Double Bars Blues

[AugustneverEnds]

I'm starting to think as time wears on more of us will be faced with the prospect of replacing torsion bars. For me it's double jeopardy this year as both have broken in nearly the exact same spot. I suspect my bars were overtorqued in the past. A few visuals:

58322
58323

I bought a passenger side bar and have a solid lead on a drivers side bar. I might be foolish but I want to try to replace them myself. I need to construct the jig and buy a quality 3/8" hex socket. A few questions: If the roof has separated should it be plainly visible? I thought before I took anything off the car it might have separated but now I'm not sure. It seems solid, no flexing. I know my car has spent its life in the Northeast and I'm surprised I don't see much rust under the T panel so I believe the roof box is intact. I also have a receipt from a previous owner stating the rear window was replaced though I'm not sure if that affects much.

Thanks for any help or tips

[dn010]

It is hard to tell in your photos but your roof does not seem to be separating and it would be obvious. Post a photo of the back roof above the glass. If you're feeling adventurous, you can remove the headliner and bolt the roof while you're at it to prevent any future separation.

You also don't need a jig to install the torsion bars yourself if you've already removed the louver. A length of pipe or the handle off a jack will work over the hex wrench. Take the piston off the door, open it as far as you can (which will be past what piston will allow, hence its required removal), hold it open with a 2X4 (good to have someone holding the door as well in case something happens) and at that point you don't need that much force when setting your new torsion bars since the doors are open beyond their normal range of travel. Thread the torsion bar retainer bolts in by hand at first to prevent them from X threading, that is an awful place to have stripped parts. Leave the piston detached, lower the door and see how far open it remains under the torsion bar tension only. Adjust the bar as needed.

If you want to really check for rust, remove the inspection cover in the center of the roof and use an inspection camera if you have one. If not, put your cellphone on video with flash on, put the phone in the roof so the camera is in the hole and see what it looks like then.

Also - that sucks! Sorry to hear about both going on you, hopefully this did not happen while you were driving.

[Rich]

For me it's double jeopardy this year as both have broken in nearly the exact same spot. I suspect my bars were overtorqued in the past. A few visuals:

58322

First, +1 on all the advice you got from Dan in his post.

Second, a question for you. Please tell us more about that "exact same spot" where both bars failed. Does it line up with any part the rear door hinge it was next to?

Lastly, if you post up another photo of the bar brackets see if you can take it directly from the rear. Would like to see how much they are rotated and also see if the box is locally deformed. As Dan notes it's hard to tell from the angled photo in first post.

[AugustneverEnds]

More pics: 58326


I'll get some closer pics of the broken torsion bars. Both broke near the rear hinges, yes.

The driver's side broke over my head when I was driving and the worse part was I knew exactly what it was. The passenger side went unheard

[AugustneverEnds]

58328

Under the T panel after I pulled off the Dunmurry tape.

[Rich]

More pics: 58326

I'll get some closer pics of the broken torsion bars. Both broke near the rear hinges, yes.

OK on that more-photos plan, and maybe you can open a door and temporarily put the front section of its bar back in place to show us where the busted end lines up vs the hinge.

Meantime I'd say your roof box looks normal in this photo. The brackets usually rotate a tad due to the mounting hole clearances - I see no buckling or deformation of the box. Other votes on that?

[SS Spoiler]

I store my car with the doors open all winter to
release pressure on the bars,,,

[AugustneverEnds]

OK on that more-photos plan, and maybe you can open a door and temporarily put the front section of its bar back in place to show us where the busted end lines up vs the hinge.

Meantime I'd say your roof box looks normal in this photo. The brackets usually rotate a tad due to the mounting hole clearances - I see no buckling or deformation of the box. Other votes on that?

I think I'm going to do the bolt down procedure anyways. The UK Club has a nice kit and it's worth the peace of mind to me. Already have everything ripped apart so might as well do it now.

[powerline84]

Your box hasn't popped bad but its definatly starting to lift. Your doing the right thing

[AugustneverEnds]

You're dead right about the box starting to lift. When I unbolted the retaining plates this morning I definitely saw movement. Time to get the headliner out.

[opethmike]

I second not needing the jig. That's only for if the louvers remain on. Frankly, I think it is safer to do it the louvers off and the breaker bar/socket directly into the torsion bar than having a bunch of extensions and the jig to worry about.

[powerline84]

Yeah man I think they all are at this point . I mean it was just adhesive. You will notice your doors close better as well . Dont be surprised if you have to adjust the hinges in the doors slightly .

