Graham Technical Corner

last update April 7, 2013


Shell Bearings for Early Graham Motors


I have owned a 1929 second series 612 for the last thirteen years. The vehicle was a fairly rough mechanical shell when I got it but it had been resprayed and could be coaxed to run so I could get some fun out of it.

After 12 years of working to bring everything else up to scratch I finally turned my attention to the motor. This was brought about by the failure of the con rod bearings and I finally realized I must face a total rebuild as the only path to a trouble free future (optimist).

Along the way I had bought a motor which originated from a model 64 1934. It is a 224 cu inches original patented Graham Paige engine. This engine was internally similar to my 1929 but had a camshaft with an extra babbit bearing and had a cam drive by chain but an external V belt driven water pump at the front of the block replacing the side mounted timing chain driven water pump and rear mounted generator of my 1929 engine. The rear of this engine block was also different as was the bell housing to the gearbox and the starter motor. This particular motor had a ‘Bohnalite’ aluminum cylinder head which was unfortunately unserviceable.

The 1934 motor was an advance on my 1929 motor in that it had modern shell bearings for the crankshaft but still had babbit bearings for the conrods. I wanted to convert the whole engine to shell bearings if I could as babbitting is expensive and it was difficult to find someone to do that work in my home state. Many experienced tradesmen who used to do this work have retired or died in recent years and it is hard to pay money without a guarantee of success.

As many of you would be aware Graham sold the plant for making Graham Crusaders to Nissan and the plant was transferred to Japan. This engine was used by Nissan in 1937 for the Nissan 70 (a Graham copy)

I am a reader of an Australian magazine called ‘Restored Cars’ and I recalled reading an article about the first Nissan Patrol sent to Australia in the late 1950’s for evaluation by the Australian Army. I think it was a 4W61. The striking thing about that Patrol was that it was obviously fitted with a Graham flathead motor although the carburettor was a Mikuni.

Further investigation revealed that apparently in the late ’50s, an American engineer was brought to Japan to convert the engine into an OHV. Conjecture was all I could find with no facts to back up this story.  The conjecture suggested that the engine born out of this metamorphosis was the Nissan P40. The P40 4 litre OHV motor was used in Patrols to 1985.

This certainly excited me as the Nissan P40 was modern and had shell bearings for conrods and mains. I took this up with my engine rebuilder and he consulted some old books he possessed which gave the technical details of both the Graham engine and the Nissan P40. Surprisingly enough the tunnel bore size for the 1929 Graham and the P40 were identical! Now that was really interesting.

I hunted around the wreckers yards and bought a P40 without the head and started to pull it apart. I already had a 1928 block that I had been given as a possible replacement for my 1929 if all else failed and I had dismantled this prior to getting the Nissan. It was near to hand and lying on the workshop floor in many pieces. I start to disassemble the P40.  Surprise surprise. All the nuts and bolts on the P40 are SAE. No metric in sight! I took off the substantial Nissan flywheel which was much bigger than my ’28 Graham flywheel. Again SAE bolts, the same number as the Graham. Not only but also. The same rebate in the flywheel as my 1928 donor and the ’28 flywheel fits straight onto the snout of the Nissan crank, same bolts and all. This is getting really interesting but the Nissan block is about 32mm shorter top to bottom than the flattop Graham block.

 I take off the Graham flywheel from the Nissan crank and take a close look at the Nissan mains. The caps appear very similar but with no hole for the oil gallery pipes that come in from the bottom of the cap on the Graham. Instead there is a neat Nissan emblem shaped like a London Underground station sign. I take off one cap and put my ’28 donor cap in its place. Fits the side registers on the Nissan crankcase snuggly and lines up perfectly with the cap bolts.

I then took off all the caps and removed the Nissan crank. A bit of a disappointment. The rear oil seal is done without a wet seal arrangement like the Graham and the rear of the crankshaft is noticeably different from the Graham crankshaft. Of course the Nissan also has balancing lobes as well but it is obvious when it is pulled out that it won’t fit straight into the Graham block because of the rear oil seal. But what about the Graham crank? Will it fit the spacing of the Nissan block? I lower it into position and it fits perfectly as far as front to back is concerned. There is no excessive end play. I put back the Nissan caps and tighten them up to test. There is a large amount of play vertically and more than you would expect even from a worn 1928 crank. A bit of a disappointment given the positive encouragement of the last half hour.

What is the difference and why? Then I remember the 1934 engine and its crank. I hunt it out from storage and heave it next to the P40. I place the ’28 next to it and there appears to be no difference. Undeterred I grunt it into place in the P40 and replace the caps and bolts. A near perfect fit running on the Nissan shell mains. This is a fantastic discovery but why won’t the ’28 do the same job?

