Jump to content

Namerow

Free Member
  • Joined

  • Last visited

Everything posted by Namerow

  1. FWIW, I found that I had to keep 'doping' my bath with the Caswell's 'brightener' liquid to keep my zinc plating coming out shiny. It worked really well.
  2. Another option (which may or may not appeal): Rubber/neoprene grommets respond really well to 'super glue' (aka KrazyGlue'). Just slice the grommet across half its width, position over the cables, apply a drop of glue to the cut surface, carefully line up the edges, and press together for 30 seconds. Job done. If you do it carefully, the joint will be pretty much invisible... and more than strong enough to withstand the distortion of the grommet when you push it into place in the hole in the firewall. If you need convincing, buy a generic grommet from the hardware store and try it out.
  3. Just did this job myself. To confirm for you, the lip faces away from the hatch opening (i.e. faces to the front of the car for the section that goes over the top of the hatch). Before you get started with the adhesive, use a paint-friendly cleaning liquid to clean both the pinch strip surface and the surface where the lip will seat. Use the same cleaner to clean the inside groove of the weatherstrip, as well as the bottom surface of the lip. Apply a light coating of weatherstrip adhesive to the inboard edge of the pinch strip. Also apply a thin smear of adhesive to the underside of the weatherstrip lip. Forget about applying adhesive "to both surfaces" (overkill). When installing the weatherstrip, start from the centre of the top of the opening and work outward to one side and then down. Then go back and do the other side. The weatherstrip will stick pretty firmly right from the get go, so go slowly and avoid creating wrinkles. Be extra careful around the upper corners of the opening. The pinch strip isn't very tall and it's easy to miss the actual contour of the corner when you're pushing the weatherstrip into place. Pushing a length of clear-vinyl tubing (1/2" diameter?) down into the groove so that it presses down on the lip will help to seat the weatherstrip properly (especially around those two corners). If you don't remove the hatch from the car, it will be a bear of a job (maybe impossible?) to install the weatherstrip properly over the top of the hatch opening. Probably a good idea to apply painter's cling-film over the roof and rear quarter panels around the opening before you get started with the adhesive.
  4. Well, not quite NLA... but close ($145!) http://www.ebay.com/itm/Datsun-240z-280z-Differential-Bar-Stopper-Mount-Bushings-Upper-Lower-NOS-/282099223385?hash=item41ae6c8b59:g:w4oAAOSwY45USAog&vxp=mtr
  5. I believe I read that the end washers are also NLA.
  6. A few more thoughts before I head off to work... In the end, it all comes down to finding something that works effectively. The Shore hardness ratings for the poly bush would be easier to interpret if we knew what they mean in terms of relative stiffness (related, but not the same as hardness) vs. the OE rubber bush. If there's someone out there who has a good-condition OE bushing and wants to do a little testing for the greater good of the rest of us, they could perform a simple deflection test by clamping the outer casing in a horizontal mount (use a bench vise and a carpenter's bubble-level) and inserting a tight-fitting long bolt or steel rod into the inner sleeve. Then hang a 2-lb weight off the end of the bolt/rod at a measured distance from the end of the sleeve (i.e. known lever length) and measure the bolt/rod deflection (at a known distance from the end of the sleeve) with a dial indicator. You could then perform the same test with your poly bushing, to generate some relative-stiffness information. If the 70sha poly bushing comes close-but-not-close-enough to the stiffness of the OE bushing, it's possible that it could be fine-tuned by drilling some end-to-end holes through the polyurethane (similar to the internal voids in the OE bushing). If the poly bushing looks promising, there will still be the matter of cobbling up a reasonable facsimile of those special end washers. As I said earlier, I think this can be done.
  7. Some additional photos to support the view that the OE design for the MB mount insulators is 'special'...
  8. Consider these points: The rubber in the OE bushing is softer than polyurethane The OE bushing design uses a voided core. The poly bushing is solid. The OE design for the MB mount uses not just the main bush, but also special rubber-faced end washers with a ribbed design. The aftermarket poly bush set-up uses plain steel washers. If you go with the poly bushings, they should locate the MB just fine. However, you'll probably experience more 'gear noise' coming through into the cabin. More noticeable under highway cruise conditions than around town, I would expect. Impossible to say just how much more until you try it. If you do, let us all know how it works out.
