PROLOGUE:


I completed my testing of the intercoolers several weeks ago and did not publish the results, because I could not interpret the data in a meaningful way. Data is data and if it disagrees with your belief, this does not change the readings. Finally figured out the cause of the skewed results and now see that the results weren’t really bad. The test set up had a fundamental issue, which I had anticipated – and got bit by.

On to the real testing. To do bench testing of the intercoolers thermal capacity I ran hot water from a water heater thru the waterside of the IC. The gas water heater was 50-gallon capacity and with the thermostat, the water was held at a relatively constant temperature. The water went from the top of the water heater thru the IC then to the 90 series pump and back to the bottom of the heater to conserve water and maintain the water temperature. I reduced the water pressure in the heater so that it was about 5-6 psig pressure as the 60 psig house pressure would have damaged things and been blowing hoses off fittings, not to mention damage the IC.

I put the pump on the outlet side of the IC so that it would be sucking the cooler water and not cause cavitation or boiling at the input to the pump blades due to the low pressure. 120-degree water will boil very easily when its pressure is reduced, it does not boil in a pot because the atmospheric pressure holds the water surface together. 32-degree water will turn to ice in space because the vacuum causes it to boil so rapidly…… See chart at the end, 2 psi causes 120 water to break surface tension and boil..cavitate.

http://docs.engineeringtoolbox.com/d...emperature.pdf

PHOTOS at bottom, H/E chart also provided

Test setup video:


The included photos show the charts I prepared, and because this is the third or fourth time I did them I got carried away and neglected to write the FLOW GPM values on the photo that says” 5.5 Static psig”. The correct values should be are 2.0, 3.0,and 3.9 GPM ( almost 4.0) top to the bottom. The reason multiple tests were run, was that the setup had cavitation issues at first and visible bubbles were present at the flow gauge. I increased the pressure from 0.0 to about 5-6 psig to try and eliminate the boiling or cavitation in the pump. Think of the picture of an outboard motor prop in water. That is cold water and about 2 feet under the surface, about one more psig above the atmospheric value ( 14.7 psig absolute ).

The readings worked out to give the number of BTU’s expelled for each water flow rating. The BTU value, is the temperature change in degrees x (times) the number of gallons per minute x times the weight of water per pound (8). A 1-degree drop in water temperature with 2 gallons per minute would be 1 x 2 x 8 = 16 btus per minute. The first test of the 90 series pump on the stock IC was 6.6 degree drop with 2 gallons per minute x 8 pounds /gallon = 105.6 btu’s per minute. In an hour the result is multiplied by 60 to give about 6336 BTU’S per hour. Like I said the air conditioner is between ½ to 1 ton of cooling which is 6000 to 12000 BTU’s per hour. My test is representative of the operation of the IC in the car but the airflow is lower, and I anticipate the IC is doing even more cooling than this 6xxx BTU test value.




Results:

The stock ( series ) IC produced water temperature drops of 7.1, 5.0 and 3.9 degrees for a flow rate of 2, 3, and 3.9 ( 4.0) gallons per minute. I calculated the Btu rates, and got surprising results in that the 2 GPM provided 6336 BTU/ hour and 3.9GPH a greater water flow gave me 4956 BTU/ hour heat rejection rate. This all seemed wrong to me and was an issue to understand. How did better water flow reject LESS heat to the air stream - - - back in a moment.

My pet parallel flow IC seemed to operate correctly in that the heat rejected followed the water flow and did better with higher flows. When I crunched the numbers it worked out that the 2 GPM flow rejected 7104 BTU/ hour, 3GPM rejected 7200 BTU/ hour and at 4 GPM we were able to reject 7488 BTU/ hour a definite gain over the stock design. I will say that the parallel IC is a definite better design and will reject more heat and flow more water.

The video will show the test bed and some detail about the operation of my test stand. I’m glad to have gotten the results and that I am basically done with all this fiddling. Phew that was 20 hours to end up getting confused.


EPILOGUE:

The stock IC rejected less heat than the parallel IC design, but the heat it got rid of seemed to Fall as the water flow went up????? I now have gotten my arms around this counter intuitive result.

The water was hot, the pump was sucking the water thru the IC and then the flow went thru the flow gauge. I am convinced that the water flow was turbulent and also restricted by the three-pass design. This lead to voids or cavitation within the IC with increasing pump speed and flow. This means that the readings were lower due to the disruptive water flow…………….. Remember the static water pressure was about 6 PSIG!!!

In the car the water is pressurized by the radiator running about 15-19 PSIG of pressure and cavitation of the IC would not be an issue. With the separated IC water tanks, there is basically no pressure ( plus less heat in the water ) and cavitation may occur REDUCING the heat rejected. If cavitation occurs in the pump and or the IC body we loose benefit by some margin. The air heating the IC may be well over 250 degrees, on a stock SRT.