I take out the ’34 crank with a great sense of elation and have a close look at the Nissan shells resting in the crankcase. I take one of them out and place in a corresponding ’28 bronze backing shell with babbit lining. The unit fits perfectly into the Nissan tunnel as indicated by the information in the old technical journals that my rebuilder had. So what is the problem? I looked closely at the Nissan shell bearing and the Graham bronze backing with its cast babbit lining. It was then that I noticed the Graham unit was much thicker than the Nissan shell although both fitted perfectly into the Nissan tunnel. I then measured the journal size for all of the crankshafts that I had. The Nissan and the ’34 were the same but the ’28 was considerably smaller. It therefore appears that to accommodate the switch to shell bearings the Graham engineers had to make up the difference in the overall thickness of the two types of crank support systems. The obvious and most cost effective choice was to increase the journal size of the crankshaft and Bobs your uncle – problem solved. The Nissan engineers simply followed on and apart from the rear oil seal configuration and the counterweights to allow the OHV unit to rev higher the rest of the bottom end stayed the same until the end of P40 production in 1985.

If you want to re babbit an early Graham engine a badly worn ’34 crank can be turned down to give a crank to original ’28 journal size.

If you want to convert a ’28 or other babbit bearing early Graham motor to shell bearing mains you will need a ’34 or similar crankshaft OR

You will need a Nissan P40 crankshaft AND alter the rear of the Graham block to match the Nissan crank and its non-wet rear oil seal. I don’t know if this block modification can be done but the rear Nissan cap fits straight on to the Graham bolt holes. I suppose there would be a lot of brazing of the block to do and then machining to take the shape of the Nissan seal. If this was done then you would have a fully lobe balanced crankshaft and it may increase the availability of potential replacement crankshafts. As long as ’34 or beyond Graham cranks are still available this potential modification may never be necessary but remember P40 cranks are becoming rare anyway. With the low revs of the Graham motor the balancing lobes on the Nissan crank are in my opinion superfluous.

For those who may still be in doubt about the origins of the P40 motor I will list the other common features that I found

·         The front timing cover is identical to the ’34 with the same bolt holes and bolts

·         The sump pattern is identical with the same bolt holes and bolts therefore the cork sump gasket is identical NB the sump itself is shallower due to the P40 having an oil gallery cast into the block

·         The three babbit camshaft bearings of the ’34 and the P40 are identical sizes and the cams of these two are interchangeable. Unfortunately as the P40 is a gear driven cam and not a chain driven cam it travels in the opposite direction and the drive for the side mounted distributor is in a different place The crank is the same diameter for the gear and sprocket drive

·         The crankshaft of the P40 gives the same stroke as the Graham engines. As mentioned above the block is about 32mm shorter than the Graham. In the redesign of the engine to OHV the cylinders were cast lower into the top of the crankcase itself so the same stroke could be maintained. To make up the difference the conrods are unfortunately shorter and cannot be used in a Graham engine. This is one area where the metric system is used and creates a situation where the conrod caps also cannot be used in a Graham. Although they are exactly the same size so they fit into the Graham crankshaft they use a metric bolt. This bolt is smaller than the SAE bolt but is located so that the inner edge of the bolt hole lines up exactly with the Graham conrod but the outer edge takes up all the difference. It is therefore not possible to just drill it out as you would then eat into the space occupied by the bearing. I don’t know if a machine shop could do that job precisely but if not then you are constrained to the Graham conrods and caps

This brings me to the next step in the modernization of the Graham engine. As mentioned above the 1934 Graham engine still had babbit bearings for the conrods but the P40 has shell bearings. I wanted to ream out the babbit, put in tang grooves and fit shell bearings for the conrods using off-the-shelf Nissan units. I had lots of conrods but only 5 from the ’34 were useable so I went to use one of the ’29 conrods. Problem. The conrods are different. The ’28 and ’29 conrod heads have babbit cast through the conrod bearing face AND the side bearing face where it meets the crankshaft journal. What I had intended to do was to bore out the babbit to the OD size of the Nissan shells. The problem with using the ’29 conrods is that I was afraid there would not be enough babbit left through the middle of the conrod to bind the babbit on the side face to the steel so it would not just fall off. This can’t happen with the ’34 conrods where there is no babbit on the edge. 

I found another ’34 type conrod from a supplier in New Zealand had the set bored and balanced and had the Nissan shells fitted. It is a perfect fit. There was one modification necessary as the hole for the squirt up the cylinder is in a different place for the Nissan conrod (the cylinder walls come down lower into the crankcase) another hole had to be made to match the Graham conrod.

