Jump to content

Terrano1992

Members
  • Posts

    178
  • Joined

  • Last visited

Everything posted by Terrano1992

  1. Code 12 (Mass Air Flow Sensor circuit) means not only MAF sensor itself, but MAF sensor wiring faults also. You have already test your MAFS (by replacement with a new one without any changes). So, you have a wiring (harness) failure... Probably, your temp gauge harness has faulty also... Main rule of diagnostics: don't replace "faulty" sensor before you examine sensor harness.
  2. Yes, exactly. One tooth. This is not a complex task. 18-20° BTDC is good angle for LPG or high-octane fuel. However, it can lead to a knocking with standard (87..91) octane. Me don't think so... Your tach was designed for very old type of ignition systems. To check modern engines, you need a tachometer with "contactless" sensor (inductive or capacitive), like modern timing lamps has (this type of sensor clamps over a high-voltage wire). You can try to remove dashboard from bezel and tighten small nuts (or screws? - can't remember) connecting tachometer to a printed circuit board, to restore electric contact. Sometimes it helps.
  3. Seems like you have incorrectrly installed distributor (rotor is turned "early"). If so, you need to reinstall distributor to get 15° BTDC "near a middle". FSM pages EF&EC-24..28 - full procedure of "Idle speed / Ignition timing / Idle mixture ratio inspection". I suspect AC tensioner Pulley as a source of whine under the hood. Alternator bearings can produce high pitched whine also. Remove AC belt to determine.
  4. This is described in EF&EC section of FSM (standard timing and mixture check procedure, can't remember exact name in table of contents). VG30E has an electronic advance control, not a vacuum. It will be better than any other oil to use on ACTP. Chain Lubricant is very thin (has a good penetration) immediately after spraying, then forms a layer of a thick grease (similar to commonly used NLGI-2). So, it can penetrate inside the sealed race bearings. The same idea as for "Spray'n'Seal".
  5. Ball bearing. It's a common failure at old (1987-95) Pathfinders. Natural wear of bearing... Me think, CCW rotation is to retard and CW to advance timing. If you are afraid to get zapped, you can stop engine, rotate distributor, fix bolt and start again to check timing. 30 deg. BTDC is too early. Are you sure that you don't made "one tooth error" when you put TB on sprockets last time? I.e. all 3 paint marks (outer side of TB) was clearly aligned to punch marks at sprockets? If you are absolutely sure about "all things are fine here", I suppose your engine had incorrect timing before TB replacement (legacy from previous owner).
  6. Doc, what do you mean "normal"? When you check ATF level, you must do it exactly as FSM prescribed for this tranny model, or you'll take incorrect readings. It's not a personal Nissan, Toyota, Ford, GM or any other car manufacturer imperfection - this is a consequence of automatic tranny construction.
  7. This is a "feature" of all auto trannies from JATCO. If you check fluid level while engine is stopped, you'll have incorrect readings (ATF level will be shown too high). So, check it exactly as described in Factory Service Manual (engine at idle, selector at "P" position). "Hot" level means 80-100 deg.C (normal working ATF temperature), "cold" - 10-30 deg.C. Anyway, excellent work and excellent thread. Thank you.
  8. Holy Permatex, finally you did it! Congratulations, DoctorBill !!! Small ToDo list before you test-drive your rig: 1. Burp cooling system, add coolant to "Max" level. 2. Check auto tranny fluid level as described in FSM (pages MA-20, MA-21), add (or drain) some quantity of Dexron-III if necessary. Do not drive with very low (or high) ATF level, if you don't want to damage tranny. 3. After test-drive, check all fluids level: engine oil, power steering fluid (the same Dexron-III), transfer case, front and rear differentials.
  9. You had to loose all these bolts before applying "Spray'n'Seal" (to allow sealant wick into the gasket). Do you remember to retighten them next day, after sealant curing? No, they did not get loose if the gasket is well compressed now. VG30i/E has cast iron engine block, not aluminium. You'll probably break the overtightened bolt before you strip the threads. If you are in some doubt, you can twist off completely 2 small (#2 and #3) M6 bolts to inspect them. Do it one by one, not both together. If they are stretched from previous overtightening (M6 bolts are relatively soft, even if they are 7-th or 8-th Grade), replace them with a new (7-th Grade minimum, see mark on the head).
  10. I agree. This is machined groove, not a wear trace of lip. So, this is irrelevant to oil leak. Try to install new oil seal carefully (without any damages of a lip edge). Use "margarine container tool" to safely slip a seal edge on a crankshaft neck, and suitable drift (a piece of plastic or metal tube with 45-48 mm outer diameter) to gently press seal into the hole. Also, you can use petroleum jelly ("vaseline") instead of motor oil to lubricate shaft, "margarine tool" and seal. I think it will be more convenient.
  11. Seems very strange. Too deep and wide to be a wear track. Moreover, it is placed too far from outside end of crankstaft neck. Standard seal has 8 mm total height - so, distance from front plane of seal to the lip edge will be 6-7 mm... You need more accurate measurements to be sure were the lip places (on a groove, or not). I can't say this is a wear track or a machined groove right now. Anyway, you can try to purchase a "thin" (7 mm height) seal (Payen NJ368 or any other 47x33x7 mm) and check how it fits.
  12. My $0.02. Place ordinary 19 mm nut over the "secret" lug nut top and MIG-weld both nuts through the nut hole. Then use standard lug nut wrench...
