Jump to content
Remove Ads

Zed Head

Free Member
  • Joined

  • Last visited

Everything posted by Zed Head

  1. Zed Head posted a post in a topic in Help Me !!
    Different years use different power control to the pump.
  2. 1976 has the 3.54 R-200. You would need the half-shafts, mustache bar, etc. to use the R-200 in your 240. I think that you'll need 12, 14, 17 and 19 mms to get the various nuts and bolts off. The standard manual transmission for 76 was a 4-speed also, so you probably won't find a 5 but you never know. And, not to be more of a downer, the 76 alternators were externally regulated too. But the alternator looks to be higher output than the 240 (I show 72 as 50 amp by the FSM and my 76 came with a 60 amp), so it might be a bolt-in upgrade.
  3. To tie it up - I replaced the transverse link bushings with polyurethane, and left rubber bushings in the back on my T/C rods (for the reasons jmortensen proposed) and kept the urethane in front. The steering is now much tighter and the wheels don't shift when I hit the brakes, like they did before. So I came out ahead, overall, after the motivation of the broken T/C rod. FWIW, even with the old rubber bushings off, it is hard to tell that they are bad. You can stick a big screw driver in the center, brace it and reef on it to get them to move, but it's not obvious. The cracks in the rubber are only visible when it's stressed. By the way, a Truecraft 27mm 3/8 drive socket, part #45227 (manufactured in the early 80s, may be different now) is a "too perfect" fit for pressing out the complete bushings from the transverse links. Too perfect because it takes as much force to press the socket back out as in, but it's easy to fixture the socket with a bolt or rod to press it back out since it is solid and you can press on the center. But it fits the outer bushing ring perfectly. Thanks for all of the input.
  4. Quoting myself above. Kind of nonsensical. Four small bolts plus one big one - I'll just take the LCA off...
  5. Thanks all for the comments. I feel comfortable now with poly front/rubber back, although the steering does feel looser. Replacing the LCA bushings will probably help that, I'm sure. ktm, I'm just telling a little story here, and trying to add to the body of Z Car knowledge, with some pictures and solid facts, for anyone to use as they choose. That's all. No intent to offend or ignore anyone else's hard work.
  6. Good point, thanks for adding. I did notice that my control arm bushings were worn also, with visible cracking, and thought that it may have contributed. Before I took the broken T/C rod off I found that I could grab it like a handle and move the control arm around easily, with play visible in the control arm bushing. The end of the broken rod actually hung about an inch below the bushing mounts because of the worn control arm bushings, so there was probably a side load on the rod also when in use. I didn't want to do the control arm bushings yet so just went with the rubber in the back for a short-term fix. Has anyone replaced the bushings with the arm still in the car? I enjoy laying under my car so might try it...:classic:
  7. Changed mind on comment...
  8. Here is a picture of one more Z car T/C rod gone. I installed MSA polyurethane bushings front and back on the T/C rods of my stock 1976 280Z about 4 months ago and have put about 3,000 fairly easy road miles on since then. I had read the various accounts of broken rods but the evidence still seemed iffy and most of the descriptions seemed to be for the later model cars with front-oriented T/C rods and not the rear-oriented Z cars. So I decided I would do both sides with polyurethane on my car and see what happened. It looks like fatigue cracking started top and bottom and just worked its way through the bar. The coup de grace may have been when I backed the front wheel in to the curb while parallel parking. About ten miles later, that evening, I heard something bumping off the bottom of the car, but had not seen anything in the road. So I took a look under the hood in the morning and saw that the rod end on the passenger side was almost worked out of the bushings, the rod was broken at the base of the turned-down end, and the big washer on the inside was gone. That was the noise. I had a spare rod from a 78 and used the rubber bushings on the back on both sides, in place of the hard polyurethane, as described around the various forums, to take some of the bending pressure off the rod end. For the record, for those still wondering.
  9. Zed Head posted a post in a topic in Help Me !!
    The fuel pressure regulator (FPR) has a vacuum line attached to regulate fuel pressure according to manifold vacuum. Yours looks like it is probably working correctly. If you want to check the FPR, you can pull the small wire from the starter (it comes off by hand, no tools needed), then turn the key to Start to run the fuel pump. The fuel pump will run (you'll hear it), but the engine will not turn over. Without the vacuum from the motor, the FPR will give a higher pressure. 36.5 psi is normal for 1976 and most years, I believe. A lot of people have the problem with rough idle after starting a warm engine. I have seen one proposal for vapor lock but no followup. Have not seen a good remedy yet.
  10. Zed Head posted a post in a topic in Electrical
    It could be that your front or rear brakes are low on pressure. Either leaking or need bleeding. There is a valve that measures the balance between the front and rear systems and activates the dash brake light if the balance is off. I had the same problem on my 76. Found out that a PO had put the front calipers on the wrong sides, making them impossible to bleed correctly. I could pump the brakes up while driving and make the light go off, but it would always come back. Here is a page from the FSM describing the system. 74 260 Brake Page.doc
