SATCFI
Frequent Flyer
Joined: Dec/10/2009
Location: San Antonio, TX
Online Status: 
Posts: 210
 Skybolt
|
   Topic: Dual Oil Coolers
Posted: Jul/17/2010 at 17:39 |
For those of you who have dual oil coolers, are they connected in series or parallel?
thanks and best regards,
Brad
PS: You don't have to have two oil coolers to comment :)
|
|
Best regards,
Brad
|
|
|
|
|
|
darylat8750
Moderator
Joined: Jun/20/2008
Location: laramie, wy, usa
Online Status:
Posts: 671
skybolt
|
Posted: Jul/17/2010 at 18:22 |
|
If they are not in series the oil would go on the path of least resistance. It could all go through one and the other not do any good at all.
|
|
Daryl in WY
|
|
|
|
Neil
Frequent Flyer
Tech' Advisor
Joined: Nov/10/2006
Location: Monroe, LA
Online Status:
Posts: 1288
Acro Sport II
 View My Gallery
|
Posted: Jul/17/2010 at 19:20 |
|
In another thread I mentioned that I thought it may be possible to air lock an oil cooler by mounting it in such a way that it would trap air that could not be purged out because both inlet and outlet are at the bottom. I have often thought this may be the reason a second oil cooler was installed since the air lock would make the first cooler terribly inefficient.
A similar condition exist in the liquid cooling system on many of the newer cars. My daughters Camaro has a port on top of the thermostat housing that must be opened to release trapped air (burp the system) or it will run hot.
Edited by Neil - Jul/17/2010 at 19:21
|
|
Born with nothing, still have most of it.
|
|
|
|
rbriege
Frequent Flyer
Joined: Aug/22/2008
Location: Calgary, Alberta, Canada
Online Status:
Posts: 84
Pitts S2S, Eagle, Acrosport 2
|
Posted: Jul/17/2010 at 22:05 |
Daryl, I think that for the flow to all go through one cooler in a parallel arrangement, the resistance to flow would have to be severe enough to be called plugged. The flow through the coolers could be modelled similarly to electronic resisters in parallel, with the current or flow in each being inversely proportional to the resistance to flow.
You are correct, if one cooler was nearly completely plugged, all the flow would go through the other cooler. But then, what would happen in the same situation, one cooler was completely plugged, but the coolers were configured in series?
I would have guessed that the coolers should be configured in parallel .....but, I must be wrong as Walter Extra configures the Extra 300's with two Stewart Warner SW 8406 R's, mounted in series.
Also for those interested, Kitplanes had an article on oil coolers back in 2009 see link below.
http://www.velocityxl.com/wiki/images/1/1a/Cool_Runnings_0609.pdf
~ Rob
Edited by rbriege - Jul/17/2010 at 22:28
|
|
|
|
FastEddie
Established Member
Joined: Dec/20/2008
Location: NW 'burbs, Chicago, IL
Online Status:
Posts: 56
Pitts Model 12
|
Posted: Jul/18/2010 at 09:14 |
Here's my $0.02:
2 oil coolers in parallel may not share the flow, in fact, I think that one will almost always get more flow than the other.
I think that the better analogy would be putting 2 transistors in parallel, not two resistors. When you put two transistors in parallel and turn them both fully on (low resistance) you can get a 'thermal runaway' situation. One transistor (oil cooler) has a little less resistance than the other so more current (oil) flows through it. This makes this transistor (oil cooler) run hotter than the other. As this transistor (oil cooler) gets hotter, its resistance (viscosity of oil in cooler) goes down, which makes even more current (oil) flow through the hotter element. Eventually, almost all the current (oil) flows through only one transistor (oil cooler).
IMHO, the main reason to run 2 oil coolers is because you cannot find or fit 1 larger one. You want all the oil to flow through as much cooling surface area as possible. The best way to accomplish this is to put them in SERIES. If you are worried about the oil cooler 'plugging up', use an over-pressure relief valve to route oil around the plugged up area. This is what is done to protect against a plugged up oil filter.
FYI, I have not 'field-tested' this analogy in an oil cooler application.
