Columbia vs Mitchell overdrive

dean333

dean333

Well-known member
I guess it’s an age old question, which is “better” for my 34 sedan with a 59AB, Columbia or Mitchell overdrive? Thanks All!
 
I believe you have got to cut a hole in the floor to clear a mitchell , then a hump over the hole
 
Ford torque tube was never designed to carry the rotational stress of a gear drive. When shifted properly, a Columbia works just fine. Using 34 housings, a spacer plate must be added. I used a Eaton vac shift button on mine attached to the gear shift handle.
 
My '36 coupe came with a Columbia, early style '36.. I did not care for the early mechanical vacuum switch and the fact that the early models were prone to problems.
Every time I note a topic on the "Barn" and/or EFV8 Forum about Columbia's I usually add a comment.
From my point of view, I like the Columbia's, I converted the one under my '36 to a '47-48 unit in 1955, I bought the car in 1952. I used a combination of '36 parts, the banjo and torque tube which is possible on a '35-36, not so on the earlier cars.
I used a '47-48 Columbia complete with the '47-48 axles and axle housing, '37-38 radius bars and '39 drums. The '47-48 axle housings have better suspension, tube shocks and provisions for a panhard bar. Yes the axle width is two inches wider than the '36, however, they don't need to be cut down because an added inch on each side is not a big deal.
The shifting mechanism for the '46-48 Columbia's is electric over vacuum, very user friendly requiting a tap on the toggle switch and a quick tap on the clutch, down shifting is a simple quick tap on the clutch with a quick release of the throttle.
The Columbia is a variation of the very common truck two speed rear axle which has been used in commercial trucks for over seventy years.
I have driven my '36 over 90K in the seventy years I have owned the car, the only problem I have ever had with the Columbia was the failure of a brass spacer within the unit that caused the right axle to briefly seize.
 
You would have to own an early columbia unit to fit your 34 rear correctly or modify a later unit to suit. It would also likely require rebuilding which is expensive. The Mitchell is good. I have fitted two of these to 1936 Fords but had to cut a hole in the rear floor which you may have to do on your 1934 sedan. As mentioned by others the Mitchell has synchros which makes changing easier. You can have cable operation or a rod and lever. No electrics or vacuum required which might be a good thing.. Two ratios are available. Regards, Kevin.
 
bbrocksr said:
Depends on what you want, Mitchell is probably better, Columbia is probably more original.
Bill
Click to expand...

This pretty much sums it up for me. I have a Mitchell in my '47 coupe. I did have to notch the floor a small bit under the rear seat otherwise it would hit on bumps, especially with passengers in the back. Hooking it up is as simple as a pull cable. There isn't really much to go wrong. You could easily spend the same or more to have a functioning columbia that would still be 75 years old and have a more finicky operating system.
 
I likewise have enjoyed the Columbia for many thousands of miles. I've used it in a '34 Cabriolet and 5w Coupe and currently building a '39 Lincoln Coupe with one. In all of these rear ends I've used 9 inch Ford axels with splined end gear. The only problem I experienced was the same as Bill with the brass washer. When these units are set up properly you're good to go.
 
I have a 37-41 version of the Columbia in my 32 roadster and love it. My rear end had 3.78 ratio gears. I replaced the cork clutch material with nylon, a kit that Connelly sells. There is much less noise when shifted and barely noticeable.


Check here on the Barn for other re-builders in addition to John.
 
Last edited:
I can't comment on the Mitchell. The Columbia I just installed works fantastic. I mounted the 37 valve body to the K-member on the passenger side and activate the main valve off the clutch cross shaft with a rod and lever. Lo/Hi are controlled by a cable with the pull handle in the dash choke hole.
 
Uncomfortable engine rpm at Interstate cruise (65) had me looking at Mitchell versus Columbia.

I went to Plan "B". A big inch engine (284) with high compression (Offy 425), plus a cam with good low end torque (Schneider 248f) and tall rear end gears (3.25). Works like a champ in my '41 Coupe. Should work even better in your lighter '34.
 
There is no comparison when a robust OD is important. The Columbia is fragile and requires operators with a soft touch. The Mitchell can handle substantial abuse.
 
JWL said:
There is no comparison when a robust OD is important. The Columbia is fragile and requires operators with a soft touch. The Mitchell can handle substantial abuse.
Click to expand...

And another big advantage to the Mitchell, it adds FAR less unsprung weight to the suspension. The weight is added to a less critical area also. A very big advantage to ride quality and handling.
 
Maybe not the most practical option, but one worth discussing to muddy the water a bit is the Borg Warner overdrive. Lloyd and probably others used to mount them in the torque tube like the Mitchell and they can be electrically operated from the driver's seat. Like the Columbia you still have some of the inherent drawbacks of an old design and probably a less rugged unit. But if you can find one or have the machinery to build your own it might be a very cost effective alternative.
 

Attachments

  • 20200608_160621.jpg
    20200608_160621.jpg
    48.2 KB · Views: 166
I am afraid that I have to take exceptions to the comments about the Columbia being fragile.
In 1954 I removed the original 21 stud engine from my '36 coupe, it had a 103 K and was getting a little tired, had to do a valve job on it so it would start on cold winter mornings.
I replaced the engine with a well built 58AB engine with a dual Offie Super manifold, Offie heads .060 over bore, 1/4" stroke, a Weber F.1 cam and H-C ignition. I street raced the car on a regular basis, and from time to time I would run the car at the local drag strip. I destroyed several transmissions, got so good at repairing/replacing transmissions that I would R&R a ****** in less than two hours.
I would make a guess that the quality of the Columbia shifts were in direct relationship to the quality of the engine and the available inches of vacuum it produced.
 
You must log in or register to reply here.
Back
Top