I'm putting a flatcad 346 in my '31 fordor. The plans are to ditch the factory pump and run an electric water pump reversed cooled, heads down. How do I wire it to turn on and how do I get a thermostat in it? I see they sell remote thermostats but I'm not sure what I need. I thought about using my fan switch to turn on the fan and the water pump at the same time and not using a thermostat, but then water isn't circulating all the time and I'm afraid of hot spots. I'm not sure how to plumb any of it. Can reverse flow be achieved with a down flow radiator or do I have to use a cross flow? It seems to me the order would be from the bottom of the rad into the pump, to the heads, out the block, into the t stat, and back into the rad. if that order is right then I could run a non full flow line from the engine side of the t stat to the pump as a byp***. would this be correct? any ideas?
Relay's relay's relay's. They are cheep & reliable will carry a load without a load on a switch. I would not piggy back the pump on the fan switch. What is an electric fan doing on a 1931 Hot Rod? In normal conditions the pumps only run with the motor running so I generally wire them with key on only. You can put an over ride switch to the relay if you want so it runs with key off. Same with elect. fans if you must run them. I hate those things in early Rods. Remote stat's work just fine but not knowing Cad flathead motors not sure what to advise there. A good photo of what you have going on might help a lot. Reverse Flow. Why do you want to do that? How do you run a cross flow in a Model A? The Wizzard
I understand the use of relays and will use them. I'm putting a 4-71 on top this 346 flatcad. I'm ditching the stock water pump because it ****s. I will run an electric fan because its great and doesn't directly rob horsepower and is easy to mount in a simple shroud to the back of the radiator. I understand its not so "traditional". I realize the water pump only runs when the engine runs. I also realize that the water pump flows water through the engine and heads while the thermostat is closed, this loop is what I'm having a hard time trying to figure out. Reverse flow because its more efficient to cool from the heads down. I asked about the cross flow not because I feel I could ever make one fit/look very well, but just as a general concept question.
Not being familiar with a flat caddy engine, how does the factory setup work? Are there 2 pumps like a Ford, two outlets like a ford ? Where is the tstat now. understanding the water flow helps a lot when you want to change it. With baffles and bulkheads in the radiator tanks you can have any flow you can think of in the radiator. Tstat should control water out. Inline in the lower hose (block side which out is pressure side) but discharge to upper radiator. Need an internal circulation loop, that's just pressure to suction with only the block in the middle- before the tstat. Most of the factory reverse cooling systems were abandoned. Pontiac tried it it the 50s, Chevy tried it not too long ago I forget who else but it never lasts too long before they change is back.
stock setup: There is one water pump, on the p*** side. There are two water outlets, one on each head. I don't know about stock thermostat, my guess is one in each upper hose.
You would have to turn the stats over and run the water pump into the top hoses as opposed to the regular pickup for a standard application. Make yourself a manifold. Then run your standard pickups to the top tank of your radiator. Reverse flow doesn't car if you run your radiator sidways, on its back or even pulled behind on a trailer. It has everything to do with how it p***es through your engine and not your radiator.
thanks, heres a pic of my exact engine. it has two outlets on the top, from each head. and the water pump(only one) has one inlet. where the thermostat is stock I haven't got a clue.
the stock stats are not in the upper hoses. maybe they are at the radiator, but I have the engine completely dis***embled and they are not in the head.
They won't be in the radiator. Someone has run it without stats. On a lot of the older flatties they used to run a restrictor as opposed to a stat. It was just a washer to make the orifice smaller for the water to p*** though. This was not factory it was a racer trick, some of the older fellas where I was at when I was comming up called it a full flow cooling system.
How does it ****? because it's old and repairable and looks funny? Or have you had the engine running and not had enough water flow? I don't get it. I started out with a pair of electric fans on my blown big block, I went to a mechanical fan years ago to keep the engine cool. Maybe, maybe not. If you can make it work, great....but if it's a big h***le to build it and get it working, then maybe it's not more efficient. How do you control the coolant flow into each side of the engine, if there's only one outlet? Circulating the coolant when the thermostat is closed is usually done with a small byp*** hose or orifice.
Squirrel I think he means that the pump only has one pickup. it no doubt goes to either side of the engine block like a later model valve in head engine. He wpuld have to build a manifold from the two points that the water pump hooks to the block or at least a stub for the hoses if he has an upper radiator with two hose barbs. if his upper radiator obnnly has one hose connection he would want to do something like my olds has on it. I think I got a pic somewhere let me take a look. Look close @ the front of my mill.
