Oh god, it's another overdrive thread...

[Mjit]

Seems to have been a run on overdrive threads recently - but none that have helped me get mine working properly in my 2000 so thought I'd see if anyone has any ideas...

So I took the gearbox and overdrive that always worked out of one of my 2000s, rebuilt the gearbox and swapped it in to my daily driver 2000 - but since then it's been hit and miss.  I can jump in the car and drive the 2 miles to the supermarket and it will work initially, then stop engaging overdrive.  I can then drive 1/4 of a mile from there to the bakery and it won't engage at all - but then when I jump back in to drive home it will engage every time on the drive home.

So far I've:

1. Replaced the seals in/refitted one solonoid that didn't 'clack' before but did reluctantly after - no difference.
2. Repleace the seals in/fitted a second solonoid that 'clacked' OK before/very positivly after - no difference.
3. Checked the oil level - still full (and when the overdrive is working it engages the inistant you flick the switch, suggesting the pump's generating plenty of pressure).
4. Checked the inhibitor switch - not loose and as best I can judge looking through the inspection hole only fiitted with 1 washer.
5. I've tried adding a test bulb inline in the switch->solonoid circuit and that came on every time I flicked the switch "In" - but the o/d never engaged like that.  Remove the bulb from the circuit and it would behave intermittently, as before.
6. I wasn't expecting the bulb load to have an impact but swapped it for an LED bulb - which again lights up every time you flick "In" but with buld the o/d never engages/remove from the circuit and it does, intermittently.

Logicially to me:

1. The fact it works instantly when it does work, and can work both hot and cold (not just when the oil's cold and thicker) suggests it's not a pressure issue.
2. The fact the test bulb works every time suggests it's not an electrical circuit issue (a temprimental inhibitor switch would cut power to the bulb).
3. The fact it does the same thing with different solonids suggests it's not them - or I've managed to get 2 that have failed in the same way, regardless of their 'clecking'.

Which leaves me...well scratching my head - and confused why the test bulb shows a powered circuit but somehow stops the o/d engaging.

Any ideas...?

[Clive]

You say you put the bulb inline, that will add resistance and restrict the current right down.

May I suggest 2 bulbs. On connected to the +ve terminal on the solenoid and earth the other (NOT the solenoid earth) This will check power to the solenoid.

The 2nd bulb you can connect to the +ve and -ve terminals of the solenoid, this will check you have a good earth. (as long as teh other bulb lights!)

I had an OD go intermittent on the way to the start of the RBRR a few years ago. Tried a few things, until I got lucky when fiddling, and saw a small spark from the middle of the earth wire. The wire insulation had gone brittle, and fractured the cable inside. I bodged a fresh earth, and all good for the next 2000 miles. Fixed properly when I got home. But check the short earth wire, if brittle/hard try replacing it. 

 

[JohnD]

Better to use a multimeter that will show the actual current used, rather than a bulb.    But care - the Type-A O/d has a two-coil solenoid, that draws 18A when going into O/d, but much less while holding it.    Many M'meters will not take more than 10A.

John

[Colin Lindsay]

If you can see the inhibitor switch, have a look at how the cam engages it. I had this problem too, years ago - the switch was so worn that engaging it was very hit and miss, and the bracket for the switch was also bent away from the cam lobe. Try to straighten the bracket is required but check the switch for wear; mine had been seriously worn away on one side.  

[Mjit]

Doh!  Forget to say in my original post "J-Type overdrive".

Bulb was added because everything always seems fine what stationary so was trying to see if there was any difference between when it worked/when it didn't.  Will try awapping to in parallel with it's own earth.

[Mjit]

For the inhibitor switch if it wasn't engaging then there wouldn't (or shouldn't at any rate) be a circuit through the o/d switch and test bulb to the solonoid - and the solid 'light on' every time suggests that's all fine.

Will re-check the solonoid connectors as they are quite exposed and always at the "I've managed to get EP90 oil running down my arm again" stage so do get refitted in the "get the connectors on and get out" stage

[Pete Lewis]

have you actually removed the small circlip and popped the shuttle out , it might   make claack when shaken but if the small 0 rings within have worn the OD will develop a mind of its own 

so sorry its unscrew it again and take the basted thing apart   just drop the 10mm circlip and shake the plunder out do not attempt o get to the solenoid body/coil zone  or you mend it to disaster

the small 0 rings are available from some decent suppliers 

eg  https://www.jamespaddock.co.uk/overdrive-seals-set-j-type-2

and  https://www.jamespaddock.co.uk/overdrive-solenoid-j-type-spanner-2

[SpitFire6]

Radio tuned to AM around 150Kc's to 300Kc's AKA LW will click when solenoid stops drawing current.
In ye olden days I would switch my radio to LW & listen for clicks.

[SpitFire6]

10 hours ago, JohnD said:

Better to use a multimeter that will show the actual current used, rather than a bulb.    But care - the Type-A O/d has a two-coil solenoid, that draws 18A when going into O/d, but much less while holding it.    Many M'meters will not take more than 10A.

