I have a 72' and I'm just sorting out my engine compartment in preperation for the engine/tranny install. I have searched around the net and seem to be having trouble finding the answer to my, what was thought a simple question ( panteraplace wont load right now). I can't seem to find an accurate diagram of the factory cooling set up ??.. Once I see what the factory intended I kinda see where I'm at... My system consists of, alum rad, SS cooling tubes and two beat up tanks, I also have a "green stripe" hose kit w/thermostat... I need some advice/guidence on how best to set the system up from here??.. The car is a sleepy 300 or so HP car in a hot climate.. Thanks for any help or advice!!...
Original Post

The OEM cooling system was a botched job, if I were to explain it to you you'd either assume I was full of bull or somebody at DeTomaso mistakenly assigned the janitor to design the cooling system. On top of that the 351C was designed with a unique thermostat that over time has always been replaced with the wrong thermostat (i.e. a Windsor thermostat), resulting in over heating every time.

The aftermarket replacement equipment available for the Pantera cooling system duplicates the original "wrong" engineering. There are two ways you can choose to go: (1) engineer the cooling system properly which requires custom fabricated equipment, or (2) use what is readily available and make the best of it.

The goals for improving the cooling capacity of the cooling system are:

1. employ the correct thermostat
2. increase coolant flow through the system
3. increase the surface area of the radiator
4. decrease the radiators restriction to air flow
5. decrease the radiators restriction to coolant flow
6. increase the air flow capacity of the electric cooling fans
7. fix the system's inoperative air bleeds

I'm going to assume you do not want to re-engineer the cooling system, few owners do. The two most prevalent replacement radiators (Fluidyne & Ron Davis) effectively deal with goals 3 & 4. I've seen many Panteras in which the original cooling pipes were partially or fully plugged with rust, so replacement with a stainless pipe kit is highly recommended to help with goal 2. The dual Flex-a-lite fans that are so commonly recommended flow about 2500 cfm, which is very good, effectively resolving goal 6. The Edelbrock coolant pump is rated at 35 gpm and is the best of the usual choices, it will hopefully improve coolant flow (goal 2). As a side note, Larry Stock sells a racing coolant pump rated at over 100 gpm (manufactured by EMP/Stewart Components); the pump is designed for Windsor motors but Larry performs the modifications needed to allow it to fit a 351C. Many people cringe at the price of the Edelbrock pump, Larry's pump is even more expensive.

This brings me to the two tanks, the swirl tank and the overflow tank. The original tanks rust internally and get thin over time, eventually a hole in one of the tanks will develop, so replacement of the tanks is recommended. Now here's the head slapper. The purpose of a swirl tank is to remove air from the coolant, any air collected by the swirl tank must be vented. Swirl tanks do not normally have radiator caps on them, a radiator cap doesn't allow air to vent, it defeats the purpose of the swirl tank. In other words, the Pantera swirl tank is non-functional. There is normally an air bleed at the high point in the swirl tank where air collects. That air bleed would be connected to a "surge tank" (also called an expansion tank) which is the low pressure point for the cooling system. Now take a look at the "overflow tank". It also has a radiator cap, overflow tanks are normally vented to atmosphere not pressurized, there is no need for a radiator cap on an overflow tank. The Pantera's overflow tank was originally intended to be a surge tank! I state that with 100% certainty because all cooling systems employing a swirl tank, or any type of air bleed, also employ a surge tank. Surge tanks are the normal place to route air bleeds because they are the low pressure point in the cooling system. The Pantera has no proper place to route air bleeds because its surge tank was mistakenly used as an overflow tank by an inexperienced engineer! The radiator's air bleed in the Pantera is connected to the swirl tank, which is a higher pressure point than the radiator outlet, its not possible for air in the radiator to vent into the swirl tank, the coolant will actually flow in the opposite direction. Air is supposed to be vented away from the swirl tank, not into it! It does no good to remove air from the coolant with air bleeds or a swirl tank if you have nowhere to vent that air to! This is why I am certain the overflow tank was originally intended to be a surge tank.

Even if you do no further re-engineering of the coolant system, I recommend you make some small modifications to your 2 new tanks which shall allow the tanks to be used as originally intended, as they would normally be used in a cooling system employing a swirl tank. Remove the radiator cap fitting on top of the swirl tank, cap the tank off with a plate (requires welding) and install an air bleed at the high point. Then install 2 connections for air bleeds in the overflow tank.

