MTFCA Forum

As if I hadn’t put enough time into this motor project already… this Boyd water pump turned up not too long ago, and it looked to be an actual centrifugal pump, and not just an agitator or circulator as many others are. After some deliberation as to how… it got a new stainless steel and aluminum front end. Next step is to drill a hole in a bucket, and see if my new pump actually pumps

Always such a pleasure to see your work.

How long 'til someone tells you "a real T doesn't need a waterpump"

A real T doesn’t need a waterpump.

Kevin,

Lovely job!

Scott (and Michael at 12mins after Scott's comment ,

Beautiful work, as usual, Kevin. With the HP you are going after, you likely need that water pump.

Thanks fellas! This was one of the simpler projects so far… but is the last piece of the puzzle before swapping the Akron-Hed over to the speedster engine. This should allow me to pull the electric water pump off

Kevin,
Super nice machine work. I wish I had a quarter your talent. I don't care for electric water pumps but have found them to be good
applications in some circumstances. Especially for those of us that that desire more from our model T.
Craig.

A real T doesn't need a waterpump . . .
A special T has ports on the drivers side . . . .

I tried to run without a pump for a while… but the rear cylinders would heat up and begin to seize the pistons after only a few minutes at speed (3:1 gears @ 2000 rpm; bout 60mph). I setup the electric pump on a thermostat, and it kicks on at bout 170 deg. At speed, the temp gauge will climb to bout 210, then plummet down to 160 when the pump turns on. Not ideal, but the pump draws more power than the alternator produces, so can’t run it continuous.

The Akron-Hed has really crowded water jackets, and I’m planning on having water flow issues. The exhaust ports are only bout 1” long, so hoping that will help to reduce heat. The next step for this pump is to cut in a fitting to route some water to the rear freeze plug via one of my new Fronty water manifolds

It will certainly pump water. I believe it will pump more volume than the engine will require, which is good. I would use the pump in conjunction with a thermostat (with some bypass) and a pressurized system. Doing that will prevent cavitation and allow for some super-pressurization of the block and head by the pump while reducing the pump's power consumption at higher running speeds. Besides adequate cooling, benefits would include quick, even warm-up and more consistent running temperature and elimination of local hot spots around exhaust ports. Bypassing some coolant via the intake water jacket would be beneficial.

A thermostat is already in use (with a couple 1/8” bypass holes), and will continue to be used after the swap. Not sure how the Berg radiator would respond to pressure… but intend to leave the system open as stock.

I tapped the freeze plug holes in my block years ago, and intend to drill pipe plugs to act as restrictors under the Fronty water pipe. Not completely sure how I’m gonna get water to the Fronty manifold yet… but right now the best plan seems to be abandoning the standard Fronty connection, and routing a pipe around the back of the engine connecting to the rear of the Fronty manifold…

Wonderful work. Thank you for posting….

I don't think the Berg radiator is intended for use on a pressurized system. The core and header plates would probably do OK, but the upper tank probably would not. I think you'd need a custom radiator designed to hold pressure and a neck designed for a pressure/vacuum cap. A pressure system offers a number of significant advantages, especially in a high performance application. Connecting the Berg radiator overflow tube to the bottom of a vented reservoir would probably be helpful. That would keep the radiator topped up and prevent coolant burping out the overflow at high RPM.

Pat, Kevin will find out if your suspicion about the radiator top tank is true, probably not to long after the first start up and run at road speed.

Craig Leach wrote: ↑

Thu Nov 17, 2022 11:47 pm

Kevin,
Super nice machine work. I wish I had a quarter your talent. I don't care for electric water pumps but have found them to be good
applications in some circumstances. Especially for those of us that that desire more from our model T.
Craig.

I echo Craig's statement, :I wish I had one quarter of your talent (and knowledge). Damn, you do such beautiful work, its too pretty to put it to use. Even though Hank Lee has moved on from Model T's, I put your and his work, on the same level of quality, precision, and beauty.

I think the worst that would happen with the Berg radiator vented normally would be for coolant to burp out the overflow at high RPM. The restriction offered by most thermostats, even when fully open, will probably prevent that, other than what might occur due to normal coolant expansion. If not, the impeller could be trimmed to reduce capacity, or, better yet, a downstream restrictor plate could be added, perhaps at the radiator upper inlet. Adding an overflow reservoir as suggested would prevent burping and would not hold any pressure on the radiator. It's been my experience with centrifugal pumps that restricting the discharge side reduces both volume delivered and power demand.

With the current electric water pump and thermostat, the pump is mounted in the same location as this Boyd pump will be, and is rated for 55 gpm flow rate. There is no noticeable water loss, and hasn’t for 10+ years with either the current Berg or the previous Brassworks radiator.

