Like Chris's thread, we to have been looking at radiator efficiency. We would like to take some pressure readings inside the ducts and at a few spots along the body leading into, and out of the radiator. Could someone here with some experience in these matters give some advice on what kind of pressure sensors we should use, the pressure range, mounting advice, where they could be purchased? I've looked at a few of the data acquisition manufacturers, but I don't see anything for this application, would Mcmaster-Carr have anything? Ideally this could plug into a mychron gauge.
The last time I tried to take pressure readings on a race car involved various types of Omega gauges/manometers sitting in my lap. Driving a race car while looking at your crotch was pretty amusing to those spectating, but not much fun for the driver. The net result was absolutely zero data, wasted money, and feelings of dread- stupidity every time I think about the whole matter. Any guidance would be appreciated.
The first question is what are your data logging capabilities? Does your system have some available analog channels that can accept user-defined 0-5V sensors? If so, you should be able to accomplish your goals without too many problems......no need for manometers or anything else in your lap while on track.
You'll can put static pressure taps in walls of the duct, but thats going to give localized information that can be impacted the packing or swirling. Your best bet is to put the smallest pitot tube in the duct you can find in the center of the duct, both in front of an behind the radiator.
You can also do instrument grade anemometers, but they are very fragile instruments and can be quite expensive to get any worth the data they'd collect. They also have more blockage so affect the readings more than a pitot tube.
Tim
__________________
------------------
'Stay Hungry'
JK 1964-1996 #25
For total (stagnation) pressure measurements, the Omega 0-20 psi absolute pressure sensor should work. If you want to measure dynamic pressures, the Omega +/- 1 psi differential sensor should be OK. To use the differential pressure sensor, you will also need to mount a pitot/static tube on the nose of the car, so that you have a valid freestream static pressure reference.
To use the differential pressure sensor, you will also need to mount a pitot/static tube on the nose of the car, so that you have a valid freestream static pressure reference.
I think most pitot tubes are double walled devices such that they have a dynamic and static orifices on the same tube, and thus two ports out the end to plumb into a single differential sensor. Thus you'd measure just the dynamic pressure. That should give you the absolute velocity in the duct. Yes, if you want to compare that to airspeed of the car then you'd need another pitot & sensor plumbed the same in the free air stream but car speed may be arguably good enough. Pressure in a duct is going to be tough either way though as you really do need the pitot aligned with the airstream to not create error (dynamic pressure on the static orifices).
I think most pitot tubes are double walled devices such that they have a dynamic and static orifices on the same tube, and thus two ports out the end to plumb into a single differential sensor. Thus you'd measure just the dynamic pressure. That should give you the absolute velocity in the duct. Yes, if you want to compare that to airspeed of the car then you'd need another pitot & sensor plumbed the same in the free air stream but car speed may be arguably good enough. Pressure in a duct is going to be tough either way though as you really do need the pitot aligned with the airstream to not create error (dynamic pressure on the static orifices).
Pressure diff sensors can work with your application by making sure the static reading is reliable, if all your doing is making sure your duct works.
The static side can be a water bottle as your base, and run the static tubes or tube if your using one, into it.
You can take a rivet, remove the nail and now you have a surface tap in several places to determine boundary layer separation and move the tube from the sensor tube to the several "tapped" places in the duct . If you use more than one sensor, requires less time(more taps are used). You could also build the same behind the radiator and move the tap, just the same
The most important thing in using diff. press. sensor is that you vibration mount the sensor as the diaphragm can vibrate if hard mounted upsetting the readings.
You can do this without a pitot tube, as the concern is not head wind speed but efficiency of the duct. Though the pitot will make it more accurate as you know the head/tail wind speed.
IMHO, the most important area in the duct are the radius's of sides, top and bottom meeting points and of course the obvious shape of the duct.
__________________
George Main
SpeedSense consulting
Last edited by George Main; May 22nd, 2012 at 1:05 AM.
Trying to get pressure measurements which show differences in flow through a varying area duct which has a radiator in between measurement locations has a number of complications. Measuring with a pitot (stagnation or total) pressure probe before and after the radiator gives the loss in total pressure, due mostly to the resistance of the flow through the radiator. One could correlate this pressure drop with velocity, as the loss is probably proportional to the velocity across the radiator squared. One can easily make pitot (stagnation pressure only) tubes out of stainless hypodermic tubing.
