Intermittent starting problem

You say that the battery tests good. Is this on a load test or howdid you test it ?

One of those $20,000 computerized testers at the dealer.

Test code 9mg274-881455-80 whatever that means.

I insisted that the battery must be defective because the truck starts with the jumper, and they replaced the battery even though it tested good. No problems for 2 weeks and now it has not started 2 days in a row.

Stumpy posted for all of us...

And I know how to SNIP

How does the battery test now? Maybe the alternator is not up to the job or a ground/bonding strap is loose, corroded or missing.

The battery has a full charge. This is acting like a low charge situation, but I don't think it is. All the cables and wires look clean and solid. This is still a low mileage vehicle.

I am considering just replacing things and crossing my fingers. If I give it to the dealer, they only get one shot every 24 hours for it to fail.

How was the battery tested? I'd bet on a bad battery or a bad connection at the starter if a boost starts it. My suspicion is a thermally sensitive bad interlink connection if the battery connections have been properly cleaned and tightened. They can LOOK perfect and still be bad.

It was a computerized load test sequence. When it's done they hand you a little printout that says "bad" or "good". Mine lists:

12.85 V 772 CCA Rated 710 CCA Temp 74 deg.

Would you get a thermally sensitive bad interlink connection at 40 - 50 deg.? It's not even freezing.

I just noticed the time of the test is 8:30 AM. My non-starts are all before 7:00 AM

I've had batteries that 10 degrees C made all the difference in the world - and the Midtronics tester could not find it.

If a midtronics type tester says a battery is bad, it is. If it says it is good, there can still be problems. 2 batteries with the same problem makes it less likely - but not impossible. Remave and clean all the battery cable connections and torque them properly on re-assembly.. The other POSSIBLE problem is high resistance on the control side - which USUALLY causes problems on HOT starts, not cold - but stranger things have happened. Can you get a remote starter button and connect it to the solenoid connection and battery cable? See if it starts better that way. There is about 16-20 feet of thin wire, as well as the ignition switch and safety (clutch or neatral) switch, as well as several connectors in that circuit - and that starter solenoid draws a couple of amps - so ANY voltage drop can affect the starter operation. Used to be a fairly common problem on the TE5 and TE7 series Corollas. I used to put a starter relay from a late sixties Mopar across the solenoid, and trigger IT with the factory wiring. Solved EVERY ONE of those (in this case hot) starting problems.

If the remote starter switch does not make it start, that is NOT the problem, and you COULD have a solenoid contact problem or a sticky brush problem on the starter. The contact kits are fairly readily available now compared to back in the seventies and could be worth while

It might be worth getting one of those remote starting switches. Never looked at one before. It's been a long time since I had to stick a screwdriver across the solenoid contacts, ~1984.

Off the wall ideas.

1) bad/intermittent connection between the negative battery terminal and either the block or the chasses.

quick check connect a jumper cable between the block then chassis and the nag term of the battery.

2) partial short/open in one or two legs of the alternator rectifier system.

I've double-checked battery connections and ground wire.

The alternator/rectifier is keeping the battery charged.

I think I want to make a remote start switch just because it's cool. If Radio Shack doesn't have a 10 amp momentary switch, O'Reilly has one for $16. Looks like I could use it to launch a rocket or something.

I made a remote starter switch with an O'Reilly 20amp switch, some heater cord and jumper clips. Wore myself out crawling around on my back, then jacked up the front left side so I could stick my arms in the side over the wheel. Broke 2 of the grommets on the rubber mud shield. I thought that I could semi-permanently install the homemade switch as a backup on top of the existing wires. It won't work. Toyota has those plastic connector covers to protect the terminals.

I'm gonna leave it at the dealer. This is too much of a pain in the ass to do myself. Too bad, because the starter pushbutton duct taped to the dash would have been an interesting custom feature.

Still does not preclude a partially functional alternator and more importantly one that can discharge a battery when the engine is off.

EXTREMELY easy to check that, Connect voltmeter set to AC volts - should show less than 1 volt.

The battery has a full charge when it is parked and still has it when I try to start it the next morning. I keep an analog meter installed on my dashboard to be sure.

I wimped out because the access was uncomfortable. I don't work well on my back or on my knees. Let Toyota fix it - paid someone else to solve my problems.

$329 for a remanufactured starter $200 for labor $30 to keep the old core

At least the service manager threw in a loaner car. The contacts in the solenoid compartment didn't look all that bad, but it was full of black, sooty residue. I'm going to get one of those $30 starter rebuild kits and keep the old starter as a spare. If I need it in the next 6 years, I'll consider it a $60 backup.

For some reason this starter required more current than the design, which mimicked a low battery when cold, worked OK after the first start of the day.

I don't see the logic of your test.

A bad diode will put ripple on the DC. The ripple is read by the AC Voltmeter. It is a test that goes all the way back to 1960 and the introduction of the alternator on Chrysler products at that time.

Might then again might not put a measurable ripple on the DC much depends on the load and the battery condition.

Well, in over 35 years I have yet to find one with a bad diode I did not catch, or a good one I called bad

snipped-for-privacy@snyder.on.ca posted for all of us...

And I know how to SNIP

Don't confuse him with actual logic and knowledge...