Starter circuit problem

[Laneman950]

I think I may have found the problem. I did a continuity test on the wire going from the ignition key to the solenoid. I had my wife move the wire around and I was getting intermittent continuity. I repositioned the wire temporarily and it started ten times in a row even while hot. It's never passed this test yet. I think the ignition wire has a break or loose crimp in the copper near the terminal. I'll crimp on a new terminal and post the results.

 

[john_bud]

I think I may have found the problem. I did a continuity test on the wire going from the ignition key to the solenoid. I had my wife move the wire around and I was getting intermittent continuity. I repositioned the wire temporarily and it started ten times in a row even while hot. It's never passed this test yet. I think the ignition wire has a break or loose crimp in the copper near the terminal. I'll crimp on a new terminal and post the results.

Spend the 1/4 cent and solder the terminal on. Then paint on the liquid electrical tape and slide a shrink wrap tube over all that. Wait an hour or two and then shrink it with a match or heat gun.


I HATE crimp connections as they uncrimp at the worst times.


jb

 

[gemini5362]

I might differ in the above advice just one little bit. take off the terminal. Slide some heat shrink tubing over the wire and get it as far as you can from the terminal end. Put the terminal end back on solder it in place then slide the heat shrink back over the terminal and use a hair dryer or something like that to heat the heat shrink.

 

[Laneman950]

Thank you, soldering didn't even occur to me. I think that is a good idea.

 

[patrick_g]

Still having trouble after starter repair. What I haven't done is replace the neg cable (which looks and tests good), the ignition switch, the starter brushes. Cables have almost no resistance, no corrosion, pos cable is new. Any advice is welcomed.

A couple feet of number 28 or 30 gauge wire will test on your ohm meter and show essentially zero resistance. OF course it will burn up in a fraction of a heartbeat if you put it in series with your starter current. Testing battery cables to see if they are good for high currents, double or triple digit amps with a standard ohm meter is: 1. almost a waste of time and 2. misleading.

Say for example that your peak cranking draw is 100 amps. Lets say your cable has 0.1 ohms. Ohm's law (E=IR) shows us that a cable passing 100 Amps and having a resistance of 0.1 Ohms drops 10 volts. Taking a very simplified look at the situation... the 2 volts left over won't crank your starter very well.

OK lets say you would have noticed 0.1 ohm on your meter.. Would you have been alarmed (assuming you would even notice) if the cable measured 0.01 Ohms? At 0.01 ohms you would still loose a full volt in the cable. Many folks don't have meters that accuratly measure resistance in the area of a small fraction of an Ohm.

A better way to measure things for trouble shooting purposes is under load, passing a realistic current (not the microamps of the ohm meter AKA Ohm Eater.)

If you measure the DC volts between the battery post (not terminal) and the starter terminal (the threaded screw, not the terminal on the cable) and then try to crank you get an idea of the actual loss in the cables and connectors. If the voltage is more than a fraction of a volt then something is providing too much resistance. The resistance can be in the battery clamp to battery post contact, the terminal to wire crimps (if any), corroded cable, or whatever. If you are not delivering nearly full voltage to the starter it will not crank correctly. Now please note that "full voltage" is NOT 12.6, 14.2, or whatever. Under the starting load the battery voltage sags considerably. You need to measure the battery voltage under load and see what percentage of that makes it to the starter.

Digression time.. a brief walk down memory lane... Back in the olden days some 12 volt cars had ballast resistors in series with the primary side of the ignition coil which reduced the voltage to the coil to about 10 volts. When cranking the starter, the ignition switch shorted out the ballast resistor to give the coil FULL VOLTAGE. Of course while cranking full voltage was about 10 volts. When you released the key it went back from the start position to the run position and the ballast resistor was in circuit again preventing the 10 volt coil from burning out running continuously on 14.2 volts.

IT is a well established fact that the battery voltage needs to be measured under load. If crankiing and measureing at the same time is a hassle (it often is) or you are concerned that al this crankinig is not good for the starter (it isn't) then for maybe $25 at Harborr Freight you can buy a battery load tester which can be substituted for the starter as a battery load while making tests. Some guys use and advocate the use of the headlights on high beam as a load test but there are shortcomings to that: 1. the lights aren't wired to the starter cable and doing so is a pain, the load choices when using the lights are just two (high or low beams) and neither is anywhere near the load of a starter or the load tester.

