36 Pontiac Cooling System (or "Tales of the DPO")

[Bloo]

There was at one time a rubber dam at the top of the radiator. It is a big no-no to let air get from the front of the car to the engine compartment without going through the radiator, if you care about cooling at all. If air can go around, it equalizes the pressure on the back of the radiator with the front, so far less air flows through the radiator. Some cars do a better job of this than others. Some are so over-engineered it doesn't matter, but I think those are in the minority. This car clearly had a dam at the top. Unfortunately there was only about a 1.5" x 2" piece of it still there. I bought 1/8" neoprene sheet on Ebay, and started making patterns from paper. It was not a symmetrical hole and not obvious. Pretty soon I had a pattern but a new problem.

 

A while ago in this thread I soldered the mounting brackets back to the radiator where they had broken loose. What I didn't notice at the time was that the bracket that holds the rubber air dam was broken loose at one end. It was in a horrible spot, too. It was not only right next to the core, a Harrison strip core that might fall apart if you get it too hot, but it was also right where the top tank solders to the side tank on this oddball cross-flow radiator. I protected the core as much as I could with a piece of copper like I did earlier in the thread. I got it soldered but heard a "pop". Fearing the worst I painted the area with lacquer, turned the radiator upside down. I poured Acetone in there, the idea being that if there was a leak the acetone would come through the hole, wash off the lacquer, and make it obvious. It took 3 tries but I finally got it soldered with the bracket attached and no leaks.

 

 

WHEW! Here is what the rubber air dam looks like:

 

 

As unlikely as it seems, that actually fits right.

 

One more thing to do, get some better timing marks. The timing marks on a stock 1936 Pontiac are on the flywheel, and very hard to see with a timing light. Pontiac did not put the timing marks on the front until 1949. It looks like this. I believe this pic is one of @cevensky's engines.

 

 

My 36 has never worked with any timing lights except one super-simple one running on a 12 volt battery. It has never worked with any tach, in fact the electrical system is so dirty it seems to interfere with and sometimes break digital electronics. Several years ago I added a timing mark to the front. It was a simple piece of bent welding rod, and I made marks to go with it. When the car is all assembled, the access is terrible. You can't really get your hand down there at all. It looked like this when I took the car apart.

 

 

It was arranged with an offset from the bolt so that as you tightened the bolt it would bump against the timing cover and always land in the same spot. It had to stick up a little like that to miss the pulley. Functional, but far from perfect. With all this wondeful access I decided to either find one of those 1949-1954 pointers like the one on cevensky's engine, or remodel this one with another rod down from another bolt pushing down, an idea copied from the Pontiac pointer. Obviously the Pontiac one is more solid. I didn't know if it would fit, but there was at least a possibility because the 49-54 engines that used it take the same timing cover gasket my engine takes. On my trip east to the Flathead Reunion this year, I was on a mission to find one of those pointers.

 

On the way back I caught up with @VW4X4 at a farm close to Pittsburgh. He has a bunch of engines that have been outside but have some good parts. There were 2 Pontiac flathead sixes out there, an Olds straight eight, some Mopar flathead sixes, an early Olds Rocket 88 v8, a flathead Mercury v8, and probably some others I have forgotten. Check out his "Field of Engine Dreams" thread:

 

https://forums.aaca.org/topic/369768-field-of-engine-dreams-pgh-pa/

 

I got this from @VW4X4. It was a little rusty, but I cleaned it up and filled the pits with brass.

 

 

Looks nice now...

 

 

But does it fit? No, and the reason is a surprise. I was imagining that those later engines probably had a larger diameter harmonic balancer, and that this would probably bolt on but be a mile away from the balancer. Nope. The balancer is the same size. This pointer doesn't stick out far enough, and it hits the pulley.

 

 

The cure was easy. Putting the pointer up on top of the generator slider and spacer, instead of below it moves the pointer out 3/8". I added a 3/8" spacer at the other end and a longer bolt. You can see the new spacer at the right. It fits perfectly this way. At this point I tumbled to the probable reason the balancer sticks out further. This car uses a really wide fan belt. I don't know if they were still using those in 49, but probably not.

