What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

SW 8432R vs 10599R

DanH

Legacy Member
Mentor
Got a curiosity question for the group.

I just ordered an oil cooler for my RV8. I selected a 10599R after studying the Stewart Warner performance charts:

http://www.oilcoolers.com/LCHX Specifications.pdf

Oil circulation for an angle valve engine is between 7 and 9 GPM. If I assume 7.3 lbs per gallon, that's roughly 50 to 65 lbs of oil per minute.

OK, look at the "Oil Side Pressure Drop" chart for a 8432R. They don't even rate the 8432R past 55 lbs per minute....and at 55 lbs you have an oil pressure drop of 15.5 psi. Compare the same chart for a 10599R. At 60 lbs per minute oil pressure drop is only about 5 psi.

The 8432R is a dual pass cooler. You would expect additional resistance to flow compared to a single pass 10599R, but 10 psi? It takes a lot of power to pump oil.....and it looks a bit short on flow capacity anyway.

Both coolers have the same air side pressure drop. Assume 5 inches water across the baffles and you have 27 lbs of air per minute.

Now go to "Calculated Heat Rejection". At 55 lbs/min oil flow (the maximum) and 27 lbs of air you can transfer about 400 BTU/min with an 8432R. At 60 lbs/min oil flow the 10599R will do 375 BTU/min. The difference is only a little over 6%, and 55 vs 60 oil flow rate is probably a fair comparison given the reduced oil flow resistance of the 10599R.

I found the charts surprising, and thus my question.

Why is the 8432R so popular? Am I missing something?
 
Not much help but a data point

Why is the 8432R so popular? Am I missing something?

I installed an 8432R after talking to POC at their Oshkosh booth. I was concerned with oil temp since I am running 9:1 pistons and fly on 100+ degree days during the summer. I did view the charts you referenced before hand. They had recommended the 8432R and IIRC it was a bit smaller. I have a non-standard installation with an A.J. Judy inspired design and it works fine.
 
8432 R

Hi Dan,

I'm running the 8432R on my IO360 here in AZ. I was running a 1" standoff box and was running 175-190 degrees on normal days under 90. On hotter days I was getting 190-200. (This is at low altitude)

A few months ago I reworked my rear baffles and did not reinstall the stand off box. Now I'm running 205 to 210 on 100+ days at 2900 ft. At altitude the temps drop to 180.

I'm not sure if you need anymore capacity or performance for the 390. I'm hoping you figure that out before I mount my 390:)

Can't comment on the other one but have been doing fine with the 8432R.

How about the Setrab (sp?) racing units. They look like they would work great and are definitely set up to handle much more HP and displacement that we have.

Keep us informed.
 
Got a curiosity question for the group.

..........I found the charts surprising, and thus my question.

Why is the 8432R so popular? Am I missing something?

It works and the size is right. But the charts would indicate the 10599R will work also and does not cost quite as much. I'm not sure the pressure drop is a big deal. The Lycoming oil pump is a monster.

Size is important. Anything much larger than 5x6 becomes a challenge installing it. The 10599 is a half inch longer than the 8432. I tried to use a left over 22 vane NDM cooler from the Subby but simply could not find a place to install it. It was about 5x8.

So far the 8432R is over kill. The max oil temp I have seen is 180. I do believe a 4" air duct to the fire wall mount is the key. It provides almost twice the flow of air as the 3" duct.
 
I don't think the cooler sees 100% oil flow ...

Please confirm this but as I understand the oil cooler only sees a proportion of the oil flow, not 100% as you have used in the calcs. The balance between bypass and the cooler oil flow is determined by the vernatherm, and the respective oil flow will be a function of the vernatherm position, the pressure drop across the bypass and the presure drop across the cooler.

The graph provides the pressure drop across the cooler. The actual flow rate through the cooler could be established by measuring cooler inlet and outlet pressures and reading the flow rate off the graph.

Doug Gray
 
<<Please confirm this but as I understand the oil cooler only sees a proportion of the oil flow, not 100% as you have used in the calcs.>>

Certainly true below a target temp of 80C (176F).