[MikeWard]

Meantime I'd say your roof box looks normal in this photo. The brackets usually rotate a tad due to the mounting hole clearances - I see no buckling or deformation of the box. Other votes on that?
That roof box has lifted for sure, if the bars were still intact it would look even worse than the picture shows. The separation of the box is the most likely cause of the bars snapping.

I think mine was caught just in time;
Bars in
58345

Bars out
58344

Re the jig, having done it myself in my opinion it's much safer to go with a jig and extension bars, but to each their own.

Flick through the pictures on the repair kit page on DeLoreanGo and it will show dimensions for an adjustment jig. https://www.deloreango.com/us/roof-box-repair-kit.html

[Rich]

You're dead right about the box starting to lift. When I unbolted the retaining plates this morning I definitely saw movement. Time to get the headliner out.

Nice. Go for it.

A little movement is normal when the torque is removed even on a good box, but it's less than 1mm. The boxes aren't infinitely rigid and there's no spec.

Good luck with the job.

When you get a chance please post up a photo of the forward piece of one of the bars installed in its front hinge - curious to see where it broke relative to rear hinge. No need to send a pic of both bars since you report they each broke in the exact same spot. This is about the failure mode.

[Rich_NYS]

There needs to be clearance between the torsion bar & hinges...do you think your torsion bars might've been rubbing?

Luke pointed it out to me, and I've heard it mentioned a few more times since.

[Rich]

There needs to be clearance between the torsion bar & hinges...do you think your torsion bars might've been rubbing?

Luke pointed it out to me, and I've heard it mentioned a few more times since.

+1

You are on to my question about this particular situation. Both of the OP's bars reportedly broke at "exact same place" at different times recently. Later there's a report that this failure location is at or near rear hinge.

This seems to be a strong pattern with D torsion bars - perhaps a vendor will chime in on their experience in this.

I contend that a properly-torqued bar (never up-torqued to compensate for a worn strut) that's cushioned from the rear hinge (or not in contact with hinge to begin with) will last indefinitely. Oh, and the bar's surface must not be nicked with tools during other work such as upper door seal replacement...

[FABombjoy]

I caught my drivers side bar just as the hinge was starting to polish it from very light contact (around 1999 or so). Left much longer and I'm sure it would have been damaged enough to break. You should be able to pass at least a business card between the bar and hinge. If not, its rubbing.

PO had a dealer adjust the bar to compensate for worn struts which accelerated the debonding and shifting process. I did the sheet metal screw through the rear method going around at the time but should probably switch to the bolt through method being done now.

[David T]

All of the broken torsion bars that I have seen have broken right in the area where it passes by the rear hinge. My suggestion is to get a small piece of old inner tube, about a 2" X 3" piece, and stick it in between the bar and the hinge. You can pull the bar away just enough with your fingers. DO NOT USE ANY METAL TOOLS THAT CAN COME IN CONTACT WITH THE BAR EXCEPT THE 3/8 ALLEN TOOL IN THE END! Unless you want all that extra work fussing with the rear louvers, the jig and the extension bar is necessary and not all that hard to use. The jig you can make, the extension can be expensive but if you cheap out and buy junk you can get hurt. There is a LOT of power in those torsion bars and you really should use quality tools. Besides a long breaker bar, I have 5 feet of pipe I slide over it to make it easy to do. When doing the torsion bar you should have everything and anything you may need handy including taps, wrenches, pliers, pry bars, files, etc so your poor assistant doesn't have to hold the bar longer than necessary. Not unusual to find a bolt for an anchor cross/threaded or stripped, difficult to slide an anchor bracket off the torsion bar that kind of thing. Sometimes the bar is cracked right where the 3/8 allen fits into the bar. Very bad. It causes the bar to expand and prevents you from sliding the anchor off. If you are messing with a torsion bar you must be ready for any eventuality. Otherwise it is best to leave them to someone with some experience. Some DMC cubs offer free torsion bar adjustments. That "free" service makes it more than pay to be a member. NEVER adjust a torsion bar to compensate for weak door struts. Over torquing the torsion bars can contribute to the roof separating, bending and breaking the anchors. Before touching the torsion bar use a pencil and make a mark on the bar and the anchor so you KNOW it's last position. Rare to have to adjust it more than 1 spline. You can adjust 1/2 spline by loosening the bolts, turning the torsion bar and tightening the bar. There is enough play in the holes in the anchor to get a little bit of an adjustment that way. When removing the bars they must be protected so nothing rubs or scratches the surface and creates a "stress riser". It WILL cause it to crack. When working on the torsion bar you must be very careful around the Tee roof. When you remove the rear plastic valence it exposes a VERY sharp edge of the Tee roof. Be very careful with that plastic valence, they are NLA so be gentle. To remove it you roll it out (rotate it) after removing the 3 screws.