A similar modification had to be made to the mains on the bearing cap shells. The retaining lug is slightly smaller on the Nissan as it is only a retaining lug whereas in the Graham motors it also serves as the oil supply source so it is a fraction bigger. The hole for the oil supply on the Nissan main is on the crankcase side of the shell because the Nissan has its cast and drilled oil gallery. The hole for this in the Nissan bearing was just left as it was.

Because of the Nissan having a separate oil gallery you also have to make a small grove in the Nissan front main bearing to allow a flow to the timing chain oil feed dripper which is fed from this source but deleted from the Nissan shell.

In the end a Graham motor with modern internals and an easy renewal if ever that is necessary again. All in all however it is a fairly easy with a result that I hope will be long lasting


Regarding the Nissan bearings I wrote to you about for the technical corner. The Nissan crankshaft rear seal flange facing the inside of the engine is the identical diameter to the 1928 Graham crankshaft. A machinist friend of mine has said that it would be possible to mill out the rear of the graham block around the old seal grove to allow the installation of a machined plate to be attached to the rear of the block to match the flat required by the Nissan rear seal and thus convert a ’28 block to a Nissan crank. The Nissan crank does have balancing lobes but these don’t project further than the main journals so should not affect the oil gallery. The Nissan crank certainly weighs a lot more than the Graham original



Submitted by Brian Barwick March 21, 2013 EMAIL


Headlamp Shields for 1938 to 1940 Grahams


Submitted by Dale Robbins

Installing Chevy 6 volt generator.

On my 1929 615 I would like to install a '54 or earlier Chevy 6 volt generator. I need more amperage to run lights, elect wiper and heater. The main trouble spot is that the generator is negative ground. Is it possible to polarize it to a positive ground and is there a regulator with a positive ground. Has anyone used a Chrysler Product one, as they had a positive ground (just thought of that)?

Andy Wittenborn 


914 941-2744 (cell)

Regarding the generator with more amperage. I took a late '40's, Plymouth or Dodge, 6 volt positive ground and installed it on my '29 615. I removed all the mounting bracket tabs so it would fit in back of the water pump. Had an adapter made ($205) to use the tube type connection from the water pump. Got a regulator ($65) from Andy Bernbaum. Added an amp gauge under the dash. Hope one day to be able to put one (gauge) where the other is in the cluster. Seems to work OK, have not tried it driving yet. WP turns at 1.2 times the crank sprocket which may be a little slower than it should be.

Had trouble with my turn signal indicator light staying on. Found it was the flasher. Got a bi-polar flasher from Snyder Auto Parts ($4.95) and it now the light comes on only when I signal. Truck Lite had me reverse the leads on the flasher because of the positive ground. Changed it back to what it was before to make the flasher work. Truck Lite made my turn signal switch.

Since I am 6' 2" and older I moved my fixed front seat back 4 1/2 inches which makes driving more comfortable. Used steel plates to hold the back cushion anchoring the plate to the seat & center post. Makes getting into the back a little more difficult.


Andy Wittenborn 


914 941-2744 (cell)


If you want to check and adjust your car's toe in alignment yourself the following may be helpful.

Start by making a trammel bar out of a straight 2 by 4, two plywood triangles and screws. Screw one triangle solidly to one end of the 2 by 4. The second triangle needs an anchor screw to act as a pivot and a second screw in a slot in the plywood triangle to make the bar adjustable.

See photos.

1.. Raise the car and place jack stands under the front axle as close to the front wheels as possible. When you let the car down onto the jack stands the suspension must compress to it's natural position as if the wheels were on the ground.
2.. Set wheels facing forward.
3.. Next use chalk to whiten the center of the tread all the way around each wheel.
4.. Then hold a nail on a block of wood or cinder block and scribe a line in chalk by spinning each front wheel.
5.. Using your trammel bar set the points of the two triangles on the two scribed lines at the back edge of wheels between the top and bottom of the tires. Then tighten the screw in the slotted adjustable triangle end of the trammel bar.
6.. Carefully move the trammel bar to the front scribe lines. These lines should be toed in by the amount recommended by the manufacturer.
7.. If not loosen both tie rod end locking bolts and turn tie rod to toe in or toe out as needed.
8.. Now check position of steering wheel. If not centered you have to lengthen one tie rod end and shorten the other exactly the same amount until the wheel is centered at the straight ahead position.

I hope this information is helpful to you guys out there and feel free to call or contact me anytime.