Take Care,
Ed
P.S. For you electronics people, I am talking about BJTs, not MOSFETs. MOSFETs have a positive temperature coefficient (like resistors) and will current-share very well.
Edited by FastEddie - Jul/19/2010 at 10:39
|
|
"...the reason flyers fly, whether they know it or not, is the aesthetic appeal of flying." Amelia Earhart
|
|
|
|
darylat8750
Moderator
Joined: Jun/20/2008
Location: laramie, wy, usa
Online Status:
Posts: 671
skybolt
|
Posted: Jul/18/2010 at 10:12 |
Originally posted by FastEddieHere's my $0.02:
2 oil coolers in parallel may not share the flow, in fact, I think that one will almost always get more flow than the other.
I think that the better analogy would be putting 2 transistors in parallel, not two resistors. When you put two transistors in parallel and turn them both fully on (low resistance) you can get a 'thermal runaway' situation. One transistor (oil cooler) has a little less resistance than the other so more current (oil) flows through it. This makes this transistor (oil cooler) run hotter than the other. As this transistor (oil cooler) gets hotter, its resistance (viscosity of oil in cooler) goes down, which makes even more current (oil) flow through the hotter element. Eventually, almost all the current (oil) flows through only one transistor (oil cooler).
IMHO, the main reason to run 2 oil coolers is because you cannot find or fit 1 larger one. You want all the oil to flow through as much cooling surface area as possible. The best way to accomplish this is to put them in parallel. If you are worried about the oil cooler 'plugging up', use an over-pressure relief valve to route oil around the plugged up area. This is what is done to protect against a plugged up oil filter.
FYI, I have not 'field-tested' this analogy in an oil cooler application.
Take Care,
Ed
P.S. For you electronics people, I am talking about BJTs, not MOSFETs. MOSFETs have a positive temperature coefficient (like resistors) and will current-share very well.
I think you meant to say "series"?
I don't know if the electrical analagy is reasonable or not. But I do believe that it would take very little difference in resistance to flow to create a large difference to volume of flow.
Edited by darylat8750 - Jul/18/2010 at 10:14
|
|
Daryl in WY
|
|
|
|
pigpenracing
Frequent Flyer
Joined: Dec/17/2009
Location: Brenham Texas
Online Status:
Posts: 148
Skybolt
|
Posted: Jul/18/2010 at 10:18 |
|
Remember you don't want it to run to cool. One cooler should be fine. I noticed my front baffling was blowing back. I got it tucked forward and stuck it down with silicon and it is 30 degrees cooler now. Mine runs between 195 and 215 which I think is perfect after research. A little higher temp will cook off any moisture in the engine. Lycoming says they could run 300 all day if the oil wouldn't break down.
|
|
|
|
darylat8750
Moderator
Joined: Jun/20/2008
Location: laramie, wy, usa
Online Status:
Posts: 671
skybolt
|
Posted: Jul/18/2010 at 11:20 |
|
It's heck getting old. I finally remember why I have an opinion on this. My wife says I seldom have an actual reason for my opinion. ;-(
A well known builder of long distance cruising boats built himself a long distance power cruiser. In normal/ cooler water temps from New Zealand through the Pacific to Alaska and down the west coast of US he had no problems. When getting to the tropics on the west coast of Mex. he started to have hyd. oil temp issues. He checked with the manufacturer of the hyd system and it turned out that the two heat exchangers were plumbed parrallel instead of series. He changed them to series and cured the problem.
I've worked my brain real hard to remember this. I'm going to rest now.
|
|
Daryl in WY
|
|
|
|
rbriege
Frequent Flyer
Joined: Aug/22/2008
Location: Calgary, Alberta, Canada
Online Status:
Posts: 84
Pitts S2S, Eagle, Acrosport 2
|
Posted: Jul/19/2010 at 09:53 |
Perhaps I mistakenly thought that pressure drop due to laminar flow through systems of pipes, hoses, and other equipment was related to a number of different factors such as friction factor, size (dia. & length), fluid density, viscosity, etc.
For any given system with a given fluid, most of these factors are constants ( the friction factor will change some with the Re#) and the pressure drop changes as the flow rate varies. I think that the pressure loss changes as a squared function of the flow rate.