  9. Here's another photo that I forgot to post yesterday that illustrates in pretty dramatic fashion how this insulator can collapse with age. Not surprising, given that the cross-section of the rubber core necks down to what looks like only 1/8" in width... (As before, my apologies for not being able to credit the owner of this photo.) If you go back and look at Photos #1 and #2 of my earlier post, you can see how the later-style insulator's design changed so as to significantly increase the width-wise dimension of the rubber block at the front. Perhaps this was an effort to reduce the insulator's vertical compliance. The crest built into the top of the rubber block indicates that insulator was obviously designed to bottom out. Still, it doesn't look like you'll be doing this insulator any favours, long term, if you use the diff casing as your rear jacking point. @HardwayOn a related point, Hardway's recent post in which he stuffed coolant hose into the insulator opening to (successfully) cure a diff 'clunk' issue suggests that his front insulator is now operating in a permanently bottomed-out condition. Waiting to hear whether he reports any significant increase in gear noise coming into the cabin as a result. If he doesn't, then that 'stuffing' fix may make replacement of a collapsed front insulator a non-issue.
  10. FWIW, there's a vendor in Japan currently offering the 240Z early-style insulator (55415-E4102) for low, low price of just 28,000 Yen (that would be about US$375!)
  11. 'Please put down your HVLP gun and step away from the rotisserie with your hands held in sight.' p.s. Where is your brake proportioning valve?
  12. Here are some pix that I've accumulated from this site and others (apologies to the owners -- I didn't record names with photos). Picture #1 shows the top surfaces (front of car points right)... 240Z early type (55415-E4100/4101/4102) on the top, 240Z later type (55415 E8300, introduced in 7106 for the MY1972 240Z) on the bottom... x Picture #2 (below) shows the bottom surfaces (front of car points up)... early type (55415-E4102) on the left, later type (55413 E8301) on the right... x Picture #3 (below) shows a third type (55415-E8301) which -- if I read the parts manual correctly -- was offered between 7308 - 7412 (which would make it specific to the 260Z, I believe) x Picture #4 (below) shows a fourth type (55415-N4300/4301) introduced in 7501 and therefore specific to the 280Z. x Going back to Picture #1 and #2, it looks to be like the 'early' and 'later' 240Z designs, there doesn't appear to be anything different about the hole spacings/positions for the two mount bolts. The location and size of the threaded centre stud looks the same, too. The upper/outer plates have different shapes, but I don't see anything that would prevent re-shaping the contour of the later style plate if that's needed to make it fit into a Series 1. Also, the width of the inner plate's 'jaws' looks similar, early vs late, so it appears that the later style mount would fit down over the early style crossmember. I emphasize the word, 'appears', because I've never read anywhere that this can actually be done. Perhaps someone else has tried this and can comment. For possible reference, Picture #5 (below) shows the early-style insulator sitting in position on top of the matching front crossmember. BTW, I don't think there would be any need to turn the later style insulator back-to-front to make it work in Series 1 application. Again, others who've actually tried this may have a different opinion. In case you're curious, the threaded holes in the crossmember on either side of the insulator are provided to take a bolt that retains an exhaust system hanger piece.
  13. I'll bet the front and rear manifold studs are snapped off too.
  14. I have never seen a rippled seal like that in a hydraulic system before. Maybe someone else has, but not me. It looks more like a washer than a lip seal. Without seeing the other side (of both items), though, I won't speculate any further.
  15. If you're going to install new seat covers, there's an incentive for replacing the foam at the same time. When rebuilding the bottom cushions for the seats from my 70 Z, I discovered that 'skirt' of the replacement bottom seat covers was considerably 'taller' than the stock seat covers (my replacement covers were manufactured by Distinctive Industries). I chose not to replace my original foams (they were in pretty good shape), so I ended up having to shim the underside of the bottom cushion foam with about an inch of added-in foam sheet so that the cushion would adequately fill the new seat cover. This was the only way I could bring the lower hems of the covers into proper alignment with retaining tangs on the seat frame tubes.
  16. Namerow replied to Patcon's topic in 510
    Only one U-joint in the 510's steering shaft?
  17. You've managed to achieve a really nice stance for the car. The match-up of tire size/profile to wheel diameter/offset relative to the wheel openings looks spot-on. I really like those wheels, too. A refreshing change from Panasports.
  18. The designs of the front insulator and the mustache bar bushings were tuned to isolate the chassis from the gear and drivetrain noise/vibration 'collected' by the diff casing. It will be interesting to see whether stuffing the front insulator's cavity with heater hose really does have an adverse effect in the form of gear howl getting into the cabin. Let us know what you discover. If a problem does set up, you might be able substitute something softer than heater hose that would be better able to meet the dual requirements of noise isolation and clunk elimination.
  19. Namerow replied to Patcon's topic in 510
