What is the proper method of securing the ball cap to the oil pan? I see in the manual it requires 2 automotive studs, 2 springs, 2 castle nuts, and wire. I also see the parts suppliers have these. However, when attaching these to the oil pan, how much tension should be on the springs? Before I looked in the manual and looked at the parts suppliers, I only had 3/8'' bolts with lock washers and wiring going through the heads of the bolts... now I'm wondering what I should do? The springs don't seem to keep much tension on the ball cap. Did I buy weak springs (don't seem to weak to me; is there a spec)?
I am new to the game, but early cars like my 1913 used holed bolts with a wire to secure them. Later they used studs with castle nuts. What year is your car?
From what I can remember the spring should not be fully compressed,so it may not feel like it is holding anything steadfast.
If there feels like alot of slop,check condition of the ball on the wishbone and check the socket on the oil pan.From what I remember you have the 4 dipper pan.
But yes,there should be 2 studs,2 springs,the cap and 2 castle nuts and wire between the studs.
If you eventually tour,someone will peek under there to make sure it is installed that way for safty and liablity reasons.
There is a accesssory cap that is spring loaded as well.
A freind of mine that restored T's back in the 50's and still drives them on tours today,wrecked in 1 back in the 50's.The wishbone came loose and it flipped end over end.No loss of life,.But that is 1 point he made clear to me was to get that right.
Brian, it's a 17. I guess I should put it back the way I found it. It will be more secure and I will feel better about it.
The spring should be fully compressed. The idea of the spring is to maintain pressure on the ball as it wears and the recess in the pan fitting wears. So you tighten the nuts until the springs are flush, no gaps. Then safety wire. There is no torque value on any poart of a Model T.
Obviously you need to grease the ball before assembly and plan to dissassemble and grease it every couple years if the car is driven often.
Guys,don't forget to figure 8 that safty wire. It will keep studs from backing out of pan.
4-dip pans have the socket mounted on a standoff, so you can use bolts clean through the threaded holes, and jam nuts instead of studs that have to be safety wired to keep from backing out. Dunno what Henry used on the 4-dippers.
Personally, I don't think that the spring's function was to automatically take up wear. The ball area will wear while the flange area, where the studs are, will not, at least not to the same degree. I think the springs' function is similar to that of the radiator springs: to allow some twisting and "give" which will help protect the pan and reduce the amount of the wear. Just my opinion.
I guess I forgot a very important step. In my zeal to put it all back together, I forgot to lube it. Thanks for the heads up. Michael
I would suggest fully collapsing the springs, as you suggest, but then backing off maybe 1/4 turn before wiring the studs. You wouldn't want to make the ball socket bind by fully tightening. The need to get the springs nearly fully collapsed is so that they can't compress any further and let the cap drop enough to let the ball escape.
.... and if the ball is worn, the period replica accessory spring cap supplied by Frank Fenton is excellent (no connection other than as a happy customer).
I agree with Jerry VanO's advice.
With respect to those suggesting the spring be totally compressed, that would not be an acceptable install and defeat the purpose of the design. The springs are there to absorb any abnormal shock encountered and allow components to return to the original position without any damage to the components. If they are completely compressed, they are useless and there might as well be no spring at all. Leaving them with a gap will not allow the ball to jump from the socket, but it will allow it to flex and move in the event of an extreme shock, thereby lessening the potential of the ball from being snapped off the wishbone.
A fully compressed spring has a tendency to break too. The ball will not come out of the socket when properly installed. The nut is run down to align with the stud hole and that's it. It's similar to the rear engine mounts and steering bracket--The side fastener is NOT supposed to be drawn down tight.
I'm still the new guy here, but just as an aside, I looked up bolts versus studs and nuts in the 'Model T Parts Identification Guide, Volume I'. Bolts with a lock wire were used from 1908 until the end of 1913. 14 and later used studs and springs. From an authenticity and functional standpoint, I would replace the bolts with the correct studs and springs for your 1917. It is a better design. Better still my 1913 runabout happens to have an original Apco spring loaded ball socket installed. Makes sense to me to keep it as it looks like good engineering. They are available from Lang’s and others and are the answer to a slightly warn ball or oil pan. The quality of the reproduction is top notch.
