined up with the seat tang welded on the frame. Thankfully my measurements and estimations were good, as the fit and function are perfect.
So I can call the seat done, finally. The extra leg room and more upright seated position on the bike is noticeable and feels much better than the stretched-out, knees-under-the-armpit position from before. It is not 100% what I would wish for, but it looks very passable for a first time fabrication, and it works well. So much went into it too that it deserves to be let go of for a while at this point and just ride.......
Ugh. This seat issue has dragged out for far too long; besides the added effort and cost, it is interfering with riding time too! To recap, the seat and pan as purchased on bike were not stock; being made of aluminum, the pan also was losing shape and not a good platform for my project. I purchased a cafe seat which looks great, but was not designed to fit with the 4" rise of the stock rear fender. Much modification later I got it to fit, but it was too low (not a lot of leg room) and did not meet the tank in an appealing way. So the next plan was to add two inches of foam to raise the seat height and cover the seat with a stock(ish) cover. But the leather (vinyl) of the V7 style cover I bought does not stretch enough to retain the foam shape I'm looking for (cafe style similar to my after market cafe seat).
 |
| stock Moto Guzzi V7 seat cover |
Next was to finish the foam reconstruction of the seat pan to raise the height 2". This further required an extension to the front of the pan, by which I cut two strips of steel and riveted each to either side at the front. The foam also needed to be built up around the front to 'cup' the rear of the tank (dark grey section you can see in the photos).
 |
| finished, with piping |
 |
| compared to purchased cafe seat |
Next is to measure the rear and sides; but before sewing these on, I'd always really liked the Moto Guzzi logo in white across the back of the V7 cover (and other models as stock). I hand cut a stencil using clear window decal paper and purchased a can of white vinyl spray paint. The first test (yes, testing first is always a good idea!) had the paint bleeding under the stencil due to the embossed nature of the leather-looking fabric. A second test showed that dusting the paint on in a single pass (or two) with 5-10 min in between passes resulted in very little bleed. Once I started painting the cut rear fabric, it would be all over for better or worse. Level, square and no bleeding were the goals.
 |
| stencil placed along center line |
 |
| stencil taped for painting |
Ok, it's about 2 weeks later, and about that time! The logo painting worked very well, with just a little micro-bleeding under the the stencil (not visible except very close; below). And after MUCH puzzling and planning and measurement I finally decided how the sides and front would be sewn, and committed to doing so. The only section I was prepared to mess up was the front, as any errors would be covered by the tank. Indeed, I had to pull stitching out of the front two times due to flawed estimations. The rear turned out to be tougher than the front, reverse of what I'd been expecting. I realize now that any change in surface angle or direction requires a separate section cut out for it; sewn interfaces are much easier to conform to turns and corners, etc, than relying on the stretch of the fabric alone. But I made it work. 'First' Final product below (not fully tacked down):
I was happy but not entirely so. The seat on its own looked fine, but was a little bulky looking on the bike itself. And the 90 degree turns in the back led to bunched up fabric that did not look as professional as I'd hoped for. So I did some more thinking and looking on the internet for examples of seat shapes. I realized that a gentle curve up near the tail would both reduce the bulk and avoid sharp bends and thus reduce bunched up fabric. Even more puzzling and thinking led to a plan. I had purchased several tack strips but had not used them; one of these would be needed. I realized that if I made two cuts on each side of the seat pan it could be bent upward gradually towards the top of the fender. A length of tack strip could then be riveted into place across the gap, and over top of the fender, to continue the smooth arc from one side to the other. Again, my trusty 4" cut-off wheel did quick work of it and soon the idea was tested (below).
 |
| curve cut into seat pan (top left) to take the line up and over the rear fender |
Once the cover was stretched over the seat pan, the latch mechanism was the last element to construct. I pulled the one off the seat that came with the bike and, after several measurements, drilled the holes for rivets. These holes had very little margin for error, as they directly dictated how the seat lined up with the seat tang welded on the frame. Thankfully my measurements and estimations were good, as the fit and function are perfect.
So I can call the seat done, finally. The extra leg room and more upright seated position on the bike is noticeable and feels much better than the stretched-out, knees-under-the-armpit position from before. It is not 100% what I would wish for, but it looks very passable for a first time fabrication, and it works well. So much went into it too that it deserves to be let go of for a while at this point and just ride.......
Subscribe to:
Posts (Atom)