UN1 rod change query

I'm committed to using a rod change so no going to cables for me.

What would be considered the preferred diameter of rod change shaft to use.
I've got a solid 5/8" OD item which is undoubtedly up to the job but it sure does weigh a lot and hurts if you drop it on your hoof!

Has anyone used a smaller sized item - either 1/2" diameter or, at a push, a 3/8" diameter item?
Other possibility would be to use tube as an alternative although stiffness may preclude that as a sensible route.

The logic behind all this is weight saving potential without compromising gear change due to unwanted flex in the rod itself.
 
A tube is much better and much lighter. I can`t remember the diameter I used but there are plenty of guys on here that can help you, I`m sure.
 

Neil

Supporter
Stiffness is exactly what you want in a shift rod. A tube is almost as stiff as a solid rod but it is far lighter than a solid rod. I use a 3/4" x 0.065" tube in one design and a solid Thompson 3/4" Case 60 ground & polished rod in another. The tube is far lighter and I would have used a tube in both if I could have found hardened ground & polished tubing. The support bearings are important; a good polymer bearing such as a Thompson is best but a Delrin or HDPE bushing is good.
 

Mike Pass

Supporter
Why not use the solid where it runs through the bearings and pivots and then use tube in between. Select some tubing which fits over the solid rod and pin or weld it to the rod.
Cheers
Mike
 

Howard Jones

Supporter
4130 soild 5/8 diameter = 1.043 lb/ft
4130 seamless 5/8 dia .095 tubing = .435 lb/ft
4130 seamless 5/8 dia .065 tubing = .389 lb/ft
 
Thanks for all the responses :)

Seems like a weight reduction process is very feasible.
What I have just realised is that the system I currently have uses a stainless rod and this will be slightly less robust than 4130 (if my memory serves ok).

0.065 wall tubing has been used successfully (although at 3/4" OD) so perhaps the other option is to stick with 5/8" OD solid but switch material to aluminium.
This would allow me to maintain all the current couplings, bearing supports etc and achieve a simple weight reduction.

Is this a step too far?
 

Mike Pass

Supporter
You can keep the diameter of the aluminium connecting tube down by turning down the ends of the solid sections to a suitable diameter.
Cheers
Mike
 

Neil

Supporter
If you want to keep your bearings, etc, just switch to 4130N or stainless 5/8" x 0.065" tubing. Its stiffness will be almost that of your solid rod but its weight will be much less.

BTW, re: "4130 vs stainless robust...": In this application, yield strength has nothing to do with choosing between the two. Both have perfectly adequate strength; the stiffness of the tube/rod is the key spec. There is almost no difference between the two materials. (Youngs modulus)
 
Mark,

You really want to get as much stiffness as you can in the shifter rod. I have a Tornado and it has a 5/8" solid rod. I had to do several things to stiffen up the whole system. I first welded on a flat plate in the axis of bending to stiffen it laterally. Notice the 3/4" plate horizontally on the shifter.

shifter stiffner 003.JPG


Next I had to position the intermediate torque rod that was mounted to the engine to be perpendicular to the shifter rod because it was pointed at a very downward angle that deflected the sifter rod down.

shifter 004.JPG
Transmission adapter.JPG

before and after -->

Next I had to put a gusset on the actual transmission shift lever.

shifter arm 002.JPG


Even though it is drastically stiffer than the original, It is still not stiff enough in my opinion.

To answer some of your questions, if you are concerned about weight, you should use a tube instead of a solid rod. Below is a normalized analysis of a 5/8" shifter tube with a variable wall thickness showing the stiffness and the weight of the tube relative to a solid rod. It also shows the stiffness to weight ratio. If you were to use a 5/8" tube with a 0.120" wall thickness (t/r = 0.38), you would have 86% of the stiffness of a solid rod with 62% of the weight. The rewards diminish with an increased wall thickness. Going to a larger diameter tube would help a lot if you could redo your bushings. I would definitely consider that. A 3/4" tube with 0.120" wall thickness would almost double your stiffness.

