Pirate 4x4 banner
1 - 20 of 92 Posts

·
Vendor
Joined
·
975 Posts
Discussion Starter · #1 ·
I think by now most everyone would agree that ring gear deflection is a large contributor to the short life of Ring & pinions. If you look at aftermarket housing they all have more webbing, thicker castings, heavier hardware and some even have load bolts. This is all done to keep the gears in meshing properly in line with each other. If you have ever seen a Strange Engineering Top Loader rear end you will see how they take a 9" ring & pinon and put it in a box built like a bunker such that they can withstand 2500 hp. http://www.google.com/imgres?start=...=117&tbnw=213&ndsp=24&ved=1t:429,r:1,s:90,i:7
So what can the wheeler do to achieve ultimate strength in the diff he can afford. Options so far have been to buy billet Main caps, & girdle covers. Now I have a third option, the do it yourself load bolt. With the testing I have just completed you will see that this single improvement will do more than the previous options to prevent ring gear deflection. 1st lets define a load bolt. A load bolt sits about .003" behind the back side of a ring gear opposite the pinion. Thus when ever the ring gear gets deflected under torque it moves until it hits the face of the load bolt thus stopping further defection. A nice picture of a Chevy HO72 3/4t rear end OEM load bolt can be seen at post 56 of this forum.http://www.pirate4x4.com/forum/showthread.php?t=771317&highlight=ho+72&page=12
Next are a few pics of load bolts I have installed and the Kit I am marketing. Right now I have them for Jana 76 & D70 applications. Although I don't see why I couldn't make similar kits for many other applications.
Auto part Fuel line


Screw Fastener Auto part Tool Metal


Auto part Rim Tool accessory Tire Automotive wheel system


Product


Update:
If you found this thread you'll also be interested in our full product line of Differential Hybrids;

HOW TO INSTALL:
Load Bolts into Stock housings:/forum/axles-tires-wheels/1059807-load-bolts-how-install-yourself.html
D70 into D60s, Jana 76;http://www.pirate4x4.com/forum/axles-tires-wheels/946421-jana-76-tm-how-install-70s-your-60-a-3.html
D80 into D70s, Jana 87;http://pirate4x4.com/forum/showthread.php?t=958266
D50 into D44s, Jana 54:http://pirate4x4.com/forum/showthread.php?t=948996
New 8.9" JKs into 44s Jana K4:http://pirate4x4.com/forum/showthread.php?t=957491
Holding it all together, Jana Cover Girdles;http://pirate4x4.com/forum/showthread.php?t=958047
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #2 ·
Inside view of load bolt install

Inside views
Auto part Metal Steel Rim Wheel


Auto part Nut Screw

To make room for the load bolt the carrier is machined down to the edge of the ring gear bolts exposing the back side of the ring gear. The ring gear is then polished to give a smooth surface for the load bolt face to ride on.
Auto part Automotive wheel system Rim Wheel Rotor
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #3 ·
How much defection?

So next thing I hope your thinking is how much torque input does it take to make the ring gear deflect .003", and how much will a ring gear defect when its under a real world load? Well I wondered the very same thing and it took about 4 weeks for me come up with a recession budget testing jig. That means I had to build it from what was lying around the shop with virtually no cash outlay. Starting with the parameter from DANA that the maximum momentary output torque for a D70 rear end is 8000 Ft lbs I made the following assumptions. To get 8000 ft lbs out, you would need to put 2000 ft-lbs in if you gear ratio was 4 to 1. I just happened to have a slightly used set of 4.56 gears, so If I could put in 2000 ft lbs I would get out 9000 ft lbs and this would be right in the range of just over the rated maximum that we all push things, and hopefully under the breaking point as did not have a bullet proof glass enclosure to protect myself. So the question is how could I reliably put in 2000 ft lbs. Well first I had to bolt a rear end down to a stout table, no problem, then I figured with a 10 ft lever and 200lb weight I could make the torque. then I found that there was so much wind up that I could not swing a 10 ft lever far enough around without running into something in my shop, like the floor or the ceiling.

Seen here a Jana 76 rear end bolted to the bench, with a front axle in the right side with a bar though a pair of holes in a 5" pipe, (welded to the axle tubes) and the ears of the the axle such that it can not rotate. We started with the arm at 2:00 and rotated it till it hit the floor, which with his weight on it ( thank you Andrew) created around 600 ft lbs input, not nearly close to what we needed for testing. You also see a dial indicator installed in the load bolt hole so that the indicator point rides right on the back side of the ring gear opposite the pinion. Thus we can directly read any ring gear deflection.
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #4 ·
Wheel of fortune

So with the big lever idea a bust the next thing was a large wheel with a cable. (Thank You Fred!) The wheel was arbitrarily made 80" in diameter, as this was about as big as I could go and not hit the floor.
Wheel Auto part Vehicle Automotive wheel system Spoke


