It's Alive!

It's been almost a month since the engine has run when I drove the car home. During that time I replaced the push rod guide plates which necessitated adjustment of the rocker arms. When the pulled the spark plugs I found one the terminals to be severely corroded. So I installed new wires and plugs gapped to .035" (the old plugs were all at .045"). While I had the plugs out I was able to crank the engine over freely so I adjusted the floats on the Holley carb as best I could. I also set the idle mixture screws at 1 1/2 turns out.

The engine started right up after just a few cranks but wouldn't idle at all. I had to crank the idle stop screw to raise the rpms up to about 1200 or else it would stall. I unplugged the vacuum advance and checked the timing. The balancer is only marked up to 12 deg before TDC but the timing light showed much higher than that, I'm guessing 20. So I backed the timing down to 12 and adjusted the idle mixture. The engine ran best with both screws out a total of 2 1/2 turns and I was able to lower the idle down to 900, but the idle was still very rough.

I hooked up a vacuum gauge under the carb and discovered the engine was only pulling 6 inches at idle. Very poor. That explains why it won't idle smoothly. I'm not sure what to do at this point. I'm pretty sure there's a vacuum leak somewhere. Hopefully it's under the carb and not the intake manifold. And I'm still not confident the carb is performing correctly. With the low vacuum I'm not getting enough airflow for the carb to meter the fuel properly.

UPDATE:

I decided to checked the intake manifold bolts and boy was I surprised. They weren't loose exactly but they weren't torqued any measurable amount. So I torqued them down in stages to 20 ft/lbs. With my fingers crossed I started her up. The idle did seem a little better but I still only measured 6" of vacuum. I was more than a little disappointed since I thought I had fixed the leak. Now I'm pretty certain the manifold must have been leaking before I tightened the bolts, and I'm guessing it must still be leaking. I've ordered new intake gaskets so stay tuned.

Instrument Cluster

I had removed the instrument cluster a while back to make it easier to access things under the dash. I learned my lesson with the turn signal switch so I made sure to diagram all the wires.

It looks to be 12 wires that need to be disconnected before the cluster can come out. I really wish there was a single connector, or maybe even two. Maybe a future project...

Here's the back of the cluster in case you were curious.

And here's the cluster back where it belongs. All the gauges are AutoMeter behind the stock bezel. The gauges sit in a plastic gauge pod that's made for just this purpose. Unfortunately the speedo sits a little too high and is not adjustable. A lot of the online reviews of these pods state the same problem. Oh well.

Let There Be Light

On my one and only day driving my car I discovered several issues with the turn signals. First, the stalk was very loose even though it was tightened all the way. There was too much free play and the detents were weak making it hard to tell if it was centered, left, or right. The second issue made this worse -- the left and right indicators on the instrument panel didn't work. And lastly, turning the steering wheel made loud clicking noises that didn't sound good. I almost forgot, the horn didn't work either.

Here's the old switch with the wheel removed. It turns out the nasty noises were coming from the horn contact plate. You can see the yellow wire is covered with copper filings.

Here's a closeup off the plate. You can see the inner contact ring is worn through while the outer contact was actually hitting the solder joint. Unfortunately you can't buy just this plate so I had to buy a whole new wheel mounting kit.

This is what I found when I crawled under the dash to disconnect the turn signal switch from the harness. Whoever installed the previous switch didn't bother to reuse the other half of the connector and just pushed the wires in. More than one wire fell right out. This is probably why the dash indicators weren't working. Luckily I was able to find the correct matching connector online.

Here's the new switch in place. I know, looks like the old switch.

Before I took the old switch out I found a wiring diagram online that showed how to wire the two harness connectors, a 6-pin and a 2-pin. Like an idiot I didn't bother matching the old wires to the diagram. To make matters worse, when the new switch arrived it had different colored wires than the old one. I figured I could just compare the two switches and be able map the new wire colors, but when I went back to the diagram, half off the wires on the old switch didn't match. Let the fun begin.

So, I then spent several hours trying to connect the new switch. Some of the wires are hot all the time so I had to be careful not to fry something. Thus, every time I tested the connections I had to disconnect the battery, crawl under the dash to move the wires around, reconnect the battery, climb back in the car to turn the ignition on and flip the turn signal, check the front signals and then the brake lights to see my if I had it right.

I probably had the wiring correct after just a few tries. But no matter what I tried I couldn't get the rear lights to work. So I kept moving the connections around. After a while I got frustrated and decided to take a different approach.