For the first time I’m seeing this and will devote thought to the idea of running a pressurized super cooler, to insure this can’t happen. I would have no way of knowing if the pump in my car is cavitating, remember the outboard motor video image.

WOW I learned more than I had expected, just by doing this “”simple”” measurment. Now I will have to see if there is a way to incorperate the better pump ( 90 ) in my car and monitor the water conditions in the IC.

PLEASE REMEMBER this has nothing to do with my using the 90 series pump in the test set up. A variable voltage motor speed controller controlled the speed and thus it was running slower than a stock pump. This is due to the larger impeller flowing the same water as a stocker or 30 series pump –around my self-imposed limit of 4 GPM in the testing program. Had the 90 series pump been running at high speed the results would not be similar to the real car conditions, and thus the readings would be out to lunch.

So parallel path IC is good, testing with higher static water pressure would provide better results on both of the IC tests. IF I do more testing I will be testing with 7 to 10 PSIG static pressure to avoid errors due to localized boiling. Gotta check my system in the car as soon as it gets warm and relay the findings back to all u all.

Wdy, enjoy

 

Woody,
Thanks for posting this result page. I for 1 found this very interesting and I am running parallel IC's and just thought it a better way to cool the charge....now I feel it was the correct way to go with your testing results. I also have a super cooler to work with the new srt6 and will check the differences between the 2 when the spring comes around and I have the surgery done on the other car.
Robert

 

Robert, I'm probably confused here, but are you running parallel H/E's, or you've modified your CAC(charge air cooler)as Woody did to be a parallel flow CAC?

Bob

 

Yes running parrallel also have another with the super cooler I will be experimenting with. I will get the new IC from Rudy to use on the new srt when it is ready to go.....$1000 is a little bit too much right now but I will be ready come spring. Still fixing that 1 up so the time and wallet are going to those repairs first.

 

Woody,
After reading your results, I'm now wondering how much small amount of pressure builds up w/ my separated system w/ the pin hole vent on the cap of the power steering tank that I'm using as a reservoir. Even w/ the tiny vent hole in the cap, doesn't the pressure build up a few pounds when the coolant comes up to max temp. after a hot lap? Otherwise, I wonder if I could find a low pressure cap(5 or 6psig)that would fit the reservoir, and if the little tank is strong enough to hold the added pressure. I'm thinking about teeing in a low pressure gauge(1-10psig), and popping the hood after the lap to have a look see. Because the car is never subjected to a freeze, I'm running distilled water, water wetter, and a little wtr. pump lube per your recommendation; so I'm concerned about this "localized boiling" that you speak of. Perhaps I need to go back to the antifreeze mixture for better protection against this?

 

Unless I'm missing something.....I'm not sure I agree here.....

For example: simplifying from your numbers (getting rid of the 3GPM)

GPM....dT water parallel CAC.....dT water series CAC
..2.................6.6F.........................7 .1F..........
..4.................2.2F.........................3 .9F..........

heat = dT x Flow

If nothing on earth changed.....if you double the water flow, the dT (temp diff) should halve.
However we expect heat rejection to improve with increasing flow....due to the much touted "scouring effect"......
So you've doubled the flow .......the dT should go down by less than half.
wow - if it stayed at 6.6 and 7.1 ....you'd have doubled the heat loss by the water.....'cause you doubled the flow...and heat = dT x Flow

This is the case above with the series...but not with the parallel.
?????????????
Or am I getting something skewif?
(not sure you use that term elsewhere - it means a$$ about!)

 

Dinasrt. Hello; Well I gotta tell you that I am concerned by my finding and have even entertained putting back my test setup to do another test. I really have not gotten over the disruption it did to the house last time yet.

This test would be to take a stock IC and do the same test, the variable function would be that I would run like 3 GPM thru the IC and hold that constant, while varying the static water pressure. My thinking is that the low pressure was causing cavitation within the stock IC making voids that reduced the cooling effect. If this is so as the water pressure increased above some critical pressure, I would see heat rejection increase up to that point. To refresh your memory, as the water flow in the stock IC increased - sucking water out of it harder, the heat rejection fell ---not increased. The lower restriction parallel path IC did not exhibit this anomoly ( sp.)

My super cooler design had a pressure cap on it and I will have to pop the grill to read its value, though it might not build much pressure due to the very large size H/E. Like you I am planning on adding a water pressure gauge to my separate IC water circuit just to monitor the pressure. This is exciting to me as I may have found a deficiency in the separated IC water circuit, leading to less than optimum IC operation due to low static water pressure - something that I at first considered to be a good thing.

Billy22bob, you can see what I am talking about, the water flow increased and I fear so much that at the lower pressure of a separated IC water circuit, that the turbulent flow exceeded the waters ability to be a fluid. It was vaporizing due to its heat and low pressure from restriction in the core and boiling - cavitating which reduced the wetted core surface and actually decreased the heat that was rejected from the IC. Valuable infromation, but totally unexpected by me...........