As I understand the system, the varitherm extends to block a passage leading directly to the engine oil gallery and thus forces all flow to the cooler. The cone shaped tip of the varitherm presses against a matched seat in the case, fully closing the direct-to-gallery bypass passage above 80C. Seat damage (meaning failure to fully seal) is a well known cause of high oil temperatures, because all the oil is not directed to the cooler. With a bad seat (or a bad varitherm) some still leaks to the bypass even when the oil is hot. The higher flow resistance of the 8432R would make this worse. See Lycoming SI 1316A.

However, until the varitherm extends enough to be fully seated, oil flow is indeed divided between cooler and direct-to-gallery; both routes are open. The division would be as you say:

<<The balance between bypass and the cooler oil flow is determined by the vernatherm, and the respective oil flow will be a function of the vernatherm position, the pressure drop across the bypass and the presure drop across the cooler.>>

Excellent comment Doug, as it illustrates something I had not considered. The higher flow resistance of the 8432R would cause more oil to take the bypass route direct to the engine oil gallery until oil temp rose enough to seat the varitherm. You should get faster warm-up with the 8432R after a cold start (a plus). Winter operation should be better also, as less oil would flow through a 8432R with both passages open. The 10599R would get more oil flow below 80C and might need an airflow block in the winter.

All of which still leaves the questions about a 55 lb/min limit with the varitherm bypass passage closed, and the large pressure drop. Yeah, I know an 8432R works, but obviously there are differences in the details.
 
I agree, provided the vernatherm locks shut at 80 degrees.

Again as I understand this (outside my domain of knowledge here) the vernatherm would lift off the seat at normal oil temperatures if the oil cooler back pressure exceeded the spring force over the valve area.

It may indeed not open at all. But on the other hand if it did open at say 10psi, you would see a divided flow between the cooler and bypass. Provided the cooler was sufficiently large (in thermal capacity given the actual flow) for the requirements I don't think the oil pressure drop of the 8432R is really an issue.

Doug Gray
 
<<the vernatherm would lift off the seat at normal oil temperatures if the oil cooler back pressure exceeded the spring force over the valve area.>>

Yes, but per Lycoming's "Troubleshooting High Oil Temperature", the oil cooler pressure drop necessary to unseat the varitherm is 60 to 90 psi. This is a defense against oil cooler blockage.
 
10599R fits well

I'm sure you're aware of this already, but the SW10599 is not a big deal to install in an -8. Here's a photo of how mine went in.

556083611_wF4rA-XL.jpg
 
Guy, I'm going to remote mount a 10599R. With an angle valve everything is tighter on the port side baffle.

Clearly both coolers will work. I'm interested in the operational details and differences between the two coolers. Like Doug said, the big pressure drop across the 8432R is probably not a big deal; people are using them successfully. However, it certainly has some subtle operational effects. One example would be the previous realization about faster warm up. Conversely, consider a failed varitherm. A 10599R would still see some oil flow and do some cooling, while a 8432R would see very little flow.
 
Got thinking about the cooler pressure drop while riding around on the lawn tractor last night. Tell me what you think of this theory. I'm guessing the big Lycoming oil pump makes lots of excess pressure, and the practical difference between the two coolers is simply the amount of oil dumped by the pressure relief valve

Full RPM, oil hot, varitherm closed, arbitrary pump number but reasonable for the examples:

8432R: Pump produces 100 psi, cooler drops 15 psi, pressure relief valve dumps another 5 psi to the sump, and gallery pressure is 80 psi.

10599R: Pump produces 100 psi, cooler drops 5 psi, pressure relief valve dumps 15 psi to the sump, and gallery pressure is 80 psi.

If true, circulation (in lbs of oil per minute) should be higher using the 10599R. The 8432R installation is mostly dropping pressure via viscous friction, while the 10599R installation is mostly dropping pressure by venting it back to the sump.

If true there are all kinds of interesting ramifications.
 
<<Please confirm this but as I understand the oil cooler only sees a proportion of the oil flow, not 100% as you have used in the calcs.>>

Certainly true below a target temp of 80C (176F).

