Custom water pump - impeller press fit suggestions

My Triumph consistenly runs warm and I would like it to flow a bit more water. The vanes on the impeller are minisule compared to other
water pumps with similiar displacement engines. In the image the 2.0 / 2.5L engine displacement Triumph water pump with impeller is on the left and the proposed donor impeller (from an Isuzu 2.3L) is on the right.
http://i290.photobucket.com/albums/ll272/aribert/machining/P1010003.jpg
How much press is needed for a cast iron or cast steel impeller? My machinery handbook gives a bit of a overview but no good suggestions. I am thingking about 0.001 to 0.0013. Your thoughts?
In trying to dismantle the two pumps, there is inadequate room to get a bearing seperator between the impeller and the pump housing on either pump. My thought is to remove the pressed on fan mounting flange and then press the shaft (with impeller attached) out of the pump housing. I got the pressed on fan mounting flange off of the Isuzu water pump but did not succeeded in pressing the shaft out of the housing (watching the deflection of the press and the increasing effort on the jack hande, I stopped adding force). I have a "30 Ton" Harbor Freight A-frame press that I suspect is capable of about 20T. Does my proposal in dismantling sound wrong? I can cut the Isuzu pump body apart if needed.
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Jul 27, 9:38 am, snipped-for-privacy@c3net.net wrote:

From your post, I assume you have already done the thermostat thing and looked for a water distribution channel in the engine and/or the radiator that may be plugged. One interesting thing I discovered long ago is that the faster the water flows through a system, the less heat transfer that occurs.
An electric fan in front of the radiator may be a better solution.
Paul
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

So maximum cooling occurs at zero flow? <g>
--
Ned Simmons

Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

No, maximum heat transfer occurs. You then have to get that hot water out to cool it and recycle it back to the engine. Look at the top/ bottom of a radiator. Many small, restricted openings to the cooling tubes.
Been too long since the college physics class.
Paul
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

I can imagine there may be specific regimes where increasing flow reduces heat transfer, but I don't believe that it's true in general.

It's been 35+ years for me since I dealt with Messrs. Reynolds, Prandtl and Nusselt, and 34 years since I forgot it all.
--
Ned Simmons

Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I was really hit with this a few years ago when we installed a 7 stage convection oven at my plant for soldering surface mount printed circuit boards. The last stage of the oven is a cooling stage cooled by chilled water. I bought a 1 ton chiller and hooked it to the oven using 1 inch rubber hoses that match the connections on the chiller. The oven had 1 inch connections, but had reducers to about 1/4 inch. It was a used oven.
Being one for more efficiency, I pulled the reducers and connected the 1 inch hoses. Guess what? No cooling, but the water really ran fast through the unit. When I put the reducing fittings back on the oven connection, a very slow flow of water came out, but lots of cooling air now came out of the unit.
Paul
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I'm not sure in which end of the process (heating the water, or cooling the water) the problem really lies. I'm guessing it's on the cooling-the-water stage.
I know that the hotter you get the water, the faster it will dump its heat in the radiator (the bigger the temp change, the faster the transfer), and more heat will be removed from the system, so it could just be the 'cool the water' stage that's inefficient.
It's not really the flow rate that matters, but the delta-T on each end, which is of course affected by the volume rate of flow.
And of course, the longer the water lingers in the radiator, the more heat will be removed from it.
Dave
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Triumph car or motorcycle? If cycle I didn't know thay had any liquid cooled engines.
Anyway, you might want to ask around some Triumph centric forums on the web before going further. It could be it needs to run warmer than you think to get the best power and emmissions.
Many time on an old truck or old car I was able to get them running cooler, by pressure flushing the radiator, and then back blasting the bugs out from between the fins.
My dad made a pressure flush adaptor that I have since duplicated that does a pretty good job. It has water and air inputs with ball valves on each. Fill with water, then open the air valve, run water until clear and repeat. Duplicate process until it immediately runs clear, and then reverse directiosn and do it again.
If the radiotr splits you needed to have it fixed anyway. It doesn't flow well even after you pressure flush it then it needed to be hot tanked anyway... in either cases its already out of the vehicle. (You can do it in the vehicle, but its really a mess under the hood and you can only do it in one direction. I wouldn't.)
Bob La Londe www.YumaBassMan.com
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I don't want to bust your bubble, but you are barking up the wrong tree. You are saying that your overheating problem is a product of poor design yet thousands of these have been made and sold and in successful service. They don't have your problems. I suggest you find the real problem and fix it. Everything can be improved.......including the Triumph cooling system, but redesigning a cooling system instead of fixing what's wrong is not reasonable. Steve

Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

steve - despite your statement, at least in my experience with a 61 plus 4 (uses a TR-3 engine), oveheating is just part of the experience - I could get the car to run OK at 55 to 60 MPH by adding an oil cooler and some baffling and removing the fan, or I could get it to be OK at 0-45 mph by adding an electric fan in front of the radiator (no room behind, the front suspension/frame member is 1/4 inch behind radiator in the middle) - but I could never get it to operate across the whole range of speeds without putting boiling water in my lap - perhaps part of the problem is the warmer temperatures here than in the UK - and the real TRs had a different radiator design that perhaps was better -
an after market oil cooler in a place that gets good air flow will help with overheating for sure. I agree that a larger water pump probably won't help
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

usually we drill a small hole in the old impeller hub along side of the hardened shaft to remove the hub stress. The impeller will come off then. You may have to cut the body of the Isuzu pump to get the bearing out. Some have plastic rings that lock the bearing into the housing.
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I'd be very careful pressing impellers on and off of shafts. Water pumps usually have mechanical seals consisting of parts made of graphite and ceramic, both of which are very brittle.
Engineman

Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I make water pumps for Lotus cars.
http://i221.photobucket.com/albums/dd236/arborigine/Pictures014.jpg The impeller we use (made the pattern, foundry does the casting) has a .500 in. hole and the shaft is .502. I never re-use an impeller, as we charge too much to risk failure. I set them in with Locktite engineering adhesive, takes heat to get them off. Looking at your unit, I would heat the whole thing to 350 degrees and then push the shaft out of the impeller and casing in one move, letting the case be the stop for the impeller, and ruining the seal. It should all move with less than 1.5 tons force. Diameter is 2.25 in. height .833 dish in back, .22 I can send you one if you think it will work, get dimensions of yours and the ceramic seal assembly. I don't really think adding vanes is going to fix your heating problem, more likely a clogged radiator, silted block, or leaking head gasket is responsible.
--
Stupendous Man,
Defender of Freedom, Advocate of Liberty
  Click to see the full signature.
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Have you looked into electric water pumps? Try here for some good info;
<http://www.daviescraig.com.au/Electric_Water_Pumps-content.aspx
Pete
--
Pete Snell
Department of Physics
  Click to see the full signature.
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Thanks for the replys - though I was not overly interested in revisiting either Thermo 1, or Heat Transfer. Specific thanks to Rick, engineman and Stupendousman for some *very*usefull information. Steve L: I am on a couple of Triumph specific lists but no one has a vehicle configured like mine.
I have owned this car (Spitefire bodied GT6) for the past 19 years, The high coolant temp issues began 14 years ago when I replaced the stock 2.0L 6 cyl with a warmed over 2.5L engine (TR6). Milling 0.135 inch off of the smog motor head could have something to do with the inability to cool things down. I had both the block and head of the 2,5L engine tanked to clean out any rust/silt 14 years ago. I also had the stock 3 row core radiator upgraded to a 5 row core at the same time. Immediately after my first boil over incident during engine brake-in, I added an electric pusher fan forward of the radiator and fabricated a metal rad fan shroud along with a duct forward of the radiator to funnel all air entering the grill into the rad. I have also added an engine oil cooler but since it sits forward of the rad, heat rejected by the cooler passes thru the rad.
I am looking at the potential upgrade of the water pump as step one in the overall cooling system upgrade. THe GT6 is a smaller car than the TR6 and the forntal grill opening and the radiator cross section are both smaller than the TR6 which typically does not have overheating issues.
I got to the point of removing the impeller off of the Isuzu water pump when it suddenly dawned on me that even though both engines (and both water pumps) rotate CW when facing the front of the engine, the vanes are curved in the opposite direction on the Isuzu impeller. THe Isuzu impeller has a very slight reverse curve at the very tip of the vanes but I need to turn down the impeller about 0.100 inch, pretty much negating the reverse curve. THings are on hold until I have a more close look at the Isuzu cooling system to figure out the reason for the impeller curves.
I hope the text wraps - looks miserable in Google groups.
Add pictures here
✖
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.