Cougar

I would have a shop do a load test on the battery to see what kind of shape it is in.

I suggest you charge the battery up and go from there.

I would check the compression of each cylinder. This will tell the story of what needs to be checked.

I wasn't saying the batteries were bad, though they can be damaged if things are not working as they should be with the alternator. I meant that if the alternator isn't charging the battery as it should then the battery can't provide the amount of stored power it would normally have. Since you have gone through so many batteries in a relatively short period it is pretty obvious that something is not right with the charging system. I suspect the alternator has some bad diodes in it. Set your meter to measure AC voltage and place your meter probes across the battery posts. Check to see if there is any AC voltage getting to the battery while the engine is running at around 2,500 RPM, check the DC voltage also. You should have less than .1 volt AC and around 14.5 volts DC going to the battery. If you measure more than .1 volt AC you need to replace the alternator. You may also want to get a load test done on the alternator to see if it can handle a high current load.

I agree with OB99W. The battery may be weak due to age or it may not be getting a full charge from the alternator due to a problem with it. I recommend you get the system load tested at a shop and and see what they say about it.

The problem could be with the alternator but I am betting the problem is connected to the lead that feeds the main power panel. The simplest way to find the trouble source is to place a ammeter in series with one of the battery leads and then start disconnecting things, like the fuses one at a time, to see which one makes the current drop.

Excessive ac noise from the output.

There may be some broken wires in the door jamb that is causing that to happen. When the door is closed, then the wires are making connection.

Since you had the timing belt replaced within the last 10k miles it seems to me the belt should be ok unless something else in the loop caused it to go bad.

Thanks for the update R2.

If the tach is still going wild after the disty replacement then I suggest you check the alternator as the source of the trouble.

The down side of disconnecting the battery for an extended time is the ECU losses its' memory. It will have to relearn the engine characteristics each time. If there is an IM test due on it there may not be enough monitors set to qualify it for testing. The surest way to check the current draw on the battery is to place a ammeter in series between the negative battery post and the ground lead. An amp-clamp will work also but you need to make sure you are on the correct lead to the trouble.

Glad you figured it out and got it back together.

As stated, see if there is some small holes that you can fit a wire into to hold the brushes back until the armature is seated again. Things like this happen at times but I'm glad you are learning on how to work on things like this. Before long, you will be able to this job in your sleep.

Looking at the results of the testing it seems to me that the battery is good. In fact, the voltage spec shown while starting seems too good to be true. I would have expected the voltage to be around 11 volts or less but that is the results I guess. I have to wonder about the drain test results because it doesn't seem correct to me unless, the test was done on the main alternator lead or the starter lead. These two leads should show no current flow with the engine off but, the battery lead running to the power distribution panel which is usually around a 10 gauge wire tied to the positive post, should show some current drain normally and this is the lead that is most likely tied to the trouble spot and carries the current to all of the accessories on the car. The testing results makes me think they didn't test that area. I would have the shop that did the testing check the lead running between the battery and the power distribution panel do another current check on it. You should normally see at least a 25 milliamp draw on that wire and no more than 80 milliamps when systems go into the 'sleep' mode. If the drain is killing the battery overnight then you will see something significantly higher. Hopefully they won't charge you for this additional test since they should have caught this the first time. To be fair, they may have really done the testing on the correct wire and then just compensated the results by subtracting the normal current draw. I don't know, but since you seem to be still having trouble and if the trouble isn't really due to something else wrong, I think this is what is going on.

I recommend you make your splices in an area away from where the flexing takes place. Add enough wire to do that. If you can get some pieces of some teflon aircraft wire to make the splice that should work well.

Glad you got the windows working Duner and good job. Thanks for the feedback.

Oops! Sorry about that. I didn't look close enough. I think there is a set of 4 manuals for the Loyale. My '88 version has 6 sections covered in 4 manuals. I think they will cover most of the things you need to know without having the supplements.

Looking at some info on the AC system it shows a grn/red wire tied to the return side of both fan relay coils and it should tie to the ECU on pin 73. This may be the same wire you are using. By tieing it to ground at least, both fans will turn on. Your meter shows 12 volts on the wire while it is open but it won't be at 12 volts when it is terminated. When you tied the low resistance of the lighter heater element in series with the relay coils it will still allow the relay coils to close. By pulling the lighter out of the socket you disrupted the return path of the relay coils and the fans turned off. I didn't see any info on the blk/wht wire. I would use an ohmmeter to see if that wire is really ground. It should be very close to 0 ohms if it is a true ground lead.

Here is a link to a set of manuals you may want to purchase. They will pay for themselves in no time. http://cgi.ebay.com/ebaymotors/1991-Subaru-Legacy-Service-Repair-Manual-Set-Vol-1-6_W0QQcmdZViewItemQQ_trksidZp1742Q2em153Q2el1262QQcategoryZ6762QQihZ001QQitemZ110182538520#ebayphotohosting

If you suspect a bad ground wire you can test that out by placing a voltmeter across the suspected bad ground point and a known good ground. Then move the switch to the UP position and see the voltage goes up on the meter. If it does then the suspected ground is bad. If you see no voltage then the ground is ok.

I suspect that the black wire isn't going to ground directly. Try tieing the lighter shell to a known ground on the chassis and then see what happens.

"Well done young Skywalker. You are a true Jedi of the Subaru."

After looking over my data (a FSM for a '88 Loyale) I think the most likely spot for this trouble is in the main switch. I suspect that the power lead inside it has a connection problem in the up position so Msteel's comment about a bad connection may be correct. You can prove this by jumpering a couple pair of wires if you want to. First, roll down the front passenger window about half way. Remove the main connector to the driver's window switch control panel. The main connector has two large wires that supply battery and ground (red and black). The wires going to the front passenger window on my data are shown to be white/blue and blue/black. By jumpering the power and ground wires to these wires you should be able to make the window go up and down. By tieing the red wire to the w/bl and black wire to the bl/blk wire you should see the window go up. If that works then the main switch is the problem. As far as manuals go I recommend you try to find a FSM that covers the wiring at least. I think they made them until '91. There were updates after that I believe.

The dimming dash lights could be due to a bad ground between the dash area and the chassis. Also check the bonding to the chassis at the battery. I recommend you clean these connections with a soft wire brush, even if they "look" ok. This will eliminate them as possible trouble spots. The problem could also be due to bad connections at the fusible links. Make sure the connections are clean and snug. The poor grounding could be causing the engine trouble. Others have cleaned the grounds and reported the engine problems cleared afterwards.