Jump to content
Remove Ads

inline6

Subscriber
  • Joined

  • Last visited

Everything posted by inline6

  1. Yes, I think I will send my propeller shaft there (https://carolinadriveline.com/) as I had planned, but now it needs to be shortened in addition to getting it balanced. I just very carefully measured my center to center distance again. I got 21 and 1/8". Looking very closely, I think it is maybe .010 more than that. That is 536.829" mm. The factory shop manual says the length from center to center of the u-joints is 540 mm. There is no way this propeller shaft is that long between the centerline of u joints. Perhaps because there is a slip yoke, factory tolerance of propeller shaft length was not strictly held. After measuring carefully, I will be removing 2 inches from the propeller shaft. I suppose if I had the pieces in the car to move the differential back to align the drive shafts, this propeller shaft might fit as is. I seem to recall coming across that before in my internet travels. I could search to confirm, but it is a waste of my time as I won't be swapping in those pieces from a later car. If the drive shaft angularity causes issues on this car, I'll swap in some CV axles.
  2. Hmmm. I am rethinking this. The driveshaft shield on the yoke is bottoming out against the 240SX transmission. And I measured the depth of that at 1.775". Just removing that shield might give me enough. I will need to have another, closer look.
  3. I am thinking about using Inland Empire Driveline: https://www.iedls.com/
  4. I made some decent progress today. I finished up the differential - here is the separate thread covering all the issues I had with it. I installed all of the oil seals and the rear cover. From there, I retrieved my "stencil" which was the result of many hours of drawing pretty much entirely free hand. I have a mustache bar with a portion of the letters and numbers in tact. Interestingly, though I have found a few pictures on the web with this same part number, I cannot find the part number listed in the parts manual. I am unsure what the difference is between 55450-E4101 and 55450-E4102. However, I think my car came with the E4102. So, I went with that number. If anyone has information to share about these two, please enlighten me. As can be seen in these pictures, the paint was sprayed on: I found a few things to correct on my stencil, so I did work on it some more before cutting the letters and numbers out. After I was done with the cutting, I needed to rest my eyes a bit - it felt like I was going cross-eyed when I finished. I made a test run on some scrap aluminum first. Being pleased with that, I aligned the stencil on the bar. On the old bar, the 4 and 1 were a bit too high for my liking, so I adjusted it so the letters and numbers would be a bit more centered. While not perfect, I think it came out nicely. Given the number of hours I have it, I am glad it turned out well. From there, I was able to mount the differential in the car. For the stub axles, I elected to glass bead them lightly and apply only some clear coat. These were not painted from the factory, so I kept that look. I found the stub axle retaining bolts to bottom out in the Quaife carrier. Luckily, I was able to stack a couple of washers to resolve that without causing interference with the bolt heads when installing the axles. From there, I grabbed the driveshaft to see if it would fit. And... not even close. I thought this was the "short" driveshaft, but it can't be. The short one fits when installing the 240SX transmission, but this one is way too long. It is the only one that came with the car (it was not in the car). So, it looks like I am going to need a driveshaft. I took several measurements. From center of the front u joint to the center of the rear u joint. It looks to me like it is in between 21 and 1/16 and 21 and an 1/8th. Since it is metric, I am going to guess it is 536 mm (21.103"). From the back face of the shaft to the center of the back u joint, I show about 1 and 3/8ths. Again, the closest metric measurement in mm is 35 mm (1.378"). For the front of the shaft, I measured two distances: from the tip of the input shaft to the front edge of the shield, and from the front edge of the shield to the base of the input shaft (inside the shield). For those I got 2.450" and 1.775" respectively. That totals 4.22 inches or 107.188 mm. Call it 107 mm. With these measurements, I should be able to get a custom driveshaft made that will fit nicely. I am not going to mess with sourcing a stock early one and rebuilding it with new joints, etc. I did that with the one I have and it was a lot of work to source new OEM joints, strip the paint from the driveshaft, install the u joints, and then paint the driveshaft. And, I would want to send it off for balancing too.