13 T runabout
25 T speedster
We have exchanged emails as I question his post on the use of bolts on a 1913. The information In Volume I parts Id Guide states this. However the copy write is 1994 and contains information of Gail Rodda's research. The Judging Guidelines put out by the International club, which includes Gail's info, copy write date of 1997 ,last revision, at least the last copy I have states that cap screws wired together, were use in 1911 and change to a stud with castle nuts in 1912 and is correct for later years. My question to the form is which is correct.
In the past 50 years I have always bottomed out the springs and safety wired the two studs and castle nuts together. I can't see allowing for any play in this connection could help anything. Others may tighten them as they see fit but be sure to safety wire through both castle nuts and studs simultaneously.
Putting cotter pins thru each castle nut might look pretty but will do nothing to keep them from coming unscrewed from the pan.
P.S. I just realized for the first time as I typed this note I have owned and driven a Model T for the last 50 years. This is a real accomplishment for a person who just turned 39 (again) last summer.
For the past 25 years I've put cigarettes in my mouth, applied fire, and inhaled the smoke. Neither the safest way of doing things, nor the correct way, but because I haven't had a problem yet it must be just fine. I encourage everyone to do the same!
Seriously, why leave the springs on there if you are going to defeat the purpose of the design, and in fact make it more apt to fail? A compressed spring is nothing more than a sleeve formed of coils adding much more stress to the bolts than they were designed to take. The ball can not move under extreme pressure, putting far more stress on it too than it was designed to take in the event of a pothole or other abnormal application of force. There is far more potential for failure with the springs compressed than with them installed correctly. You're better off with them not there at all than with them bottomed. That ball ain't coming out of that socket, until something breaks or the nuts come off. The springs absorb any excess energy that might otherwise break something.
Ted, you've driven a T longer than I've been alive, so please don't think I'm trying to be disrespectful to you or trying to argue. You might not see any reason for the design, but its not "play" (unless there is a lot of wear). The springs on the suspension aren't play, they are there to allow movement and prevent over-twisting of the frame and body. Same with the engine mounts as Ken mentioned. The rear mount springs are there so the 3-point engine mounting system can work correctly, allowing movement when more force is applied than the components are designed to receive. Skyscrapers have to flex, or they break. Henry understood that steel can only flex or be pulled so far, and past that point it must have a failsafe in order to absorb any further force.
I'm not trying to change your mind, you are free to do it as you wish and I respect that.
Basically I just want to make sure new T owners or tinkerers don't misunderstand and think it's safe and OK to do it that way. "I can't see allowing for any play in this connection could help anything" might be a good enough reason for Ted, and he is free to do it that way. Just like I haven't quit smoking because I don't see any reason to...I haven't got cancer yet. If you understand the reason they are there, or simply realize they are there for a reason, just install them correctly. Tightened down until the holes can be accessed with tie wire, then figure-8 both studs together as others stressed.
Mr Anderson is correct, the springs have nothing to do with taking up for wear. The anti-rattle caps have an internal spring, and that is designed to take up for wear. Later Fords didn't have springs on the cap, but they had something the T doesn't have; much beefier wishbones!
Let's think about this logically.
If the ball and the socket and cap ARE NOT badly worn, the surface of the cap around the stud holes will not contact the surface of the socket.
If you have the springs tight (but I agree with backing off half a turn), then as the ball and cap wear, the springs will expand and maintain ball/socket contact without a rattle for some time - certainly until it's time to re-grease everything.
If the ball, cap and socket ARE worn so that the cap is touching the socket around the stud holes, the springs are of no use. The best of a bad job is to have them solid so the cap cannot move away from the loose ball at all. If things are this bad, you should at least put a spherical shim inside the cap (in the UK a 2 pence coin hammered into shape is excellent) , and/or file a little off the cap surface. The when you have a gap, set the springs almost tight, as above.
If you fit the excellent APCO device, its cap may as well be hard against the socket because its own spring (which is quite stiff) will stop any unwanted movement. In fact, I think that tightening the centre spring will probably compress the stud springs to 'solid' anyway. So, although the instructions don't suggest it, you might as well have no springs on the studs if you have the APCO. But the nuts still have to have figure 8 wirelocking!