  11. Hey Jennys280Z, does your car actually run like it used to now, and run well? I'm curious since your AFM resistance readings were the same as mine, which was definitely bad. Just trying to learn something. Thanks.
  12. Zed Head posted a post in a topic in Suspension & Steering
    Post #3 in this thread, Chris's writeup with pictures, will help you guys out a lot. The pictures show where the clips are and how to get them out. The clips are on the inside, unlike most American-made cars. They are probably covered in old grease and dirt, hard to see. http://www.classiczcars.com/forums/showthread.php?t=27855&highlight=u-joint
  13. Zed Head posted a post in a topic in Fuel Injection
    I live in Portland Oregon, fairly cold up here, and have been driving my car all winter with the cold start valve fuel supply disconnected. It just takes a 1/2 second longer to start but once it's going it's fine. I will not be reconnecting it. I think that it may have been leaking a little. The left over fuel supply line makes a good spot to install a fuel pressure gauge.
  14. this just came up today on hybridz - http://forums.hybridz.org/showthread.php?t=127993&page=7
  15. I used the word crude and tried to direct attention to the R/T mount because my fix is rough and has only been on the car for a week. My car is essentially my own little experimental test ground to play with, and a reason to get greasy, tear things up, put them back together then go drive around. The new "diff arrester" is a piece of 1/8 x 1 1/2 inch steel that has been twisted and drilled to use the large arrester strap mounting bolts (the ones with the 22 mm heads) to hold it in place. The snubber is a 1 1/2 inch diameter bump stop from Energy Suspension, mounted in the middle of the metal strap, that has been trimmed down to barely touch the top of the diff. It may be susceptible to metal fatigue though so I really can't recommend it unless you feel comfortable with taking a chance and like to crawl around under your car to check it out regularly. Full acknowledgment - I did read Arne's post about using the R/T mount as a snubber, along with hundreds of other diff mount ideas, and that fed some of my thinking. The two other popular ideas with the tabs and the bolt or the L-shaped bracket are essentially snubbers also. One other odd finding - I tried to preload the snubber a little bit (more is better right?) by adding a thick washer under it and levering the mounting bolts in, and found that it transmitted a loud drumming noise in to the car at takeoff. So I went back to the original tiny air gap (about a metal ruler's thickness) and things have been very tight with no added noise or vibration. I've also noticed that we pay a lot of attention to the diff nose rising but that there is very little space for the diff nose to drop before the mount hits solid metal. The stock setup allows the nose to rise a lot but gives a hard landing on the way back down. One more thing to think about. Sorry for the excess words. I like to write.
  16. It pulled away fine if you left quickly, the faster the better. It was the attempts to drive slow and smooth that gave the "chatter." I did not resurface the flywheel (should have pointed that out), so the weak diff mount might have been multiplying the effect. Either way, if someone has a similar "chattering" issue, wherever it's coming from, the diff mount might be part of the problem. That was my basic point. Just trying to add something I haven't seen described before. Anyway, thanks for reading, and the alternative view. I should have taken a video (I recently saw one here or over at Hybridz) to try to nail down what was happening.
  17. I fabbed up a crude snubber for the front of my R-200 differential, since I knew that the diff mount was showing signs of end-of-life (visible debonding when pried on with a screwdriver off the car), and found that it eliminated a bad clutch chatter problem that I was having. I assume that the weakening original rubber diff mount was allowing the front of the diff to bounce up and down at low acceleration levels, causing the clutch to stick and slip, ie chatter. Just putting this out there for anyone that may be having a clutch chatter problem. Something to look at. My clutch and pressure plate are brand-new, properly installed and broken in, but still chattered if I tried to leave a light or stop sign slowly. The snubber I made is fairly solid and positions a urethane bump stop about 1/16 inch above the diff at rest. In action, the diff rises up and contacts it, stopping the bouncing (my guess, not actually seen in action). The R/T mount or similar would probably have the same effect.
  18. Zed Head posted a post in a topic in Help Me !!
    It's not clear if you meant fuel was flowing in your fuel lines, or you just saw that lines were full of gas. Residual pressure remains in the lines after they get pressurized, so the fuel pump may not actually have been energized. Maybe you know this but you could pull the wire to the starter and listen for the fuel pump when you turn the key to Start. It's the small wire with the female connector made to be pulled off and pushed on easily. The electric fuel pumps can usually be heard from inside the car. Pull the wire (make sure it's not touching a ground), turn the key to start, listen for the pump.
  19. Shiny. Does it run right yet?
  20. Zed Head posted a post in a topic in Engine & Drivetrain
