Monday, April 13, 2015

Garmin 765 Loss of Power Fix

Back in 2011, my wife bought me a GPS to help me find my way around the countryside.  I needed it.  I get lost easily, it seems.  I shopped around and found what I wanted, a Garmin nĂ¼vi 765T.  You can read all about it here, and in later postings called out below, under "Previous GPS Postings." 

The device has been fine, mostly, but the power connection has been troublesome.  I was frequently getting this message while en route. 

Fiddling with the power connection in the back sometimes made it recover, but finally, there was no amount of wiggling that would make the power come back on...and the internal battery only runs the unit for 10 or 15 minutes, so that is not an option.  This type of failure has happened twice now. 


Others have had this trouble too, and have written about it:
GeekDad66 even tried to fix the problem. 

I decided to do some first-hand investigation on how to remedy this nuisance. 

DO NOT attempt repairs if you are not familiar with safe mechanical or electrical work practices, or if you don't know how to properly solder electrical circuitry.  
I take no responsibility for anything in this posting used by others.  There is a possibility that something described here or misinterpreted from my description could cause the GPS or other electrical systems, including the vehicle in which it is used, to malfunction.   

It turns out that there is a component called the cradle (Garmin Model: 010-10823-07) that attaches to the back of the GPS and contains the 18-position receptacle that, among other things routes 12 volt power to the GPS itself.

It is that receptacle that usually fails.

Here is how I found out.  Lets look inside the cradle.

Remove the single Phillips-head screw.   

The two halves of the cradle still won't come apart because there are four latches inside around its periphery. 

I couldn't force the halves apart with my hands, so I used a small wood chisel applied adjacent to the pushbutton that unlatches the cradle from the main body of the GPS.  It only takes a light bump with the palm of your hand on the chisel to pop open the cradle.  Be very careful not to cut yourself. 

There are two springs inside that will pop out if you are not careful as you open the case.

There are four latches around the inside of the halves of the cradle.  One of them is circled.  One of the springs is visible, too. 

There are two more latches down here.

The inside story:

The 18-position connector:

The body of the receptacle breaks right off the board due to the stresses of the power cable, and not long afterward the tiny wires connecting it fracture too. Mine just fell out when I opened the cradle. 

In most cases, that would mean that the cradle is junk, but a new one costs at least $21, and I am on my third one, counting the original.  This is getting expensive, and I'm a tightwad.

I decided to press onward with a fix.  

The circuit board is marked with the polarity of the 12 volt supply, and there are large enough holes in the board centered in the Positive and Negative pads that can accept wires. 

 The polarity is even marked!

The plug on the power cable I use on the bike (Garmin Model 010-10747-03), and that mates with the 18-position receptacle, has only four active terminals, two for the positive and two for the negative feed from the 12-volt source.  I suppose they use two to make the connections more reliable.  They are what are called bifurcated connections. 

I decided to engineer a two-wire cable connection using 1/8" (3.5mm) miniature phone plugs and jacks (sockets).  The cradle circuit DOES NOT change the input voltage.  It is a simple pass through -- 12 volts in, 12 volts out.   

I went to Radio Shack and bought a mono (2 conductor) audio cable with male plugs on both ends.  Since the cable has two male ends, it is enough to fix two cradles -- just the number I have that need fixing.  This plug will provide the 12-volt source.

Radio Shack
Model: 42-2420
Catalog #: 420242
One cable, two usable plugs

I also bought two in-line jacks, and two more plugs.

Radio Shack
Model: 274-333
Catalog #: 2740333
Radio Shack
Model: 274-287
Catalog #: 2740287

The idea is to cut my Garmin power cable near the plug that mates with the 18-position receptacle, and attach a jack there. 

A plug that I will add to the GPS cradle will mate with it, eliminating the troublesome receptacle.

Since I have a functioning cradle with the receptacle intact, I will also attach a male plug to the end of the power cable I cut off, so it functions as it did originally.  Pictures later. 

First I cut off about five inches of the Radio Shack audio cable, stripped the center conductor, and separated the shield foil and wire, so I can attach it to the circuit board inside the cradle.

Then I prepare a hole just to the right of the receptacle slot for the new cable to exit the cradle. 

It will look like this when assembled. 

The wires in the new cable are attached to the positive and negative holes in the circuit board.  The center conductor is positive, the shield is negative. 

The shield wire, being soldered very close to the circuit board helps prevent the cable from pulling out of the cradle when we are all done.  The undersize hole that was added in the side of the cradle also helps this.  

The positive (+) side is connected to the tip of the phone plug.  The negative (-) to the sleeve.  The ones I used are all mono, two-conductor plugs and jacks, not stereo, three-conductor. 

A thorough description of phone connectors can be found here

I put the parts back inside the cradle, routing the new cable through the new hole.  I glued a small piece of felt over the old receptacle slot to keep the mud daubers and dust out. 

Don't forget the springs.  If you lost one or both of them, you can make new ones from shortened ball-point pen springs.  (I know this from first-hand experience.) 