I agree with the many here who think that fully compressed springs defeats the purpose of why they are there in the first place, which is to provide a flexible ball joint connection at the pan to lend stability and support to the front axle. To tighten them down will result in the conversion of this flexible connection to a rigid connection which could, actually result in a catastrophic failure of the whole front end, as well as, possibly ripping a hole in the pressed steel pan. The springs should, absolutely, without question, be tightened down, then backed off equally (1/4 to 1/2 turn sounds fine to me) until there is space between the coils, then secured with wire. I have learned to never try and second guess Ford's engineers. Jim
to spring or NOT to spring; that is the question! So, which is it? Tight w/o or loose with springs?
I agree with Royce, who posted way above here on this thread that the springs are there to take up the slack when wear occurs. That's how it was explained to me by my Model T gurus, and no one has yet changed my mind on that.
When you put the assembly together with a liberal amount of grease, everything is fine. But I think that the guys who designed this setup were smart enough to know that the average T owner was not going to disassemble that connection and reapply grease every now and then. Without grease wear will occur, and there will be some slop in that connection. The springs are there to take up that slack.
Has anyone ever seen a ball joint cap drill and threaded to accept a grease fitting? It would make greasing the ball of the wishbone much easier and thus, more apt to be greased. Jim
Tigger. Without question, semi-loose WITH springs. Jim
Mike,I must respectfully disagree. The wear occurs in the ball,not on the shoulders of the cap. If they are tight against pan ,a 10 ton jack isn't going to move them to tighten up anything. This is a floating connection for that very reason.
Good one, Uncle Jack! I didn't know the shoulders of the cap were supposed to be sitting against the upper support on the pan. I've never seen one put together with all new parts.
I dont have the Ford book with me, but i recall it stating to fully compress the springs. One way or another i would check the book for its recommendation.
I've been misunderstood. Figure 88 in the shop manual says to reinstall it with studs and nuts and safety wire it. Doesn't say tighten it all the way down. If you tighten it down it can not float and in my thinking,it needs to be able to float just a little. It 's part of that three point front suspension thingy. Your milage may vary.
Page 31, Par. 104, Line:
"(d) Place ball cap springs "C" over ends of studs and run down the two radius rod ball cap stud nuts "D" and wire the nuts together"
To fully understand the meaning of "run down" you need to be familiar with the terms used throughout the manual. The key word not included in this instruction is the word "tight" or "tightly"--As in "run down tight". Other instructions throughout the manual use these terms to describe how the bolt or nut is to be assembled. The tight and tightly descriptors are noticeably absent from the radiator installation, rear engine block mount and the steering bracket block mount.
In my opinion, the instruction is meant to convey that the nut is run down far enough to insert the wire and no more.
Same here Ken.
If the ball and socket are not perfectly round, and the joint tight, the working portion of the joint is the neck of the ball. It WILL fail. I can provide a photo. :-(
The ball is large. You can see how far you need to open up this connection for the ball to come loose. It is a long way. If not replace the ball or cap.
One of the cars at home has a grease cup mounted into the cap. It does get occasional greasing.
I think Jack and Ray summed it up very well. A spring can only work as a spring if it has room to move. A coil spring tightened up until the coils touch each other becomes in effect a spacer, not a spring. You may as well stack washers on those studs instead. As Jack said, where the shoulders of the cap contact the pan is not a wear point. The wear point is the ball and the inside of the cap. I keep mine "run down" and wired, but you can see light through the coils! Now, how tight is the front crank snout motor mount supposed to be? Mine is tight as heck, and now I'm thinking I need to loosen it!!! Anybody know that one?
Service Manual, Page 32, Paragraph 105 says to run down the bolts tight and there should be no play. If there is, the beraing cap should be filed down in the same manner as a main cap.
I have a NOS engine mount bearing and cap, and it is has a honed bore. It's not just drilled, but bored and honed for what appears to be a wear surface. The bolts should definitely be tight, as running loose will cause wear on the mating surfaces.I'm going to say (with no authority or expertise claimed other than 20+ years designing and building hot rod chassis) the cap should be snug on the pan but the pan should be allowed to rotate if need be. I will offer my reasons for believing this but welcome any disagreement with an explanation;
(1) if the pan cannot rotate, the nose of the pan will be twisted loose due to the amount of torque that can be applied by the frame. This can break the solder joints loose and cause a leak.
(2) The pan is not a structural piece of the chassis. It is not there to add strength and prevent flexing of the frame. If it were, it would be mounted rigid at all three points (actually there would ideally be four points if it were intended to add rigidity). With any 3-point mount system, one point must be able to move in order to prevent stress breaks. Therefore, being not a structural member, it must allow the frame to move and flex without hindering that movement.