Measuring with pitot-static tubes (giving the dynamic pressure by the difference between total and static) gives the velocity. A single pitot-static probe on the centerline would give a measure of the flow rate through the duct. This probably is the basic information you want. Connecting the total pressure and static pressure from a single probe across a differential pressure transducer will give you an analog output proportional to dynamic pressure, and hence velocity squared. One inch of water dynamic pressure corresponds to roughly 70 ft/s. Of course, the flow through the duct is not always uniform.
Measuring wall static pressure differences with wall static taps (little holes in the wall) is more problematic. The static pressure is uniform across a duct if the streamlines through the duct are straight and parallel. If the duct is quite curved, like an elbow, the static pressure varies from the inside wall to the outside wall. The static pressure, even when uniform in the plane across the duct, changes with velocity and loss of total pressure. Thus if you measure static pressure upstream of the radiator where the cross sectional area is 0.8 ft squared and downstream where the area is 1 ft squared, the velocity will be lower downstream, and, if there wasn't a drop in total pressure, the static pressure would be higher. Throw in a drop in total pressure across the radiator and things get a little more confused.
If you just want to know flow rate through the duct, I'd use a pitot-static tube somewhere on the duct centerline where the flow is expected to be well-behaved - no separation.
- Frank Chambers
Last edited by Frank C; May 22nd, 2012 at 8:59 AM.
The origninal poster said that his goal was to measure radiator "efficiency". So perhaps the first question should be......how do you define radiator efficiency, and how do you propose to measure this parameter using pressure sensors? In other words, exactly what do you hope to measure?
Do you want to measure the total pressure drop across the radiator? The velocity change across the radiator? The changes in wall static pressure upstream and downstream of the radiator?
Once you have answered these questions, then you can select the appropriate hardware and sensors for the job, keeping in mind the capabilities of your data acquisition system. I think that selecting and installing the sensors is the easy part of the job.
Thanks everyone for your help. It's amazing how much information can be had here in such a short amount of time. I'm sure this forum has kept many people in the sport because of the help that is available to those in need.
This question brought about a trip into some long suppressed memories, and how you went about things before the internet came along. Maybe some of the more seasoned racers might remember how you tackled a question like this back in the dark ages...
Problem: car that cooled just fine, is starting to overheat. Must be mechanical, take everything apart cooling related, check, flush, clean everything, and put it back together. Hope that just taking everything apart, and putting it back together will cure the problem. (as it often does with race cars) No car stills running hot. Maybe it's some aero thing brought about by the changes you've made to the bodywork? Need more information on aerodynamics: get your Classic Motor Book catalog, and order everything related to automobile aerodynamics. Maybe comprehend 10% of what you just read from all of your new additions to your racing library. Next, go to the local college engineering library, and check out all books, and SAE papers relating to automobile aerodynamics. Maybe comprehend 5% of that material. Armed with a vague idea of what the problem might be, you need sensors to verify if you really understood anything that you just read. Take a trip to the local public library to that magic row of green: the big green books (Thomas Register) This would be the equivalent to a Google search today, for all of you young folks. Write down a bunch of phone numbers, and start calling, hoping to get someone that would be knowledgeable and willing to help, most of the time the person on the other end of the phone realized that you where just some clueless hack, looking for some free information, and a few dollars worth of parts. Most of these phone calls would be made from "your" payphone at work, the well hidden one, that everyone knew was off limits on Monday mornings because you be there all morning ordering parts for the next race week end. This also produced a big spike in your next months phone bill because of all the long distance calls, and a few tense moments with your wife/ girlfriend over the large long distance phone bill. Next you mount all of your new goodies, and pay up for the next open practice day before a race week end. Make sure your data logger is in good shape (drink a few less beers the night before) Hit the track looking for verification of you new found knowledge, and maybe some new found speed...nothing, gauges do nothing, maybe theirs not much pressure differential on the short straights, maybe it just happens at the end of the long straights, again, not a good time to be looking at you crotch. By the end of the day, you've almost wrecked your car numerous times, scared your self s__less numerous times, and learned nothing. Go to plan B, tape yarn, and smear oil all over your car, and have some buddies take pictures as you drive your car down the road. A few more tense moments with the wife/ girlfriend when she finds out what your doing. Take your buddies out for beers as payment for taking part in illegal activities, and get yourself plenty drunk to forget the past few months of wasted effort.
I hope I'm not the only one that used to go through this sort of drill?
Our goal is to measure the pressure drop across the radiator, and if the exit is in a low enough pressure area.
The easiest way to do this is with a single channel differential sensor such as our SEN-10DS, range is +/- 130" of H2O.
