1993 Honda Accord car wont start

  • 2.2L
  • 4 CYL
  • FWD
  • 155,000 MILES

I have a 93 honda accord I aquired it around the summer/fall time of last year. It never struggled to turn on, or seemed "dead." As of late, its been real cold in NJ, so little by little I have noticed my car sturggles to turn on, and when it finally starts, the rpms will start on around 1, and then just graduately fall to zero and stall. When I start the car again it will stay on and act fine for the rest of the day. So it seems to be only the first crank of the day. A couple days ago it snowed and I needed to go outside but my car was completely dead. I was able to get a jump and again the car stuggled to start but turned on. I went to get gas, so I had to turn the car off and when I tried to turn it back on, it wouldnt. Again I needed a jump. I got one and drove home. I parked up and let it run for a little bit. Turned it off and when I tried to turn it back on. It was dead. Today I tried to jump my car because I havent been able to really look into it because I have been working, but I tried to use my wifes camry to jump it and it sounded like it was struggling to turn on, but then just went completely dead, and started making clicking noises every time id turn the key to try and jump it. Maybe because it is extremely cold outside and im only having this issue when its cold. Im really tight on money and dont want to go spend money on a battery when it turns out to be something else. I have also heard that if I do jump the car and let it run if I remove the battery connections and the car conltinues to run then its the battery if not its the alternator. I was wondering how true that is and is it okay to do to a car. If you were in my place what would you do?

Do you
have the same problem?
Tuesday, January 7th, 2014 AT 8:10 AM

1 Reply

  • 26,445 POSTS


Tell the person who told you that to stop telling that to people or he will be responsible for a lot of destroyed computers. That was a trick done many years ago by mechanics who didn't understand how these simple circuits work or how to diagnose them, and they did it on cars that didn't have computers. Today a mechanic might get one verbal warning if he is caught doing that, but for sure he will be fired the second time. It's really that big a deal.

AC generators, ("alternator" is a term copyrighted by Chrysler when they developed it), put out three phase output, and as such, there is a lot of "ripple" voltage, meaning it varies by a couple of volts hundreds of times per second. The battery is the only component that smooths that voltage out and makes it steady. Without the battery in the circuit, if the voltage regulator responds to the high points in the output voltage, it will cut back on how hard the generator is working and the engine may stall. That would falsely tell you the generator is defective, so what good is that test? Most of the time the regulator responds to the low points in the output voltage by bumping up the output. Then, that higher voltage goes right back into it to run the electromagnet which gets stronger, so you get a higher output voltage, and it's a vicious circle. I did this every year for my students to show them what can happen when there's no battery in the circuit. It is real easy for system voltage to reach well over 30 volts. That will destroy every computer on the car and burn out any light bulbs that are turned on. The only thing that might save you is all generators are very inefficient at low speeds. At idle the engine will usually just stall.

It is also possible to have one defective diode of the six inside the generator. That will result in a loss of two thirds of its output current capacity and ripple voltage will be really high. 30 amps from the common 90 amp generator is not enough to run the entire electrical system under all conditions, especially if you have the head lights or heater fan on. The battery will have to make up the difference until it runs dead. That can take from an hour to a few days.

The proper way to test the system is to measure battery voltage with an inexpensive digital voltmeter. With the engine off it will read 12.6 volts if it's fully-charged. If it's near 12.2 volts, it's okay but discharged. Suspect the generator. If you find around 11 volts, the battery has a shorted cell and must be replaced.

Next, with the engine running, the battery voltage must be between 13.75 and 14.75 volts. If it is low, suspect the generator or voltage regulator, but if it is okay, the next half of the test requires a professional load tester. Your mechanic will test the generator for maximum output current. If it has a bad diode, you'll only get one third of its rated current and ripple voltage will be very high. It's not practical to replace blocks of diodes. Just replace the generator. With a bad diode, very often you'll find the battery voltage is a little above 14.75 volts. That is a clue but the battery is still the key component in helping the voltage regulator hold system voltage down to a safe level.

The battery must also be professionally load-tested. Just because a cell hasn't shorted yet doesn't mean most of the lead hasn't flaked off the plates. That happens naturally over about a five-year period, and that lowers the amount of current it can deliver to the starter. The battery must maintain 12.6 volts, but it must also be able to deliver enough current to the starter.

If you find you need a new battery, there's two places you can get one for half the cost of a tank of gas. Salvage yards typically sell one to three-year-old batteries from smashed cars for around 20 bucks. In my city we also have a battery store that sells unadvertised "reconditioned" batteries for $25.00. These are brand new but typically have been sitting on the shelf for more than a year. Those are all I buy for all of my vehicles. You won't get a warranty but you'll pay one third the cost of a battery from an auto parts store.

Was this
Tuesday, January 7th, 2014 AT 1:51 PM

Please login or register to post a reply.

Related Engine No Start Questions