As I understand the system, the varitherm extends to block a passage leading directly to the engine oil gallery and thus forces all flow to the cooler. The cone shaped tip of the varitherm presses against a matched seat in the case, fully closing the direct-to-gallery bypass passage above 80C. Seat damage (meaning failure to fully seal) is a well known cause of high oil temperatures, because all the oil is not directed to the cooler. With a bad seat (or a bad varitherm) some still leaks to the bypass even when the oil is hot. The higher flow resistance of the 8432R would make this worse. See Lycoming SI 1316A.

However, until the varitherm extends enough to be fully seated, oil flow is indeed divided between cooler and direct-to-gallery; both routes are open. The division would be as you say:

<<The balance between bypass and the cooler oil flow is determined by the vernatherm, and the respective oil flow will be a function of the vernatherm position, the pressure drop across the bypass and the presure drop across the cooler.>>

Excellent comment Doug, as it illustrates something I had not considered. The higher flow resistance of the 8432R would cause more oil to take the bypass route direct to the engine oil gallery until oil temp rose enough to seat the varitherm. You should get faster warm-up with the 8432R after a cold start (a plus). Winter operation should be better also, as less oil would flow through a 8432R with both passages open. The 10599R would get more oil flow below 80C and might need an airflow block in the winter.

All of which still leaves the questions about a 55 lb/min limit with the varitherm bypass passage closed, and the large pressure drop. Yeah, I know an 8432R works, but obviously there are differences in the details.

I do not agree with this line of reasoning at all because my oil temperatures are running very low. In fact too low.

On the last flight it was 160-170, below vernatherm extension. There's a lot of oil flowing through the cooler. The highest oil temp I've seen so far is 180 and that's been with some pretty hard running to seat the rings. I have tested the oil temp probe and it was within a couple degrees of the digital thermometer in the can of hot water so that's not an issue.

At this point I would not trade the 8432R for any other cooler no matter what the engineering charts say.
 
Last edited:
<<On the last flight it was 160-170, below vernatherm extension>>

Not necessarily. Again refering to Lycoming's "Troubleshooting High Oil Temperatures":

4. To test the valve for operation, immerse the valve in light oil and heat, at about 150 degrees the valve will start to grow. At 185 degrees the valve should be fully extended (.160 minimum travel).

The vernathern is not a switch, fully retracted at one temperature and fully extended one degree later. At 160-170F your vernatherm may or may not be extended enough to fully close the bypass port. Even partial extension routes some oil to the cooler.

BTW, I've used a target temperature of 80C (176F) in this discussion, but there is probably some variation among production vernatherms of the same part number. There may also be different extension and temperature specs for different part numbers, which have changed a few times. The above reference may not even be current information. Heck, even a thinnner vernatherm gasket or a new, unworn seat would close the port sooner.

<<At this point I would not trade the 8432R for any other cooler no matter what the engineering charts say.>>

Obviously not a "bad cooler". I'm curious about the differences in operating details created by the different pressure drops for the two coolers.
 
There is such a wide variation in oil temps (as well as CHTs) between different airplanes that there are probably other factors that we are ignoring. One of the benefits of doing the analysis and calculations that DanH does is it moves us past anecdotal evidence into real engineering and helps eliminate the differences due to variations in different applications.

In this case if Dan's logic and caculations are sound, the outcome should be predictable. The 10599R should reduce oil temps. If Dan makes the change and oil temp does not go down then either the logic or calculations are incorrect.

Some of those other factors may have been discovered by Larry Vetterman in his work in and around the vernatherm. I neeed to search the posts and reread what has been posted and maybe give Larry a call.

BTW, is the proper name of the device in question "veritherm" or "vernatherm?" Both words are being used in this thread. I Googled vernatherm and found the company that makes them http://www.rostravernatherm.com/ and references to Lyc and Superior and Aero parts. A search on "veritherm" brought up mens clothing and a heater.

I think vernatherm is the proper term and I stand corrected as I was earlier using veritherm.
 
<<...it moves us past anecdotal evidence ..>>

Right. The payoff is in knowing why it works, not merely that it works.

<<In this case if Dan's logic and caculations are sound, the outcome should be predictable.>>

So far just logic, no calcs. When considering a puzzle, it really helps to bounce logic around with your peers.

