Columbia not engaging

[danliveshere]

Richard can you post a photo of the switch, the arm under dash and the valve on the firewall. The switch on the dash does 2 functions,
1 it switches the vacuum between the 2 pipes,
2 there’s a little arm that goes to the back of the speedometer ratio adjuster.
We need to see exactly what you’ve got to make sure you have the correct early system. Again I’m no expert on these, just learning myself to set my 35 up correctly.
Best to get onto John at the following website and he’ll guide you through. Dan
https://www.columbiatwospeedparts.com/index.cfm

[danliveshere]

Richard this is what you should have if the correct pedal valve and selector valve has been installed. See the switch is really just a u-shaped rod that rotates the selector valve through the firewall.

[glennpm]

I suspect that the rod to the vacuum switch has been adjusted too tightly. The pictures are from my 1940 columbia.


"ADJUSTMENT:- Control Valve Clutch Connection
– Adjust clevis on valve operating rod at control valve end so that valve plunger
is depressed 1/8" minimum, ¼" maximum with clutch pedal depressed to floor. Valve travel must not exceed ¼".


The 33-34 rotates to switch vacuum between hoses. same comment through regarding , in your case, rotational movement of the control valve

[Ricosan]

OK. Following the linkage from the dash board I don’t find any thing out of place. The linkage goes directly across to the firewall and is connected to the brass valve on the engine side of the firewall. No leaks. Two copper tubes disappear under the car heading to the Columbia OD at the rear.

This morning I started the car and ran it still on the jack stands for about 10 minutes until it was warmed up. I could easily turn the OD selector from extreme right and then back to the center position without any problem. I shifted this back and forth several times and I believe I could feel it “gently clunk” into OD and back out. Unfortunately I was by myself and couldn’t watch to see if in fact the speed of the wheel was actually increasing. I’m going to see if I can get someone to help me this afternoon.

Still don’t know why the selector won’t move all the way to the left. It may be designed this way as this linkage doesn’t go by the speedometer. This may be the way it’s supposed to function as it’s not an original system.

Richard

[solidaxle]

One good thing you have going for you, your car is clean and rust free. It's just a matter of getting the bugs out of it.


Do you have a tachometer, you can tell if it shifts if the RPMs drop

[V8COOPMAN]

Sponsored Links (Register now to hide all advertisements)

Originally Posted by Ricosan

Still don’t know why the selector won’t move all the way to the left. It may be designed this way as this linkage doesn’t go by the speedometer. This may be the way it’s supposed to function as it’s not an original system.

Richard

Richard ....With the recent explanations that both Glenn and Dan have posted here, and additionally with YOUR recent photographs, I've begun to get a warm and fuzzy feeling for YOUR system.

If I understand all of this correctly, the ONLY time that vacuum actually comes into play is when the clutch pedal is depressed. When the clutch pedal is depressed, momentary vacuum is directed to suck on the firewall-mounted valve.

Hence, the firewall-mounted valve directs MOMENTARY vacuum to whichever of the two ports that has been selected. Your dash-mounted switch, of course, makes the selection of which port that vacuum will be applied to. Therefore, vacuum is either selected to ENGAGE, or to DIS-ENGAGE the overdrive function.

From details I've picked-up from a couple of the pictures seen herein, it appears that the FIREWALL valve requires a full 90º of rotation to function accordingly. By looking at the rear side of that firewall valve where the selecting rod connects under the dash, can you ascertain that the valve is ACTUALLY moving through a full 90º rotation? IF THAT VALVE IS ACTUALLY ROTATING through the full 90º, then that valve is working properly, and the vacuum signals should be directed to the rear end to shift it.




If the above scenario is what you have, then it is likely that the rod between the selector switch and the firewall valve has somehow become mis-shaped as far as the bends are concerned. One last thing to verify is whether or not the SELECTOR SWITCH below the dash is actually rotating a full 90º from one stop to the other.

I believe that I would visually check for full 90º rotation at the firewall valve before going any farther. That sounds like the obvious culprit to check at this point.

Coop/AMERICAN

.