 

Now it is going to need some timing marks on the balancer. I might make a separate thread about that since the subject seems to be coming up fairly often lately.

 

More to come....

 

Edited November 27, 2022 by Bloo (see edit history)

[Bloo]

18 hours ago, critterpainter said:

Just a quick note regarding the green stripe Gates radiator hoses.  Bobs Automobilia DOES use this hose (or at least did as of 4 years ago when I was there last).

I looked into getting some of that. The green stripe hose is tough to find in 1-1/2", and it is expensive. Wow. I found a chunk of similar NAPA wrapped hose on Ebay. According to the parts book, the original was a piece of bulk hose cut to 7-1/4".

 

 

Someone suggested wiping the modern marks off with acetone. that didn't work on this hose, still I don't think it is going to show much when installed.

 

More goodies from @Kornkurt :

 

 

 

 

[EmTee]

8 minutes ago, Bloo said:

Someone suggested wiping the modern marks off with acetone. that didn't work on this hose...

Nothing that a dusting with some flat black paint won't fix... 

[Bloo]

I did make a separate thread about timing marks. It is here: https://forums.aaca.org/topic/388360-how-to-get-a-top-dead-center-mark-on-a-flathead-engine/ 

 

 

Now that I have reliable timing marks, I think it's time to get the radiator shell back on this thing and put it away for the winter. The beige slime is LPS-3, like Cosmolene more or less. I felt it needed some rust protection after having eighty plus years of sludge removed.

 

Before:

 

 

The headlight ground wires exit back forward and wind up in front of the radiator.

 

 

Where they will be completely hidden by a shield plate that has not been installed yet in this pic, then they pass through into the engine compartment, just to the right of the nose mount.

 

 

Where they pop out on the driver's side of the engine compartment.

 

 

And run under the engine mount cradle. Then they pop out and bolt to the engine mount plate (green). This is the same piece of metal the generator bolts to.

 

 

Completely invisible.

 

 

And it's time to put it away for the winter. We have snow.

 

The engine runs 185 degrees now, 10 degrees cooler than before. Time will tell if it is a big improvement but it seems to be. It is too cold outside to tell for sure. Ed's evaporust treatment will have to wait for spring now. I doubt it needs it, but I am going to do it anyway.

 

[EmTee]

Nice to she her all back together again! 

 

So, which water pump did you install?

[Bloo]

The one with the modern bearing. It's the only one that's together. The late 36 pump is next.

[EmTee]

I need to remove the water pump from my '38 Century sometime this winter (another job I get to do-over).  It appears that whoever re-built it pressed the fan collar on too far.  The belt wants to climb up the forward edge of the pulley.  The crank and generator appear to align, so the pump is the outlier.  I tried shimming the pulley forward with one flat washer for each bolt, but it's still too far back...

[Bloo]

My guess is a 38 has a modern bearing, and you should be able to press it, but it might be easier to use a bearing splitter and a puller to get it apart. I'd be tempted to take it all the way off and put it back on with a drop of loctite just in case, in fact I tried it, but the fit was nice and tight and I don't think any loctite actually stayed in. The trouble is if you take the hub clear off, you will have to have the back off of the pump to press it back on. You need access to the center of the shaft.

 

On the other hand if it's just a bit too far on you could pull it out just a little. This thing works great.

 

https://www.harborfreight.com/bearing-separator-and-puller-set-62593.html

 

Use the small bearing splitter upside down so you get a flat surface, and leave it just loose enough to miss any radius that might be on your hub.

 

Edited December 1, 2022 by Bloo (see edit history)

[EmTee]

4 minutes ago, Bloo said:

The trouble is if you take the hub clear off, you will have to have the back off of the pump to press it back on. You need access to the center of the shaft.