<<The 10599R should reduce oil temps.>>

Maybe, maybe not. Remember, temperature is controlled by vernatherm modulation until (and if) the oil gets hot enough to fully close the bypass port. From that point oil temperature is a function of maximum cooler heat rejection, which is driven by air and oil mass flow and cooler design.

The 8432R rejects more heat per pass than the 10599R given the same oil and air mass flows per unit of time. You can read that much off the charts. What the charts don't tell you is the oil mass flow for a given pressure.

Remember, the engine holds two gallons of oil, max. Lycoming says oil system circulation is 7 to 9 gallons per minute....thus the entire oil supply is circulated about 4 times per minute. "About" is the operative word. Logic says the the lower flow resistance of the 10599R will allow more oil flow at a given pressure, ie more passes through the cooler per minute.

So, is it better to make more passes with less rejection per pass, or less passes with more rejection per pass?

<< makes the change and oil temp does not go down then either the logic or calculations are incorrect.>>

Sorry, my own case is a new installation, no baseline.

<<I think vernatherm is the proper term and I stand corrected as I was earlier using veritherm. >>

Me too...whoops!
 
Flip side?

Remember, temperature is controlled by vernatherm modulation until (and if) the oil gets hot enough to fully close the bypass port. From that point oil temperature is a function of maximum cooler heat rejection, which is driven by air and oil mass flow and cooler design.

The 8432R rejects more heat per pass than the 10599R given the same oil and air mass flows per unit of time. You can read that much off the charts. Logic says the the lower flow resistance of the 10599R will allow more oil flow at a given pressure, ie more passes through the cooler per minute.

So, is it better to make more passes with less rejection per pass, or less passes with more rejection per pass?

Dan,

Don't these two factors come in to play?

1. Wouldn't more residence time, (to a point) allow the heat transfer to fully occur. Too much circulation can be a bad thing?

2. Isn't it true that pressure drop may not be all bad? Isn't there associated cooling involved due to the pressure drop of the fluid whether it be air or oil side?

These have interesting implications as the vernatherm modulates the oil flow through the cooler. Are these part of the reason for the 8432R's somewhat surprising performance from a smaller package?
 
<<Wouldn't more residence time, (to a point) allow the heat transfer to fully occur.>>

I believe so.

<<Isn't there associated cooling involved due to the pressure drop of the fluid whether it be air or oil side?>>

Isn't that just for compressible media?

<<the 8432R's somewhat surprising performance from a smaller package?>>

The drawings show the 8432R to be taller by 0.040" and narrower by 0.020". The primary package difference is the port locations.
 
Dan,
I have been having the same questions as you regarding the 10599 vs. 8432. Considering the pressure loss through the highly regarded 8432, why is cascading two 8406/20002 or similar coolers a bad idea? (other than the extra hoses etc.)
 
Diagram Needed

Does anyone have a schematic diagram of the oil system for a Lyc opposed 4 with oil cooler and filter? I think I need to see what it really is instead of what I think it is.

The last two days I have been doing a lot of flying (yeah). Sunday 2 hrs DeKalb, IL to Burlington, IA and back plus some sightseeing. Monday DeKalb to Burlington. Tuesday Burlington to Olathe Kansas. Wednesday Olathe to Hicks/Ft Worth. My CHTs are staying in good shape, although I sometimes have to richen things up a little, but my oil temp is running 230F. OAT is around 80 to 90 F so it's hot but not really hot. Oil pressure is running at 62 to 64 so a little low although within Lycoming specs.

I'm concerned that maybe my vernatherm is not opening fully, or that the pressure being a little low is not putting enough oil through the cooler. That's why I need the schematic. I want to understand what the oil pressure reading is really telling me.

My oil cooler is the standard Vans EA Cooler II (8409?). With about 500 hours on it the flanges are cracking so I need to change it out and improved the installation. I had been planning to use a Setrab but this thread has me thinking. That's always dangerous!
 
<<I'm concerned that maybe my vernatherm is not opening fully,..>>

Closing fully. Vernatherm extension closes the bypass port, forcing all flow to the cooler. System diagram and text here:

http://www.airwolf.com/LinkClick.aspx?fileticket=YVB+2vwdCLQ=&tabid=86&mid=384

<<Considering the pressure loss through the highly regarded 8432, why is cascading two 8406/20002 or similar coolers a bad idea? (other than the extra hoses etc.) >>

Mike, I don't figure it's a bad idea, although God is in the details. Thinking out loud:

Oil flow in series: the first cooler will drop oil temperature by some amount, so the second cooler will be working with less air-oil deltaT. I think series flow is the usual method.