[dn010]

Again, if you open the doors as MAX as you can, you're really not dealing with nearly as much torsion. If you leave the door piston installed and mess with the bars at this point then you will deal with a lot of torsion that is unnecessary if you'd just open the door all the way with piston removed. You don't even need to remove the entire piston, just pop it off the door and then rotate the piston towards you and lay it down. Max position with the T-panel installed is when the door contacts the panel, with it removed you can go even further. The plastic valances are NLA but stainless is readily available. The louvers are already removed so there is no need for a jig, not only that it's 4 nuts holding the louver on and you can even leave the struts installed, just mark the louver hinges with some nail polish or similar so you can easily line it back up when reassembling.

It was suggested above that you may need to adjust the door/hinges afterwards. I would suggest NOT to adjust the door at the hinges unless for some reason you really need to. Some people get lucky - they unbolt the door and then are able to quickly align everything to their standards, for those people I'd love for you to come adjust my driver side door. Myself and others have not had luck with "quickly" adjusting the doors, it is a PITA to get it perfect and can do more harm than good if you get it wrong. There is a lot of room for adjustments at the door hinges-in towards the middle of the car, out away, hinge all the way up towards the roof, all the way down, angled up, angled down, etc. and there are TWO hinges to adjust, and then you've got the pins.

Good luck and let us know how it goes.




All of the broken torsion bars that I have seen have broken right in the area where it passes by the rear hinge. My suggestion is to get a small piece of old inner tube, about a 2" X 3" piece, and stick it in between the bar and the hinge. You can pull the bar away just enough with your fingers. DO NOT USE ANY METAL TOOLS THAT CAN COME IN CONTACT WITH THE BAR EXCEPT THE 3/8 ALLEN TOOL IN THE END! Unless you want all that extra work fussing with the rear louvers, the jig and the extension bar is necessary and not all that hard to use. The jig you can make, the extension can be expensive but if you cheap out and buy junk you can get hurt. There is a LOT of power in those torsion bars and you really should use quality tools. Besides a long breaker bar, I have 5 feet of pipe I slide over it to make it easy to do. When doing the torsion bar you should have everything and anything you may need handy including taps, wrenches, pliers, pry bars, files, etc so your poor assistant doesn't have to hold the bar longer than necessary. Not unusual to find a bolt for an anchor cross/threaded or stripped, difficult to slide an anchor bracket off the torsion bar that kind of thing. Sometimes the bar is cracked right where the 3/8 allen fits into the bar. Very bad. It causes the bar to expand and prevents you from sliding the anchor off. If you are messing with a torsion bar you must be ready for any eventuality. Otherwise it is best to leave them to someone with some experience. Some DMC cubs offer free torsion bar adjustments. That "free" service makes it more than pay to be a member. NEVER adjust a torsion bar to compensate for weak door struts. Over torquing the torsion bars can contribute to the roof separating, bending and breaking the anchors. Before touching the torsion bar use a pencil and make a mark on the bar and the anchor so you KNOW it's last position. Rare to have to adjust it more than 1 spline. You can adjust 1/2 spline by loosening the bolts, turning the torsion bar and tightening the bar. There is enough play in the holes in the anchor to get a little bit of an adjustment that way. When removing the bars they must be protected so nothing rubs or scratches the surface and creates a "stress riser". It WILL cause it to crack. When working on the torsion bar you must be very careful around the Tee roof. When you remove the rear plastic valence it exposes a VERY sharp edge of the Tee roof. Be very careful with that plastic valence, they are NLA so be gentle. To remove it you roll it out (rotate it) after removing the 3 screws.

[AugustneverEnds]

Again, if you open the doors as MAX as you can, you're really not dealing with nearly as much torsion. If you leave the door piston installed and mess with the bars at this point then you will deal with a lot of torsion that is unnecessary if you'd just open the door all the way with piston removed. You don't even need to remove the entire piston, just pop it off the door and then rotate the piston towards you and lay it down. Max position with the T-panel installed is when the door contacts the panel, with it removed you can go even further. The plastic valances are NLA but stainless is readily available. The louvers are already removed so there is no need for a jig, not only that it's 4 nuts holding the louver on and you can even leave the struts installed, just mark the louver hinges with some nail polish or similar so you can easily line it back up when reassembling.