First thing was to look at the wiring diagrams. Right off I noticed that the one of the wires from the switch went to the turn signal flasher and in turn to a terminal on the back of the ignition switch. Then it hit me -- I remembered finding a loose wire on the back of said switch. I didn't tighten it at the time because it's very inaccessible and I was busy with something else at the time. This sounded promising. So I climbed back under the dash and tightened up the connection.

But when I tried the turn signals again, everything worked except the rear lights. So I used a multimeter to check the voltage coming off the switch for the two wires that I thought were for the rear lights. While the turn signal was activated I could see the voltage cycle on and off. Bam. So at least I had the wiring right. Then I went back in the trunk and checked the connection back there. Again, I could see the voltage cycling on and off. I turned on the parking lights and the license plate light, which shares the same wire as the tail lights, came on. However, the tail lights were still dark.

So why don't the lights come on? The tail lights are totally custom and are sealed to the body. I know, who does that? I have power going to the lights but no light. The housings are plastic so maybe the ground is bad. But the only way to see what's going on inside the lights is to cut them off. It's strange that the tail lights stopped working after I took the turn signal switch apart. I don't see how I could have fried something by hooking up the wrong wire. 12 volts is still 12 volts. I'll have to investigate the grounding for the lights, hopefully without removing them.

Anyway, since the switch appears to be working correctly I buttoned things up a bit. Here's the wheel all cleaned up and back where it belongs.

Third try's a charm. As I said before, the tail lights are solidly sealed to the body. The only wires going into the housings are two sheathed wires.

The white wire (connected to black from the harness) is also tied into the license plate light, which works, so I know that wire is good. The black/red is for the backup lights. One of the sheathed wires carries three conductors. White is for the parking lights and the red and black are for the left and right signals. All three of these test good. The other sheathed wire runs along the left side of the trunk towards the front. I removed the door sill plate to see if it ran through there.

As you can see only two conductors were in this sheath. The black wire is grounded to the sill--so much for that bad ground idea. But what about that red wire? Then it hit me. It must be +12v power. I couldn't figure out how the rear lights could sequence when the turn signal wire was only sending a momentary signal. There must be some kind of "brain" in the rear lights that requires constant power. So I checked the voltage at this splice and got zero volts. That's good. The lights weren't working because they weren't getting power, not because I fried something.

This is where the red wire goes, wedged right into the fuse box. Nice. I had seen this wire before but I assumed it went to the radio or something. It's not even protected by the fuse as it's on the hot side.

Here's the back where you can see the always hot black/yellow feed come in. Not wanting to zap myself, I disconnected the battery so I could check the continuity in the red wire. That's when this happened.

The red wire fell right out of the connector. Bingo! Turned out to be a very simple fix. Just to be safe I wedged the terminal on the other end of the fuse so it would be protected. If you're gonna hack something, at least do it right. I'll look at upgrading this fuse box to something more modern in the future.

And here's the final result result of several wasted hours, all because of a loose terminal. You can see how dim they are, even on an overcast day.

You're Fired!

I decided to pulled the spark plugs to make it easier to rotate the crankshaft while I was replacing the push rod guide plates. It's a good thing I did, mainly because one of them was loose.

Cylinder number 4 couldn't have been firing too well with some of the exhaust gases blowing out around the plug. The entire plug is covered with soot except the part covered by the boot. The rest of the plugs look pretty good with a nice healthy tan color, although 1 and 3 are kinda light. A few of them have fresh oil around the base and threads which means the valve cover gaskets are leaking a little so I ordered new gaskets along with new plugs.

My new push rod guide plates arrived; they're the correct 5/16" instead of the the 3/8" somebody else installed. You can read my previous post concerning the push rod guides here. I set one on top of an existing one to check the fit (yes it's upside down).

Looks much better. If I centered one rocker over the valve stem the other rocker aligned almost perfectly, something that couldn't happen with the bigger guides. Here's what the oversized guides look like:

It's a bit of work to replace the 8 guide plates, mainly because you have to adjust the rocker arms, which in turn requires you to rotate the crankshaft until each cylinder is at top-dead-center. The harmonic balancer has a marking for TDC but that only works for cylinder #1 (and also #6). I needed markings for the other six cylinders.