John

Better to use a DC clamp meter.
Or measure mV drop on supply cable/

[Pete Lewis]

did   we mention the wires up the gearstick  very prone to shorting and general misconduct

and it wont be the first time a rebuild has left the woodruf key out of the pump cam  but  think thats more likely it  never works 

the temperamental  operation  apart from inhibitor switch setting still has a ring of solenoid   internal seals leaking 

when driving have you tried the old trick of holding the gear stick to the left and right  to check the inhibitor setting  might need a tweak  

Pete

 

[Mjit]

Yep, solonoid seals have been replaced (in 2 solonoids - would be in a 3rd but piston shows no interest in coming out of that one!).

Gearstick wiring all looks good and I don't get electric shocks (as I did when it was shorting once) but will try swapping to the wires from my other car and running them outside the gearstick.

No, pushing/pulling the gearstick doesn't make any difference, either to o/d engagement (doesn't drop out when working/in when not) or test bulb display (always stays on clear and strong).

[NonMember]

5 minutes ago, Mjit said:

test bulb display (always stays on clear and strong).

Is that with the re-wired in parallel one? Did you also add one from solenoid -ve to a good earth, to check the earth connection, as clive suggested?

If you've got good voltage across the solenoid - as indicated by a nice bright top-side bulb and no illumination at all of the bottom-side one - but the overdrive doesn't engage, then it's a mechanical or hydraulic fault. Probably the solenoid itself but there are some points in the J-type hydraulic circuit where a bit of stray crud in the oil could stop it working.

[Mjit]

Didn't get a chance to mess around with the car yesterday (and the way work's going might not today either ).

[Mjit]

So...

1. Solonoid/earth cable replaced.
2. Cabin/solonoid cable replaced.
3. Geastick switch cable inspected (looks good and running inside cable cover so could only be short-circuiting the switch and not earthing out).
4. For testing solonoid/earth cable extended inside cabin (so runs solonoid female spade connector/female bullet connector inside cabin, male bullet connector inside cabin/male spade under car, then female spade to earth).

All the same

• Initial run with standard (cabin loop) wiring -> started engaging OK, then had a slow engagement, then no more engagement during test run.
• Swapped tell-tell LED into earth circuit -> LED goes on/off with the switch but no O/D.
• Popped off switch, pulled switch cables out of column shaft and reconnected switch 'hanging free' -> LED works fine but no O/D.
• As above but with LED removed from circuit -> No O/D.
• Bypass switch, hard-wiring O/D "On" -> No O/D (just huge fear of forgetting and then trying to reverse ).

The solonoid piston is nice and free - though I will pop it back out and give it an 'oil side' clean (what in?) - and when working (both hot and cold) engages instantly when it does engage.  Starting to sound like I need to waste another can or 2 of EP90 draining down and cleaning something else in the O/D sump - but what (considering I'll be doing this outside, under the car, not on the bentch)?

Or do people still think it's electrical and have some more tests I can run?

[NonMember]

I don't understand your terminology for some of the electrical tests:

• "swapped tell-tale LED into earth circuit" - do you mean you unplugged the solenoid earth and put an LED between it and ground? If so, OF COURSE THE SOLENOID DIDN'T WORK. The LED cannot carry anywhere near enough current to operate the solenoid. The earth side test is to put the LED in parallel alongside the earth wire, to confirm that the earth wire works. The LED should never come on in that configuration.
• "As above but with LED removed" - I assume (hope!) this means you put the earth wire back!

If you are getting good volts across the solenoid (a 12V bulb wired in parallel glows properly brightly, not dim!) and the OD isn't engaging, then THIS IS NOT AN ELECTRICAL FAULT. It's also not a problem with switch fittings. If there's volts across the solenoid and no OD then the problem is mechanical or hydraulic.

So the question is: When the voltage is applied to the solenoid, does it clonk? This test can be done with engine stopped and using a test wire. You want quiet! With it all warmed up to the point where it doesn't work, apply 12V to the solenoid and listen for the clonk. Same test when cold (and thus working) should give you a reference for what to listen for. No clonk means the solenoid is failed. And yes, they can fail "only when hot".

[Pete Lewis]

agree a simple hook up to bypass all and run a jump wire from a good feed direct to the solenoid   add a switch if you wish

you have done a lot of compromise wiring and found little so far  its certainly looking elsewhere 

solenoid windings and many other coils have problems of intermittent if the thing is wound too tightly and heat stretches connections internally to breaking 

so as Rob say  get a clonk   some clunk   any clunk     

[Mjit]

So tried a brand new solenoid and...no difference (well other than the fact it never seems to engage now, rather than working perfectly some times/not at all others - but that started with my last round of testing. not with the additon of the new solenoid).

Blub still lights solidly each and every time when plumbed in to the circuit, so electrons getting happily from the battery +ve, through the switch and solenoid and back to the battery -ve OK (the purpose of the test bulb).

Can't say I can hear any clonking from the solenoid when power switched on/off but then does a J Type solenoid actually make much noise?  Next chance I get I'll try running some jump wires direct to/from the battery while the car's up in the air and see if I can hear anything with my ear right next to the solenoid, rather than the wrong wide of the 2000's soundproofing.