There should be no play fore and aft either, since the front bearing is what controls thrust of the rear axle. Any force applied from the rear axle is transferred through the drivetrain to the front mount. Fore/aft movement over time will stress the rear engine mounts and cause a leak or total failure.
Consider This, Henry never wasted anything. As posted above a fully compressed spring is nothing more than a spacer. Henry's engineers would know this, and you know they would see no need for the studs to be that long. I'm sure if it was meant to be tight, those 30 million + studs would be shorter with a lock washer rather than cross drilled for a cotter key. If the spring is allowed to work as a spring, I believe it's function is to be a tensioned flexing joint that would actually decease wear on the ball. If compressed and as wear began that ball would bang around in a solid chamber. With the weight of the engine, the pile driving effect of high pressure tires against chuck holes and all the the other obstacles available to those early drivers, and with wear beginning it wouldn't take long for it to start sounding like machine gun under there. I think spring loaded flex was necessary. Model A's still were spring loaded and Model B's and V-8s had a hollow rubber ball that fit over the wishbone ball as another version of a cushioned flexing joint. Just my take on it.
I forgot to mention the savings on 30 million+springs. I read somewhere that upholstery installers were told to space tacks farther apart to save . I have a box of hemi-spherical washers for wishbone ball joints. the box doesn't say what car or piece of equipment they are for. They will fit a Model T. I used two of them until I located a better wishbone. They worked very good.There were several makers of spring loaded caps. They were adjustable just like a tie rod end or drag link. I use them and they really save those ball ends. Lubrication is necessary for maximum wear control.
Ray, I think you posted to the wrong topic. Yours goes in the snout bearing thread. This is the radius rod thread. ;)
Regarding the comment about using a lock washer instead of having a drilled studs. There is a reason for the drilled studs. As mentioned in an earlier posting, the two studs are safety wired together. Why is this, one may ask? Because these studs are threaded into the bracket on the pan, and they could get loose. The safety wire minimizes the risk of a loose stud falling completely out. Remember - they weren't using Loctite back then....
These cars were designed by engineers, who actually knew what they were doing.
Does the ball and socket need to flex and float? I doubt it. The ball and socket merely acts as pivot for the wishbone. Flexing and floating happens at the front end and along the frame rails, etc. The wishbone will move up and down with a tight socket.
If the spring is fully compressed, it is still acting like a spring - not merely a spacer - and will take up the slack. Also, keeping it tight will keep the front axle from moving "fore and aft" and assist in keeping caster in check.
Also, why do so many think that setting it tight will create more wear? Sloppiness will probably accelerate wear.
You state, "That ball ain't coming out of that socket....".
Seen it happen with my own eyes. Neither the socket nor the ball was severly worn. (Not stating they were NOS either, but how many cars are running around with un-worn parts, i.e. typical T.) The guy was backing up on grass field, the wishbone dropped down, the ball imbedded in the grass and the front of the car leapt about 3' in the air. Very impressive.
Erik, I disagree. A fully compressed spring can no longer act like a spring, nor retain its' shock absorption qualities. Compressed down tightly and fully, it simply becomes a thick, 3/4" spacer with the ball cap locking the ball of the wishbone firmly in place. The ball is unable to move, so the entire ball cap assembly is force to move where it is secured to the front of the pan below the hogshead, which puts undo stress on the relatively weak pressed steel pan onto which the assembly in riveted. Jim
I never think to look in the book until I'm on this thread, but I'll ask the obvious question. What does it say in the book? It must not say or this wouldn't have gone on so long. It's been a while since I looked at mine, but can you actually fully compress the spring and the castle nuts still have enough engagement to catch the wire at the end of the studs?
Also, I came from the Model A world (Not so very different in this particular aspect). As I recall the general consensus there is to tighten the nuts only enough that the slots in them line up with the holes.
I guess I'm missing something here. A fully compressed spring, to me is still nothing more than a spacer. As that ball and socket wears, the ball becomes smaller, the cavity becomes bigger. I can't see where a fully compressed spring can do anything to help improve this condition. It would have no way of expanding to do anything. The studs were saftey wired. Something I always wondered about, were they always of that design? Since there is a difference in the O.D. of the stud and the I.D. of the spring, there is space for a tightly secured cap to suffer more severe wear,since it would be forced back and not center it's self until a different direction of force would maybe return it to exact center, maybe not. A flexible ball joint would be much more forgiving. I can't see anything between the front axel and the wishbone that could flex or float to ease the load on that tiny ball joint. Again this is just my opinion for the sake of discussion.