The firing order for the Ford 289 Windsor is 1-5-4-2-6-3-7-8. Being a four-stroke engine each cylinder fires once for every two revolutions of the crankshaft. With all eight cylinders firing once every 720 degrees of revolution you get a single cylinder firing every 90 degrees. So I needed markings on the balancer for 90, 180, and 270, measured from TDC. The balancer has a diameter of 6.375" (ok, I looked it up) giving a circumference of almost exactly 20". Wow, that makes things easy. All I needed were marks every 5" around the balancer. I even labeled them so I could be certain I had the right cylinder.

Just turning the crankshaft to TDC for cylinder #1 doesn't mean it's in the correct position to adjust the rocker arms. It could be 360 degrees off. To ensure both the intake and exhaust valves are closed the distributor rotor needs to be pointing at the #1 spark plug wire.

Since replacing the guide plates was going to be tedious I set up a little routine so I wouldn't overlook anything.

1. Start at TDC for cylinder #1.

2. Remove the rocker arm retaining nuts and then lift off the rocker arm for the two valves. I set the left rocker arm on the left side of my parts tray and the right one on the right side so they would go back on the original side.

3. Use a breaker bar to remove the rocker arm studs. They must have been installed with thread locker instead of thread sealant. I kept the left stud on the left...

4. Clean the threads on the studs where they go into the head.

5. Remove the old guide plate.

6. Clean the threads in the head (as best as I could).

7. Place the new guide plate in position.

8. Coat the stud threads with Teflon (PTFE) paste thread sealant before hand tightening each stud. When I removed the studs most of them came out with dry threads but a few were wet. You can see in the picture that these studs do indeed screw into a water jacket so thread sealant, not thread locker, is required.

9. Torque the studs to 60-70 ft.lbs. Before tightening I pulled down on the guide so it wouldn't be cocked to one side since there is some play.

10. Oil the push rod end, stud threads, and valve stems in case you wiped them clean.

11. Install the respective rocker arms and retaining nuts.

12. Adjust each rocker arm. Tighten the retaining nut until the rocker arm becomes snug on the push rod. Then add 1/2 to 3/4 of a turn more to set the pre-load on the lifter.

13. Squirt some oil on both ends of both rocker arms as down around the pivot areas.

14. Rotate the crankshaft clockwise 90 degrees and repeat the steps for the next cylinder in the firing order.

The rocker arms on the right have the new plate installed and align with the valves nicely. The two on the left still have the wrong guides. Yikes.

While I was doing all this work I also wanted to check the rocker arm geometry to make sure the push rods were the correct length. This is easily done by coloring the top of a valve stem with a marker, assembling the rocker arm, and cranking the engine a few turns to check were the rocker touches the valve stem.

Above is a closeup off the #1 exhaust valve stem. After I adjusted all of the rocker arms I turned the crankshaft another 90 degrees to put the #1 cylinder back at TDC. When I had first adjusted #1 I had colored the valve stem tip with a marker before assembly. You can just barely see the horizontal line where the rocker arm contacted the valve stem while I was hand-cranking the engine through 2 revolutions. It's not super important that the wear pattern is centered on the valve stem as long as it's close. What is important is that the thickness of the line is as small as possible, meaning the minimum amount of friction. If the push rods weren't the correct length the wear pattern would have been wider.

Holy Cowl!

Early Mustangs are notorious for having rust in the cowl area, usually so bad that rain water drips inside the car and rusts out the floor pans. I can't see any evidence of a leak but then again this car probably hasn't seen any rain in many years. It did drizzle a little on my lone trip but not enough to make me roll up the windows. From the outside the center cowl area looks clean with a texture on the bottom which is probably some kind of sealer, but it's impossible to see the sides where the fresh air "hats" are located and where rust usually occurs.

The driver side vent was working very stiffly and looking up under the dash I could see it had some surface rust. I decided to remove the vent and spiff it up while at the same time get access to the bottom of the cowl.

Here's the nasty looking vent. The rust is mainly at the top were it bolts to the cowl, but at least the rubber seal is still pretty pliable.

This is the bottom of the cowl looking up through the hat. Water is supposed to run around the hat and drain down the side of the car behind the fender, while fresh air blows through the hat down into the vent assembly. At first it didn't look too bad for being 50 years old but when I scraped off the flakey rust it got worse.

Using a small mirror I was able to look on the other side of the hat at the horizontal surface of the cowl. It looks like the the rust is limited to the hat as the flat parts look really good. Fortunately this is a common problem and there is a repair kit available.