Oil flow in parallel: probably needs some custom fittings to split and rejoin a balanced oil flow. Equal deltaT. In 100 years of aviation probably somebody has done it that way.

Air flow in parallel: compared to a single cooler, more total cooling mass (air) through the cooling system for the same baffle drop, more cooling drag.

Air flow in series (stacked coolers): compared to a single cooler, less total cooling mass through the system for the same baffle drop. I doubt the reduction would be significant. The second cooler would again be operating with less deltaT; it not only receives partially cooled oil, but also receives heated air.

Which scheme or combination is best? Darned if I know. I am pretty sure dual coolers weigh more and require more physical space.

Back to the 8432R. The Kitplanes article referenced in another thread makes an interesting point; the system is driven by a positive displacement pump. The author asserts that pressure drops in the system (like the 8432R's higher restriction) thus don't result in a reduction in oil volume, but rather an increase in oil velocity. I suspect that's not completely true in the real world (viscous friction and all that). Hydraulic engineers, comments please?

However, if even partially true it may offer an explanation as to why the 8432R is more efficient. If increased velocity reduced the stagnant boundary layer inside the 8432R's oil passages, it should transfer heat better.
 
Last edited:
Oil flow in series: the first cooler will drop oil temperature by some amount, so the second cooler will be working with less air-oil deltaT. I think series flow is the usual method.

I too figured the delta T would make the second pass less efficient, but going back to your initial thoughts about more pressure drop in the 8432 and making some assumptions about how the oil pressure regulator works; If we assume that (less vol., lower pres.) oil exiting the 8432 is colder than (more vol., higher pres.) hotter oil exiting the 10599 (due to each unit vol. of oil passing more surface area) , could the system be letting more colder oil into the engine vs. dumping it back to the sump? I'd really like to see real-world inlet vs. outlet oil temps for these coolers and get a better understanding about the oil pressure control and flow volumes/destinations. Too expensive a science project for me.

Interesting thought about the boundary layer, but don't the oil cooler passages have turbulators(sp?) in them to address that? I know Setrab does.
 
Thanks for the link, Dan. After I read that and consulted the Lycoming overhaul manual and the parts manual and the Superior manual I have learned a lot.

Superior (RIP) used a Vernatherm, a oil filter bypass valve (in case the filter clogged) and an oil pressure relief valve to dump oil to the sump to regulate oil pressure to the engine. Lycomings use either a relief valve or a Vernatherm, and rely on the Vernatherm to handle any clogged oil cooler or high pressure drop issues.

I think it is interesting that the Lycoming diagram shows an "Oil Cooler By-Pass Valve" as well as an "Oil Relief Valve" in their diagram even though either one or the other is but not both are used. There diagram does not show an oil filter or oil filter relief valve, just a "Pressure Screen."

My oil temp is running about 230 and oil pressure 62. I thinnk it is time to pull and test the Vernatherm and inspect the seat. Perhaps it is lifting at a lower pressure than it should and therefore bypassing the oil cooler.

This also sheds some light onto Larry Vetterman's thinking on Vernatherms. According to this post by rocketbob http://www.vansairforce.com/community/showpost.php?p=287401&postcount=4

"Guys Larry Vetterman has cracked the code with complete control over oil temps, by removing the vernatherm and installing the oil cooler bypass spring and plunger in the filter housing, a ball valve on the line to the oil cooler, controlled with a vernier from the cockpit. The part numbers for the oil cooler bypass are SL62415 for the oil bypass plunger and 69436 for the spring."

I am still convinced that this is not something esoteric like boundary layers in the tubes. There is so much variation in oil temps between builders with the same RV model, oil cooler and (approximately) the same engine that we must be missing something simple.
 
Does anybody know if the Vernatherm used by Lycoming and Superior are the same? I am wondering why Lycoming uses the Vernatherm as both a thermostat and a pressure relief valve while Superior apparently uses it only as a thermostat and lets a separate relief valve control pressure.
 