  5. I was measuring runout incorrectly. I was measuring on the outside, angled edge of the crown gear. When I measured the backside of the gear, I got less than the .0031" factory specification. So, here are the wipe pattern pics: Drive side - first pic is directly painted teeth, second pic is the paint transfer to clean teeth. What I see here is that the contact area is nearly centered both from a top and bottom of the tooth and toe and heel of the tooth. The total surface area of the exposed metal (exposed by paint displacement) is much greater than my prior efforts where I was changing the pinion shim thickness. Paint is still present on the tooth surface above and below the exposed metal... and on the toe and heel of the tooth as well. This looks very good to me. Shifting now to the coast side of the gear: This is not as ideal. The exposed metal is mostly on the toe side of the tooth. It almost runs off the toe edge. However, I see a little bit of paint on the toe edge. And the surface area of exposed metal is quite wide and centered top to bottom on the tooth. I'd like to see the exposed metal area shifted more toward the center (with reference to the toe and heel) of the tooth. Looking now at teeth that were not painted: The pics above show paint that was transferred from the pinion gear to "clean" teeth. In these pictures it looks like the contact area is a bit more extensive. The width of the contact from top to bottom of the tooth is quite large. And the exposed area of contact appears to cover about 1/2" of the tooth between the toe and heel, mostly on the toe side. I am happy with the results I have achieved here. So, I will proceed with installing the oil seals and installing it in the car. It will be interesting to see how much noise this differential makes. From when I had it installed previously, I recall it made a low but noticeable amount of noise on deceleration, but it was quiet when under load. I had a 3.9 in the same car for a bit and it was incredibly quiet. I sold that one some time ago though as it wasn't the ratio that I needed. Net/net (summarizing this thread), I have installed the Quaife LSD carrier. I have replaced all of the bearings in the differential. I have reused the existing pinion shim and pinion spacer. In order to get the backlash within specification, I had to get the left side bearing retainer turned in a lathe to remove .5 mm of material, so that the retainer could recess that amount more inside the left side of the differential (when differential is viewed from the back). The .5 mm (about .020") which was removed was "replaced" with shims. I added back a .3 mm (.1181") shim to the left side bearing retainer, and put a .2 mm (.00787") on the right side bearing retainer with the rest of the shim stack there. It would seem that replacing bearings on the pinion with new ones was actually "no big deal" with regard to changes to the pinion and ring gear meshing. Swapping in the Quaife carrier caused the ring gear to move away from the pinion gear vs. the stock carrier. Also, replacing the side bearings in the side retainers did not appear to change the relationship of the carrier to the case either. I am confident in this because I checked the bearing preload on the side bearings at one point - the pinion gear was not installed, and I installed the carrier and checked rotation of the carrier. My check wasn't measured with tools, but by hand; however, the preload "felt" noticeable, but not "heavy". Anyway, I think I landed in a good place. I will be able to test that out now when I install the differential.
  6. Ok, I am back at this thing! I sent the left side flange off to have .5 mm removed by lathe operation from the area indicated in the picture in the prior post. Tonight I found the time to mock things back up again and check backlash. With .5 mm removed, I put a .3 mm shim on the left side. Thus, I added a .2 mm shim to the existing set of shims on the right side flange (.3 mm .4 mm and .5 mm). In theory, the distance between the bearing races of the left and right side flanges is what it was before the lathe turning, though I suppose it could be a few "tenths" of thousands different. I don't think I can know for sure. Assembling and then measuring backlash as I have done before, I got .005" of backlash. Hallelujah! I did a quick and dirty version of checking the wipe pattern and it looks good to me. I will post pictures of that a bit later. One thing I did check and find out though is that while the Quaife unit has no measurable runout (checking with a dial gauge (units in .001") on the part of the carrier unit the crown gear bolts to, the same cannot be said for the crown gear. I ran the dial gauge against the gear and measured a bit more than .004" of run out. I am going to loosen the crown gear bolts and try to adjust that so it is within specification (.0031"). Stay tuned!
  7. inline6 posted a post in a topic in Interior
    Should I acquire Pirelli strap material? The strap material I got from MSA is something else.
  8. inline6 posted a post in a topic in Interior
    I am starting to rebuild the seats for my 6/71 240Z. I have this seat re-strap kit from Motorsport Auto, but I don't have any idea how much tension to put on the straps when I attach them to the bottom frame. The strapping material is rubberized, so it can be stretched. I am wondering if anyone has done a seat rebuild using this strap material and has any guidance? Once I determine the amount of stretch to apply, I will be marking the non-stretched length and the stretched length on a template for cutting so I can stretch them equal amounts.