Thanks for the tips.  I did have the back off the pump a year ago when I had removed it to get the radiator out.  I wanted to look inside based on comments from Ed, since I was chasing an overheating issue and couldn't assume that the radiator was the only contributor.

[Ben Bruce aka First Born]

I would just buy and install shims behind the pulley.   Available from Summit and others. Various thickness. Works for me. They are meant for fan spacers but I don't see why they would not work.

 

  I have never learned a "perfect" way to determine which pulley is out of place/alignment. Trial and error?

 

  Ben

[37_Roadmaster_C]

8 hours ago, Ben Bruce aka First Born said:

  I have never learned a "perfect" way to determine which pulley is out of place/alignment. Trial and error?

 

  Ben

My approach is to assume ( Yep, there is that word again) that the crank pulley is right and go from there. This is by no means a fast and true rule. Anything could be in the wrong place, but without knowing, you have to start somewhere.

[Bloo]

Yeah I made that asssumption when I published the numbers earlier in the thread for 36 Pontiac. I have no reason to believe my crank pulley is wrong, but I don't know it isn't wrong. What inspired me to nitpick it is the fact that I had three water pumps and all three were pressed to different heights.

 

On the Pontiac, depending on which water pump. you could only do so much with shims. On the pump that is installed on the car right now, the one with the "modern" cartridge bearing, the hub has barely enough sticking out to index the pulley and fan.

 

[EmTee]

Before I take anything apart I need to try to get a decent measurement of the actual misalignment so that I know how far the hub needs to move to bring it into alignment with the crank and generator.  I'm still thinking-through the best way to do that from under the hood with everything in place.  The other issue I want to address is the distorted flanges on my fan pulley.  It looks like it has been dropped, pried-on or otherwise physically abused many times over the last 80 years.  It will take some careful 'tweaking' to get it back into shape.

[Bloo]

I imagine you saw this.

 

 

It's just an aluminum rod tied to the pulley with 2 tiny bungee cords. The idea was to get a measurement. After screwing up what should be simple math more than once, I was doing this with a water pump and pulley in place, and just checking for error. It would be close to impossible with the car all assembled, but I'm thinking you would have to do some variation of this... unless a spec exists for the flange height and you could just make the pump right with it off the car. Such a spec exists for 49-54 Pontiacs, but I never found any specs for older ones, even after digging deep in both AACA and POCI's literature collections.

 

[EmTee]

Yes, thanks.  I may try something like that.  I need to look more closely to find a solid reference point that I can use to measure the crank/generator and crank/fan pulley offset distances.  Should be doable, but I'm currently in up to my elbows with torque-ball seal replacement...

[Bloo]

Yikes! That ought to keep you busy for another day or 2....

 

[Oldtech]

Yes, if you have a puller should be easy to just pull the hub a bit. the bit to be determined. I have had to move them. no problem. 

[Bloo]

Regarding the "Is the postition of the lower pulley correct?" question, when I was checking the water pump, I also checked alignment to the generator.

 

I did pry the crank pulley looking for endplay and didn't see any. I didn't put a dial indicator on it. Obviouslly there had to be some endplay but not much. At first I thought the generator was a bit off.

 

 

That green round thing just above the 3 silver bolts is where the front plate of the generator bolts on. As it turns out, the generator agrees with the crank pulley. It lines up perfectly using no shims.

 

The slider bracket on top was a wee bit off, but that is just a stamped part. I stuck a washer on that for a shim to make it slide nice.

 

[Bloo]

It's been snowing for two days straight. I don't even know how much came down, but there is at least a foot of accumulation in my front yard. I spent all of yesterday morning digging my car out after the city plowed it in solid. I think I see another storm rolling in. Bah. Let's talk about water pumps.

 

These are the 3 victims, as posted earlier in the thread.