8432R doesn't reduce oil pressure

Like every other engine I know of, the Lycoming has a positive displacement oil pump driven with a fixed speed ratio relative to the crankshaft. Since oil is incompressible, that means that the flow rate from the pump is constant at any given RPM; independent of any restriction. If one were to completely block the pump's output, the pressure would go to infinity with an ideal pump. In the real world the engine would stop or something would break in that case.

The pressure relief valve is downstream of the cooler and filter. Since the flow rate is constant, the 8432R's extra pressure drop just raises the pressure at the inlet port and the pressure at the outlet port is unchanged.

What I've been wishing for is a vernatherm replacement which completely blocks oil to the cooler below the set point. It seems that it wouldn't be that hard to design a valve system that could:
divide flow between two outlets based on temperature,
shut off flow completely to the cooler the temperature is below some cold limit,
divert all flow to the cooler when above a hot limit,
select some medium division of flow for temps between the two limits,
and have no valve position which prevents oil from flowing to the engine.

I suspect there might be able to find an oil thermostat with those properties in the car racing catalogs but I haven't seriously looked for one yet. It would be nicer if the valve fit in the same space as the vernatherm to avoid the extra plumbing and weight of an external unit. Until such a thing is marketed, I guess many of us will be stuck with accepting low temps or manual temperature control via air (or oil) valves.
 
This post is not meant to pick on kcameron but just to understand what the system is doing.

From kcameron: "The pressure relief valve is downstream of the cooler and filter."

I don't think this is true if a Vernatherm is installed except in Superior engines. In my reading of the documents Lycs have either a relief valve or a Vernatherm. The Vernatherm acts as a relief valve if the oil cooler delta-P gets too high.

From kcameron: "Since the flow rate is constant, the 8432R's extra pressure drop just raises the pressure at the inlet port and the pressure at the outlet port is unchanged."

I agree with the first part but not the last. You cannot raise the pressure at the inlet of the cooler without raising the outlet pressure. The delta-P is dependent on flow rate not inlet pressure so if the flow rate stays constant and the inlet pressure goes up the outlet pressure goes up too. Also, since the Vernatherm and Oil Cooler are in parallel, once the pressure is sufficient to lift the Vernatherm the delta-P across both will be the same.

I'm not trying to get into any hair splitting but we are all struggling to understand what is going on and why there is such a wide variation in operating temperatures.

The way the system is arranged now has the following failure modes:

1. Oil cooler blocked - Vernatherm opens, oil flows, gets too hot but pilot has a chance to reduce power and make an emergency landing
2. Vernatherm fails open, too little oil flows through oil cooler, same outcome as above.
3. Vernatherm fails closed - oil flows through the cooler, perhaps with nobody even noticing until the next cold startup when oil temps do not come up as expected.
4. Relief valve fails open - pressure drops but I think these valves are sized so that full flow is less than the oil pump so there will still be some oil flow through the engine.
5. Relief valve fails closed - pressure is high, but oil still flows through the system. Pilot may notice unusual high pressure and chose to make a precautionary landing.

Very robust! Note that the last failure mode is really nothing to be concerned about because if I understand the manuals correctly, if a Lycoming has a Vernatherm it does not have a relief valve therefore this is a normal mode of operation not a failure mode.

I also not in favor of a system with a component that could fail and leave no path for oil flow. Separate components that could block or allow flow to the oil cooler or around the cooler give the additional failure mode of both failing closed, in which case no oil flows and the engine seizes. That is not a desirable outcome.
 
John,
The pressure relief valve Kevin spoke of is the ball valve in the right main engine oil gallery, just forward of the dipstick. It regulates final pressure delivered to the engine regardless of upstream installation differences, so long as pressure delivered to the gallery is higher than the desired regulated pressure.

The "either valve or vernatherm" valve is the "Type 1" valve illustrated in the "High Oil Temperatures" Lycoming document. In the absence of a vernatherm installation it allows oil to bypass the cooler in the event of cooler blockage. Apparent Lycoming installations don't install a vernatherm and a Type 1 bypass in the same engine because they can physically touch and interfere. I assume that's not the case with the clean sheet redesign Superior engine.