It was suggested above that you may need to adjust the door/hinges afterwards. I would suggest NOT to adjust the door at the hinges unless for some reason you really need to. Some people get lucky - they unbolt the door and then are able to quickly align everything to their standards, for those people I'd love for you to come adjust my driver side door. Myself and others have not had luck with "quickly" adjusting the doors, it is a PITA to get it perfect and can do more harm than good if you get it wrong. There is a lot of room for adjustments at the door hinges-in towards the middle of the car, out away, hinge all the way up towards the roof, all the way down, angled up, angled down, etc. and there are TWO hinges to adjust, and then you've got the pins.

Good luck and let us know how it goes.

Thanks for your advice. I'm waiting for the kit I ordered to bolt the roof down and after that's done I will install the torsion bars. I do plan to follow your guidelines and disconnect the door struts. The T-panel is off so no worries there, the louvres are off and they need new struts so I ordered those too. My doors alignment was acceptable before so I don't plan on touching the hinges unless something shifts radically. It's been cold (even for upstate NY) around here lately, I still need to take a pic of the broken torsion bars for everyone to see.

[DMC-81]

Sorry to hear about this problem. It seems you got good advice so far on the adjustment. The DCF Club did mine a couple years ago and used a jig. Here is what it looks like in position:

http://dmctalk.org/attachment.php?attachmentid=46426&d=1475370344

When you get the new bars in and adjusted and if you choose not to apply heat shrink tubing as mentioned, you can opt to protect them with vinyl tubing after they are in place. I used a 3/8" i.d., 1/2" o.d. clear vinyl tube slit horizontally. I found this thickness allowed enough clearance between the bar and the rear hinge. The overall length was 26 1/4". I oriented the slit at the bottom, and fastened them with 2 small zip ties at the ends. This way, the tubing can be easily removed for bar cleaning, or whatever:

http://dmctalk.org/attachment.php?attachmentid=54628&d=1511211476

Good luck with the job.

[HollywoodF1]

I store my car with the doors open all winter to
release pressure on the bars,,,

It's not the static stress on the bar that ages it, it's the working. Every time you cycle the bar, you age it by one tick. You get to enough ticks and it fails. Assuming it were protected from corrosion, you could leave a torsion bar stressed for tens of thousands of years and nothing would happen. But if you cycled it day and night, it would fail in short order. Metal fatigue is strictly a function of the number of stress cycles (load-unload) and the amplitude of the stress cycles relative to the material's yield point. Trust me-- I'm a structural engineer. You can feel free to keep the doors closed this winter.

[AugustneverEnds]

The more I think and investigate the more I believe the roof box separating contributed to my bars failures. My T-panel was popping up for the past few years and the passenger side would make a creaking noise when it opened and closed. From where the bars broke it's possible they were coming into contact with the rear hinges especially when the doors were closed. I remember reading somewhere Grumman promised the bars would be good for 50,000 opening/closing cycles. Of course I have no idea how many times the doors were open and closed. Presumably the drivers side should have more use and should be the first to go. Maybe we should all install counters for the doors like an odometer or lambda counter. :hmm:

[Rich]

Grumman's lifetime estimate may not have included any provision for DMC installing the bars in such a way that they rubbed against the rear hinge. Some of them interfere with the hinge, some don't and none of them should touch the hinge.

In case your roof box deterioration helped either of the bars hit its respective rear hinge then, yes, the roof box deterioration could be linked to one or both failures.

Had your bars both been protected/cushioned before they hit the hinge too many times they may both may still have been in one piece. Nothing you can do about that.

Recommendation: After your rebuild see where each of the replacement bars is. And cushion whichever one is against the rear hinge.

[David T]

It's not the static stress on the bar that ages it, it's the working. Every time you cycle the bar, you age it by one tick. You get to enough ticks and it fails. Assuming it were protected from corrosion, you could leave a torsion bar stressed for tens of thousands of years and nothing would happen. But if you cycled it day and night, it would fail in short order. Metal fatigue is strictly a function of the number of stress cycles (load-unload) and the amplitude of the stress cycles relative to the material's yield point. Trust me-- I'm a structural engineer. You can feel free to keep the doors closed this winter.

The "heat treat" (cryogenic) and twisting that Grumman did was supposed to induce more stress into the bars then they would see during their design life. The fact that all of the bars I have seen broken all happened at the same spot would seem to indicate the formation of a stress riser in that area and the most likely possibility is that rear hinge. Any surface defect (scratch) will cause the stress to concentrate in that area overstressing the metal, leading to failure. I don't believe they fail because of too many cycles (openings and closings). If that was so you would not expect all of the failures to occur in the same general area of the bars.