This is how the cradle looks with the new cable sticking out. 

By the way, the power comes out of the cradle and enters the GPS unit using the following pins in the other connector of the cradle:

More bifurcated connections there. 

Now, to attach the female jack to the cord that I cut the plug off of. 

The opening for the cable in the jack housing is too big for this cable diameter, so I put some layers of heat-shrink tubing over the cable to make it fit snugly in the housing. 

Ready to solder.  Don't forget to put on the heat-shrink tubing and the jack housing onto the cable before soldering!  Correct polarity is red to center (tip), black to sleeve. 

 Soldered connections and heat shrink tubing visible. 

Buttoned up. 

Now, to make the end I cut off the power cable functional again.  It would be used with a cradle where the 18-position receptacle is still intact.

This plugs into the female jack we put on the power cable and converts it to mate with the 18-position receptacle, just like the original power cable worked. 

So here are all the pieces:

The red and the black male plugs mate with the black female jack so you can use the power cable either way -- with the newly-added cable on the cradle or with the plug for a cradle with a functional 18-position receptacle.

Whew.  Did you get all that?  

Rear view of the new cable coming out of the cradle, on bike. 

And a rider's-eye view:

The jack is wiretied to the handlebar clamp. 

This ought to last a lot longer than the troublesome receptacle that came in the cradle.

  • If you leave the 18-position receptacle on the board in the cradle, and if it is still functional, DO NOT connect a power cable to the cradle through the 18-position receptacle.  That will make the tip of the added phone plug have a positive 12 volts on it.  If the tip touches something grounded in the vehicle, a fuse will blow, probably in the power cable.  If you need to connect power to the functional 18-position receptacle, then the added phone plug MUST be insulated.  A jack with no cable attached to it could achieve this.   
  • The traffic function of the 765T ("T" stands for traffic option), will not work except with the original power cable (Garmin GTM 20) plugged into the functioning 18-position receptacle in the cradle.  That cable has a skinny rectangular receiver in it (at right, below) that mates with other terminals in the cradle. 

    The power cable I use on the bike does not have the traffic receiver, only the power connections. 

Previous GPS Postings:

Let me know if you find this fix to be useful in solving the power problem on your GPS. 


Thursday, April 2, 2015

Pooped Pump....No Fuelin'...

OK, here's the rest of the story. 

On my way back from my first ride this year to the Blue Ridge Parkway a couple of weeks ago, within about three miles from home, the engine on the bike quit suddenly.

Almost exactly like this bike is doing:

Original video courtesy of YouTube user invisiblemon. 
He was still 16 miles from home at the end of the video, but did make it there a mile
or two at a time as the fuel pump was failing.  He later tested the pump and found
that its output pressure was low.

Not good. 

This can be quite disconcerting if you are in the middle of a turn or in traffic.  At any rate it can be disconcerting anyway, 'cause visions of having to call for help begin dancing through your brain when these things happen.  And, I hate to ask for help.  I think this is built into the wiring of a man's brain.

Anyway, I coasted to a stop, fortunately in a safe place, and pressed the starter button.  The engine fired right up again.  I revved it and it seemed OK.  About the time I was going to get back on the road, it quit again.  But it started right up.


Plenty of gas in the tank.  Now what?

I let it sit for a few minutes, and it started and ran for a while, so I took to the road again.  It stalled again not a minute later.  Again it restarted, and I revved the engine while slipping the clutch very significantly.  When it ran, it ran strongly. 

It bucked and almost threw me a couple of times as the engine came to life at large throttle opening, died, then came to life again.  People around me thought I was a beginner or a stunter, I'm sure.  It was close to wheelying a few times.

Finally at the road to my house, it stopped and wouldn't restart.  I paddle walked it enough to get it moving downhill toward my house.  I coasted into the driveway as quietly as a mouse.

I decided to do some diagnosis.

I raised the tank enough to fish the fuel pump cable out to the side.  I probed the pump side of the connector with my voltmeter, and found steady power.  I started the engine.  It ran fine.

After about ten minutes, though, the engine missed slightly.  The voltmeter still showed steady, however. 

I put my ear to the fuel tank.  I could hear the fuel pump running, but it had a sort of intermittent screaming sound to it.  When the scream became more significant, the engine stuttered, then, after a little while longer, finally stalled. 

I decided that the pump must be failing, although I had never listened to the pump prior to this, so I wasn't sure what it is supposed to sound like.  I looked online to see if I could find anyone who had also had this problem.

Sure enough, the Ninja 650R forum mentioned a problem, and the fact that a new fuel pump part number replaces the original.  The original Fuel Pump Assembly P/N 49040-0010 has now been superseded by P/N 49040-0026.  The pump includes a filter, a pressure regulator, and the low fuel float switch.  None of these is serviceable individually. 