Wood Floor Hardwood Vehicle Table

The wheel was made out of 1 1/2" schedule 40 pipe, muffler tube & and an old truck rim. To apply the proper torque a 1/2" wide slot was cut in the OD of the wheel, thus allowing a cable to be wrapped around inside the resulting groove of the tube. To apply the proper torque the cable then went through a snatch block, so that I could back my car hauling trailer winch in the shop and pull on it for torque. With the large wheel it turned out that only 600 lbs of pull were needed on the cable to produce 2000 fl-lbs input torque. You will see the cable running through the snatch block to a chain that is then attached to the hydraulic steering ram off of my Jeep. The other end of the ram is then attached to the winch on the front of the car hauler. The rams hydraulic fittings are capped of with one end having a pressure gauge in it. I calibrated the gauge such that I could read it in pounds of force. ( rather than psi) As it turns out with a 40" lever arm 75lbs force causes 250 ft-lbs torque. Then I made a chart to record the defections of the ring gear starting at 250 ft-lbs and going up 250 ft-lbs at a time until 2000 ft lbs was reached.
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #5 ·
short summary

Short summary, I wanted to get all the results and video of the test posted tonight but Its just to late. Throughout the testing I got some strange results, some of it was from axles twisting, some from my jig breaking and some from bending aftermarket spools. After I got the bugs worked out I found some stuff I never expected. I tested with and without a load bolt, and I tested with and without a Jana Girdle cover.

Test result: no load bolt, no cover.

In the forward direction with out load bolt the ring gear defected only .001" at 1000 ft lbs torque, going to .002" deflection at 2000 ft-lbs. No problemo right? But in the reverse direction it gets really scary, at 1000 ft-lb there was .010" and a total of .030" deflection at 2000 ft-lbs. No wonder why reverse cut gears were designed for front ends. There is more to post and I'm sure there will be questions, for now just PM me please until I get the rest of the info up, then I will make a post that says questions please, as I hope to answer most questions before that.
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #6 ·

·
Vendor
Joined
·
975 Posts
Discussion Starter · #7 ·

·
Vendor
Joined
·
975 Posts
Discussion Starter · #8 ·
bent spool

When I began the testing I had put the ring gear on a 35 spline spool, and after running a couple of torque tests, with the same parameters I was getting different results. What was happening was that the spool was permanently deflecting throwing my readings off between tests. I bent it .004" I eventually installed a stock Dana 70 spool and after several test found I had warped it at well but only a .002 thousandths. Load bolt warped spool flange - YouTube
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #9 ·
data

Test results: Although not what I expected what it shows is that there is a tremendous amount of movement going on under different circumstances.
 

Attachments

·
Vendor
Joined
·
975 Posts
Discussion Starter · #10 ·
Kit Install

Now you want to see how easy it is to install a load bolt. Basically I supply a kit that holds the threaded load bolt boss onto the carrier in the correct place for welding. All you do is lay out the location of the hole and drill thru the housing with a 1/2" step drill bit. This allows you to verify your lay out. If you are a little off you can die grind the hole onto center. Then you drill out the hole to 7/8" with a step drill, and finally die grind to fit the boss using a supplied tapered guide for fitment. Once the boss bolts to the guide tool without binding in the hole in the housing you tack weld it in place, then remove the installation tool, fully weld in the boss and then re-tap the threads as it will shrink up a bit when welding. On the 60 housing with 70 gears the unit is installed 6.5" down from the end of the pinion and along the side of the housing rib.
Glass Gas Metal


Ruler


Pipe Soil




Pipe Auto part
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #11 ·
Kit install continued

More Install:
I supply a tapered bushing that is used as a guide to match grind the 1" hole, by putting some gear marking compound on the taper you can mark the inside of the hole where to grind. The tapered bushing slide on a guide pin connect to the fixture on the carrier.
Pipe


Auto part Pipe


Roof Pipe Ceiling Metal


Rust Metal
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #12 ·
Ok

Questions Please.
 

·
Registered
Joined
·
617 Posts
How much for the kit?? Will this be now included in the Jana76 install kit?

I'm about to pull the trigger on two setups to put some D70 7.17's in my D60's.

:smokin::smokin:
 

·
Vendor
Joined
·
11,822 Posts
Short summary, I wanted to get all the results and video of the test posted tonight but Its just to late. Throughout the testing I got some strange results, some of it was from axles twisting, some from my jig breaking and some from bending aftermarket spools. After I got the bugs worked out I found some stuff I never expected. I tested with and without a load bolt, and I tested with and without a Jana Girdle cover.

Test result: no load bolt, no cover.