Jim, can we just cut to the chase and let me exclaim that I am not running the stock mechanical pump and I will run an electric fan. I've never had a problem with an electric fan ever. I size all my fans appropriately and mount them in a near sealed shroud. The stock water pump is prone to failure(mine is broke now and has all ready been rebuilt), and has a terrible efficiency. back to the original topic, To reverse cool or not is debatable and I'm open for suggestion. I'd like to run reverse but I don't have to. later I'd like to run a set of fancy aluminum heads and I feel reverse cool would be beneficial for this. How easy it is to setup doesn't concern me if a performance advantage is to be had. I welcome challenges. heres a picture of a tank setup. I ***ume the auto application to be similar. The thermostat is located at the radiator in the upper hoses.
http://www.jalopyjournal.com/forum/showthread.php?t=430439 Look here. 38flattie did it. Pictures and all. Didn't get far enough to find out if it worked or not or if he changed the set up.
the big hole in the block is where the water pump sits stock. and would be the "in" for the coolant stock. The top "outputs" come out of each head to the thermostat to the radiator. so with an electric water pump, and reverse cooled... coolant would be drawn out the radiator to the pump. from the pump to a spilt(manifold) to each head, then out the block to a remote thermostat housing, then back to the radiator. This is ***uming the thermostat is open. Should it be closed than I will need a slightly restrictive line that connects the engine side of the thermostat to the heads. By slightly restrictive I mean more restrictive than the radiator so that coolant flows through the rad. when the thermostat is open. does this make sense to anyone else besides myself and is it correct?? also does the coolant in a rad need to flow top to bottom or visa versa? or which ever way works best for plumbing? my fill(cap) for the whole system will be on the top of the rad. I'd have to sit and think about how this plays a role in flow direction.
OK you are going to need to run a manifold for the top and the bottom to make it work. You can build a manifold for the water to come out of the original inlets in the block for the stock water pump location, biuld yourself a little plenum in it and run your T stat right there then up to your top tank. Now run your picup from your bottom tank to the electric pump; I am making the ***umption here that you are running an inline pump. Build a manifold to split it and got to both heads. pretty simple if you are a plumber. I have been around a while and from experience there is little if anything to be gained bu reversing your coolent flow. It has been done in the past but pretty much went the way of the reverse flow flathead, more of an oddity than anything else. The theory is sound but in practical application it was just about a wash. How do you plan on huffing it? Unless you are going to move your exhaust ports you are going to be heating an already heated intake chagre if you run a roots type huffer on top. You can run a potvin type of a setup with better sucess that a standard belt driven compressor on it. You won't get as much boost or at least not have the under/overdriven options but it would work better for you in my opinion. Also a whipple style charger with a blow through setup may also work well for you. just thinking out loud here. I am sure that you alrready have that worked out.
buddy would be the one to talk too Makes sense, you'll need to make sure the hoses to the heads stay full of pressurized water. And very important .... That the flow into the heads thru the block can match what's going to come out when the tstat opens. Everthing is reversed and the upper hoses wouldn't need to be full at all times under as designed operation. water would just burble up when they opened and find itself in the radiator eventually. Clear as mud right? Is also guess that the byp*** loop is built into that hole some how. Might need a work around for that. Radiator flow, if you had a multi core radiator you could have it flow up one, down the next and so on for every core. You also could build it so both hoses were on the bottom with a fill at the top. Cross flow, vertical flow, single p***, multi p***. What ever you need or want if you can spend the cash.
I realized that a minute ago I didn't need a remote t stat housing and I could put the stat in the block. when the system is on byp***, how does an electric pump react? does it cavatate a bit to compensate for the lower flow rate? I think to keep things simple I will run just a regular down flow rad. and feed from the top. This makes bleeding all the air out easiest. I didn't think about multiple cores but I don't think its necessary, although its cool. Buddy is hard to get info out of. He is doing that huge and expensive build so it is understandable. the huffer goes on a manifold from a guy out in Seattle is making. Looks like this...
Instead of a thermostat, why not just have a temp sensor and PWM the electric pump? Although this would probably work better if you left the flow direction as nature intended.
The pump shouldn't cavitate, That will give you problems. The flow rate shouldnt change, but The flows path will change. You (reversed) always want flow from pump to heads to block. Always. Just some times, like when the tstat opens you want the flow to go from the block to the rad thru the core and back to the pump. Constant flow thru block, sometimes thru radiator.
A little hard to do with an electric, they don't have a by p*** like a mechanical pump. I would run a restrictor instead of a T stat if i were going to run an electric pump and have a cinstant flow. That or I would run my pump on a stat-switch and only run it when the t stats were open. Like an electric fan.