  9. inline6 posted a post in a topic in 432 & 432-R
    Hi Kats, I know you look at eBay in the US. Are these too expensive? https://www.ebay.com/itm/235682782552?_nkw=datsun+240z+brake+master+cylinder+cap&itmmeta=01J67QJ9PY0AS770M2QXW1T476&hash=item36dfc9dd58:g:K2MAAOSwjqNgPFM8&itmprp=enc%3AAQAJAAABAHoV3kP08IDx%2BKZ9MfhVJKkgEI8xlrOrXR35ZVEauWhWy3b8zqM5sCMdsPS%2B6W3J1gShEGzW56wS9%2BnAJ57eSUYPiXhT2D6XTQLZmKj9x85i31Kt6IooZKn4Sx3mT5vUNWaVGsIRnMyhcxGwYZJ2Hgnww%2FiWahettnb9CA7%2BQHysIe6DnRn683anTugQjDTwOkVbycEemnRtOZKwFIDGcZAAI09Gc%2B9EnVkTT9OZras%2BdFMv04JujhXmtUyosCFGeZI68fy62idnDVpZbRC2KqRdOH%2FDSaNqZ1--0AQZ%2B4PuGZJOCidpcoNO3tRNpvtPXV%2BUW0Td8LqreQYYFC2GITM%3D|tkp%3ABk9SR8ibyfexZA
  10. Working on the car today was not fun. I put in about 8 hours. First problem I ran into was the SU nozzle issue above. When compared the nozzles side by side, interestingly, the shorter of the two had not actually been shortened. The nozzles have a little groove around the circumference. You can see it in the picture below. Aligning the grooves on the nozzles, the distance from the groove to the top of the nozzle was the same for each. I decided that the nozzle was going to have to be replaced. But, I couldn't set it aside. I had to mess with it. So, I ended up removing this crimp collar (using a hammer and flat punch, it came off pretty easily. After that came off, I was able to remove the brass nozzle portion from the plastic portion. From there, I experimented with putting the brass nozzle piece back into the plastic, but not as far in. And then I pressed the collar back in to place. It took me 3 tries (remove the collar, adjust the height of the brass portion, reinstall the collar) before I got the 'short' nozzle to be the same length as the other one. It seems to be held in place very well when the collar is put into place. However, I don't trust it. My guess is that with heat and use, it may "drop" down further into the plastic portion of the nozzle. So, ultimately I wasted a bunch of time messing with it - time I really should have spent making progress on the car to try to get it ready for Z-Con. From there I worked on the rear hatch glass stainless steel trim. I removed more small dents and stepped my way through 320/400/600/1000/1500/2000/2500 grit sand papers and then used metal polish on them. Each one took a lot of work. Then, I TRIED to install them. I had no idea it was going to be so hard. Literally impossible with all my expertise and tools on hand. I ended up cutting/slicing/gouging the rubber hatch glass gasket in so many places that it is trash. I am so frustrated. I really thought that it wouldn't be that difficult. So, now what? I have to remove the rear hatch glass and replace the seal. While it is out, I can try to put the trim in place and then reinstall the rear glass. But what about the front windshield. I don't want to destroy that seal... and I don't want to take the winshield out. What am I going to do? How am I going to get the new stainless trim I bought from Japan installed on that one? Combining the events of today along with the fact that the parts I sent to be plated are not going to be plated in time and my seat belts and small parts sent off for re-chroming aren't going to make it back in time, I'm not going to be able to make it to ZCon. The business I sent a few remaining items to be plated has closed. My package was stuck for 7 days in limbo, but thankfully, I got someone at USPS eventually, and I am having them send my package back to me. So, at least I didn't lose the parts.
  11. Started to check everything on the carburetors today. They were rebuilt by Paltech - I think they are out of business now. And one of the things on my list to do is to check the basics before attempting to fire the car up. Unfortunately, I ran into and issue. I screwed the mixture screws all the way in on both carbs: Front carb - and back carb respectively: The top of the nozzle on the front carb looks odd. It has a slight angle/chamfer that you can see in the pic. But more importantly, with the mixture screw all the way screwed in, the top of the jet nozzle is not flush with the "surround" - whatever that piece is called. That is not good. I removed the jet nozzles and measured their lengths - this length. The measurements are front: 1.405" and back 1.445". These are 4 screw SU's. Can some folks measure the lengths of any nozzles so I can compare with what I have? I am going to take a closer look at my nozzles. But, I fear someone ground the end of the front one down or something.