Stiffness and Weight of Circular Tube.jpg


You certainly don't want to use aluminum. Scott's suggestion of carbon fiber rod is a good one if you are really concerned about weight; however, it requires a lot more engineering analysis of how to attach the steel insert.

-Bob Woods
Tornado GT40 in Texas
 

Attachments

  • Stiffness and Weight of Circular Tube.jpg
    Stiffness and Weight of Circular Tube.jpg
    90.1 KB · Views: 196

Shaun

Supporter
This is really interesting stuff, I have dry fitted my gear rods but am not happy with the general sloppiness of the whole thing not in a small part due to the actual shift rod coming out the UN1 it seems quite loose in its housing even thought its fully rebuilt, Bob I like the gusset you have on the shifter rod, did you make that? Also what's the cable you have clipped to the side of the box is that part of the system?
 

Ed McClements

Supporter
Hi Mark

This is what I used when constructing mine recently:-

16mm stainless tube, 3.5mm wall thickness

$_57.JPG


RS Pro Universal joints:- https://uk.rs-online.com/web/p/universal-joints/7906719

My own shifter, lockout lever and top plate

50514166537_fc9759530f_o.jpg



I had to re-do the whole thing, after starting with smaller-diameter solid rod (12mm or 1/2"...I can't recall) which was far too flexible. The first iteration also used some "value" UJs from Car Builder Solutions, but the slight play in the two UJs, and the whippiness of the rod all added up to a lot of lost motion, and getting all gears was next-to-impossible. The current set-up is very, very positive, with a narrow shift gate & pleasingly mechanical feel to the shift action. I have a centre bearing through which the front rod slides & rotates - this is mounted very solidly to the chassis behind the driver's bulkhead.

I found that the tiniest amount of play in the UJs & bearings multiplies together to cause unacceptable play at the lever and gearbox end, which is why I started again.
 
Last edited:

Ed McClements

Supporter
This is really interesting stuff, I have dry fitted my gear rods but am not happy with the general sloppiness of the whole thing not in a small part due to the actual shift rod coming out the UN1 it seems quite loose in its housing even thought its fully rebuilt, Bob I like the gusset you have on the shifter rod, did you make that? Also what's the cable you have clipped to the side of the box is that part of the system?

The cable is (I think) the reverse lockout system. You can see the stepped sleeve on the shaft coming out of the gearbox, plus the pin which is retracted by the cable to allow extra motion to the Reverse plane.
 

Shaun

Supporter
The cable is (I think) the reverse lockout system. You can see the stepped sleeve on the shaft coming out of the gearbox, plus the pin which is retracted by the cable to allow extra motion to the Reverse plane.
Great thanks for the clarification, a good idea methinks, not best to get reverse by mistake!!! Interesting re the play angle, a little on an amount of parts seems to add up to a sloppy shift
 

Mike Pass

Supporter
If there is an issue with the resistance to bending of the rod or tube then this is not really a material problem but a design problem.The rod or tube should only experience tensile, compression and torsional forces. If the linkage relies on bending strength it needs to be redesigned. The gear linkage rods or tubes should push, pull or twist but the absolute minimum of bending loads (over the shortest possible length).
The use of aluminium is not a problem if used correctly. The aluminium should be used as a suitable alloy. The Porsche 917 had an aluminium space frame, which used straight tubes, which carried the tensile and compressive loads. Aluminium alloy tube is also used in many aerospace applications. As aluminium is prone to galling, its use in the bearings and pivots is not a good idea.
Eddy’s point about “lost motion” is a good one so good quality bearings and universal joints are essential so that the gearstick movement is accurately relayed to the gearbox input shaft. This issue is relevant to the comparison of rod change versus cable change. As in most road cars the engine and gearbox are mounted on rubber to reduce vibration. This means that they can move relative to the chassis where the gear lever mechanism is mounted. This movement will be worst when changing up or down when the torque reaction on the engine is the greatest which is when the gear change mechanism is in use. A cable change does not have this problem as, any relative movement is taken up by the long, flexible cables. In a racecar this issue does not arise because the engine is rigidly bolted to the chassis. The gear change that works well when stationary may not be so good when the engine and box are moving under torque reaction. Some attention should be paid to controlling this movement with a rod change.
The issue of the weight of the gear change mechanism I believe is not that important in the context of the overall weight of the car. If anyone can detect the difference of 10Kg placed in the sill area of a 1,200Kg car I will be amazed. How much does 9 gallons of fuel weigh? Maybe one less pie a week would be a better bet – certainly in my case!
The mass of the linkage will have an effect on the inertia of the mechanism but whether this is noticeable is open to argument. I have tried many rod gear change systems in GT40s and race cars and been unable to detect any inertia issues. The main thing seems to be how easy or hard it is to change gear without missing a gear.
 