If we get some Va snows Ill be trapped inside and forced to rerun this test..........Enjoy, WDY.........Durned hard to type with my siamese on the keyboard and my wrists, hee hee

 

Just plug the little hole in the cap and see if the iat goes down.

 

Yes, that would be an easy and practical test, and I'll be doing that. Woody has me curious now what the pressure would build up to after plugging it; as that #dictates what boil protection we get w/ just distilled water. Thus, I'll be installing a low pressure gauge so I can test/monitor what the pressure does from startup to heatsoak. This will happen when the autox season starts up in a couple of months. Woody, thanks for sharing all the great testing your doing to improve our weak links in the IC system. Taking notes here.

 

Although it would be nice to evaluate the pump ....maybe not ....
Given your concerns, think about simply putting the hot water service line into the CAC via a throttle valve and have your flow meter, tempIN and 0-10psi pressure gauge up stream as well (at the same level/height) and then
have a tempOUT and the outlet of the CAC to atmosphere ie: 0psig (bucket?).

Its important any static elevations are taken into account and hoses are >1".
500mm = 0.8psig and a 1" hose with 5gpm is still a low enuf velocity to prevent any measureable dP.

Feeding/capturing the waste water and feeding it back to the loop (if you must) needs to be done after this (use now spare johnson pump?)
That way you can effectively measure your dP and dT across the CAC without fear or favour.

I think your numbers may have provided some valuable info just the same - will post shortly

 

So, from what you are saying, maybe having the IC separator from NW may not be a good thing? I will be keeping an eye on this thread, very interesting. We all appreciate your work Woody, hope to meet you at Carlisle.

 

I dont want to get the horse behind the cart, I firmly believe that the separated IC water circuit is a step in the right direction and a GOOD thing.

The findings that I have been reviewing, indicate that there may be a deficiency in the approach due to low water static pressure. This means you may be able to do more with a little static water pressure. The stock car IC had flow about 1.5 to 2 GPM and the upgrades 30 series pump pushes that up. Better H/E designs ( supercooler) increase the flow a bit more, possibly enough to cause localized boiling or cavitation. We may be able to gain lower water temperature by doing mot so much. +++++++

My hot water studies were run at a static pressure of about 6 PSI- - - ever wonder why?? My water flow meter is a clear plastic tube and without this pressure the flow meter had bubbles. THese were not bubbles of cavitation because it was at the output of the pump, on the high pressure side of the pump. The bubbles were the the gasses in the water that were boiled out in passage thru the IC and pump. These are the gasses that boiling drives out of tap water. These dissolved gasses are significant, its how fish live - breathing the oxygen in the water. If you had fish and boiled the water to kill bacteria, you know that you have to let the water sit before adding it to the fish tank so it can take up oxygen.

Nifty, Woody


Carlisle..........thats outside????????????????Oh yeah that is when the sky gets warm...... not now

 

Yes, that's when the sky gets warm, can't come soon enough. This water flow, cavitation stuff is unfamiliar territory for me, so it may take a bit of research and a little pondering about this but I will get up to speed, oh, and reading your excellent dissertations. Thanks.

 

Not that this info may be useful, but I have a sealed tank setup for my IC system. Even when running the crap out of it in southern Georgia have I never had pressure built up when I removed the cap. Also the water would get hot, but where I could stick my fingers in it and it's not uncomfortably warm... how's THAT for a technical analysis? lol

It will probably be another month before my car is back together, but I guess that water temp sensor I put into the IC coolant reservoir might prove useful.

FYI, my setup:
Stock HE, Ford Lightning IC pump, stock CAC, and 1-gal aluminum tank w/sealing cap. I also have a 2-pos switch that allows me to switch the pump on, with the car off, for cycling ice water at the track.

 

When did Geilenkirchen, Germany get moved to southern Georgia ?
And which Georgia are we talking about ?

 

Sorry, guess that could be misconstrued as the country of Georgia...

However, I lived in Macon, GA for 7-years before moving to Germany.

 

got it. Thanks.

 

That is useful, but discouraging for me because I may have just wasted money on the little 0-15psi pressure gauge I just ordered from Summit. But I'll report results when I get it done. I would like to see 6psig(per Woody's observation) at full temperature.

 

Cavitation bubbles are always on the output side of the pump. The bubbles are the result of boiling at low water pressure. As you go up a mountain the temperature to boil water drops, go high enough and water would boil at very low temperature. The bubbles of cavitation are cold steam and gases in the water. When these bubbles implode on the surface of the impeller say, they do severe harm due to the forces released.
To solve cavitation impellers need to be larger and rotate slower and be designed accurately. That's what I have read anyway.

 

Would these cavitation bubbles be the same,lessened,or non-existent with antifreeze mix vs. straight water?