I feel a shower in EP90 in my future to drop the O/D sump and try to work out what can be safely unscrewed and cleaned...  Or is it time to invest £60 in a J Type pressure gauge - can you fit one to the O/D on a 2000 in situ?

[NonMember]

Right, just to confirm:

• The bulb is wired directly across the solenoid and ONLY across the solenoid. It's not connected, say, from the switch to a good ground, or anything meaningless and irrelevant like that. THIS MATTERS!!! Look up "Three Mile Island nuclear disaster" if you don't believe me.
• And it's not wired in series, either, because that's not a test, it's a guarantee of the solenoid not working.
• Have you done the clonk test as I described, with a wire directly to the solenoid? Yes, you do have to crawl under the car to do this but you have changed the solenoid so you've been under there. I can assure you that a J-type solenoid makes a fair clonk when held in your hand or mounted to a dry gearbox, but it would probably be well masked by the soundproofing that stops you hearing the noise of the gearbox when driving.

[Mjit]

So the temporary solenoid wiring is:

1. Loom +ve feed -> female bullet in cabin (standard).
2. Male bullet -> one side of gearstick switch (standard).
3. Other side of gearstick switch -> female bullet in cabin (standard).
4. Male bullet -> female spade on one side of solenoid (standard).
5. Female spade on other side of solenoid -> female bullet back up in cabin (testing addition).
6. Male bullet in cabin -> male spade under car (testing addition).
7. Female spade -> earth ring on overdrive mounting bolt (standard).

This is the standard wiring, just with the "solenoid female spade -> earth ring on overdrive mouting bolt" temp. extended to look back up inside the car where I can quickly and easily fiddle with it on a test run.

The test bulb is then connected to a pair of fly leads, one ending in a male bullet, the other in a female bullet.

• For "Is the overdrive working?" testing I run it exactly as outlined above.
• I can then swap the test bulb between:
  • 1. and 6. to confirm I have power coming in from the loom, both static and on the move.
  • 3. and 7. to confirm the switch is working, both static and on the move.
  • 1. and 2. or 3. and 4. to confirm I have a circuit from 1. through to 7. when switched "In", again both static and on the move.

I'm not specific trying to/expecting to have a working overdrive with the test bulb in the circuit, just swapping it in when the O/D isn't working to try and identify an intermitten supply/switch/earth issue, so proving the electrical side of the circuit is sound.  In these tests I can run the same short road loop, flicking the switch (no bulb) and have no O/D, then swap the bulb in and run the same loop flicking the switch as normal and get a solid light each and every time.  Swap the bulb back out of the circuit and repeat and it's no O/D again.  This leads me to belive it's not an electrical issue.

 

'Life' means evenings are out for car time so I only have lunch hours (when I get one) and weekends - and the car's outside so that's weather permitting.  Today's lunch hour was only time to get everything out, the car jacked up, the solenoid swapped, the car back down and everything put away in time for a quick drive around the test loop.  Not sure I'll get a chance to tinker tomorrow so will try listening up close and personal to the solenoid Friday/on the weekend

[NonMember]

I really, really don't understand why you think it's sensible to arrange the test bulb so that it can't be in circuit when you want the overdrive to work. That just seems utterly bonkers, given that it's the overdrive NOT WORKING that you're trying to diagnose.

Connect your test bulb directly across the solenoid - you can do so with a nice long pair of wires so the bulb itself is in the cabin, but you MUST leave the wiring of the solenoid in the condition you expect to work and ONLY ADD diagnostic probes that don't interfere with that.

The tests you've done are not testing for all the possible electrical issues. They fail to check for resistive connections, intermittent contacts, tired wires and other stuff. Do the test properly, as I've described, and you actually can say, for certain, "there's power to the solenoid but the overdrive's not working". From what you've done, you can't.

[Pete Lewis]

i agree with Rob   you seem to be getting into deep water for a simple  bulb test jumped into the sol  pig tail connection 

i would add use a decent bulb like 21watt so your test bulb also pulls a reasonable current while testing 

just use a long lead and run the wires into the engine bay stick the test bulb to the screen to view as you drive 

Pete

 

[Mjit]

OK, so I think I may have cracked it - it was of course the biggest PITA to access component in the system, the inhibitor switch.  Despite having tested the switch on the bench before fitting as I know what a swine it is to change on the car, having swapped it for a new one the O/D seems to be behaving (so far) and testing the old one on the bench now shows intermittent contact depending on exactly the angle it's pressed at/how far/just randomly.

On the one hand I have a working O/D again (subject to more miles to check this WAS the issue and it's not just a good spell on another intermitten issue) and also a seatbelt warning like that comes on when I'm in o/d (switches inside the seat belt buckles are dead and not sure I want to try splitting them open to fiddle around).
And on the other a renewed exasperation that Triumph's engineers managed to put 2 holes in the 2000 transmission tunnel, one for the gearstick and I'm guessing the other something for autos - but neither of them in a useful position for accessing the inhibitor switch!