 

 

The one on the right is all done. It was the rusty mess with a bolt glued in it that I found in the trunk when I bought the car. It is probably 1950s vintage, but made to fit 1933-1936. It uses a modern cartridge bearing and a carbon seal. I had no intentions of rebuilding it ever unless the original type pumps proved too troublesome to use, but time ran out and it was the easiest. With a rebuild kit from ebay, and a new impeller from KornKurt it came right together. It is on the car now and stored for the winter.

 

The one on the left came from Ebay, an old NORS rebuild. It matches the pictures in the service manual, and it is what I intended to use. It uses oil for lubricant, and has 2 flip top caps like you might see on a a machine tool. The oil soaks through sintered bronze to get to the shaft. After a teardown, It looks a bit hacked up. I've pushed it to the bottom of the list for now.

 

The one in the middle came off of the car. It had zerk fittings on it. I thought it was the same as the one on the left, but hacked up to use grease instead of oil. As it turns out not very much of that is true. Hacked up? Yes. Not the same as the one on the left though. It turns out this is... well was... a factory needle bearing pump used in the last half of 1936 production. The factory pushed these hard for replacements on 1933-1936 cars as well, but sometime before 1940 they were superseded by a pump similar to the one on the right, using a cartridge bearing and carbon seal.

 

When you get these apart and laying next to each other it is clear something is up and they are not the same. The pump on the left has the oil cap holes tilted toward the driver's side. This is sensible for good access. The oil cups for the generator and starter are on the driver's side as well.

 

 

The pump in the middle has the holes pointing straight up, and a tall threaded boss around the rear hole. This is what it looked like coming off of the car.

 

 

At the rear, that simply shouldn't be a zerk. As it turns out it should be an oil reservoir cup, and a big one. No doubt that is why the factory re-cast the body with it sticking straight up. I was expecting maybe the rebuilder had made some changes inside, but no. It was just full of grease that couldn't go anywhere.

 

At the front, when you have both pumps to compare, it is also clear something is up. On the earlier pump that used an oiled sintered bronze bushing in front, there is a massive reservoir encircling the bushing.

 

 

But, on the pump in the middle, no reservoir.

 

 

Not only is there no reservoir, but the bore is stepped at the back. The rebuilder used a bronze bushing and drilled a hole in the bushing for grease to get through. That much worked fine. Unlike the rear bushing, it had lubrication. Progress on this pump ground to a halt when it was discovered that some previous rebuilder had reamed this outer bore from 3/4" to 13/16".

 

There was no bearing to fit the hole that would also fit the original shaft size. In fact, the shaft size is non-standard, neither SAE nor Metric, and there is apparently only one bearing that fits. That bearing has an O.D. of 3/4". There are no sleeves commercially available this thin.

 

@Bhigdog came to the rescue. He made me a repair sleeve. In fact, he made two of them with a slightly different fit.

 

 

These were intended to be held in with loctite sleeve retainer, so they would not distort from pressing. Meanwhile @37_Roadmaster_C made me a "plug" to check concentricity of the bores. Not knowing who reamed that bore oversize or how, there was some doubt.

 

 

They are concentric! (Whew!). Now the only holdup is waiting for the Loctite 640 to get here. More to come when it arrives. It may already be here.

 

 

Edited December 11, 2022 by Bloo (see edit history)

[Bloo]

The Loctite 640 sleeve retainer got here, and after curing in front of the heater for a day, I took it over to @37_Roadmaster_C 's house and poked a hole in the side for the grease port.

 

 

And then a little cleanup and deburring with a dremel tool....

 

 

And it looks really, really nice!

 

Next I will need to test fit the rear bushing and the impeller to make sure they still fit like I think they do, and check the hardness of the shafts. I have some Rockwell files on the way.

 

Then, back to the front:

 

 

 

493298 is apparently a Torrington NB-10 bearing. This one is marked NB-10-X. At some point Torrington stopped putting the oil hole in the side of their drawn cup needle bearings by default, and afterward you had to specify the hole if you needed it. I think that would be NB-10-OH. The hole is needed here.