In your re-engineered system, the bottom nipple of the surge tank (former overflow tank) shall be connected to the suction of the coolant pump. The Edelbrock coolant pump has a spare pipe threaded connection that can be used for this purpose. This tank will be run about half full of coolant, an air space shall be maintained in the tank to compensate for the condensing & expansion of coolant as the system temperatures vary. There is no need for an overflow tank in a cooling system employing a surge tank. The new air bleed on top of the swirl tank will be connected to one of the 2 air bleed connections on the surge tank. The air bleed connection on the radiator outlet tank shall be connected to the other air bleed connection on the surge tank. This small re-engineering of the coolant tanks resolves goal #7 and fixes one of the design errors of the Pantera's cooling system.

Finally, if you are employing a 351C as your Pantera's motor, you need a Robertshaw 333-180 thermostat. Your thermostat should look like the thermostat in the picture below. If it doesn't, buy the correct thermostat and install it. Take the other thermostat back to the guy who sold it to you and tell him George said he should shove it up his ...... Smiler Goal #1 resolved.


Images (1)
Last edited by George P
Here's a diagram of the basic cooling system flow from the tech manual. Keep in mind that this is a diagram only and does not reflect the exact locations of the pipes.


Images (1)
Wow, thank you very much for your advice and "tuning me up" on this. I will indeed take your advice and make the suggested mods, with the acception of.. I have a flow cooler water pump with the anti cavitiation backing plate. I did purchase the SS tubes, I had Bryan McCulloch (spelling?) make me a radiator which is a beautiful looking core ( as seen in a post http://pantera.infopop.cc/eve/forums/a/tpc/f/5650045562...360079675#9360079675 ). I also leaned the rad forward to improve air flow which I may move back again should I decide to relocated the condensor. As far as the two tanks go, I did find a pinhole in one tank ( see burn marks at bottom) along with a few old repairs. I will have consider exactly how to approach the cooling tank (s) issue ($). I guess I will start pricing new tanks as I attempt to do a nice air tight weld on the old one. I will confirm the thermostat also. So just to be clear, ( I'm not only ugly but I'm kinda' stupid too) When you say "swirl tank" you are refering to the smaller tank, yes? Iliistration 11, #28, page 35 p/n 11023A "Tank Pressure" ??... You want this cap welded up and an air bleed at top?... Thank you for your expert advice, I can see now new tanks are in my future...


Images (1)

Please clarify for me, and maybe others.

You used the terms swirl tank, surge tank, and overflow tank. and your modifications switch the duties (and thus the names) of the tanks in the process. That got me totally confused.

I am going to pose my questions using only 'shorter tank' and 'taller tank'.


By swirl tank you mean the shorter of the two tanks? Right? De Tomaso referred to this as the 'pressure tank'.


By overflow tank, you mean the taller of the two tanks. Right? De Tomaso referred to this as the 'expansion tank'.


The shorter tank has two 1 3/8" nipples (top and bottom, has a threaded boss that was the original location for the temp sensor, and a radiator neck that accepts the cap and has a top nipple for a hose that will bleed excess pressurized coolant into the bottom nipple of the taller tank when the radiator cap on the shorter tank opens - right?

The taller tank has a bottom fitting connected to the pressure bleed of the shorter tank, and an overflow hose that will dump excess coolant to the ground, right? Fine, got it.

Now where you sure lost me was when you started outlining the modifications to the tanks.

You say weld up the shorter tank radiator cap fitting. In this action, you remove the entire radiator cap fitting neck, including the small overflow hose nipple, right? Fine, got it. I'm assuming the radiator cap is then moved to the taller tank, yes? But you know about 'assuming'. Wink

You say the small bottom nipple of the taller tank will be re-plumbed to a coolant pump intake port. Fine, got it. The other end of that small hose was attached to the previously removed (and welded) radiator neck nipple on the shorter tank; so that section of hose is no longer used. Fine, got it.

You say install an air bleed at the high point of the shorter tank (which now has a welded top cap). You mean a nipple for a hose, right? As opposed to a petcock for manually releasing air. Fine, got it.

You then say add two small nipples to the taller tank; one for the new top mounted air bleed on the shorter (welded) tank and the other for the air bleed hose that starts at the radiator and runs rearward under the car, right? Fine, got it.

