R21 Solid Shifter Shaft (not cable) Problems

Well, GTD40 has an R21 w/ solid shifter shaft and I'm not super pleased at how it shifts. Basically, its difficult to slip into/out of gears. I don't know who manufactured the solid shifter shaft. It is a left hand shift/left hand drive GTD40 (Was Brian's McCarthy's car before he sold it to Ellis Hubbard then to me). Maybe someone remembers the build?

I'm thinking the heim joints or rod ends need better angles. Any suggestions? Let me know what other details are needed to help with diagnosis.

Hi Matt I have an RF with rod linkage and have to keep it lubed I use Triflow, If I wash the engine compartment and don't wipe the shift shaft and lube it it will let me know as shifting gets quite stiff. You may need to install some kind of oiler on the bearing block by the transmission, Are there bushings in that block ???? Also make sure the handle pivots are free, The handle on mine was binding on the panel under it and needed raised up to clear. Wally

Ian Anderson

Lifetime Supporter
Two things spring to mind
First is there any flex in your rod as any flex or bow in it results in reduced movement at the gearbox end.
What are your engine mounts? If rubber all movement at engine results in positioning of the rear of the gearbox also moving so you are then trying to hit a moving target

Ask me how I know about both these problems!

It would be nice to see the whole setup...I am looking specifically at the height of your linkage routing and seeing where it passes thru the paneling and wondering how the front end of the linkage looks.
I can tell you that as was mentioned engine flex can cause big problems, and also routing issues somewhat similar to what you have such as angularity which can translate as lost motion and forces bushings and bearings to take side loads.
Also take a look at some of the setups here on the forum for the Renault boxes that have the gate selector mounted to the rear of the box and run the shift rod closer to the centerline of the driveline.
Ian hit the nail on the head..any flex has to be minimized as large movements at the shifter make smaller movements at the gearbox so any flex or lost motion may make it difficult or impossible to select a particular gear, or adjustments may make the setup toggle back and forth favoring one or another side of the shift pattern.
I had a tiny bit of slop in one U-joint (Sloppy Hole) which prevented getting reverse and when adjusted over would not select 5th.
Lastly, check your rod ends....they have limited angular travel so you may have a bind in there somewhere that is giving you fits....there are high angularity rod ends if this is the problem.
I think if you post the complete linkage there is knowledge here on the site to make it possible to get it right.
By the way that is an awesome car...I remember when Brian had to sell it..has all the goodies if I remember.
That's a Roy Smart construction, or rather the rear part is for sure. I do recall him making one for Brian although to be fair all the one's he had made otherwise were RH shift versions so whether something changed in the fitment or not I don't know.

As I regularly drove and competed in Roy's GTD, I do remember that when set up correctly the gear change was very positive, sweet in operation and problem free. However, IF any of the Heim joints came loose, or if the linear bearing was not kept suitably lubricated then it could stick or miss-shift.

Another thing I recall was that you had to get it set fairly accurately in the first place and not have too much engine movement on the mounts, else the lever would tend to move back and forth when hard on / off the power which if your gate was not accurately located, could knock it out of gear.

I am probably not doing it justice by saying that because when it worked, it was sublime but like most things, when it didn't, it was a PITA.

Finally and for the records - The FASTEST shifts we ever recorded, were with a standard GTD Cable Setup, which when accurately setup, also worked perfectly well and data logging showed that it could 'out shift' even the best changes on a rod-system.

Good luck - hope you get it how you want it... :)
This is how it was done on a local Countach replica, he had a double heim joint in the centre of the pic between the motor and the chassis to stop forward and aft movement, as he said every time you accelerate or brake the motor would not only move left and right because of the torque but it would also move back and forth and was a bit like rowing a boat, and make selecting gears a hit and miss its a R21t box also.


Ahhh - yes, I remember we had 2 double heim joints too.

One stopped fore-aft movement, like yours shown although it was a little further back and another linked from the rear of head (cyl 8) to nearby suspension turret.
If I remember correctly the original supplied GTD rubber doughnuts for the tranny mounts were about 2 inches thick and soft (even softer when they got hot). I replaced these with a much thinner and harder rubber doughnut from an early Pontiac. This reduced the moving about a lot. They are much harder rubber and about half as thick.

Bill Musarra

I am no engineer, but I can tell you what I know and what I see.
What I know:
I know that any cushion you put between the motor and frame, and the trans and frame will allow the motor and trans to move around.
What I see:
First, your rod linkage is mounted to the trans which is good and bad. The good is that when the trans moves, the mount moves with it. The bad, is the rod linkage does too. Your rod should come parallel to the setup(or as close as it can to that). The way you have it, when you move the trans(torque) your rod will move in two directions at once, sideways and front to back(because of their mounting points). If it is parallel, it will only move sideways. and you will have very little movement of the shift lever in the car. The rod actually moves in an arc, but the moment of moving is very slight(I think I said that right). What it boils down to is very little movement of the gear shift lever.You will probably feel the gear shift lever move around whenever the trans moves as it is.
Second, because your rod comes in at an angle, the heim joint has to move in two directions to get the shift linkage to move in one when it goes side to side. That means your neutral gate is at an angle in the car, not truly side to side. I am making an assumption here that the neutral gate is the lateral movement of the shift linkage on the trans. I would think that the way it is now, would put a bind on something, moving like that. The movement of the rod should be the same movement of the shift linkage. All of these mis alignments, seem to me, to put a lot of pressure on parts that shouldn't be there(the pressure).
What I know:
I learned with my cable linkage that alignment is everything. Even cables can get messed up if not aligned properly. Cables that are stretched out straight act pretty much like a rod if supported right.
What I know:
If you eliminate your movement of the engine and trans, you will eliminate the movement of the rod(above). I mounted my trans and motor solid to the frame. I feel virtually no vibration at all when driving on the street. Under full power is another story. At red lights at idle, the car actually rocks a little due to the rough idle and torque being applied by the motor. I call it a dance and it gets considerable notice at red lights by other drivers. Remember, torque and vibration are two different things. Now my motor is internally balanced, but other engines apply counter weights that smooth out the motor pretty well. They put rubber cushions to eliminate all the little vibrations and smooth out the torque when applied. Remember, we aren't driving luxury cars.The suggestions made earlier to control the rock of the engine do the same thing as my hard mount.
Anyway, that is my take on it. Hope this makes sense.

Last edited:
During a discussion some months ago on a Lambo replica forum a member was having problems with his car destroying the BMW OEM oil filled rubber mounts and it was costing him a fortune to replace them , he is running a BMW v12 and Audi 016 trans, what he found was that the short distance from the engine mounts to the trans mounts, multiples the Torque well above that seen in the original bmw donor, as the mounts died they caused shifting problems, he tried all sorts of mounts but the problem seemed unsolvable, until he fabricated some mounts using the hard rubber of ice hockey pucks bolted together, he said it did not make any difference to the noise transmitted to the frame at all, and years later they were still in A1 condition. And all his shifting problems were gone.