  12. inline6 posted a post in a topic in Open Discussions
    Yes, @Captain Obvious Date on the door tag for my car is 6/71. Here are mine:
  13. Thanks for all the replies! I sprung for this set: https://www.jauce.com/auction/b1143807861 A complete OEM set - that is $289 plus shipping I also found this little gem: https://www.jauce.com/auction/d1147578027 That was $34 plus shipping. Now, how to get them here from Japan in 2 weeks!
  14. Here are a couple of pictures of the restored center console after installing it. Anyone recognize this piece of vinyl and where it goes? I am thinking it is glued to the top of the transmission tunnel under the ash tray? I started working on stainless trim today. I was able to polish the rain rail pieces without too much trouble. To polish them, generally, I used 1000 grit, then 1500 grit, then 2000 grit, then 2500 grit, followed by a soft rag and some metal polish. All of that was done by hand. I did both sides and installed them. Then, I moved on to the front windshield stainless trim. There were some rock hits, like this one in the trim pieces: To address these, I used a small metal hand punch and a hammer to push the dents out from the backside of the trim. Some experimentation with amount of force assists you with figuring out how aggressive to be. Once I confirmed that I had pushed the metal slightly outwards instead of inwards, I used 320 grit on a paint stick to "mow down" the high spot I had created. If you look closely, you can see when the 320 grit has leveled the high spot. From there, I double checked my work by using 400 grit, then 600 grit. At that point, I took another picture of the spot above: You can see the tiny nick still present, but there is no longer a dent. From here, I may try one more time to raise that nick and go with 320 again. Or, I may just move on to 800, 1000, 1500, 2000, 2500 and then polish. I figure I have about 2-3 more hours of work to get the front windshield and rear hatch stainless trim pieces done. Also, I put up a classified ad for the top center piece, as I just realized that was missing from the car when I bought it. If you have a nice one you will part with, let me know please.
  15. View Advert WTB: Windshield stainless steel trim Windshield stainless steel trim piece - top piece needed - 72811-E4100 Item number 3 in the url: https://www.carpartsmanual.com/datsun/Z-1969-1978/body-240z/windshield. I didn't realize I did not have this part and I am trying to get the car ready for Zcon in September. Can you help - do you have nice one? I have one, but it has a few dings and would require a lot of work to make nice - I am running out of time. Let me know if you have a nice one that won't need more than maybe removal of one light ding and some polishing. Garrett Offered by: inline6 Date 08/19/2024 Price $80 Category Parts Wanted Year 1971 Model 240z
  16. I didn't have much time this weekend to work on the car. Main accomplishment this weekend was getting a suitable shift boot installed. I didn't take pictures, but basically I used a stock 240SX rubber shift boot and a metal ring to seal up the hole on the top of the transmission tunnel. The 240sx transmission fits nicely in the car, and places the shifter in the stock location (unlike the 280ZX transmission). To seal the tunnel opening, which I left stock (uncut), all I had to do was bend the metal sealing ring at the edges. The diameter of the boot and the associated ring is wider than the top surface of the tunnel. So, by bending the metal ring on the sides, it folds those downwards. It took a while to get it shaped properly. To install it, I drilled holes in the top of the tunnel sheet metal and installed rivnuts. This is one of the very few (and insignificant)sheet metal modifications on the whole car. However, removing the 240sx transmission and installing the stock boot, etc. can easily be done. In addition to finishing the center console, I am in including some pictures from earlier work restoring the map light and cigarette lighter. Interestingly, the little picture of the cigarette is not painted on. If you remove the little disk from the lighter, you will see it is two pieces of plastic, one white and one black. The cigarette with smoke rising off of it is actually part of the white backing piece. So, you can sand and polish the disk without worry of "erasing" the cigarette. I also polished the map light lens. The map light as disassembled, electrical parts polished, and then reassembled. I took apart the center dash controller, cleaned and lubricated everything, then reassembled it. For all of the cables, I removed the metal wire from the plastic sleeves and straightened any kinks, polished the wire, and reassembled them. Center panel is an aftermarket reproduction part. I found that I had to file the opening for the fan switch a bit (off of center) from its original location to get the fan switch to align with the off, low, med, and high settings. Stripped old foam and paint off of this piece, cut new foam and glued that in place. There is a metal strip that gets glued to the center console on the front inside surfaces. The old glue was still present and helped me "locate" the strip in the correct location after I glass bead blasted and painted it. The glue I used was a 2 part epoxy (which incidentally has a similar appearance to the glued used originally). For the starter cables, I replicated what I did for the center panel cables. I removed the hard wires from the cable housings, carefully straightening any kinks, and then used metal polish to clean the wires. For the cable housings, I sprayed WD40 inside of them and then blew them clear with compressed air. I then rubbed lithium grease on the hard wires and reinserted them in the cable housings. The console fit nicely when I finally installed it today. I didn't get any pictures showing the installation of the reproduction shift boot, but that fit well. It looks stock, which is nice. The center console was one of the big projects left to do and it feels good to have that done.
