Stuck at gas station: dead battery

Looks like it's time for a new battery and or regulator.

After sitting for ~2 mos, the bike didn't start last night (no surprise). After charging all night and topping off the battery, it started right up. And then stranded me at the gas station 5 miles from home.

I checked before leaving:

13.7-13.9 Vdc at idle
13.3 Vdc at 5krpm, which was iirc, the value for this regulator when new (electrosport)

Luckily we were close to home, so my wife went back for a car and jumper cables.

 

Test the voltage under load. See if it stays above 10V. If it doesn't, it's probably shot.

 

R/R output seems OK, so replace the battery first.

Charging the battery once a month is the least expensive way to keep the battery healthy and reduce the load on the charging system, and for me it's required maintenance.

 

Gotta ask Woody, did she have a bike or did you make her walk? lol

 

I would check the dreaded plug between the stator and the R/R as well.

 

+1 on TC's post about the 'dreaded connector'. On pretty much any bike the connectors in that part of the charging circuit can degrade over time, especially if the bike sits for any length of time. Corrosion sets in, and especially in this circuit (between the 3-phase stator and the R/R), it causes charging problems and will ultimately destroy the R/R.

 

Who designed that setup, anyway, not a gradute of the finest Japanese engineering universities, huh ? Somehow, with better basic design and materials this common problem could have been better engineered into a zero-failure item, unfairly lowering Honda's quality reputation.

 

They don't design the products to sit in the elements (as many - even most - motorcycles do). If you're really anal and proactive about a storage process (and I know of nobody who fits that description) you would take apart every connector in your wiring harness once every couple of years to clean the connections. Nobody does that. No amount of design (within bounds determined by common sense and cost constraints) can eliminate this problem.

 

yes... :crazy:

She was on her HawkGT. We'd stopped off to top up the tires and tanks. So she rode back home and came back with the car to jump the bike with. It started right up, and ran fine all the way home.

 

Well, this particular bike has been garage-stored it's entire life. And rarely ridden in wet weather. Aside from the few scrapes it's picked up the 3(?) times it's been laid down, it's amazingly clean. Not a single oil leak from the engine to muck up the engine cases, either. It's not as clean as a show bike, by any stretch (too much crap thrown from the front tire and all the bugs it's been ridden through), but it's a joy to work on compared to my wife's Hawk (before we rebuilt it).

The hawk had been parked outside it's whole life, and had the PCV on the rear head replaced with a plastic tube, which had collapsed from the heat, and then blown off due to crankcase pressure. The back of that engine was a total mess.

 

I'll spend some time tomorrow with the bike, after I get the other car (WRX) running. It has a cracked hose in the boost mgmt system for it's turbo, very bad juju if not repaired.

I'll just go through the standard trouble-shooting after I charge the battery up again tonight. However, this time I'll do so with enough water in all the cells, one was only 1/2 to 2/3 full. Then I'll tear the charging system down and see how all the connectors look, and how the regulator checks out.

I hadn't realized that these are SCR based... what a god-awful way to design something. Given that power mosfets have been around for a long time, I'm amazed at the ancient technology used in these R/Rs.

Luckily, since I last replaced the R/R (when I bought it back in 2006), mosfet regulators are the norm, and readily available. If the R/R does get replaced, I'm going with a mosfet one of modest size (since this bike really doesn't have much of a high-power need for something like a 50A regulator). This write-up is quite nice:

Charging System Diagnostics - Rectifier/Regulator Upgrade - Triumph Forum: Triumph Rat Motorcycle Forums

and uses these connectors:

R/R Connectors

Also, these kits are pretty nice (although I'd rather use a real ATO fuse as suggested on the triumphrats site), and honestly, for $100 for a 50A rated mosfet regulator... nice.

http://roadstercycle.com/Shindengen FH012AA Regulator upgrade kit.htm

And it looks like the Shorai batteries have been coming down in price nicely. This is their recommended replacement for the stock Yuasa:

ShoraiUsa - LFX14A5-BS12

And at a retail of $159, it's 3x the cost of a replacement Yuasa, but these have been getting amazing reviews all over the place. Aside from one write-up about a fire this last spring. Shorai covered the cost of the repair (parts and labor) to the bike, and were very, very open about why it happened, and how they now have protection circuits in the batteries to prevent a failed R/R from applying over-voltage to the battery. Frankly, that's a really hard combo to beat (Shorai Lithium-Iron and a MOSFET R/R)

 

Agreed on the R/R and Battery upgrades; but if you don't take apart that connector (and the rest of them in the charging circuit) you will eventually have issues. I know I said 'exposure to elements'; but just being parked can be enough, and it depends on where you live and where the bike is (and has been) parked. It's okay if you don't want to heed the warning, just be aware. (I didn't heed the warning, and ended up with a dead battery and fried R/R - along with the connector)

 

So, found some suspect things:

1) Battery directly off the charger is 13.0 Vdc, no load.
2) Running the lights on the bike (headlight, ambers, etc), engine off: 11.3 Vdc
3) Turning over the bike: 5-7Vdc

Fully charged off the charger, it will start the bike right up. With the bike running (idle):

13.0 V at the battery
13.4 V regulator output. 0.25 V dropped on the positive side, and 0.15 V dropped on the negative side, between the regulator and the battery.

At 5K rpm, voltages drop to ~12.8 V at the battery..

The stator checks out ok. 20Vac across each phase at idle, and ~70 at 5000.

diode tests in the regulator all pass, but they did on my last R/R that died, too.

When I disconnected the stator from the R/R (bike off) the connector was hot enough to make the mix of dielectric grease that I'd previously packed it with, and the WD-40 that I used to clean it out yesterday, to start smoking lightly... Very, very warm. I'm certainly replacing this connector with a Metripack 3P280 from Eastern Beaver, as well as insert heavy-gauge wire direct from the R/R to the battery.