You make your own by p*** for the block. A T off of the pressure side, and a T on the suction side before the tstat. hard part is getting it to flow into and around the block. Factory mechanical pump has the same set up, one side of the impeller is pressure the other side is suction. What's it look like inside of that big hole for the water pump.
I've thought about this before. I'm not sure how the pump would like PWM. This also seems the "engineers approach" aka the most complicated solution to a simple problem. I might could make the PW very, very large so the pump runs for say 10 seconds every minute for a degree range, then 30 seconds for every minutes for a higher range, eventually getting to running for a minute for a minute, aka 100%. I was going to neglect to mention this, but I am very good friends with an embedded circuits engineer and I am well versed in simple software/hardware design... so I was going to make my own fan controller with multiple other features that tie into a cluster F of an efi system based around a megasquirt controller. the flow rate has to change( drop) in the byp*** or when the thermostat opens the water won't flow through the rad. Its simple fluids. Path of least resistance. In a BBC or SBC, the byp*** is much smaller(about 1/3 dia, measured by visual inspection) than the full flow line. what happens when the flow is restricted on a electric water pump? If I knew how the pump would react, then setting up a byp*** and a t stat wouldn't be difficult. I see the pump doing two things... cavitating and allowing pressure to rise a little and flow rate to drop a good bit or building a ****load of pressure and bogging the pump down drawing a bunch of amps trying to push the same flow rate as the 1.75" hose @ ~14psi through say a .5" hose. inside the big hole is the number 2 piston. Its straight into the water jackets which are huge.
I might add that I'm going to be running an oil cooler as well to help with cooling. stuck in Austin traffic on an elevated interstate at last year's Round Up really showed me how much cooling these higher horse engines need. I was the only one in our group to not get really hot, stayed about 215.
The flow rate is a constant @ Based on GPM. Velocity may change based upon the size of the p***age. Those laws aren't mine but since the volume will stay the same, the pressure will drop as the fluid p***es thru a constriction. High pressure will flow to low pressure or pressure to suction, with increased speed. You will not build much measurable pressure with the pump because the system should be a continuous loop. Only slightly higher on the pressure vs suction sides. If you build a dead head into the system, well that will be incorrect and probably a source of many problems. T stat closed = exclude radiator from the loop & T stat open = include radiator into the loop. Most OEM byp*** routes incorporate the heater core, also a continuous loop; Out of the head and back into the block and the water pump is responsible for the flow. The pump shouldn't cavitate unless you let it **** air some how. The radiator hose may be 1-1/4" or 2" but the opening in the T stat isn't any where near that size. And that's the size of the p***age.
RAD, I commend your efforts and thinking here, but this is my take on the reverse flow subject. I have done intensive testing over the years on both reverse flow cooling as well as reverse flow oiling systems. I dont personally see a large gain here simply because of the flat head design in and of itself. Now on a typical OHV set up cylinder head, the combustion chamber is within the cylinder head itself as well as the heat generated by the many recipricating parts that also develope heat. Therefor there is more to be gain by transfering that heat downward into the block, rather than upward from the block into the already high temp area of the combustion chamber. I might develop something to take advantage of the " heats " natural ability , for lack of better term, to rise upward. The flathead cylinder head, by design poses a positive opportunity for a engineering type individual to create a more efficient way to cool from the top of the head and opposed to transferring the heat into the block. Keep the flow as stock, and think a little about what I just said. I built my career by thinking outside the box. TR
thats makes sense. I don't think of the system as a closed "hydraulic" unit but more of a very very ****ty quality steam with air voids. One of us is right but I think it doesn't really matter. The goal of plumbing is figured out. I thank you for your efforts and knowledge. My selling point for reverse cooling is the differing materials when I do swap to aluminum heads. currently that is further down the line for me but it could happen before the engine ever runs. its not rocket science and I don't want to beat a dead horse but aluminum conducts heat much better than steel. I feel the heads will naturally want to run cooler than the block but because they are on top and they are of a material that conducts heat better than steel, their temp will want to be higher and I should cool them first. I'm also concerned about keeping the heads as cool as possible to keep the head gasket in place. It's hard to o-ring a flathead for boost and I'm not going to. If I get enough of expansion and movement I have faith I will spit a gasket. I guess I don't see any real cons of reverse cooling but many pros. Am I being ignorant? I have no real world experience. And really it's not much extra work to reverse cool than to traditionally cool since I am running an external pump anyways. On that subject...are there inline universal belt driven pumps? Quick search yielded me nothing.
The Jalopy Journal