I doubt this Boyd pump will move water at the same rate, and judging by the current temp gauge movements, it doesn’t need to either. If the impeller does need to be cut down, the front plate design will make it an easy job

That's a LOT of water. I'd use a restrictor plate at the radiator inlet, if necessary. That could be easily adjusted as needed.

TXGOAT2 wrote: ↑

Fri Nov 18, 2022 1:50 pm

That's a LOT of water. I'd use a restrictor plate at the radiator inlet, if necessary. That could be easily adjusted as needed.

You wouldnt want to create much water pressure in the block.
The idea of modifying the pump to work less efficiently sounds right.
Kevin is clearly a bright mechanic & talented machinist. Kudos.

TXGOAT2 wrote: ↑

Fri Nov 18, 2022 1:50 pm

I'd use a restrictor plate at the radiator inlet

That’s right where the thermostat is

This pump housing has an increased internal diameter around the impeller, so won’t be anywhere near “positive displacement”. I didn’t make the impeller fit too closely either as I figured a bit of leakage would be needed. We’ll see how my bucket test goes, then I’ll decide if additional effort is necessary

Well… it works!

My 2000 rpm drill pumped about 4 gallons out of the bucket in 20 seconds! With the pulley ratio, that should be about 12 gpm at 1300 rpm
I carefully excluded my testing rig from this pic as it was mostly constructed of duct tape

That sounds about right to me. That should give a capability of at least 8-10 gallons per minute at 1300 even allowing for some discharge head. That's 4 or more changes of the entire capacity of the block and head every minute. I'd think that's plenty. Too high a rate can cause an excess pressure drop on the suction side giving rise to cavitation and steam bubbles.

TXGOAT2 wrote: ↑

Fri Nov 18, 2022 5:20 pm

That sounds about right to me. That should give a capability of at least 8-10 gallons per minute at 1300 even allowing for some discharge head. That's 4 or more changes of the entire capacity of the block and head every minute. I'd think that's plenty. Too high a rate can cause an excess pressure drop on the suction side giving rise to cavitation and steam bubbles.

Can a Model T radiator swallow & digest that much that fast? Is he intending to pressurize the system?

Based on his experience with the electric water pump, the radiator can flow a lot more than that, and do it without developing any significant pressure in the upper tank. Keep in mind that the pump is pulling on the lower radiator outlet and putting the same water back into the upper inlet. Any restriction between the pump discharge and the upper radiator tank inlet will reduce the pumping rate, not pressure up the radiator. Any pressure rise would be limited, and would occur in the water jacket, where it would be an advantage. Bypassing some coolant through the water jacketed intake manifold will provide another path for the water, too. A clogged radiator would be another matter.

Flow is completely regulated by the thermostat located under the water outlet. A centrifugal pump will not produce much pressure, and so will basically “slip” when a pressure threshold is reached. I have not seen my thermostat in years… but most have an orifice about 3/4”-1” in diameter and only opens about 3/16”. My gut tells me that the pump will outperform the thermostat…

All that’s left now is to figure out how to tap into the water pump to supply the Fronty water manifold. I just ordered an assortment of fittings and tube to see what I like…

1960s MOPAR slant 6 engines had a very similar water pump arrangement, and I think that one heater hose connected to the water pump housing with the other connected to the upper water outlet. I've seen one flathead V8 installation where the hot water was taken from one cylinder head near the outlet and returned to the lower radiator hose with a tap arrangement.

Simply beautiful.
Glad you showed this pump, been working on a similar, it was called a 5-1/2" pump.

Duey_C wrote: ↑

Tue Nov 22, 2022 2:28 am

Simply beautiful.
Glad you showed this pump, been working on a similar, it was called a 5-1/2" pump.

Let’s see it Duey! I’m always amazed how ineffective the design of most of the T pumps are, yet I believe that the cooling system is so oversized to function with thermosiphon, that even a gentle current will cool bout any stocker. Will be nice to see other variations of “effective” pumps

Kevin Pharis wrote: ↑

Fri Nov 18, 2022 5:06 pm

Well… it works!

My 2000 rpm drill pumped about 4 gallons out of the bucket in 20 seconds! With the pulley ratio, that should be about 12 gpm at 1300 rpm

608EDDDD-92A2-4830-86D4-7B67B6A11EA3.jpeg

I carefully excluded my testing rig from this pic as it was mostly constructed of duct tape

Are you making a fire pumper!?

Beautiful work!

Kevin Pharis wrote: ↑

Tue Nov 22, 2022 11:29 am

Duey_C wrote: ↑

Tue Nov 22, 2022 2:28 am

Simply beautiful.
Glad you showed this pump, been working on a similar, it was called a 5-1/2" pump.

