Garage Door Spring

My garage door spring broke and also trashed a lift cable.
this is the style with the spring on the torsion bar.
I took it all down, installed a used spring, found out the big box stores don't stock anything for this style door opener cause their too dangerous and most people call in the pros, so I had to make my own replacement lift cable.
The door only trashed one cable so I had a cable to duplicate.
The problem I'm having now is that when the spring is loaded to the max and I lift the door, the spring is fully unloaded but the door still has a panel and a half of travel to go and if I lift further, the lift cables turn into a birds nest.
So I reset everything and loaded the spring to the max (almost breaking point) and the same thing happened.
The old spring was 2-1/2" diameter, 30" long and the new/used spring is 2-1/4" diameter and 41" long. I thought cool, not as fat but longer is stronger.
I even brought home the drums for the new/used spring thinking that they were larger and would take up more cable but found out they are no bigger that my originals. The one difference is that the door the new/used spring came off of was a 12'x7' and mine is a 16'x7'.
Short of finding the breaking point on this new/used spring or a setting where it doesn't stop unwinding before the door is open, am I correct in thinking that this spring just isn't rated for this door regardless of it's size difference to the broken one?

Good assumption, you need to match the spring exactly, not just size. You can build a jig and measure with a torque wrench. how much force does it take to streach a set distance.
If you can't match the old one find two that are very close to the old and are the same.

Your awesome nealtw.
You just gave me an idea.
The springs I have don't hook into the hubs, they just screw in until they meet with the obstruction of the set screw shoulders.
I can now remove the hub off the short side of the broken spring (approx. 3") and reattach it to the remaining spring and install it on the other side of the torsion bar.
This should buy me enough time to save my pennies for new matched springs and lift cables (especially since I insulated the door myself with 2" pink foam board and rebalanced the door after insulating, several years ago). The new springs I get will be rated for an insulated door so I don't have to stress the spring into an early failure.
I just got done resetting everything and applying 8 turns or 32 cranks on the spring. However, the more I raise the door manually, the more the shaft rides to one side and eventually binds lift drum into the bearing flag.
Hopefully, the second spring will offset the lateral torque of the first spring.

When I had a shop in a complex, a bunch of us decided we should insulate the metal doors. Before I got around to doing mine, there were three or four broken springs in the complex. The door service people loved it

Much to my surprise, the spring only costs around $50 and new lift cables are under $10 for a set.
However, giving them the length, diameter and wire size is not enough for them to order a replacement. They say I have to bring it in and they will match it up.
I say, I've given you everything you need to know to order a new cable but, if I want a new spring, I have to play the game.

They can have the same look and be very different.

Ok, I decided I had better do this right.
I went out and purchased two springs rated for my door w/insulation.
No sense trying to get mismatched and under rated springs to work.
If most people knew how garage doors worked, they'd never walk under them.

kok328 said:

when the spring is loaded to the max and I lift the door, the spring is fully unloaded but the door still has a panel and a half of travel to go

Click to expand...

What's your drum diameter and door weight? I assume the lift is 7' and your drum diameter is constant and not tapered.

A torque wrench is a good idea for checking the spring before you buy.
The lb-ft required to twist the spring just one turn will tell you a lot. For a 200 lb door each spring will see 100 lbs and could be wound about 7 turns when the door is closed. 100/7 = 14 lb-ft per turn, per 360 degrees. This is the "spring constant."

The thing with all springs the life expectancy is a function of number of deflections and percentage of total deflection. When I design a cutting die that the springs are cycled millions of times we can only deflect the spring maybe 10% to get the life we needed. If you add weight to the door and then take the spring deeper into its maximum deflection range you have to expect shorter life.

In the case of a door the weight needed to be counterbalanced changes as the door goes up. The spring also has a liner curve of its force to match the door weight. So the force goes down as the required counterbalance goes down.

If I had a garage and if I had this problem I'd make a portable spring testing jig to take to the store.

Then, I'd preload the candidate spring 1/4th of a turn and then measure the torque to get to the next 1/4th turn, so by now the spring is twisted 1/2 turn. Subtract the two lb-ft values and scale up from there.

The Web is full of advice on this subject and one link had a calculator that works for their spring inventory and a 4" drum diameter.
A drum with more than one diameter is little messier to figure.

The more numbers you have, the more measurements you take, the sooner you will know if your current strategy for this project will likely work out.

Sorry, but the calculations have made me dizzy!

havasu said:

Sorry, but the calculations have made me dizzy!

Click to expand...

The next stage is that your brain explodes. . .:hide:

I just went to my local OHD and told him what I had and that I install after factory insulation. Ended up upgrading to a double spring for only $15 more than a one spring application. I also inquired about my making my own cable and they said that's how they make theirs only they have a crimping tool versus a hammer.

Skinny cable is pretty strong, page 2 of

http://www.stormcable.com/uploads/Tensile_load_support_for_cable.pdf

For personal safety, this cable including its end terminations should be at least 10x stronger than its normal load of 100 or 200 lbs.