  13. These chemicals are intended for old, "petrified" oil seals renewal (softening of rubber is a main principle). In your case (when you have a leak through a new, already soft seal) they can't effectively help. Also, they do nothing with leaky cardboard gaskets.
  14. Dirty oil contains abrasive particles... Also, do you remember that you can polish old copper (i.e. mechanically remove oxide layer from surface) with a clean soft cloth or felt? Take a piece of thin steel wire, bend one end to "L" form and use it as a feeler. I think yes, but i don' know specific names and numbers for US market... AJUSA and Victor Reinz are European manufacturers. It's possible that Payen (Federal Mogul) NJ280 (47x33x8 mm) or NJ368 (47x33x7 mm) are "repair-type" seals, but i can't verify it.
  15. No. It's a gear-tupe pump, not a centrifugal. Look here and see how "Gerotor Pump" works. Before you install new seal: check crankshaft neck for any traces of a wear (small groove under the lip of old seal). If groove exists, new original oil seal will leak too. Example photo: grooved camshaft (look to end of a neck, near a conical part of camshaft) To effectively fight with it, aftermarket spare part manufacturers produced special "repair" seals. The only difference of these seals is lip placed with "offset" (displacement) relatively to original (so this lip sits at different place of a neck, non-grooved). All other dimensions are the same as original. I can't find the better picture to illustrate these words, but the main point is clearly illustrated here (text comments at a picture are on Russian, don't worry about it): This picture shows standard oil seal installation with displacement. "Repair" seals displacement provided by their construction. Good example of "repair seals" are AJUSA 15011700 (Front Main Seal) and AJUSA 15037500 (Rear Main Seal). I had to install both (there are visible grooves on my crankshaft necks, approximately 0.05-0.1 mm in depth) - and these seals are absolutely dry now, no leak at all...
  16. WD21 rear defroster has 140 Watts power. So, it must have approximately 10 Ohm resistance.
  17. Seems as a good method. Do it. No!!! Do not touch gasket! Or you will had to replace it with a new one...
  18. Never saw it before too. But this thing looks very promising. Yes. Degrease whole area around the gasket with any suitable solvent, paint (or tooth) brush and clean cloth, loose oil pump bolts, spray this thing directly to a crack with supplied extension tube. Allow to dry (30-45 min), then retighten pump bolts. Test for leak again. Again, yes. Pump movement can damage pump gasket even greater. Also, you can destroy sealant seam between oil pump and oil pan. So, don't try to pull oil pump case from engine block. Simply loose all pump bolts (it will not hurt any part) and spray Permatex along the whole length of pump gasket.
  19. Not necessarily. For example, "100% silicone" clear windshield sealants are not a hi-temp. Only 175..235 deg.C (typical engine blue/grey/red/black sealants are between 315 and 345 deg.C).
  20. I think it will be suitable only for relatively thin liquid (like a "Gasket Shellac"). You can't create high pressure (sufficient to squeeze thick silicone sealant into the crack) by your fingers. You need "seal gun" to fill crack with automotive silicone sealant. Most of glues are not oil-resistant until they are not set... AC Compressor and alternator sits on a separate bracket, bolted to the side of engine block. So - no, there are different bolts... Yes. Yoshkar-Ola city, Mari-El republic of Russian Federation.
  21. Hey, Doc, me think you have a last hope to fight a leak without oil pump removal. Loose all bolts holding oil pump cover to engine block. Prepare a tube presser loaded with a tube of liquid sealant (you can try liquid silicone sealant or "gasket shellac"). Cut and bend tube spout so it will meet this job requirements. Pump sealant into leak zone with maximum pressure you can produce (fluid must penetrate into crack between cover and engine block and moisten a fault area of gasket). Wipe off excess sealant, then tighten all bolts to a specified moment. Allow sealant to set (3-4 hour minimum, overnight will be better) and test for leak again. I can't guarantee it helps, but you can try. It will be more easier instead of front axle and oil pan dropping... A key to success: pump sealant into the crack (applying it to mating surfaces), do not apply it over the crack.
  22. It is not a common failure. Check oil pump bolts tightening. Link to FSM ("Lubrication and Cooling" - hope it helps). Don't overtorque these bolts - loose and tighten to specified moment one by one. Then wipe dry all suspected areas, install TB (without front covers) and HB. Disconnect coil (or remove central wire from distributor and connect free wire end to engine ground). Crank engine with starter (until red oil lamp goes off) and look for fresh oil leak traces, as k9sar said. Oil pump gasket replacement will be a royal PITA... FSM says, you need to drop oil pan before oil pump removal...
  23. Loose these bolts first (with brake pressed), then remove caliper and hub. Easy way...
  24. Sorry... Part number 13510-V5000 (and 13510-D0100) was superseded by 13510-88G00. Different part numbers for the same part - it's ordinary for Nissan...
  25. Just for your reference, Nissan part number of Front Main Seal is 13510-V5000. Remove t-belt and crankshaft sprocket... Tube rolled from thin plastic sheet (margarine container, plastic bottle, etc.) and covered with one layer of adhesive scotch tape (to prevent unrolling) is a very suitable tool... Insert in into seal, press seal at place, then carefully pull out tube, slowly rotating it inside the seal (to prevent seal lip deformation). It's not so hard task. Now you have so me experience. Do it again, with more accuracy. Clean seal hole and crankshaft neck thoroughly before installing new seal, apply motor oil to neck, seal and "margarine tube". I wish good luck to you, you can do it.
×
×
  • Create New...