 

I've never seen the 493299 oil cup. I have a couple of potential substitute oil cups with the right thread here, so finishing the pump up and making it work is no problem. I'd like to see what they actually used though. There was a 1936 Pontiac for sale in Florida recently, and it appeared to have a pump like this one.

 

 

See the oiler? Let's zoom in....

 

 

I emailed the seller and asked for a better picture, no response. Oh well. I wonder if anyone in the forum here bought the car?

I have these.

 

 

I suspect the one on the right is closer to the correct size. It's diameter is about 5/8". The trouble is, it is a wick oiler. That is the wick I pulled out of it at the right.

 

 

I would have to figure out how to gut it. The standpipe inside looks silver brazed at the top. I have no idea what they did at the bottom.

 

 

Edited December 18, 2022 by Bloo (see edit history)

[EmTee]

Could you use a brass nipple and cap in place of the 493299 oil cup (permanently or temporarily)?  What size are the threads on the oil cup?

[Bloo]

It's just 1/8 pipe. I took some pictures of my car with those 2 cups I have stuck in about the right spot with masking tape. The smaller one:

 

 

The bigger one:

 

 

The car in Florida:

 

 

It sure looks like the smaller one is about the right size. I think I am going to try to unsolder the pipe inside.

 

[Randiego]

Bloo,

You have enlightened me regarding grounding the headlights.  My 36 is in the middle of dismanteling.  I have the engine rebuilt and all of the front "clip" (hood, radiator surround, headlights and grill all off.  I will be removing the body from the frame this summer, sending it off to the body shop.  Then I can attack the frame and get it done.  

 

I also want to get the headlights refinished as you described to Kingdometravele.  Even though the car will be stored inside, the thought of disassembling the headlights down the road and scratching the paint in the process does not appeal to me.  SO....... do it right the first time, eh?

 

Randy

[Bloo]

I have been running Evaporust in the cooling system again. It's been 3 weeks or so, and like last time, the car doesn't like it. It foams and pushes. I was able to keep most of it in there for about 2 weeks before It got diluted this time. It certainly does something, as my sock filter filled right up, bad enough I didn't know if I was going to be able to get it out of the radiator. I posted pics over in the @Grimy's Sock thread.

 

 

I'm not really trying to solve a heating problem. It was never that bad, and runs cooler now that the water tube is new. One reason to keep doing this was to get the temperature gauge bulb out. I am wanting to pull the head to check cylinder taper, as this engine needs rings. I need to do it without clobbering the temperature gauge. These are notoriously hard to get out without damage. I have been spraying the threads with heat riser solvent since last year, and I hoped the Evaporust would clean up any crud that might be holding the bulb from the inside. I never tried to remove it, just noticed last year that the nut was stuck. Well the nut unscrews with no effort at all now, and the bulb almost fell out. It looks like it was put in last week.

 

 

So with it out, it was time to test it, and it turns out it's not very accurate. The few times this car has got hot, the gauge was believable. Well, the scales on these gauges are not linear, and are vague on purpose I think. You can't even really tell how to divide up the scale. It turns out actual temperature is really compressed near the top, and it tricked me into thinking the gauge couldn't be that far off. Now I know. The lower the temperature goes, the worse it gets. At center scale it is off by a ridiculous amount.

 

 

 

 

 

Ok, to be fair I don't know that the meat thermometer is accurate. In fact I bough another one but haven't tried it yet. Still....

 

 

 

Also notice the thermostat. It's marked 155 degrees F. I think originals were 151F(?). I'll have to look that up. Nevertheless this one is closed at 150, as it should be, and the gauge is reading 179F.

 

 

 

Ok so now we are at 155F, the thermostat is cracked open a little as it should be, even though I didn't catch it in the picture. Gauge says.... 180F.   But, when the water is 180F.....