Question - WHERE should these two new, added air bleed nipples be located on the taller tank? Bottom, top or not important? I'm assuming top (above coolant level), but again, you know about 'assuming'. Wink

So the engine is started, coolant is cold. Coolant pump is spinning, but with thermostat closed, coolant is retained in block. When the thermostat opens, coolant is then free to leave the engine through the hose attached at the thermostat and enters the top of the shorter (welded) tank and exits the bottom of that tank towards the radiator. Any air leaving in the coolant from the thermostat exits the system at the new top-mounted air bleed of that shorter tank into the taller tank. If any air is still making its way to the radiator that air exits at the radiator air bleed and feeds into the taller tank.

As the system warms up and pressurizes, the shorter tank should be totally filled with coolant and no air. The taller tank will have some coolant (you said keep it at about one-half) and some air, all under pressure.

QUESTION – When the system is totally pressurized, can any more air can be bled off to the taller tank? I can see at start-up, the coolant pump induced pressure would force out air that had risen to the top of the shorter tank and radiator. But at full operating pressure, what would cause air to migrate to the taller tank? I mean, the whole system is seeing the same pressure, right?

The coolant pump is plumbed to receive coolant from the radiator return line AND the taller tank's bottom nipple.

QUESTION - What source will the coolant pump be pulling coolant from? The radiator return line alone? How?

Isn't it possible for the coolant pump (designed to supply cooled coolant to the engine) to draw HOT coolant from the bottom of the taller tank? I know the hose sizes are vastly different, and I'm guessing the path of least resistance is going to keep the pump receiving most (if not all) of its coolant from the larger radiator return line. In view of this, will the coolant in the small hose from the taller tank see no (or very little) movement into the pump?

When the engine is turned off, it will start cooling down. The volume of the coolant will reduce. Am I correct the pressurized coolant in the taller tank will now be pushed back into the system to fill any air voids?

And then the cycle starts over again.

If I've finally got this right, please let me know.

If not, please let me know where I've gone wrong.

I'm liking what you've proposed, just not sure I've fully grasped what is actually occurring.


Last edited by lf-tp2511
What he ^^^ said...lol... Is there a chance for a photo of a moded tank set up??...
Last edited by plt-1
Originally posted by PLT-1:
... When you say "swirl tank" you are referring to the smaller tank, yes? ... You want this cap welded up and an air bleed at top ...

Yes to both questions.

I've never "sprung" this idea on you guys before, so I figure its over due. The fact is, without a surge tank, the swirl tank is useless, it doesn't accomplish anything. The Pantera swirl tank & air bleed system has never worked, that's going to be a hard pill for some people to swallow. But the fix is rather simple.


Images (1)
Thank you master... I will report back w/photo for your approval... When you say "air bleed" a 1/4" or other simple hose fitting is all that is required?... Thank you very much for your help and advice!! I will weld it up tomorrow and submit a photo...
Larry, you wrote your post after I had already begun mine.

Lift the hood on any modern car and you will find an air bleed system just as it is plumbed here, this is nothing new. Open up Carrol Smith's books "Prepare To Win" chapter 8 or "Engineer To Win" chapter 10 and you'll find a race car cooling system with swirl tank, fully described.

The air bleed hoses are small diameter, 1/4" ID is plenty big, so the amount of hot coolant recirculated from the swirl tank to the "surge tank" (formerly the overflow tank) is minimal. I've specified pulling the air from the radiator's outlet tank, so that coolant has already been cooled.

The cooling system has three pressure zones: the highest pressure is the pressure in the block, the intermediate pressure is the pressure between the thermostat and the radiator, and the lowest pressure is the pressure between the radiator outlet and the coolant pump inlet. The swirl tank is in the intermediate zone, the surge tank is in the low pressure zone, so coolant will readily flow from the swirl tank into the surge tank. The radiator outlet tank is in the same zone as the surge tank, it doesn't seem to me that coolant would readily flow towards the surge tank from the radiator outlet. But I've checked under the hood of several cars and have found them all plumbed this same way. Carrol Smith also specifies the radiator air bleed plumbed this way. I can only guess that since air rises, that as long as the surge tank is higher than the radiator the air will migrate towards the surge tank.
Oh ...

The coolant in the surge tank can only flow out of the surge tank (into the suction of the coolant pump) at the rate that coolant is flowing into the surge tank from those two 1/4" bleed connections.
so coolant will readily flow from the swirl tank into the surge tank. The radiator outlet tank is in the same zone as the surge tank

Okay, this still has me concerned.

Swirl (shorter) tank is coolant at its hottest. And you are saying this hot coolant will readily flow to the surge tank. And coolant in the surge (taller) tank is being continually sucked into the coolant pump and then directly into the block.