[Nicholas R]

The "heat treat" (cryogenic) and twisting that Grumman did was supposed to induce more stress into the bars then they would see during their design life. The fact that all of the bars I have seen broken all happened at the same spot would seem to indicate the formation of a stress riser in that area and the most likely possibility is that rear hinge. Any surface defect (scratch) will cause the stress to concentrate in that area overstressing the metal, leading to failure. I don't believe they fail because of too many cycles (openings and closings). If that was so you would not expect all of the failures to occur in the same general area of the bars.

It seems like what you're saying is contradictory. The generation of a surface defect comes from repeated cycling as HollywoodF1 said. Rubbing the hinge once wont do anything; but rubbing it 2000 times may. As he said, statically loading the doors isn't what ages them, it's the cycling (the point being that having your doors closed wont pre-maturely age your bars).

[David T]

I don't think what I said was contradictory. if you continually damage the surface then every cycle will do more damage and concentrate the stress. My theory is if the torsion bar rubs against the rear hinge it damages the surface creating the problem. If the bar never rubs or gets scratched by the rear hinge it should not fail prematurely. Another point. The number of door closings and openings does seem to have an effect, we see more driver's side bars broken then passenger side. That would make sense because you would figure the driver's side door would be opened and closed more often than the passenger side. My strong suggestion to every owner is to protect their bars by getting a piece of rubber in between the bar and the rear hinge. It can't hurt and may save them a bar! Rubbing (scratching) the hinge once may be all it takes but if you continue then every time you open and close you are making the scratch worse and reducing the strength of the bar.

[Farrar]

When you're done with the roof cage, you may need to adjust the door striker pins. Take your time and be patient. You've got this. :)

[AugustneverEnds]

When you're done with the roof cage, you may need to adjust the door striker pins. Take your time and be patient. You've got this. :)

Thanks a million! I know your car is giving you grief but you have already come so far with it so now every day is one more closer to it being all squared away. We shall overcome :)

[Farrar]

I know your car is giving you grief but you have already come so far with it so now every day is one more closer to it being all squared away.

Thanks, I appreciate that. It's hard not to get frustrated and doubtful, but it's good to try to treat every setback as something to learn from rather than something to be angry about. And if not, well, it can wait. The best lesson I've learned is to never work on the car when I'm in a bad mood. Take that for what it's worth. LOL

[AugustneverEnds]

58455

The broken off hex ends of my car's former torsion bars

[DMC-81]

Thought this was interesting: I was just watching this Pennebaker documentary and at 30:06 they discussed a torsion rod failure ( presumably in a preproduction car) at 41,000 cycles and JZD asked the right questions including a request for a metallurgist's opinion. The presenter said that Grumman cycled them "into the millions ", which agrees with what was said earlier.

http://youtu.be/sueqDrq9VBw

Plus a worker was filmed installing a bar on a car at 37:57.

[AugustneverEnds]

Thought this was interesting: I was just watching this Pennebaker documentary and at 30:06 they discussed a torsion rod failure ( presumably in a preproduction car) at 41,000 cycles and JZD asked the right questions including a request for a metallurgist's opinion. The presenter said that Grumman cycled them "into the millions ", which agrees with what was said earlier.

http://youtu.be/sueqDrq9VBw

Plus a worker was filmed installing a bar on a car at 37:57.

I noticed that worker with his breaker bar and it reminded me that human beings at Dunmurry installed those torsion bars in the first place so I can do it too :)

Just finished bolting the roof down and waiting for the sealant to cure. Hopefully in the next few days I will have the torsion bars installed and my D will have working wings again.

[AugustneverEnds]

58681

The double bars blues has been cured! Thanks to everyone who helped

[CFI]

58681

The double bars blues has been cured! Thanks to everyone who helped

Nice job!

[HollywoodF1]

I don't think what I said was contradictory. if you continually damage the surface then every cycle will do more damage and concentrate the stress. My theory is if the torsion bar rubs against the rear hinge it damages the surface creating the problem. If the bar never rubs or gets scratched by the rear hinge it should not fail prematurely. Another point. The number of door closings and openings does seem to have an effect, we see more driver's side bars broken then passenger side. That would make sense because you would figure the driver's side door would be opened and closed more often than the passenger side. My strong suggestion to every owner is to protect their bars by getting a piece of rubber in between the bar and the rear hinge. It can't hurt and may save them a bar! Rubbing (scratching) the hinge once may be all it takes but if you continue then every time you open and close you are making the scratch worse and reducing the strength of the bar.

I think we have a consensus here.