  17. Revisiting progress on my list: Machine shop to remove a small amount of material from the left side flange of the differential Part has been shipped off for machining. Hoping removal of .5 mm will go without issue. Send driveshaft off for balancing, but... I have to put the differential in car and check drive shaft fit first. I may need to remove some of the shielding on either the rear of the transmission or the driveshaft. Waiting for the machined left retainer to come back so I can do final shim adjustments to get backlash into specification. Tell Snake Oyl to proceed with the restoration of the seat belts I sent them in June even if reproduction date tags cannot be sourced (they have delayed for weeks because they haven't been able to confirm that they can get the tags from "their vendor". Payment and additional webbing for the luggage straps sent and received. Work to restore the seatbelts should be in progress. Buy carpet in bulk (still have to decide which). Cut to fit the car and have local company put correct finished edging. Or, purchase Auto Custom Carpets, Inc. kit from RockAuto. I may purchase this kit for day to day use... and have a custom set of carpets for show. I received and installed the Essex pile version of the carpet set available from ACC (Auto Custom Carpets). Removal of carpet jute that came on carpets is not possible without damaging the carpets. I found that out when attempting to remove it from the under seat pieces. Get gas door lock and ash tray grill chrome plated by local company - quote received for one gas lock (two pieces) and two sets of ash tray parts was $200-$300. I shipped the parts off last week and they should arrive tomorrow. I was quoted 4 week turnaround, so they will likely not be back in time for Zcon. Horns - these have to be re-plated before I can put them back together - horn back plates and misc. other parts shipped to be re-plated - they should arrive tomorrow. Hoping for about a 3 week turnaround. I also have the following fairly large lift items: Assemble the seats - fit new support straps, foam and upholstery covers. Repair/restore center console - finished all the repairs and cosmetic work necessary for the center console. Ordered a 240SX shift boot ring which is the last piece I need to install the center console. Test gas tank for leaks - Put two gallons of gas in the tank a few days ago, and no leaks so far. Find source of the electrical short circuit in the windshield wiper circuit Found it with @SteveJ help. I pulled out all of the seat parts and made room on my work areas to get started. Interestingly, the old seat head rests are much harder than the new seat foam. I feels and sounds like there is some kind of core which is inside the foam of the original headrests. Also, originally they were separate parts from the seat back foam. I will take pics and show the differences. I've been searching for "rivets" that have large heads like the originals. I have found a couple of options on McMaster-Carr. Once I have suitable rivets in my possession, I can start on the seats.
  18. The next episode is out. I searched on an off for days and weeks to find an undercoat that would look like the original stuff - I wish I could have known about what he used before I did my car. Anyway, read the description also - interesting info there.
  19. It is very sticky. For most of the failed attempts, the tape and adhesive came off together. On one failed attempt, I didn't realize that some of the adhesive had stayed behind, and when I put a new piece on, it showed underneath the new piece. Removal of the adhesive using 3m adhesive remover was difficult. The 3/8" width I bought is perfect. Where I did not apply it perfectly straight, you could see a tiny sliver of extra tape off of the edge of the part that is supposed to be silver. I trimmed that by running an X-acto knife like this along the edge and then pulled the sliver of excess off. I only needed to trim one of them for a span of a couple of inches. I found some more information about the tape. It appears it may be able to handle hot car interiors: https://www.amazon.com/dp/B079C5PYV9?ref=ppx_yo2ov_dt_b_product_details&th=1 This product appears to be the same: https://store.tapesandtech.com/mylar-polyester-tapes/mmyp-1-metalized-polyester-film-tapes.html
  20. I continued working on the center console this week. I used an 800 grit sponge sanding pad and went over the texture. Here are several pictures before painting. The texture doesn't look like the factory one so much, but it is a decent approximation. From the last picture in this group of three through the rest of the pictures, the console has semi-gloss black paint applied. Another little project I completed was these interior trim panels. The silver/chrome had come off of them. I don't know what happened to my pictures of the "before", but the "chrome" stripe was blue. I found this Metalized Polyester Mylar Film Tape with Acrylic Adhesive and decided to give it a try: It took about 7 tries to get it on the panel to my satisfaction. I was getting some air bubbles on some of the attempts. And it was a bit crooked on some as well. Thankfully, you get 72 yards of it on one roll, so I had plenty to spare. 🙂 Anyway, I am very pleased with the final outcome. I hope it stands up to heat, etc. well as time progresses.