    Check the U-joint. Interesting that Nissan offered u-joint rebuild parts for the 510, but did not offer them for the 240Z. If the 510 parts are NLA at Nissan, it's possible that ez-to-source Kawasaki ATV halfshaft u-joints that fit the S30 will also work for the 510. $30 complete (spider, cups, bearing needles, seals, circlips, and grease fiiting) and relatively easy to install. A 64th inch of slop in the u-joint will be greatly amplified at the steering wheel and makes straight-ahead highway driving 'more involving'.
  20. I've now received my rebuild kit from MSA and I've taken a few pictures for reference. As you'll see, the three seals in my kit are all different -- two are lip seals, the third is an O-ring. One of the lip seals (the one on the left in my pix) is lipped on both the inner and outer circumference. The other (in the centre) is lipped on the outer circumference only. Both are about the same in height. The seal on the left with the inner & outer lips is pretty clearly Item 5 in FSM diagram BR-8. By default, that means that the single-lip seal in the centre has to be Item 3. Obviously, placing them in the right locations and getting each one correctly oriented (lips face inboard, flat sides face outboard) is going to be critical to making the valve work properly. I continue to be confused by the way the upper seal (Item 3) is depicted in FSM diagram BR-9. The shape of the seal's section on the left makes sense. However, the sectional shape on the right looks like a completely different seal! Am I missing something here?
  21. Namerow replied to Patcon's topic in 510
    Sorry, but what kind of car is this?
  22. The FSM diagram that Zedhead posted is from the 280Z manual, so it's possible that the bolts used for the 240 were stepped rather than tapered. The fact that Nissan's engineers chose to use a stepped/tapered design rather than a plain old bolt says to me that there's a lot more going on here, NVH-wise, than first meets the eye. The m-bar 'bolts' are more like pins with threaded ends. It's an expensive design (compared to a bolt), so it was used for a reason (just like those rubber-ribbed end-washers) It's a little hard to tell from the FSM sectional diagram, but it appears that only the upper 'special washer' (as the parts manual calls them) bears directly on the shoulder of the pin in the radial direction. If I'm right, it means that the 'eye' of the m-bar is solidly located (in the radial direction) only at the top, while the bottom 'floats' (in the radial direction) on the rubber of the lower special washer. It also means that the thick rubber of the big bushing only comes into play (in the radial direction) when the m-bar deflects far enough to bring the inner metal sleeve of the bushing into contact with the pin. My guess is that this is all about isolating high-frequency differential/gear noise from the chassis under light-load/cruise conditions (via the ribbed special washer), with the bushing only coming into play (laterally) to isolate against cornering-induced vibration (when the diff loads up laterally against the pins). Kind of a two-step isolator for lateral loads. In the vertical plane, the special washers and the bushing look they work in unison as a two-mode isolator (that is, they're both at work simultaneously, in series, but have different stiffnesses and natural frequencies and therefore serve to isolate different vibrational frequencies). Another interesting point (to me, anyway): the FSM refers to the m-bar as a 'transverse leaf spring'. That suggests that the bar itself is being used as a spring. The m-bar is only truly flexible in the fore-aft direction, so perhaps it comes into play when the diff is subjected to braking loads. One final thought: All of this suggests to me that it's not a good idea to jack up the rear of the car using the diff casing.
  23. here you go...
  24. from Wikipedia... Hydrogenated nitrile butadiene rubber (HNBR) is widely known for its physical strength and retention of properties after long-term exposure to heat, oil and chemicals. Trade names include Zhanber (Lianda Corporation), Therban (Arlanxeo [5]) and Zetpol (Zeon Chemical). Depending on filler selection and loading, HNBR compounds typically have tensile strengths of 20–31 MPa when measured at 23 °C. Compounding techniques allow for HNBR to be used over a broad temperature range, -40 °C to 165 °C, with minimal degradation over long periods of time. As a group, HNBR elastomers have excellent resistance to common automotive fluids (e.g., engine oil, coolant, fuel, etc.) and many industrial chemicals. The unique properties and higher temperature rating attributed to HNBR when compared to NBR has resulted in wide adoption of HNBR in automotive, industrial, and assorted, performance-demanding applications. On a volume basis, the automotive market is the largest consumer, using HNBR for a host of dynamic and static seals, hoses, and belts. HNBR has also been widely employed in industrial sealing for oil field exploration and processing, as well as rolls for steel and paper mills.
  25. Happy hunting, CO! Still hoping that someone can post those three measurements for the Nissan OE bushing so that we've got a reference point to work from...
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.

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions → Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.