What should a good battery voltage be?

Been trying to diagnose an electrical problem for the last couple of week and I'm down to thinking it's the battery (it's about 2 years old ). I've got an '01 SV650S and first it died getting into work one day (instrument lights blinked first and then the engine died just as I came to a stop). Battery was dead flat so fully charged it and ran through the tests in the Haynes manual. No-load unregulated voltage on the coils was 65V+ on all 3 pairs, but the regulated voltage over the battery was only 12.2V and didn't increase much with RPM. So thinking it was the reg/rec I bought a used one off eBay. With the replacement reg/rec the voltage does increase a little with RPM but at 5k it's still only about 12.8V so still falls short of the 13.5V to 15.5v as per the book. Decided to try riding it without an overnight charge and made it to work and back but shortly after starting out the next day it died the same way as before. The leakage test as per Haynes indicates no leak (short) in the electrical system. After a full charge I get 13.2V initially and then in a few minutes it drops to about 12.5V. Haynes doesn't tell much about how to test battery condition, it just says that if it drops below 12.3V it needs to be removed and charged, but I've read on this forum that a tired battery could cause a low reading, so I'd be grateful if anyone could tell me what the voltage should be on a good battery or how to test the battery condition. Or maybe I'm on the wrong path entirely and someone has run into this before and can point me to the likely culprit!

[timwilky]

Reads like somebody sold you a duff R/R

[Spanner Man]

Good morning.
Have you tested the alternator for earthing? Using a multimeter check that none of the three yellow alternator output wires go to earth, resistance should be infinite.
If your alternator tests ok it's probably down to the rec/reg unit or the battery. Do you have a mate with whom you can swop components with? As electrical components are notorious for testing ok but letting you down when you actually require them to do their job.
One component often overlooked when diagnosing charging problems is the starter relay. As stated, substitution is the best way of eliminating it.
Rec/reg units on SV's are a bit of a weak point, so it's possible that your secondhand unit has the same problem as your original one. For a decent affordable unit try Electrex on 01491 682369. Their units are brand new, and come with a 12 month guarantee. The SV unit is around £70 inc VAT.
When all is functioning correctly, you should see at least 14.5 volts across the battery at 4-5 thousand rpm with the lights on.
A quick battery drop test is to fully charge it, & with a meter clipped across it start the bike. It shouldn't really drop below about 11 volts. You have to be watch carefully for as soon as the engine starts obviously you have to release the button. If it drops to say 8 volts on starting it's had it!
N.B. Some multimeters do not respond quickly enough for this test.

Cheers.

[johnnyrod]

Definitely do the earth test (thoughusually the alternator's burnt out if you get earth continuity and yours sounds fine at last check). Otherwise duff RR.

Did the alternator test tonight and all 3 showed infinite resistance so we're ok there. The start test showed a drop in battery voltage to about 10.9V when hitting the button so I think we're ok there too. I guess that only leaves the R/R and the wiring. I don't have a mate to swap parts with so I'll give Electrex a call and see if they have a R/R in stock and bite the bullet for a new one (should have done that the first time ).

Thanks to all for the suggestions. I'll post the outcome here when I get the new R/R.

Cheers!

[Spanner Man]

Good morning.
It sounds as if your alternator & your battery are ok, which as you state only leaves the wiring. as a last test check for continuity between the red output wire on the rec/reg & the terminal on the positive battery lead, any significant resistance would indicate that possibly the starter solenoid is suspect.
Do this with the rec/reg & the battery lead disconnected.
Electrex usually have units in stock, & are pretty quick on delivery. Let me know how you get on.

Cheers.

Well, after doing the continuity check for suspect starter solenoid and that checking out OK, I ordered a new rec/reg from Elecrex. It got lost for a week en route but is now installed but I don't think it has solved the problem. At 5k rpm and under full load the voltage still only reads 12.4V max. I'm starting to think it may in fact be the battery even though it holds a charge and tests out OK. Any other ideas? I'll try riding it in the mean time and see if the battery is at least holding a consistent charge. I'd like to be able to test the charged level but don't know of a way except to observe how long it takes to recharge to the float point.

From your description something def not right. When charging from alternator my voltage across terminals was around 15v. Some web stuff I'd been keeping which may assist.

HTH

martin

PS Tried to format the columns below better but the forum engine wont have it!!


From YUASA's battery charging page
5ball hydrometer Sp Grav. Voltage

4/5 Balls Floating 100% Charged 1.265 12.60v
3 Balls Floating 75% Charged 1.210 12.40v
2 Balls Floating 50% Charged 1.160 12.10v
1 Balls Floating 25% Charged 1.120 11.90v
0 Balls Floating 0% Charged < 1.100 <11.80v

From another page on batteries:

Battery Charging Stuff 2
For those of you that are in a hurry, a rule of thumb is "terminal voltage 12.5 V or higher, and density in all six cells 1.24 g/ml or higher means that the battery is very probably in good shape".

Charge state I: Voltage A fully charged 12-volt lead-acid battery has a voltage of 12.6...12.8 V at 20 °C. The exact terminal value will depend on the individual construction, in particular the alloy of the electrodes - Pb, Sb, Sn, Ca - and the concentration of the acid used. The voltage for sealed lead-acid batteries - SLA - is usually at the upper end of this range.

On the other hand, a battery is considered "empty" if the voltage reaches 11.6...11.7 V (same for SLA). In this state, the battery will still be able to deliver quite some current, however, the voltage will drop rapidly - i.e. the "internal resistance" is very high. If you don't recharge in this state, the battery "sulfates", and this will affect its performance.

Thus, one of the simplest approaches to determine the charge state is the open-circuit voltage of the battery:

First remove the "surface charge" that builds up on the plates - if you do not remove the surface charge, you will measure values that are somewhat higher and thus falsify the result. For car batteries, the usual procedure is to either turn the headlights on for 3 min, or to crank the engine for some 10 s (this draws some 100 A). For a motorcycle, I suggest to leave ignition and headlight on for about 1...2 minutes (depending on its battery size).
Then, use a good (read: accurate, calibrated) digital voltmeter and measure the voltage at the battery poles.
12.6 V or better (at 20 °C) is fully charged, and every 0.2 V less means roughly 25% less capacity. Thus, 12.2 V is only 50% charged, and 11.8 V or less is an empty battery. - For SLA, use an upper value of 12.8 V and count approximately 0.25 V per 25% step.

Originally Posted by martin_SV

From your description something def not right.

I agree. But here's the results from my battery tests (with the new R/R). After removing the "surface" charge the voltage stabilized at 12.75V. After starting the engine the voltage dropped to 12.1V running at idle. At 5k RPM it increased to 12.3V. After 10 minutes of running the idle voltage was down to 12.06V and 5k voltage down to 12.24V. After stopping the engine and turning off the ignition the voltage recovered to 12.6V. Turning on the ignition at this point drops the voltage to 12.0V. I don't know if that's normal, but it seems odd that it only increases to 12.1V (idle) to 12.3V (5k) with the engine running and a brand new R/R installed. At this point I'm at a loss to know what could be causing this.

SV speedr, I just typed a really long reply and it's disapeared!!. I have some ideas, and am PM'ng my contact number. Ring me anytime,
martin