Readers Digest version: Resistance readings may lead the unwary astray. Use voltage readings under an appreciable load to make determinations about the various components' ability to carry current to the starter. If the T-shooting takes a while use a battery charger on the battery to keep it up.

If you want details on any specific issue or for any other reason you can PM me.

Pat

 

[SPYDERLK]

Much of what Pat says is absolutely correct. He is dead on regarding the need to use a voltmeter to determine the potential across a current carrying wire to find out how efficiently it is conducting. Since it must push current thru the wire and measure it accurately, an ohmeter on its most sensitive scales has grave problems with contact quality to the wire being tested. A fraction of an ohm corruption of the contact made by your test leads comes back as that much error in your reading. A sensitve voltmeter draws a much smaller current than that used by the Ohmeter. This is due to the fact that the Vmeter has several megohms input resistance. The small but inconsistant contact resistance of your leads to the circuit when using a Vmeter makes very little difference when grouped in proportion to the millions of the meter.

With a sensitive Vmeter the idea of appreciable test current can be reduced to values that may seem ridiculously low, without losing real world accuracy in the tractor testing realm. Literally, the lights are plenty of draw, and are ideal if power for them comes from a connection made at the solenoid input terminal. With a low drain, say 5 Amps, from this terminal to ground you can leisurely go about your diagnostics with a Vmeter. I have a superbly cheap digital VOM I got from Horrible Fate for $2.99 as I recall. It will discriminate down to 0.1mV. So you turn on your 5A load to check your batt cables. Touch your sharp pointed probes to each end of a cable - not the terminal lugs - jam the points into the cable. Perhaps you get a 1.0mV reading. That cable will conduct 5000A with a loss of just 1V --a very good cable for a dozer even. Next you check from the battery post to solenoid post... or the frame, depending which cable youre doing. These points show a difference in potential of 100mV. TROUBLE. 50A will lose you a Volt. Check from batt post to clamp... 0.4mV -good. Clamp to adjacent cable wire... 20mV -to be fixed. Cable wire to terminal lug... 5mV -wish it was better. Terminal lug to frame/ or solenoid post... 73.6mV -the killer. Took a minute or two. Buy a digital meter. Your starter will thank you.
larry

 

[patrick_g]

Larry, I bow to your superior breadth of experience. I do not know from personal experience what proportion of things you might want to test in a battery/starting/charging/and-everything-else-electrical-on-a-vehicle can be tested by turning the lights on and off. I have had some experience with situations that weren't ferreted out with the draw of the lights but succumbed to a variable load tester when the load was cranked up.

I have been unfortunate to have had some weird problems that may not be at all representative of typical failures experienced by most folks. You know how it goes... behind my back whether I know it or not mother nature is laughing at me along with Murphy!

Like nature abhors a vacuum, I feel that way or worse toward intermittents.

Soldering automotive wiring connections that are subject to weather or vibration is a good idea. properly crimped connections (certicrimp tool or similar) are good enough for aircraft work in a vibration environment but not everyone REALY knows how to crimp properly and the cheap Chinese terminals so common these days aren't the best (nor are the cheap crimping tools.) Soldering seems to be a dying art as well. not many folks really know how to do it right. I have seen lots of garbage joints that were either cold joints insulated by barely melted flux or overheated joints that embrittle the copper wire so it breaks much much easier.

Assuming you get a good joint, then protecting it from the elements is less important than protecting a crimped joint from the elements but is still a good idea. If you are really **** about protection there is heat shrink that oozes a sealer when heated and makes for a waterproof hermetic seal of the protected joint. You can DIY with silicone rubber. Don't heat the shrink tubing and then put on the silicone rubber afterward. Instead squeeze the silicone rubber into the space between the tubing (before it is shrunk) and the joint to be protected. Then shrink it.

If you buy a trailer or have after market accessories installed such as a trailer light/brake plug-in or such you can be assured the odds are that everything was wired using insulation displacement squeeze connectors that will corrode and fail and can produce intermittent brake and lights on your trailer or malfunctions of the arftermarket equipment. It would head off future problems if you redid their chintzy connections and soldered or used proper crimp connections and water proofed them.

Pat