  21. I continue to attempt to make progress. I was able to get in touch with the place that I had in mind to re-chrome the gas door knob and the ash tray parts. For the gas door knob (two parts) and two sets of ash tray parts (grill, plate and cigarette lighter trim ring - 3 parts X 2), they quoted between $200 and $300. I have been working on restoring the center console this week. The plastic trim plate had a crack in it. I was able to swap the one from my track car. I cleaned it up and used a "chrome" paint pen to return it to how it looked originally. The two rear-most mounting holes in the console were oversize, allowing the screws to slip through those openings. Thus the screws were not securing the console to the car. The console is fiberglass. So, I ground away as small amount of material on the underside. Then I mixed up some resin and hardener and put some fiberglass mat pieces in place to rebuild the part around the holes that had broken away. As the resin was hardening, I drilled new holes. Drilling is easier when the hardener is not fully hard and I feel I can control the location of the hole better. After the hole had been drilled in the fiberglass mat, I mixed up a small amount of fiberglass/polyester filler and applied that to the topside surface. While it was still unhardened, I installed the screws. They have a tapered head. I let them sit until the hardener kicked off, but removed them before the fiberglass filler fully cured (third pic here). In these pics, you can see the ring of filler around the holes after sanding the small amount of fiberglass filler applied on the top side. I used a counter sink bit in a drill to open these up a bit further, however, the factory recesses around the bolts were a bit larger than what I ended up with. The console had many cracks in the surface. To address those, I sanded with 80 grit first, then switched to 240 grit. Instead of using polyester filler, I used glazing spot putty. Sanding this is much easier. The benefit is that when sanding, it is easier to keep from sanding the original part. With polyester filler being much harder, as you try to sand it flat, you end up sanding the core part a lot more, ending up with more surface unevenness. After sanding and prepping the surface, I sprayed SEM texture coating 39853 on the surface of the console. Varying spray distance, you can end up with different textures. After this dries, I plan to use 800 grit sanding pads to "flatten" the texture I have here: Hopefully that will look reasonably close to the original factory finish. The final step will be to paint the console with a semi-flat black paint.
  22. Update time. I put the stock pinion spacer/shim back in and took a fresh set of wipe patterns. These wipe patterns look "better" to me in that the amount of surface area where the paint has been "cleared" is much larger that what I was achieving by adding thickness at the pinion spacer/shim location. This time I put drag on the crown wheel as a rotated the assembly to get the marks. With the .3 mm shim on the left side bearing and the .4 mm and .5 mm shims stacked on the right side bearing, I measured .021" plus some of backlash. The pattern above was with that combination. I moved the .3 mm shim to the right side and measured backlash again. This time it was .012". I somehow forgot to get a wipe pattern after doing so. I will do that and take some more pics and post them here. With less backlash, the wipe mark should move further away from the heel end and toward the toe end (on the drive side of the tooth). With no more shims to move from the left to the right, the best way I can think of to less the backlash is to machine the left side bearing retainer. A picture: Removing material from this face will result a change to the the carrier bearing location vs. where it sits now when the flange is bolted to the case. So, to fix the excessive back lash, I want to shift the carrier to the right some small number of thousandths of an inch and I want to achieve this by removing a small amount of material from the left side flange. Doing that will bring the ring gear closer to the pinion gear. Additionally, whatever amount is removed from this retainer has to be matched in additional shim thickness on the right side bearing retainer so that the preload on the carrier bearings does not change. As originally assembled with the original carrier, there was a .3 mm (0.0118") on the left and a .4 mm (0.0157") and a .5 mm (0.0197") on the right. That combination adds up 1.2 mm. Side shims are sold in the following mm sizes: .2, .25, .3, .4 .5. If, for example, I estimate I can get the backlash within specification by removing .2 mm = 0.00787" from the left side flange, then that amount would be removed from the left side flange and I would add a .2 mm shim to the existing stack of shims on the right. Or, another option which would achieve the same thing would be to machine .5 mm off of the left retainer, and then add a .3 mm shim to the left side and a .2 mm shim to the shims already on the right side (add .2 mm more to the .3 mm, .4 mm, and .5 mm shims). Again, whatever the amount that gets machined off of the left side retainer, I will need to add back into the total shim thickness. Ultimately, I need to get backlash to .004"-.008". It may be best to remove .5 mm from the left side retainer because then I have many sizes of shims to work with from an adjustment standpoint. Also, removing .5 mm covers the possibility that I run into a problem of removing .2 mm and that not being enough to get the backlash to .008" or below.