I don't think the ball socket was designed to be a shock absorber nor do I believe that the Ford engineers intended the ball to "float" within the socket. I believe the socket was to designed to be a pivot point - no more, no less. The function of the springs is to take up slack. In order to take up slack as the ball and cup wear the spring will expand. Even with the springs fully compressed I doubt that it's possible to lock the ball in place - the wishbone will still be able to pivot as it should.
Think about it this way. The front end of a Model T is virtually the same as the rear end of the Model T. Each has a transverse spring and an axle. The front wish bone is really two radius rods that meet at a pivot under the pan, creating a triangle. There are two radius rods at the rear end of the car - they meet at a ball at the end of the drive shaft forming a triangle (actually two triangles if you count the drive shaft). This ball goes into a socket in the rear of the pan at the fourth main. This creates pivot point no different that the pivot point at the front end. This driveshaft ball and socket arrangement must be tight - no play/slop there. Even though it is tight, it still pivots.
This is an interesting discussion.
Bud - the fully compressed spring is not a spacer between the socket and the cap. The spring is between the nuts and the cap. Here is the order of assembly:
1 wishbone ball inserted in socket in pan
2 install cap
3 install springs on studs
4 install nuts on studs
5 wire studs
Can you picture the above? As the ball wears (gets smaller)the socket wears (gets larger), the cap continues to exert pressure on the ball. In otherwords, over time the cap migrates closer to the socket.
One other observance. If anyone has had a properly re-built or NOS ball, an NOS pan socket and NOS cap, you will see that the fit of the mating surfaces of the ball against the socket and cap are snug. To me, this is additional evidence that the ball was not designed to float in the socket but merely pivot in the socket.
Please don't take this the wrong way but I'm sure I am not alone when I say, I cannot fathom your logic. Before this series of posts, I respected your opinions and, even now, am wondering if you are serious or are pulling off a big joke just for laughs. If this is a joke, it is very irresponsible of you to continue giving such potentially harmful advice to those that come on this forum trusting that the advice they are being given will not kill them. With this in mind, I have pondered all day, whether or not I should even waste anymore time responding to your posts regarding this particular thread, because it seems like everything you are saying is flying in the face of logic, common sense and a most basic comprehension of mechanics, but then I think, if there is just one person out there that agrees with you, I must make one last attempt to make him see reason and change his mind and if you are serious, perhaps change yours as well.
Ironically, you are correct in that the front and rear axles are similar (not virtually the same), partially for the reason you say but more for the reason you deny, which is because they both have floating ball joints to provide flexibility to both axles. The rear axle is well supported at five locations. Two (2) massive u-bolts holding the rear spring securely to the chassis cross-member, two (2) radius rods for lateral support from the differential to the motor and the securement of the differential to the motor and thus, through to the front cross-member by means of a hefty drive shaft housing, which, by the way, also has a floating ball joint surrounding the universal joint, to provide flexibility to the rear end, just as the floating wishbone (front radius rod) ball joint is designed to provide flexibility and support to the front axle.
The front axle has only two points of support and so is not nearly so well supported as the rear. It has one u-bolt clamp securing the spring center to the front cross-member and it is secured to the engine pan via the wishbone ball joint assembly riveted to the engine pan. That's it! So it is crucial that these two points function as they were designed to function, because, unlike with the rear end, if just one of the front end components fails, there is no reserve components to take over the job of the failed component and catastrophoic failure is sure to follow.
The two wishbone arms are securely fastened to each side of the front axle and cannot give at those points. The only point it can give at is at the ball and this ball must be able to rotate freely within the ball joint housing without possibility of binding. If it is securely bolted down as tight as it will go, securely clamping the ball in place between the two socket halves and fully compressing the springs, it will either break off the ball at the neck of the fork, or sheer off the rivets of the ball housing assembly from the pan.
In light of all the convincing and overwhelming evidence to the contrary, for you to continue to say that a fully compressed spring is still a spring and continues to retain the properties of a spring, escapes me and honestly, makes me a little angry, because what you are continually and confidently proposing has the capability of killing someone who is convinced you are right and decodes to follow your advice.