I'm going to double-check the battery voltages tomorrow with the engine running, given that the regulator was only making 13.4 voltages at it's outputs. If that still is crap, I'll swap batteries around between the bikes and see what changes, but with a 20Vac input, there's no reason that the regulator shouldn't be able to output 14.5 Vdc. Although I didn't test the loaded AC voltage of the stator, only the unloaded voltage, so it might be the limit. Hence the need for more testing tomorrow.

But I'm 100% certain that I'm going to replace that dreaded stator connector and add in the heavy gauge wiring to the battery (dropping 0.4V between the R/R and the battery is just bad). And about 50% on the R/R at the moment.

 

1) Battery off the charger is lower than normal, but doesn't tell you much...
2) Running about 10A, it shouldn't drop that low.
3) battery is dead. It should never get to 5-7VDC, even when cranking.

Definitely replace it, but completely check the RR before you replace and make sure it's not overcharging, or not working at all.

 

More poking at the bike today.

Battery fresh off the charger, 13.1vdc, where the charger has been keeping it at float/refloat.

That works out to 52.4 Vdc, which is roughly where my 48V deep cycle array is kept at by the charger. That system is photovoltaic and grid-tied, and sells off power over 54.4 Vdc. That's the "UPS for the house", 1.5-3 days of backup during storm conditions, and indefinite if it's clear out.

Anyway, back to the bike:

I started with a the battery at 13.1 VDC. Ignition on, headlamp on, bike not running, 11.3 VDC again. I let it sit with just the headlight on for a bit, and measured the voltage at the bulb (10.0...) Since I'm goign to be ordering stuff from Eastern Beaver, I might as well confirm how much of an improvement the headlight harness will get me.

Went to start the bike. It cranked for a while, but it did not want to start (it's extremely hard to start this bike when cold). The bike turned over well for about 5-10 seconds, and then quickly fell off and stopped turning over. Back onto the charger it went while I looked at some other things. It was still about 11.0-11.3 V when charging started, charger reading about 30% capacity. I let it charge it up over 12V, 70% capacity, and it started the bike right up.

Battery output right after starting was 13.1 V again, at 2500 rpm.

One of the questions that I've had is what the loaded voltage on the stator is. I'd hooked up a current clamp, and was measuring that and the voltage across each of the windings in the stator (at the stator connector). Turns out I was getting:

10 VAC
10 VAC
7 VAC

Imbalanced voltage? that's... odd. But the voltage is a function of the load, so it just means something is odd.

The currents were about 19 Amps AC in two of the conductors, and the voltage drop across the connector was 0.011 Vac for two, and 0.007 for the third. Clearly, lower current on that one. By this time, the connector was too hot to hold firmly (so I could get the probes to the back of the crimped on connectors).

By my math, at 19Amps and 10 V, that's a bit over 300VA. With an output voltage of 13.1, and (incorrectly) assuming 100% efficiency, that's 20amps output!. At 60% efficiency, 12Amp output seems more reasonable given the loads on the bike (I think).

Now it gets fun.

After a while, my fiddling had taken long enough for the bike's fan to come on. At this point, I fully turn off the choke, and let the bike drop to normal idle. It immediately jumps from 13.1 to 14.5 VDC. I've seen this before.

I used the choke as a "fast idle", and swept the bike from idle at 1200 up to ~5000 rpm. The higher the revs, the lower the voltage at the battery, down to barely 13 volts at 5K rpm, which is where this bike runs roughly 90% of the time when I'm riding it (4500-6000), especially when on the freeway. Yes, there's an increased spark load as rpm climbs, but the stator is also outputting more and more as rpm climbs.

At this point I'm pretty sure that the battery is toast (or even if it's not, I want something with more capacity to better handle restarts), and the R/R with it. That dropping output voltage with RPM is exactly what the last R/R did when it failed.

And given that I really, really hate being stranded by my bike, I think I'm going to go the full route:

- LiFePO4 battery from Shorai, 12+ AH rating
- FH012AA MOSFET R/R
- new stator connector (Metripack 3P280)
- new R/R harness and direct battery leads
- headlight hardness with relays

The connectors and wiring from Easter Beaver, the R/R from roadstercycle.com, and the battery from Shorai themselves (the form factor used by our bikes isn't available in the distribution channel yet).

Not really want I wanted to do on the bike this winter, but c'est la vie.

What I wanted to dig into were the carbs, because I don't think I should be able to use the enrichment circuit like a fast idle...

 

How old was this battery you had in there?

 

4 years? Based on the threads on here I searched through last night. I know I replaced the R/R along the way, but I apparently didn't ever mention that in any threads.

But I'm pretty sure that the 14.5 V at idle, dropping to 13.0 at 5000 rpm isn't good. And the flowchart pdf doesn't really talk much about the voltage dropping heavily as RPMs rise, but IIRC, the FSM does. I forgot to double-check that last night.

It seems like once I manage to get enough charge into the battery, it does well enough. But if the voltage is only 13.0 when I'm out riding, it's never going to be charged up enough.

 

Sounds like you've already decided to go 'whole hog' on the charging system - not necessarily a bad thing on a twenty-five year old bike; but I have a quick question for you. When you made the measurement that uncovered the 'unbalanced' output from the stator, where (physically) were you taking that data?

 

The stator-side of the stator connector, DMM probes against the crimped-on connectors. Drop across the connectors (crimped terminal to crimped terminal) was only 0.01 Vac.

 

Hmm... that could be bad news for the stator; but I would get the rest of the upgrades done and redo that measurement. It could be that unbalanced loading is what is causing that symptom.