  23. Congrats! I like the color choice too.
  24. From my list: Machine shop to remove a small amount of material from the left side flange of the differential - need to put the stock pinion shim back in and then check back lash with original side shims in their respective places. Then check backlash again after moving the one left shim to the right side. Compare the difference and estimate amount to have removed from left side flange, given side shim sizes available. Send driveshaft off for balancing, but... I have to put the differential in car and check drive shaft fit first. I may need to remove some of the shielding on either the rear of the transmission or the driveshaft. Tell Snake Oyl to proceed with the restoration of the seat belts I sent them in June even if reproduction date tags cannot be sourced (they have delayed for weeks because they haven't been able to confirm that they can get the tags from "their vendor". They will be able to get the labels. Cashier's check and additional parts are going out via UPS to them tomorrow. Buy carpet in bulk (still have to decide which). Cut to fit the car and have local company put correct finished edging. Or, purchase Auto Custom Carpets, Inc. kit from RockAuto. I may purchase this kit for day to day use... and have a custom set of carpets for show. I Purchased an Essex pile version of the carpet set available from ACC (Auto Custom Carpets) https://www.rockauto.com/en/moreinfo.php?pk=6070062&cc=1209158&pt=1264&jsn=10419&optionchoice=1-A1168-0-1 Get gas door lock and ash tray grill chrome plated by local company - email with pictures of parts needing to be re-chromed sent to get quote Horns - these have to be re-plated before I can put them back together I also have the following fairly large lift items: Assemble the seats - fit new support straps, foam and upholstery covers. Repair/restore center console Test gas tank for leaks - source and install a new tank from S30 World if it leaks Find source of the electrical short circuit in the windshield wiper circuit Found it with @SteveJ help. So, some progress, but it still looks like I am going to run out of time to me. My windshield wipers were not parking as they should. They were parking higher than their range of sweep. That is not right. They are supposed to have a range of sweep that is higher than the park position. After reading related posts, I realized that I had not paid attention to the reassembly of some of the parts on the wiper linkage. In order for the linkage to park lower, the offset cam piece in the wiper linkage has to be in the position to make the linkage arm longer. Some pics: As found with the issue: cam/offset in the short position: cam/offset in the long position (note that you can see the edge of the of the back plate, and the linkage bar that attaches to the motor is extended to the left a bit more: After removing the clip and the top washer: After flipping the cam/washer piece with tang over - now with the tang at the top position, the link bar is in it's shorter length position. So the pic just above is CORRECT. This is the position of the washer with the tang when the motor is operating. When the motor reverses, the offset and tang will rotate 180 degrees. This will move the link bar to the "long" position which will extend the effective length of this arm just a touch. And it will result in a park position on the windshield where the wipers will be lower on the windshield than the normal sweep range. Also, I found that the best park position was set when the link bar (the piece that bolts to the back of the wiper motor shaft is in alignment as in these pictures. In other words, the ideal park position is achieved when the link bar is in perfect alignment with the long linkage bar it is attached to. Like this ------ -------. Not angled either up or down from the long bar, but simply a linear extension. I had to install the motor with the bag hanging off to the side so I could mess with the round cap on top of the motor that adjusts final resting location of the motor shaft. Then remove the motor again, put the bag on and reinstall the motor.
  25. Were these clear zinc plated originally, or chrome plated?
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.