Having had a Model T for 38 years, I am confident of what I say and continue to say and that is, the front radius rod ball cap springs are there for a reason and that is to provide constant but flexible tension to the ball of the wishbone so as to allow it to move easily and freely while, at the same time securing the ball in the two halves of the ball joint assembly.
Michael Bunner, please do not listen to Erik, for to do so may put your life in danger. As an answer to your question:
1. Put the wishbone ball in the ball joint socket.
2. Slide the cap over the two studs.
3. Slide the springs over the studs.
4. Tighten the two castle nuts down tight.
5. Back the off approximately 1/4 to 1/2 turn. 6. Wire the nuts in place with safety wire.
The tight, but not fully compressed springs will, then, be allowed to flex and thus, do the work they were designed to do and that is to hold the ball cap socket firmly against the ball, while allowing the independent up and down movement of the wheels.
Jim Patricks advice is exactly what I do. Bring the studs down until the space between the springs is gone, but the nuts are still not bottomed out.
Thanks Gentlemen, I have been successful in getting it into place, not locked down tight, and with the proper stud,spring, castle nut, and wire in a figure eight. I can get a finger nail in there and wiggle the spring up and down. Thanks again.
When I use the term "pivot" it is the equivalent of your term "freely rotate."
I don't like the term "float" because I don't think the ball should float around in the socket. Pivot or freely rotate - definitely. Floating to me means knocking about inside the socket which exaggerates steering instability and accelerates wear on the ball, cap and socket.
If I fully compress the spring, that doesn't mean that I'm over-torquing or bottoming out the nut (bottoming out the nut could bend/stress the ears of the cap and I don't want that). If I fully compress the spring, the ball will still rotate/pivot - it won't freeze in place.
The driveshaft ball needs to pivot or freely rotate. But I don't want it floating around (i.e., knocking about inside the socket). To me, that is excess slop or play which will actually accelerate wear of the respective components. If it is floating instead of merely freely rotating or pivoting, then it's time for a sheet metal shim.
Now do I make sense? I really do think that we are both on the same page. The only difference is that you back off the nut 1/4 of a turn. I don't think that makes a difference. Even well lubricated with grease, there is going to be immediate wear on the ball, socket and cap the minute I start driving the old flivver those two springs will begin to de-compress.
Royce Peterson's original post said that he fully compresses the springs. Doesn't that agree what what I said? I think Ray Elkins, who appears to have a great amount of expertise, had best explanation with a good reference to the Service Manual. I don't think what I have said is any different than what he said.
I'm not joking but in hindsight, my choice of pivot as opposed to floating have created controversy.
Here goes some more blatant advertising of a product made by people on this forum but it does address the problem being discussed and which has already being mentioned. It's the APCO radius ball cap being reproduced by Frank Fenton and being sold by most of the T vendors. It first of all is bolted "tightly" to the pan without springs. Although the radius ball is now slightly loose inside at this point it can't be pulled out as Jerry has seen happen if the springs were not kept tight as with the original cap.
The second part of this unit is the spring and cap socket which are then placed inside the cap body from the bottom to make contact with radius ball and then the threaded plug is screwed in until the slot in the plug and drilled hole in the cap body align to allow the cotter pin to be placed through to lock the plug in place. At this point the ball is held firmly in the pan socket yet free enough to "pivot" or turn since this heavier spring makes for a stronger contact then the original springs did without having to be completely compressed.
These original APCO's were out there only the pot metal plug that threaded into bottom swelled up with age and won't screw in making them useless. Frank got the spring, plug and cap socket remade. Only this time the plug is steel and the cap socket is AMPCO bronze making for a better bearing surface then original steel one used. You probably can guess me part of this project. Bob
Well I am just a back yard mechanic.But logic tells me the spring in the Apco replaces the springs on the studs in my humble opionion to allow some movement without risk of falling out.Less maintance and worry.Bolt it tight,safty wire it,and let the adjustment press against the ball to retain relievable pressure against it.
But some here will just tighten the plug to flatten the spring and make a plug out of it too. Some folks just don't get it. They think if it's not tight it might fall off.