 

 

 

Yes 180F. What is that mark? It isn't 180F. If the bottom half goes 100-140-180, maybe the gauge gets more spread as you go up. It sure looks that way and it would be more useful, but the opposite of what this gauge actually does. The little mark next to the "water" legend looks different. I used to think that one referenced the boiling point of water. Maybe it's true. 180-200-220 with an extra mark at 212F could make sense if the gauge nonlinearity wasn't backwards I guess. Or it could be 180-193-207-220.

 

 

 

200F water... and the expected temp is either 207F or 212F. Yeah, it's off but not an extreme amount for a gauge like this. It lulls you into thinking it is OK, but it isn't.

 

 

 

203F and its still at about that mark. So compressed up here.

 

 

 

Boiling water is reading 209F on the meat thermometer. Pretty much spot on for the altitude here (800 Feet). Maybe a degree off. It had to boil a while to get there though. That surprised me. I would expect the boiling point of water to not take so long to register. It still hasn't moved much from that mark. It does more or less represent the boiling point of water.

 

The thermostat (155F) is fine. Here it is open at 180F.

 

 

Takeaway? Back before I changed the water tube and I thought it liked to run at 195F with a thermostat or without.... That was really 175-180F. Yeah. 15 or 20 degrees off.

 

Any of you guys ever attempt to calibrate one of these? Behind the face I believe it's the same as a mid to late 30s Buick.

 

[EmTee]

Interesting.  It makes me wonder about the gauge in my '38 Buick.  All I can say is mine seems 'in the ballpark' as after a drive shooting the thermostat housing with my IR thermometer yields a number close to the gauge value.  But, the temperature bulb is back by the firewall, so I expect that to be the hottest point.

[Bloo]

Welp.... I pulled the head.

 

The reason had nothing to do with cooling. As mentioned earlier my rings aren't too good. I wanted to measure the bores and see if it just needs rings as these old cars did occasionally, or if it really needs a complete rebuild with boring, new pistons, etc. I couldn't plan for what to do next without knowing. Now we all get to see inside the water jacket.

 

In the 1990s I bought a tool that had a tiny grain-of-wheat bulb at the end of a long skinny wire that you can bend around like Gumby. It is great for looking inside things like water jackets. It would have been handy to look inside the block when I did the water tube. But, as cool as it was in it's time, it is incandescent. We could do so much better now. I still have it and it still works but I can't find it. What I can find is a bag of white LEDs I bought on Amazon during lockdown. I needed one then, and I don't even remember what for. There's probably about 99 of them left.

 

So, without further ado, allow me to present.... The Electric Chopstick.

 

 

Yes, that's a 3mm white led on a chopstick, surplus wire, and some heatshrink tubing. These LEDs were so cheap they don't even have a datasheet. An LED needs a current limiting resistor. A typical normal current is 20mA on a lot of LEDs that do have datasheets, so that is what I shot for with a 9v battery. The 300 ohm resistor is hidden under the red heatshrink tubing at the battery end. The chopstick draws 21mA. Not bad.

 

 

More coming.....

 

 

[Bloo]

 

As near as I can tell that really is the bottom. I'll try to vacuum in there a little but I don't think I'll get much.

 

 

There are wires that held the sand core still in here, and unlike the scraps that fall out of most blocks they appear to be right where they were when it was cast. They don't even look pitted. It is the same in the head, but I don't have pictures of that.

 

 

Between a couple of cylinders where there is a lot of space, probably due to the location of a main bearing.

 

 

And between a couple where it is tighter.

 

 

 

More wire.

 

 

And over on the other side where you can only look down on the top of intake and exhaust ports.

 

 

More Ports.

 

So, no real sludge buildup or leftover sand that I can see. The color is deceiving. It isn't rust colored. In person it is bright yellow in there, even the parts that look blue in the pics. I wonder what it is, but suspect it might be yellow dye from the yellow antifreeze. Maybe calcium deposits? I am going to take some of it that I caught with the sock and see if it reacts with an acid, but antifreeze is quite alkaline, so I doubt it will tell me much. I imagine it will react anyway.

 

Edited May 15, 2023 by Bloo (see edit history)