Plumb the high tap upstream and the low tap downstream and you get a direct differential measurement without the complications of having to establish a steady reference pressure.
The greater the pressure difference the greater the flow.
__________________ When you need answers AND results... not just data
Thanks, John. The Cosworth/PI notes are interesting, but I don't think the little section on radiator efficiency on p. 42 is going to be very useful to the poster. They do provide rules of thumb, but more information really is needed. The radiator pressure coefficient they define is pretty much a constant once the radiator core is chosen. They state a rule of thumb that the velocity ratio is best when large, as is sort of obvious, but they do not mention that there are preferred ranges for the face velocity at the radiator surface. This also is sort of misleading, as the velocity at the surface of the radiator is expected to be less than the free stream (vehicle) velocity. The radiator duct usually acts as a diffuser. This can be confirmed by the fact that the air speed sensor they offer for radiator flow measurements on p. 36 has a maximum velocity capability of 20 m/s. Their definition of radiator drag also seems pretty simplistic for application to a radiator in a duct as a system.
- Frank Chambers
Last edited by Frank C; May 23rd, 2012 at 9:23 AM.
Thanks everyone for your help. It's amazing how much information can be had here in such a short amount of time. I'm sure this forum has kept many people in the sport because of the help that is available to those in need.
This question brought about a trip into some long suppressed memories, and how you went about things before the internet came along. Maybe some of the more seasoned racers might remember how you tackled a question like this back in the dark ages...
Problem: car that cooled just fine, is starting to overheat. Must be mechanical, take everything apart cooling related, check, flush, clean everything, and put it back together. Hope that just taking everything apart, and putting it back together will cure the problem. (as it often does with race cars) No car stills running hot. Maybe it's some aero thing brought about by the changes you've made to the bodywork? Need more information on aerodynamics: get your Classic Motor Book catalog, and order everything related to automobile aerodynamics. Maybe comprehend 10% of what you just read from all of your new additions to your racing library. Next, go to the local college engineering library, and check out all books, and SAE papers relating to automobile aerodynamics. Maybe comprehend 5% of that material. Armed with a vague idea of what the problem might be, you need sensors to verify if you really understood anything that you just read. Take a trip to the local public library to that magic row of green: the big green books (Thomas Register) This would be the equivalent to a Google search today, for all of you young folks. Write down a bunch of phone numbers, and start calling, hoping to get someone that would be knowledgeable and willing to help, most of the time the person on the other end of the phone realized that you where just some clueless hack, looking for some free information, and a few dollars worth of parts. Most of these phone calls would be made from "your" payphone at work, the well hidden one, that everyone knew was off limits on Monday mornings because you be there all morning ordering parts for the next race week end. This also produced a big spike in your next months phone bill because of all the long distance calls, and a few tense moments with your wife/ girlfriend over the large long distance phone bill. Next you mount all of your new goodies, and pay up for the next open practice day before a race week end. Make sure your data logger is in good shape (drink a few less beers the night before) Hit the track looking for verification of you new found knowledge, and maybe some new found speed...nothing, gauges do nothing, maybe theirs not much pressure differential on the short straights, maybe it just happens at the end of the long straights, again, not a good time to be looking at you crotch. By the end of the day, you've almost wrecked your car numerous times, scared your self s__less numerous times, and learned nothing. Go to plan B, tape yarn, and smear oil all over your car, and have some buddies take pictures as you drive your car down the road. A few more tense moments with the wife/ girlfriend when she finds out what your doing. Take your buddies out for beers as payment for taking part in illegal activities, and get yourself plenty drunk to forget the past few months of wasted effort.
I hope I'm not the only one that used to go through this sort of drill?
Does putting the car back to the original spec, cool the car again? Whatever change you made COULD be causing the problem. Or it could be a mechanical problem that has nothing to do with the changes. Going back can determine this....
Was this one big change or a lot of small ones?
__________________
George Main
SpeedSense consulting
I've had good luck with a simple clear flexible polymer tube with one end in front and one end behind the radiator with a vertical loop on the steering wheel with colored water in the loop. I've measured up to 8" of water column (small diffuser pressure differential from atmospheric pressure) while driving and watching the water level DIFFERENTIAL.
Pretty effective albeit not continuous (keep you eyes on the track!) and you'll get quizical looks from folks not in-the-know!
I use GM 1 bar MAP sensors with a data logger to measure air pressure on aero surfaces. They are easy to calibrate in mbar, very responsive, and reasonably priced.
General Formula Car Discussion
Air presure measurements