Kevin,
So you feel "same oil volume, higher velocity through the cooler" is the correct operating concept for the 8432R?
 
Thanks for the clarification, DanH. I'll read up on it tonight in the manuals. Here's the schematic from the link in an earlier post.

567603475_qfLWv-M-0.jpg


So the Type I or Type II refers to 2 different "Oil Cooler By-Pass Valves" but regardless of which of those are used there is still the relief valve.

I need to search the forums to find out how to adjust the relief valve to raise the oil pressure. I read the Lycoming manual last night but it seemed to be talking about the bypass valve.
 
My engine is a Lycoming IO360-A1B. It has both a vernatherm and separate oil pressure relief valve in the position Dan described. Every engine I've examined closely has that configuration though I don't claim to be an expert on aircraft engines.

John, I hate to disagree but I don't think you're looking at it right. All else being equal, the inlet (and oil pump outlet) pressure will be higher if one replaces a 10599R with a 8432R. This is dictated by the constant flow rate and higher fluidic resistance of the 8432R.

Dan, I hadn't before thought about the 8432R creating a higher flow velocity within the tubes. I would state it differently, though. Compared to the 10599R, the 8432R effectively splits the incoming oil flow across half as many tubes. Therefore, the average velocity must increase by a factor of two at any given total flow rate.

It makes sense that higher fluid velocity would decrease the boundary layer thickness and therefore decrease the average thermal resistance between the oil and tube. I had been assuming the increased efficiency came from the longer tube length which reduces heat leakage (through the metal of the tube itself) from the hot end to the cold end. This reduction in heat leakage allows the cooler to more closely approximate a reverse flow heat exchanger (which can approach 100% efficiency). I guess the advantage is some combination of both effects. I just hope that my tube length theory plays a significant part.

Last fall, I installed an SW 8432R in my RV-4. The previous AeroClassics cooler was more or less similar to the 10599R but I don't have the part number handy. Before the swap, the oil temperature would quickly rise to at least 180 degrees and stay there even on the coldest days in the California bay area. Normally, the vernatherm would keep it at 180 but, climbing out on 100+ degree days after a quick fuel stop in the central valley, the oil temp would quickly rise to 230 or more and force me to reduce power in order to keep it under control. Since the swap, I haven't had a chance to try that kind test again but the temps often failed to rise above 150 or 160 last winter. The highest I've seen with the 8432R is 195 after a bunch of touch-and-goes. I'll know better when I'm able to go out on a hot day again but, so far, I believe the 8432R is much more efficient.
 
Kevin - don't hesitate to disagree with me. Most of what I have learned has come from people who disagreed with me. It causes me to examine my knowledge.

I reread your post and think I understand things better now. The positive displacement pump will put out whatever pressure is necessary to move the required volume of oil. The higher head loss of the 8432 requires a higher inlet pressure (and higher pump discharge pressure) than the 10599 to move the same volume of oil.

The outlet pressure is controlled by the Resistance to flow in the rest of the system which is flow rate dependent. Since that flow rate is the same because of the pump it's head loss is the same and the cooler outlet pressure will be the same regardless of the cooler design.
 
<<Dan, I hadn't before thought about the 8432R creating a higher flow velocity within the tubes.>>

Don't feel bad. I started the thread thinking of the system like it was driven by a pressure head (constant pressure, volume decrease with resistance to flow), not a positive displacement pump (pressure rise with resistance to flow, constant volume). Big difference. Didn't register until I read that Kitplanes article. Thanks for the confirmation.

<<Compared to the 10599R, the 8432R effectively splits the incoming oil flow across half as many tubes. Therefore, the average velocity must increase by a factor of two at any given total flow rate.>>

Looks that way to me.

<< I had been assuming the increased efficiency came from the longer tube length which reduces heat leakage (through the metal of the tube itself) from the hot end to the cold end.>>

Interesting detail, but would not the 8432R be worse in that regard? The inlet and outlet are in close proximity, not at opposite sides of the cooler like the 10599R. Conductive leakage would be straight across the end cap, as well as across the fins between the adjacent first and second pass tubes. And conductive leakage aside, we still have the problem of reduced air-oil deltaT during the second pass.