Thus one would expect that when two similar coolers were configured in parallel and one cooler was a little fouled or had slightly more resistance to flow than the other cooler, the increased flow through the cleaner cooler would quickly drive up the pressure drop across that cooler (squared function of the increased flow rate).
I had expected that this function would assist in balancing the flow to the coolers and not send all the flow down the path of least resistance.
|
|
|
|
FastEddie
Established Member
Joined: Dec/20/2008
Location: NW 'burbs, Chicago, IL
Online Status:
Posts: 56
Pitts Model 12
|
Posted: Jul/19/2010 at 10:40 |
Oops!! Yes, I meant to say series. I have edited my post to correct this.
Thanks for catching this.
|
|
"...the reason flyers fly, whether they know it or not, is the aesthetic appeal of flying." Amelia Earhart
|
|
|
|
darylat8750
Moderator
Joined: Jun/20/2008
Location: laramie, wy, usa
Online Status:
Posts: 671
skybolt
|
Posted: Jul/19/2010 at 12:39 |
|
"I had expected that this function would assist in
balancing the flow to the coolers and not send all the flow down the
path of least resistance."
I think on a theoretical level you may be correct. That would have to assume that the coolers were identical and had the same number and type of fittings ect. Also once the flow was less in one cooler the oil might get more viscus and further slow the flow. With a single weight oil that effect may be significant and get into a cause and effect loop. Or not. We should find an engineering student and talk him into doing the experiment and writing a paper on this. Anybody in college or have a son, nephew or friend that might like a project like this? I think Brads coolers are different brands, different mounting positions ect. In a perfect world one bigger/better/more efficient unit instead of two would be lighter and less cluttered in the cowling.
In the case of the boat, both inter-coolers or heat exchangers were identical and the pluming was very close to equal. Yet run in parallel it was not satisfactory and when changed to series it worked well.
|
|
Daryl in WY
|
|
|
|
cwilliamrose
Frequent Flyer
Joined: Nov/10/2009
Location: Florida
Online Status:
Posts: 169
Pitts
|
Posted: Jul/19/2010 at 13:58 |
|
I can think of only one reason mounting the inter-coolers in series would solve an overheating problem. If there were trapped gases in the coolers maybe the increased flow rate would clear those gases and expose more of the interior to the coolant. Other than that, it makes no sense to me.
The old adage "path of least resistance" only applies to things with intelligence (like people). Fluids, electrons, etc flow in proportion to the resistance. An example is a bucket with two holes in the bottom. One hole is 1/2" in diameter and the other is 1/8". Water will leak from both holes, not just the larger one. The flow will be proportional to the area of each hole (which is a squared function)...........Bill
|
|
|
|
darylat8750
Moderator
Joined: Jun/20/2008
Location: laramie, wy, usa
Online Status:
Posts: 671
skybolt
|
Posted: Jul/19/2010 at 21:17 |
|
"In the case of the boat, both inter-coolers or
heat exchangers were identical and the pluming was very close to equal.
Yet run in parallel it was not satisfactory and when changed to series
it worked well."
I should have made clear that the parallel operation worked fine until the ocean water temp. (the cooling medium) got well above what most would consider normal. At some point the hyd oil temp. was running higher than desired and the change to series improved the situation enough to be back in the "green". It is not that the parallel operation didn't work at all, just that the series ops were at least somewhat more efficient. The parallel operation had thousands of sea miles from NZ through the tropics to Hawaii, Alaska down the West coast of USA (I think more than 1500 hours) before getting into the warmer water.
Edited by darylat8750 - Jul/19/2010 at 21:22
|
|
Daryl in WY
|
|
|
|
cwilliamrose
Frequent Flyer
Joined: Nov/10/2009
Location: Florida
Online Status:
Posts: 169
Pitts
|
Posted: Jul/20/2010 at 04:59 |
|
That kinda' fits with my theory about the heat exchangers running at partial efficiency. If they were only 80% full they might work OK until the temperature delta got too skinny. Warm sea temps would do that. Perhaps if the outlets were at the top and all the air could be purged even at modest flow rates, it would have worked as a parallel setup. As I said, I have no explanation for the difference in performance other than not having 100% of the heat exchanger's area working..........Bill
|
|
|
|