Patrick Caron

Phone (860) 722-1630 Days
Phone (860) 742-6539 Evenings


Three Planets Method of Setting Timing:

1.  Remove number one spark plug and hopefully you will be able to see the two valves and the piston.
2.  Also find the timing mark on the flywheel
3.  Set the manual spark advance to fully advanced (Normal position)
4.  Crank the engine using a socket wrench on the bolt of the crank shaft and when you get the timing marks lined up check to see that you are not between the point where the exhaust and intake valves have just closed and just about to open respectively.
5.  If you are between the valves closing and opening then you are two strokes out cycle and need to advance one more full revolution. Crank the engine another 360 degrees and repeat step number four above.
6.  Adjust distributor setting so the points are on top of the cam.
7.  Once you have successfully completed the distributor rotor should be pointing directly at the Number 1 spark plug wire terminal on the distributor cap. Plus or minus a couple of degrees.
8.  Timing light should be able to clean up the rest.


Graham-Paige Rear Wheel Puller

For more information contact:

Patrick R. Caron


I received this article October 14, 2001

I thought your club members may like to read the following letter I had recently sent to Dodge Brothers Club. I am almost positive that the device I that I have described and made would be appropriate for those members with Graham-Paige antique autos from the 1930's and 1940's. Once they read the letter and see the photos they will know if this procedure would work on their cars. Would enjoy a reply from anyone interested. 

Don Lohr 

I have a 1931 Dodge Brothers Sedan. Last year I completely did over the brakes on the car including having the drums turned and having new brake linings bonded to the brake shoes. When I went to install them I could not figure how to adjust the bottom anchor pins correctly.  My handy dandy manual said that a "special precision tool" will be necessary. I tried in vain to locate such a tool but had no luck. One source said he had seen such a device; it looked like a brake drum with a section cut out so a measurement could be made while on the car.  What I did was to try different adjustments then look at the wear on the linings and then try another adjustment. A man at a brake shop said to put lots of sidewalk chalk on the surfaces and then take off the drum and see it they were wearing uniformly. Drove me crazy; too much work and I could not say for sure the adjustments were correct.  I put out a "distress signal" on the Internet and a wonderful gentleman by the name of Frank from the West Coast told me what he had done to make the adjustment very simple. He had also gone through hours and hours of trial and error before he came across the idea below.  He actually made his device out of metal and drilled and tapped 1/4 inch threaded hole for the metal rod. I simply used a piece of maple stock for my device and was able to drill and tap the 1/4 inch hole in that. When I threaded the rod into the block I also filled the hole with Elmer's glue. The wing nut on the device gives you a quite precise measurement; the screw in the smaller block gives you a more precise measurement.   

With the Dodge jacked up I turned the top ADJUSTING NUTS until the shoe just touched. Then I took off the brake drum and swept the brakes shoes with the device using the distance at the top of the shoes as the radius. I had to loosen the bottom anchor bolts on each side and turned them individually until the device just touched all the way around. I then knew that the shoes were in the correct position because they were forming a perfect circle. I did this all the way around the car. Excellent brake pedal and excellent brakes.  I assume that there may be other Dodge and Chrysler owners out there that may be interested in this idea. Please pass it along. I have made several extra devices if anyone might be interested. I was very lucky to have a drill press and a 3/4 inch spade drill bit. That size holes really fits snugly on my axles; the original nut holds the device on. Because my device was about 7/8" thick there was not room enough on the back axle shafts for the nut to be put on; I simply countersunk a hole so more threads were sticking out. Because the nut on the back was much larger than the front, I temporarily use a nut from one of the front wheels to hold the device on. I would not be too surprised if many of the other models and years of Dodges, Chryslers, etc. probably have that same 3/4" diameter measurement.  Let me know what you think. You will probably will write back and say many others have thought of this idea. I just had never run into anyone that had.  

Don Lohr

Link for the Graham technical corner - Removing rear brake hubs all Grahams that I know of

Scott Maynard

This regards the flywheel resurfacing for my l929 G-P 615 and I guess similar flywheels. No one was able to give me a definitive answer on removal of the clutch plate mounting pedestals (there are 6 of them). They just push out with a drift pin and hammer (or in a press). Be sure to measure height of pedestal top to flywheel surface before resurfacing. The pins (pedestals) do not have a shoulder.
Andy Wittenborn


These seal used in my '29 615 Graham-Paige.

Rear axle outer seal- Timken 472150. OD is slightly smaller, but have not tried to install it yet in the original seal housing. 
Lovejoy one way shocks. Seal for arm was originally cork, I've replaced it with a modern seal that works very well. SKF 9934 (front) and SKF 8648 for rear. Did all work my self after some trial & error.
Can assist anyone who want to do it.

Andy Wittenborn

Got something Graham related and technical?

This is the place for it. Email it to me and I will post it.