These replacement hats fit inside the originals and will never leak if sealed correctly. I will NOT be using the silicone sealer that comes with the kit as it contains an acid to help it cure but can be corrosive to steel. I ordered some body sealer to use instead. I debated whether to cut out the old corroded hat or leave it like the kit instructions say. I think I'll cut it out since it's not going to be a benefit to the repair, and more importantly, I'll be able to seal the rest of the cowl area with it out of the way.

Before I started I coated everything with Corroseal Rust Converter just because it's fun watching rust turn black. Then I used a 3" cutoff wheel in my pneumatic die grinder to cut the hat off. I tried to cut as flush as possible without damaging the horizontal flange so as to minimize any lip that could trap water.

Btw, in order to work on this area, I have to lay on my back with my head and right arm under the dash, my feet in the back seat (I had already removed the driver's seat, steering wheel, and pedals). It's extremely awkward just taking these photos let alone actually working on anything.

Three songs later (about 10 minutes) and it was over.

You can see the rot was much worse on the side I couldn't see (it's upside down). I used a mirror to assess any remaining damage.

This is the horizontal surface on one side where water sits and rust-through would have been a serious problem. Not too bad.

Here's a view looking towards the outside opening of the cowl. As you can see there is very little rust. The next step was to clean and prep everything I could reach before treating it with the rust converter. I took my time here as I wanted to get rid of as much rust as possible before coating the surfaces with body sealer. Then I can seal in the new hats.

I dry fit the new hat before sealing it in place. For some reason that one stud was bent so I had to straighten it out.

Here's the fresh air vent after I spruced it up.

And here's the results of all my labor. I know it's not much to look at, but you can see the vent in place. I'll eventually get around to doing the passenger side cowl, but since it requires removal of the heater assembly I'll wait until I install the A/C.

Rust Never Sleeps, or does it?

When I treated my brake and clutch pedals I followed the directions closely. First, I brushed off all of the loose rust with a wire wheel in my drill, then the wiped them down with some diluted Simple Green.

Here are close ups of the two pedals before applying Corroseal.

And here they are after one coat of Corroseal before I painted them.

With the pedals fitted in the new support bracket I noticed the brake pedal stop (the part with the rectangular hole) was hitting the bolt for the clutch pedal stop. Somebody had bent it over so it wouldn't rub on the side of the old bracket, which it was sure to do with the crooked shaft.

I was curious about how the rust conversion would work with little to no surface prep and I had the perfect test piece, my old pedal support.

On the clutch side, with the elongated hole, I spent a couple of minutes with a wire brush and then wiped off the "rust dust."

On the the brake side I left all 50 years of patina in place.

The application of the rust converter was a little more difficult since the surface wasn't nearly as smooth as it could have been had I used the wire wheel. I also noted that the cup I was using to dip my brush was becoming contaminated much more quickly. After one coat here are the results.

Not bad for maybe 10 minutes of effort, but still not acceptable. Some of the rusty areas were fully neutralized but a lot of rust still showed. So I applied a second coat which went on very easy and only took a few minutes.

Now we're talking! To really put this rust converter to the test, I'll top coat half of it with spray paint and then leave it outside for a while and see what happens. Stay tuned...

Clutch Support

Déjà vu. This is the clutch pedal shaft where it go through the support bracket. My previous Mustang back in 1980 suffered from the same problem as Ford used an inexpensive pot-metal bushing with a plastic liner that just wasn't up to the task. Once the bushing fails the shaft starts grinding away as the bracket, even cutting a groove in the shaft.

With the shaft being so crooked, the pedal is also crooked making the linkage bind. Since the brake pedal shares the same shaft it has the same problems.

Here are the clutch and brake pedals after I did a quick brushing with the wire wheel to remove any loose rust, then wiped them down. I then applied Corroseal Rust Converting Primer.

It's water based and goes on with a brush. It kind of looks like watered down Elmer's Glue. In just a few minutes the magic starts to happen.

Any surface rust gets converted to magnetite which protects the metal and keeps it from rusting any further. After they dried I sprayed the two pedals with black trim paint.

Here's a closeup of the new pedal support showing the problematic bushing. I'm not trusting this one bit so I picked up an improved bushing set that uses actual needle bearings.

I first had to remove the stock bushing using my Dremel to grind off the 10 little tabs.

A few good whacks with my hammer and the bushing fell right out. Here's the new bearing next to the old bushing.

And finally, two shots of the finished pedal assembly.