[glennpm]

Hi Richard,


You mention the "center position". Does this mean the operating lever on the dash is straight up whereas it should be free to move all the way CCW to the "Lo" position?


You can adjust it by rotating the stud passing through the firewall. You may be up against the stud stop, preventing you from turning the lever enough CCW (CW when you're looking at the firewall in the engine compartment).


You could also put a twist in the rod from the dash switch to the inside of the firewall switch slot.

[Ricosan]

Coop, the selector linkage goes straight to the firewall valve. I checked for any kind of obstruction and I could find none. This linkage doesn’t appear to be original as it doesn’t dog leg over to the speedometer. The selector switch will go from the extreme right which is disengaged from OD then when I turn the selector back ccw, it will only go back to the center.

Glen, for some reason the selector will not go past the center (straight up and down) to the Lo position on the selector dial. This afternoon, I had a friend depress the clutch and turn the selector to the center from extreme right. I watched the wheels speed up with these settings.

Tomorrow morning, a road test.

Richard

[V8COOPMAN]

Originally Posted by Ricosan

Coop, the selector linkage goes straight to the firewall valve. I checked for any kind of obstruction and I could find none. This linkage doesn’t appear to be original as it doesn’t dog leg over to the speedometer. The selector switch will go from the extreme right which is disengaged from OD then when I turn the selector back ccw, it will only go back to the center.

Glen, for some reason the selector will not go past the center (straight up and down) to the Lo position on the selector dial. This afternoon, I had a friend depress the clutch and turn the selector to the center from extreme right. I watched the wheels speed up with these settings.

Richard ....I can visualize what you are saying about the full range of rotation by the SELECTOR switch. Visually, it only goes back CLOCKWISE to about the CENTER position, correct?

One thing I would like you to verify. Looking at Glenn's picture and RED notes BELOW, can you take a SHARPIE marking pen and put a tiny black dot on the light-colored center piece, next to one of the four notches seen on the brass part of the firewall valve on YOUR car?




Once you mark that center piece with a black dot, attempt to rotate the SELECTOR switch inside the car. We are trying to determine whether the FIREWALL valve itself will actually rotate 90º. If it has, your dot will line-up with the NEXT notch on your valve. We are trying to prove that the valve is fully functional, or is limited in travel for some reason. Can you check that?

Coop/AMERICAN

.

[Ricosan]

Good morning Coop,

That’s a great idea! Yes, I’ll put a dot at one of the four marks on the firewall valve and note the position of the selector. I am planning another test drive this morning right after breakfast to test the overall reliability of the car which seems to have greatly improved with the recent addition of rebuilt carburetors from Charlie NY.
Once I feel confident the car is not going to quit on me, I’ll venture out onto the highway where I can engage the OD.
I’ll take some photographs.

Richard

[Bored&Stroked]

If it was mine, I'd disconnect the rod from the manual dash switch and then use a big screwdriver or other "tool" to see how the firewall valve operates. You'll then know how much travel it is supposed to have to send the vacuum to one side of the cannister versus the other.

In the end, it is a very simple manual system . . . basically a clutch operated vacuum switch and a two-way valve that directs vacuum to one side of the vacuum cannister or the other.

[rockfla]

I just want to thank all you guys for sharing your knowledge and the process of diagnostic on the Columbia.....I've been following this thread even though I don't a Columbia....Well my Dad has a "late" Columbia he was just talking up selling BUT NOW I'm not sure I want to let it get away?? SO thank you guys for the learning experience.

[Ricosan]

Coop, it looked a lot different than I thought. I think this is what you suggested but without the magic marker.

Here is what’s happening. Here are some photos that show the stop pin snug up against the shoulder of the valve. This is with the selector on the dash pointing towards the passenger side door.

Second photo is with selector as far left as possible, which is pointing up and not to the left. It looks like about 1/4” from the opposite shoulder.

B&S, working upside down under the dash is one of my least favorites. I’m going to have enlist the help of a friend. I’ll get it done.

I did this with the engine not running.