Four years go, another guy (known in the RiderForums as MCRider), then 70 years old, and who lives in the state of Washington, described how his 650R lost its fuel pump the day he started out -- with 71,000 miles on the clock -- on a month-long, 8,000+ mile ride to the East Coast and back on his identical 2006 Kawasaki Ninja 650R.  The symptoms sounded almost the same, though my bike has only 48,000 miles on it.  (As of his last posting in late 2014, his bike has over 110,000 miles on it.)  

I checked all of the on-line sources that I know of, including Blue Ridge Performance.   Gary, the proprietor, is a man of few words, but he usually can get what you need for 650Rs and his prices are as low as any I found, so I ordered up the pump and the seal O-ring that goes with it, P/N 92055-0185. 

It is best to siphon out all fuel before starting this work to minimize weight and reduce leakage.  All of this work should be done outside and away from fire and other ignition sources. 
I take no responsibility for this description, or its interpretation.  Always follow the proper service manual's instructions for your bike along with good mechanic's practices.  

When I received the package I started to work changing the pump.  Here are the parts, still snuggled in their packing materials. The new pump is manufactured by Keihin, as is the original. 

The teardown: 

I take off the seat (easy, peasey), and then the black plastic side covers on either side of the tank are removed.  Then the two seat bracket screws holding the rear of the tank.  

Be careful that the tank doesn't slide sideways off its mounting position if you are working with the bike on its side stand.  A rear stand is better here to help avoid that situation. 

Hold onto the tank so it doesn't slide, now! 

Raise the rear of the tank up on a piece of 2 by 4. 

Reach under the tank and disconnect the fuel pump electrical connector. 

And the fuel hose.  (There may be pressurized fuel in it, so wrap a rag around it before pulling it off the pump.

Next, pull off the rubber fuel filler overflow hose beneath the left side of the tank. 

It has a very small spring clamp holding it.  Don't lose it. 

Arrange two 2 by 4s on the ground to set the tank on, such that the fuel pump does not bear the weight of the tank. 

While you are there, you might as well clean the air filter.  Slide the large hose above it off its nipple and remove this screw to pull out the filter. 

Set the tank in a couple of sturdy corrugated cartons to hold it on its nose. 

If there is any fuel in the tank, some may leak out of the fill opening. 

Remove the five screws holding the pump into the tank. 

There is thread sealer on the screws, so they may be difficult to remove. 

The two pumps look the same on the outside. 

I do notice that there are MANY blue paint dots on the outside of the new pump. 

In the automotive industry, such paint dots indicate that some check is to be done, and the presence of the paint dot indicates that the check was [probably] performed.  I say probably, because most such checks are visual, but people get "snow blind" and miss defects but mark the part anyway.  I have experience with this very problem in a manufacturing operation, I'm afraid.  The factory that built this one must have lots of quality issues, judging from the plethora of blue dots.  

I remove the O-ring from the tank and inspect the seal area and screw bosses.  Everything looks OK, so I insert the new O-ring and position the pump.  There is only one way all five screw holes line up, so you can't go wrong. Don't forget the thread sealant on the five screws. Tighten the screws in a star pattern, as is good practice. 

After I get the new pump mounted, I notice two more blue paint dots on it. 

Whew.  I hope they were really, really paying attention to details when this pump was built. 

I carefully slide the tank onto the front rubber mounts, and position the 2 by 4 under the rear of the tank again.  Remember to watch that the tank doesn't get away from you. 

I install the fuel line onto the new pump and slide the red lock piece home.  Don't force it.  It is designed to slide into its locked position only when the fuel line connection is fully seated.  I then plug in the electrical connector and connect the overflow hose. 

The seat bracket screws that hold the rear of the tank are next, then the plastic side covers on either side of the tank.  The seat is last to go on. 

To prime the fuel system, I turn the ignition switch to the on position to run the fuel pump for the initial three seconds it is programmed for.  I repeat that.  Then I press the starter button.  The engine fires up after cranking a few seconds getting the air out of the line between the pump and the injectors. 

I run the engine until the cooling fan cycles on two or three times.  It does not stumble.  Now for a test ride. 

I venture out with the idea that the pump might not have been the real problem.  If it wasn't, then I could stall somewhere again.  That wouldn't be good.  I pick a route to the south of nearby Greenville that is all secondary roads.  I ride along at a sedate clip and everything seems fine.  I reach the far side of Greenville, and decide to head north to Lake Lure and Chimney Rock, taking a leap of confidence that the problem is fixed. 

I don't particularly like to ride to these two places because they are very crowded spring through fall.  Today, it is not yet tourist season, so the traffic is light. 

I make it to Lake Lure and take a lunch break.  The bike has performed well so far.  I go on a little way to Chimney Rock, and then head toward home.  The temperature range today has been between 40 and 80 degrees, so he thermal effects on the pump ought to have been tested, too.  As I roll into the driveway -- under power this time, unlike last time -- I find that I have ridden 202 miles with nary a lick of a problem.

I'll call that a success.  It should be good for more than another 48,000 miles. 

I'm sure glad the pump didn't fail when I was miles from home either that first time or on my test ride.