As has been stated many times above, the spring is there to act as a cushion, or shock absorber. If you fully compress the spring, you may as well remove it and replace it with a stack of washers, since that is essentially what you have turned the spring into by compressing it completely. A spring with some space between the coils is a spring, or cushion. A spring that is tightly compressed is nothing more than a solid spacer.
I hate to keep bringing up Model A's, but the front suspension design is very similar. Heavier, but very similar. And it seems that the A is better documented than the T. Here is what I found in my Model A stuff regarding this subject.
In Les Andrews' book "Model A Ford Mechanics Handbook", page 1-234, it says:
"The radius ball cap assembly must be corrctly positioned and properly installed to provide the correct caster alignment at the front axle (5 degrees). If the radius ball is worn or loose in the ball cup, the front end caster cannot be maintained. Liekwise, if the ball is held ridgid in the ball cup, due to incorrect installation, the radius rods will bend or twist when driving on rough roads or hitting pot hole in the road. The radius ball must fit snugly to the ball cap but must be allowed to flex on the mounting bolt springs."
For those of you unfamiliar with the Model A ball cap and socket, it has a replaceable socket made of pressed steel that fits into the transmission socket and a similar pressed steel cap that captures the ball. It also has two spacers to keep the springs from being fully compressed. The assembly instructions in the above mentioned book are as follows:
"Reassemble the radius ball in the cap assembly as shown, placing the cap with the hole on top (allows oil to seep into the ball cup) in the flywheel housing cavity, the radius ball then placed in the cap, the spacers placed over the studs and the lower cap placed under the radius ball sliding over the spacers. The springs are placed over the spacers, against the lower cap. Screw the nut down until it just contacts the spacer and insert the cotter pin. This will allow spring movement of the lower cap, allowing it to move up and down the spacer when driving on rough roads."
And if Les Andrews' word is not good enough, the Ford Service Bulletin for February 1928 page 218 has this to say:
"When connecting the ball end of the front radius rod to the clutch housing, it is very important that the ball cap bolt sleeve (A-3435) is in place on both radius rod ball cap bolts. The sleeves prevent the ball cap springs from being fully compressed when the ball cap nuts are tightened. If the sleeves are not in place, the radius rod ball will be clamped solidly between the ball caps thus preventing it from having its natural motion with the action of the front springs. This condition throws a heavy strain on the end of the rod where the ball is attached and may cause a fracture at that point." It goes on to describe the assembly process which is essentially the same as Les Andrews' but it says to add grease to the assembly and to "...run down the two ball cap nuts sufficiently far to permit locking them in place with cotter keys." rather than "...until it just contacts the spacer..." as in the Les Andrews had said.
So, now that I've bored the crap out of you with Model A stuff, I have to say that I believe that the ball caps of the T were intended to be left with some amount of motion to them, not just some spring force to take up any play due to future wear. I do not mean to perpetuate an argument, but feel that there are enough similarities between these two vehicles that one can assume that what is good for the goose is good for the gander.
If you fully compress the spring, the socket and ball will wear but the ears will not, they are tight up to the ball socket mounted on the pan. So, fully compressing the spring will be just a spacer and do no good!
"The radius ball cap assembly must be corrctly positioned and properly installed to provide the correct caster alignment at the front axle (5 degrees). If the radius ball is worn or loose in the ball cup, the front end caster cannot be maintained."
I don't know this guy, but he probably plays the lottery, too, because he failed math.* An inch up-down movement at the ball will change the caster by only about three per cent, or the same as the front end of the car moving up or down an inch. 3% of 5 degrees = 0.15 degrees.
*Lottery = tax on people who failed arithmetic.
I don't know him either, personally, but he is a pretty well respected authority on Model A's. However, I don't buy the small amount of wear significantly affecting the caster, either. There is an aftermarket item sold for Model A's that have a worn radius ball. It consists of a rubber ball and a new ball cap. You put the rubber ball over the worn radius ball and secure it with the new oversized ball cap, leaving out the upper ball cap socket. Lot's of people say that this will affect your caster, but I don't believe you'd ever notice the difference, and for the same reason you cited.
All that aside, I still believe Ford intended the ball to move up and down a fair amount against the pressure of the springs, or else they would not have put the spacers on the A.
And then there is the penny that is dished with a ball peen hammer and added to the ball socket. This certainly corrects a loose socket. Of course if the wear is great enough for this repair, perhaps the ball will just pull out.