So there will be a steady flow from the 1/4" hose of fully hot (engine temperature) coolant in a closed loop right back into the engine.


What effect that small a flow really has is sure not known to me. But it IS a closed loop of full-heat coolant.

And for years, we have preached to always use the correct thermostat, because without it not all the coolant is making its way to the radiator. Which is exactly what is happening with this closed loop of hot coolant; that water never sees the radiator. Right?

George, are you saying this closed loop is found in the books you reference?


Keep in mind the Cleveland was unique in its design that closes off the recirc when the engine gets hot. Every other V8 manufactured during that era, that employed a standard thermostat, like the Ford Windsor, the Ford FE, the SB Chevy & BB Chevy, Chrysler, etc ... ALL, no exception, constantly recirc coolant (bypassing the radiator), even when the motor is hot. There is no method provided in their design to stop the recirculation of coolant. Those engines have a much larger hose for recirculating coolant than the little 1/4" hose we're talking about. For instance, the recirc hose on a Windsor V8 is 3/4" ID.

Pull the hood on any modern transportation car, you shall find them plumbed with air bleeds the same way. Inspect any modern race car closely, you'll find cooling system air bleeds all over the place. This is not an issue.
This is not an issue.

Thanks, George.

You've got the books and I can tell you have been doing a lot of thinking on this. I'm new to it, and just trying to wrap my head around it all.

For another in depth method at cooling system modifications, go to the Space City Panteras POCA chapter's website.


First article under "Pantera Technical Articles"

This a pdf file you will download and then view by opening.

George - I would like your comments on what John has done and how they mesh with your method.

I am very familiar with Johns cooling system mods, I know them by heart, didn't need to read the article. John & I have compared thoughts on fixing the Pantera cooling system. We are simpatico on this issue.

He eliminated the swirl tank, employed a single pass radiator, streamlined the plumbing and substituted a remote BMW thermostat manufactured by Stant. John believes the BMW thermostat is less restrictive than the Robertshaw thermostat. All of these mods increase coolant flow. Good stuff. Goal #2 of my original list.

I am in agreement with John, the swirl tank is not needed to remove air. Evans Cooling advises against the use of a swirl tank too, because it impedes coolant flow. To Evans Cooling, coolant flow is everything, the faster the better.

He employed a surge tank & installed air bleeds at the engine outlet & both heads. More good stuff, goal #7.

He didn't install an air bleed from the radiator. Carrol Smith recommends the radiator air bleed. It makes sense to me that there should be an air bleed at each pressure drop, because entrained air will escape at each pressure drop or drop in velocity. Since every Pantera built was delivered with a non-functioning air bleed system, John's lack of a radiator air bleed is obviously not the end of the world. His Pantera has 3 functional air bleeds which are 3 more than what it had when it was new.

The remote BMW thermostat allowed John to modify the heater core plumbing. Getting those tubes out of the tunnel and routing the coolant to the heater core through the forward firewall makes good sense, I plan to do the same thing with 6018.

John utilized two high capacity electric fans (goal #6), sans an air shroud as an air shroud would restrict air flow at high speed. Air shrouds improve the fans ability to move air, but restrict the free airflow through the rad. Its a trade off & an owner should be honest with himself & decide which is most important. If you race open road races, etc, then free airflow is more important. If you use the car for in-city transportation, then I think the fans need the shroud. I had a shroud built into the radiator I had custom built by Evans Cooling, even though Evans Cooling advised against it.

Johns modifications are very extensive, more extensive than the average owner will want to do. As a side note, the mods I'll be doing to 6018 are in some ways more extensive than John's.
Last edited by George P
Is this kinda' what you wanted??.. Please don't count off for the crappy welds, I'm a rookie... I used a few pieces of brake line this time... Once I get this system up and running I will do a nicer set up on my new tanks... You could use AN fittings or hard brake lines etc... Would you like the ports higher/lower??.. Is this kinda' a visual of what the system should look like??.. Thank you very much for your help!!...


Images (1)
Yes. As long as the tubing is 1/4" OD. It will be difficult to slide 1/4" radiator/fuel hose over those flares!

A for effort, WooHoo!
Great!!.. Thank you for your technical advice! I will fool around w/it a bit more and report back... Hope this helps the other guys...
Kim found this website for me one day while I was looking for a thermostat for my truck.
The parts look like high quality pieces. I havent had a chance to call them to inquire about the 351 pumps they sell, and they do list Robert Shaw thermostats.
Link copied to your clipboard.