    I don't think that a longer rod gives you more throw. Piston movement determines throw. A longer rod might actually use up some throw by preloading or setting a short starting position for the slave cylinder piston. Might be that you just need to bleed the system, with the original rod, adjusted to the proper amount of play. Air bubbles in the system would use up throw (piston travel) and cause your symptoms. I think the clutch is very easy to bleed, compared to the brakes. Just some thoughts, as to how I understand the hydraulic clutch system. Anyway, good luck.
  21. Zed Head posted a post in a topic in Engine & Drivetrain
    If you could watch the clutch fork while you press the pedal (a mirror beside the car maybe?), or have someone watch it, or film it, or put an indicator on it - then you could isolate the the hydraulic side of the clutch system from the mechanical side (inside the bell housing). Then you would know if it's the clutch disc or pressure plate vs. the master or slave cylinders. More - if the fork moves but it doesn't release, it's pressure plate or disc. If it doesn't move, cylinders or pedal mechanism More #2 - or you might be able to get under the car, lever the clutch fork with a wrench or some pliers and see if it disengages. A wheel or two would need to be off the ground and the car in gear, while you moved the clutch fork and turned the drive shaft at the same time. Yoga.
  22. I did not that realize the 68.7 reading you were worried about was through the fuel pump and the air regulator. There should be resistance there, the air regulator is a heater and the fuel pump is a motor. If the air regulator was bad, it would just affect your idle, either too low when cold or too high when warm. It bypasses the throttle, designed to raise the idle at cold startup. Your car runs so the fuel pump is probably fine. The Deoxit cans have an adjustment on the top. Low, Medium and High. I think they are set on High as-bought. I think that your main problem is in the AFM. As the FSM says, the values should be close to 180 and 100. 226 and 126 is quite a bit off. The ECU determines part-throttle injector open duration based on the AFM signal. Page EF-12 has a pretty good diagram of the AFM innards. EF-51 has a picture of the bottom of the AFM, with the connection point, and more thorough testing that you can do if you remove it.
  23. It looks like someone may have already been inside the AFM, I don't think that they came sealed with red sealer. But I could be wrong. But it would not affect your bad resistance readings anyway. The black wire is a ground wire. The best way to clean it would be to take it out. Loosen up all of the hose clamps and hoses. Take out the three big Phillips head screws that hold the metal frame the AFM is mounted on. That's the bottom of the mount you're looking at in the second picture. Work off one of the hoses, I think the front one is easiest, and lift that end of the AFM up clear of the hose. Work it out of the other hose, but not too far, you will need to reach under it and get the wires clear of the mount point on the car. Then you should have room to turn it and see the connector, and be able to disconnect it. I think the connector is a little difficult to get off, it may have a wire clip or something holding it. It's been a while. The black wire you saw connects the AFM to the frame so will come with the assembly. The black cover just comes straight off of the AFM. There's really not much holding it on except the sealer. You might need to cut the sealer with a blade and carefully pry the cover off. Inside you'll see the black contact pad and the sliding contacts and the weight and spring. Rotate the weight or move the valve inside the air path and you'll see how the potentiometer works. No one is talking probably because if yours was working correctly, this would be "Not a good idea." But since you'll have to take it out anyway if you have it replaced, you might as well go for it. Have fun. It is a strange device.
  24. Well, now you have the experienced guys on the case. Call the guy in CA. I would like to know if these are fixable without a rebuild too. If the core charge hadn't been so big I might have kept my old one to dink around with (maybe MSA knows that). They seem simple in concept. To be clear, the MSA rebuilts cost $169 and MSA did not give me any problems returning the core charge. As long as the aluminum body is not cracked or broken, the charge will be returned. But it is a risk, with a lot of money. To add to sblake01's suggestion, if you try the Deoxit, make sure you let it dry out before testing it. I tried something similar and tested it while still moist after and the numbers were all over the place. That contact area is really sensitive. After it dried though, they went back to the start values (the bad ones:cry:). But you might have some luck, it's worth a shot. Since you are working at the ECU connector, you should probably clean the connector at the AFM too. And that 68.7 reading could be a corroded connector too. Good luck.
  25. Those are almost the exact numbers I had on my bad AFM. I had 227 and 126. The car would idle well, but popped and surged when you tried to drive it at part throttle. If you hit the gas though, it ran fine. Of course, I couldn't drive around at full throttle all the time, in my neighborhood anyway, so I replaced it. I got a rebuilt one from MSA (thezstore.com) and the car ran great after that. Until an injector died, but that took a while and is another story. MSA will either rebuild yours after you send it in, or send you one off the shelf if they have it, but with a $400 core charge. Apparently they are running out of cores to rebuild because people don't send the bad ones in, hence the huge core charge. Someone out there might have a used one if you want to save some money.
Remove Ads

Important Information

By using this site, you agree to our Privacy Policy and Guidelines. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.