Shaun,

I just used a 3/16" triangular plate welded to the existing shifter arm at the transmission.

Eddy is right, the cable is the reverse lockout mechanism provided by Tornado. However, it is so hard to get into reverse anyway, that I don't think it is necessary. I am limited by the side-to-side motion of the shift lever at the hand since there are two side walls on either side. The shifter hits both walls (for reverse and for 5th) even after I shortened the lateral throw of the hand shifter arm. I have to hit it pretty hard to get reverse. You do run the danger that you forget to pull the cable and try to force it into reverse you could bend the lockout mechanism or mount. I have considered removing it.

Mike,

I totally agree with you that tubes should not be put into bending as a good design rule for frames and any place to avoid unwanted compliance. In this situation, we have to move the shift mechanism on the transmission in-and-out as well as rotate it. To do that with one shift rod requires that the rod be loaded longitudinally (for rotation of the transmission shift mechanism) and and it must rotate (to move the shift mechanism in-and-out). Thus, the shift rod has to twist and react the torque at some point. This does put a bending load on the shift rod. As you say, there should be a bushing or bearing much closer to the torque reaction point instead of three feet away. If the shift rod was running parallel to the transmission then there would be no lateral motion of the shift rod when rotating. Then a bushing next to the torque reaction arm would be ideal, but that is not the case with the current design.

Aluminum has its place when weight is a premium; however in this case where we need stiffness, realize that for the same dimensions, aluminum will have 1/3 of the stiffness as steel. I tend to agree for a street car the difference in weight of a solid rod versus a tube isn't significant. In a racecar, weight is important. As Carroll Smith once told me, "If you have a 1000 pound car and you would like to reduce the weight by 50 pounds, you won't find a 50 pound component that you could just remove; however, you might be able to remove 5% of the weight of every component". So if you have the choice, why not make each part lighter.

I also agree that the inertia of the shifter is really not significant in this car. However, just an interesting side note, in the Newman-Haas CART car back in the day, they found that the solid aluminum shifting knob would actually shift the transmission under hard braking. One of the engineers (my former student) was asked to design a hollow aluminum ball for the shifter.

-Bob Woods
Tornado GT40 in Texas
 

Randy V

Moderator-Admin
Staff member
Admin
Lifetime Supporter
Good info and interesting story....
I remember struggling all one winter to put one of my racecars on a diet. Finally whittled 35# out of her - much was in the wiring, some in components like seat mount brackets and battery mounts and battery itself...
This is the shifter ball I've used in a number of my racing cars..
 

Attachments

  • Alloy Shift Ball.jpg
    Alloy Shift Ball.jpg
    45.9 KB · Views: 158

Ed McClements

Supporter
Uploaded some pics of the gearbox end of the linkage:-

50859960957_aae5a57173_o.jpg


50859960762_dabcf1ab92_o.jpg



Excuse the scabby paint on this last pic...I was still modifying the mounting points:-

50859150368_1b0c71e6c5_o.jpg
 

Ian Anderson

Lifetime Supporter
Eddie
that looks like the first iteration on mine.
Problem happened when it all got hot and the rod pinched tight in the tube = no gear change, let it cool it worked again.

it now runs in two spherical joints.

I’ll take a photo later and post

Ian
 
Back
Top