In the forward direction with out load bolt the ring gear defected only .001" at 1000 ft lbs torque, going to .002" deflection at 2000 ft-lbs. No problemo right? But in the reverse direction it gets really scary, at 1000 ft-lb there was .010" and a total of .030" deflection at 2000 ft-lbs. No wonder why reverse cut gears were designed for front ends. There is more to post and I'm sure there will be questions, for now just PM me please until I get the rest of the info up, then I will make a post that says questions please, as I hope to answer most questions before that.
How much of that was only ring gear deflection and how much was carrier/housing deflection?

Meaning I know that carrier preload will also pre load the housing, so that in situations like this the carrier itself can't push away from the pinion as much.

So I'm curious if your numbers take that into account or are only looking at ring gear deflection, because of where the measurement is taken from, ie point A and B.
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #16 ·
How much of that was only ring gear deflection and how much was carrier/housing deflection?

Meaning I know that carrier preload will also pre load the housing, so that in situations like this the carrier itself can't push away from the pinion as much.

So I'm curious if your numbers take that into account or are only looking at ring gear deflection, because of where the measurement is taken from, ie point A and B.
As Einstein would say, its all relative. The motion on each indicator is relative to where it is located and how it is mounted. If the indicator was mounted to the table you would see how much the housing & axle tubes flexed when he torque load was applied. It is hard to say how much is carrier deflection and how much is carrier bearing walk, it probably a combination of both. Looking at most aftermarket lockers I would say they have less flex than the stuff I used, just because of the mass and construction of each. When I put the diff together I preloaded the pinon and carrier bearings to the max. the pinion is at 30 in-lbs, the carrier has .015" preload and the back lash started at .006. These are Ideal conditions, and I have noticed that after about a year of running those preloads are greatly reduced, thus I should run another test with much lower preloads, to simulate real ongoing life in a diff. (Just what you all wanted to hear, that every 12 wheeling trips you need to reset your diff preloads).
I have been looking over my set up. Even if there are better ways to reference the deflections what is shown is that everything is moving a lot, and the load bolt stops a lot of motion. I am also going to have to remake the fixture. What we realized is that the way the cable is pulling on the wheel also induces a side load on the pinion that is not normally seen from the input of a drive line. This side load is only 600lbs at max so I do not think it will influence the results much. But to get rid of that variable I will move the diff to the other side of the table and then mount the wheel on a separate set of bearings and connect the two with an actual drive line, then the pinion will only see a pure torque input, as in real life. Although I may now be simulating the side load of what your diff sees when the pinion is grinding over a big rock.
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #17 ·
Load bolt kit pricing

How much for the kit?? Will this be now included in the Jana76 install kit?

I'm about to pull the trigger on two setups to put some D70 7.17's in my D60's.

:smokin::smokin:
Load Bolt installation kit $150
Load Bolt & Threaded Boss $100
Then you have a choice of returning the kit for a $100 refund, Basically you rented it for $50. Or keep the kit and use it over and over with your buddies and or customers by just getting more load bolts.
 

·
Vendor
Joined
·
975 Posts
Discussion Starter · #19 ·
The warped spool was it a 60 spool drilled for a 70 gear?
Yes and come to think of it, I had machined about an 1/8" off the flange when I was trying to get the D80 gears in the 70 housing, so it was weak to start with. Just the same if I were to run a spool, I would make sure it was thick or had some gusseting on it. Course with a load bolt it may be a moot point. Guess I'll have to true it up and then run the test again with the load bolt, then if it does not warp I know I got a winner.
 

·
Registered
Joined
·
4,777 Posts
Got darn it Karl. I think you might be well on your way to something big here.:D

Kudos to you for finally demonstrating the deflection that a typical gearset incurs rather then just BS talk.

While measurements can be misleading if not interpenetrated correctly. It would seem that you are taking as many steps as you can to simulate real word conditions.

The numbers you are seeing are of no surprise. In particular the amount of deflection a RC gearset run in a rear drive application. As you know I have been a loud advocate of reducing deflection as much as possible with studs, caps, covers and girdles. All of which address the problem but do not solve it.

I will eagerly watching this thread to see how your testing evolves. Wonder if you can come up with a way to measure case distortion via the forces applied to the axle tubes while the gearset is being loaded and deflection measured.

Thinking along the lines of welding a hook of sorts to the spindle end of each tube. Then adding a cable/turnbuckle to load the axle into a simulated vehicle weight. Measuring the before/after load and deflection at the ends of the tubes and then then deflection of the case itself. With and without a cover. I think you will be surprised just how much the tubes themselves move around.

Betting the gear pattern goes all to hell and gets exacerbated when you start adding rotational force to the pinion. Which will prove the need for a truss to stiffen the not only the up and down bending forces of the axle tubes but to spread the forces over the greater area that a truss will provide.

Good job Karl and thanks for the PM.
 
1 - 20 of 92 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top