Richard

[solidaxle]

This maybe helpful

Senior Member



Join Date: May 2010
Posts: 4,070


Re: Columbia tube routing beneath hood '36 Ford
Here are the diagrams. Yes they are of poor quality
Attached Files 35-36 Columbia installation Instructions 6.pdf (386.2 KB, 46 views) 35-36 Columbia installation Instructions 7.pdf (414.8 KB, 22 views) 35-36 Columbia installation Instructions 8.pdf (595.5 KB, 16 views) 35-36 Columbia installation Instructions 9.pdf (624.9 KB, 19 views) 35-36 Columbia installation Instructions 10.pdf (493.9 KB, 20 views)

[V8COOPMAN]

Originally Posted by Ricosan

Coop, it looked a lot different than I thought. I think this is what you suggested but without the magic marker.

Second photo is with selector as far left as possible, which is pointing up and not to the left. It looks like about 1/4” from the opposite shoulder.

BINGO!

That's not the ANSWER to your switch problem, but that sure is the "WHY".




Now, you need to determine WHY that valve is not rotating any farther. Obviously, it sure does match the partial rotation of the switch on the dash.

At this point, I would suggest that you NOT GO FOR A TEST DRIVE, until you can determine why that valve is stopping just barely past half way. It MAY be going just far enough to send a partial vacuum signal to the rear end, but it needs to be a good, solid vacuum signal to ASSURE a proper shift.

I believe that I would at least disconnect the ROD that goes between the selector switch snd the valve assembly. As someone suggested above, I would carefully attempt to rotate that valve fully in the COUNTER CLOCKWISE (CCW) direction with a screwdriver or such, as seen from the cockpit. If you can't persuade it to rotate fully with some gentle massaging, the valve assembly needs to come out for some further evaluation. Who knows, there MAY just be a burr behind that tiny pin. I'd be surprised if you can't manage to get it freed-up and working properly. Or, there could be an internal problem, requiring a "Plan B".

Those drawings that "solidaxle" posted from "Terry, OH" sure do have a wealth of information, especially on the fine-tuning of the clutch rod set-up. I know that you hate going under that dash, but at least we NOW know that there is a problem with that FIREWALL valve. Please keep us informed! Coop/AMERICAN

.

[glennpm]

Combined pdfs into one file.

[Terry,OH]

As mentioned above take the linkage off of the vacuum selector valve and use a large screwdriver in the valve try to rotate the valve so it goes fully from one stop to the other. There are spray lubricants you can use on the valve without damage to the O-rings. Such as LPS #1. It may be the mechanical linkage binding preventing full operation.

[Ricosan]

Terry, OH & Solid axel, thanks for the drawings. A wealth of information for those of us who are just beginning to understand how the system works.

This morning I got on my back under the dash. I removed the linkage arm to the back of the vac diverter valve on the firewall. The valve would move from shoulder to shoulder easily. I hooked the linkage back up but it still wouldn’t move all the way to the left.

While under the dash I found linkage attached to the selector rod that goes straight to the vac valve on the firewall. There is another rod welded to this rod at 90 degrees. That rod goes to the speedometer. I grabbed this rod and and (push/pulled) it in the direction of the speedometer. At first it didn’t want to move much but after much back and forth it began to loosen up to the point that I could almost get the selector all the way to the left. So it looks like the problem may in the speedometer connection.

While I was under, I took some “blind” photos hopping I would capture something useful. In pics of the back of the speedometer where the rod attaches and actuates the change in speed once an OD selection has been initiated. Just at the end of this rod is a mass of bundled wires. This may be the soft resistance that I’m encountering when I try to move the selector to the left.
I guess there is nothing left to do but go back in.

Richard

[Bored&Stroked]

Nice to know it is something under the dash - other than the fact it is a total pain in the butt to work under the dash. I usually pull the lower seat bottom out and put some nice thick padding on the floor to flop my carcass on . . . not fun!

[glennpm]

Good so now you need to pull that wire bundle out of the way with some strategically located zip ties.



The speedometer SW transmission may need to be lubricated too since you mentioned that it had seemed to be stuck when you started this morning. You will have to disconnect, remove it and then split the halves of this transmission to pack it with grease.