Let’s see it Duey! I’m always amazed how ineffective the design of most of the T pumps are, yet I believe that the cooling system is so oversized to function with thermosiphon, that even a gentle current will cool bout any stocker. Will be nice to see other variations of “effective” pumps

No sir, sorry, I cannot show it here and it's heavy enough to make a T list a bit if it's on the left... Way OT. 5-1/2" diameter impeller.
What I can say is the impeller you built looks very much like the impeller on this larger OT pump.
I have an idea. I could show some bits over on the OT section. On the morrow.
Oh, the two curl-vaned model T door stop never let me down/boil either as it moves water but that's not for here either.

After several phone calls earlier this week… turns out that a water pump pulley is not available off the shelf. A drawing has been made, and will be ordering tooling next week. This will be one of the easier projects so far… but an unexpected task

Kevin

that's such a simple part, I think your CNC will be insulted.

this is the beauty of a manual lathe: no special tooling, no drawing on the computer...just a sketch of basic dimensions, a sharp tool, and 30 minutes of your time

With all the festivities behind us, I started lookin at the water bypass fitting. I ordered a couple 1/2” pipe “bungs”, the idea was to connect one to the Fronty manifold with a short piece of hose, and weld the other to the water pump. Wasn’t sure if I would be able to get a 5/8” OD hard line bent to fit around the front of the motor, so ordered a flex hose as a backup plan. Didn’t realize it would be bright yellow…!
The bung is stainless steel and was TIG brazed with silicon bronze filler rod. Oh yeah… got bored yesterday and started polishing the water manifold

I’m runnin out of excuses…

I plan to do a somewhat similar project and am considering the idea of putting tubes around the push rods and then running water through the block ports to even out the cooling. This would be in addition to what you have done. Still considering.

Of all the possible complications… the water pump inlet is more than 1/2” lower than the radiator neck! There is enough slop in the mounting bolts to get it within 1/4”… but the shaft becomes waaaay out of parallel with the crank pulley, and I fear never ending belt problems.

So am considering either cutting the pump housing back a few inches to allow the hose room to align, or wedge cutting the casting to angle the pipe into alignment with the radiator…

I'd be inclined to modify the pump inlet to the correct angle. If you can get set up so that the pump will develop some overpressure in the block and head, it will be a good thing. It won't develop enough pressure to do any harm, and the more pressure the pump discharge sees, the less HP it will draw. Overpressure in the block and head will act to limit steam bubbles forming around the exhaust valve area. You do not want low pressure on the suction side of the pump. It may cause bubbles to form in the hot coolant. Throttling the coolant flow coming out of the engine via a thermostat or other restrictor will act to hold some overpressure on the block and prevent excess low pressure, or "suction", at the pump inlet. Ideally, you'd have a radiator capable of holding at least some pressure.

If you find there is too much flow - It is standard practice on industrial centrifugal pumps to trim the impeller diameter to obtain the desired pressure/flow required. It would be easy for you to make a spare impeller to experiment with. Centrifugal pumps are also pretty tolerant of restrictions in the discharge piping system. Somewhat counter intuitively, for any given centrifugal pump at a fixed speed it requires less horsepower to deliver a reduced flow at a higher pressure than a larger flow a a lower pressure.

As others have said multiple times, wish I had a small portion of your capabilities and talent. Well done

Thanks for the input fellas, I am far from a pump design engineer and based the impeller off the impeller that was included with the pump. The cooling system will remain basically stock, and intend to throttle back the pump if needed. Increasing the pulley diameter and cutting the impeller down are both options…

Slowing the pump down, if needed, would offer the opportunity to speed it up again if the need arose. Slower pump speed would also reduce any tendency toward hot fluid cavitation insiude the pump, while reducing wear rate. Throttling the pump at the engine water outlet, if needed, is probably the best way to regulate the flow rate.

Kevin Pharis wrote: ↑

Fri Nov 18, 2022 5:06 pm

Well… it works!

My 2000 rpm drill pumped about 4 gallons out of the bucket in 20 seconds! With the pulley ratio, that should be about 12 gpm at 1300 rpm

608EDDDD-92A2-4830-86D4-7B67B6A11EA3.jpeg

I carefully excluded my testing rig from this pic as it was mostly constructed of duct tape

Red Green would approve.

I like to think he would have been proud🥲

At least women will find ya handy, eh?

A little cutting, grinding, some heat, and a bit of silicon bronze TIG brazing rod…

Waaaaay better!

You are an artist. While I would never consider installing a water pump on my T, I might make an exception in your case cause it is just so purty. Jim Patrick