Speaking of tube length, the popular explanation for the increased efficiency of the two-pass cooler has been "the oil stays in the cooler for a longer time." If we accept the premise of "half as many tubes=doubled velocity", that explanation is clearly wrong.

Process of elimination...unless something else comes to mind, the only thing left is reduced boundary layer.

More later, need to do some reading.
 
A little heat transfer theory...

I just read through the last couple pages of posts and since I'm an engineer and designed heat exchangers for 8 years, I figure that I should contribute.

A two pass cooler increases heat transfer (vs. a single pass cooler with everything else being equal) because of the higher oil velocity. Heat transfer coefficients always rise with higher velocities. The downside of course is that the pressure drop skyrockets. The oil is traveling at twice the velocity and it must travel twice the distance. I can't comment on if and how much the oil flow will be reduced with the two coolers in question however.

As far as comments about needing to provide "residence time" or needing to slow down the flow for best heat transfer, this is a common misconception. The heat transfer rate always increases with a function of mass flow rate and velocity of either fluid (oil or air).

To go a little more in depth - oil tends to quickly form a laminar boundary layer in all practical velocities and flow channels, which has a much lower heat transfer coefficient than a turbulent boundary layer. The internal turbulators help to continually break up the laminar boundary layer. They also serve two other purposes: 1) to provide additional surface area and 2) to provide strength to the tube under pressure.

Dan - you mentioned in a previous post about which configuration to use if you were using two coolers and you showed some good logic. First, you would definitely want the air flow to be in parallel due to the heating of the air. I'm pretty sure that you would want the oil to be in series however for maximum heat transfer. The reason being that velocities and mass flow rate would be much higher with series flow, and hence much higher heat transfer. Again, assuming that the pressure drop wasn't an issue. You are correct that the downstream cooler would have cooler oil and less temperature differential, but I think that the temperature through the cooler drops a relatively small amount and hence it would not be drastically reduced.

OK, I just wrote a lot and I have two thoughts:

1) I hope I don't come across as a know it all
2) I need a beer
 
Eric - You come across as the kind of guy I'd like to have a beer with!

Thank you for adding to the discussion.
 
<<I'm an engineer and designed heat exchangers for 8 years, I figure that I should contribute.>>

Ahh, yeah, I figure you should too ;)

Eric, if you will, a few questions on fine points.

Regarding our positive displacement engine oil pumps, would you happen to know a ballpark value for efficiency? So far in this discussion we're treating them as 100% efficient, but they don't have seals or zero clearance, so some high-to-low side leakage must be present as temperatures and pressures rise.

Regarding paired oil coolers, you said:

<<I'm pretty sure that you would want the oil to be in series however for maximum heat transfer. The reason being that velocities and mass flow rate would be much higher with series flow, and hence much higher heat transfer.>>

Can you walk us through why velocity and mass flow rate would be higher with series flow through two coolers?
 
Eric, if you will, a few questions on fine points.

Regarding our positive displacement engine oil pumps, would you happen to know a ballpark value for efficiency? So far in this discussion we're treating them as 100% efficient, but they don't have seals or zero clearance, so some high-to-low side leakage must be present as temperatures and pressures rise.
Unfortunately I can't help you here. I know very little about pumps.

Regarding paired oil coolers, you said:

<<I'm pretty sure that you would want the oil to be in series however for maximum heat transfer. The reason being that velocities and mass flow rate would be much higher with series flow, and hence much higher heat transfer.>>

Can you walk us through why velocity and mass flow rate would be higher with series flow through two coolers?
What I meant was that the velocity will be much higher through two series coolers compared to two parallel coolers. Assuming that the restriction between the two coolers/lines are equal, a parallel set up will deliver half the flow to each cooler. Of course the velocity through two series coolers would be the same as velocity through a single cooler.

Another point - although higher velocities and/or mass flow rates will deliver better performance, two parallel coolers will definitely provide more heat rejection than a single cooler. Two parallel coolers are equivalent in heat transfer to a single cooler with twice the tubes (assuming everything else being equal and that the single cooler gets twice the air mass flow rate).
 
<<What I meant was that the velocity will be much higher through two series coolers compared to two parallel coolers.>>

Understand, thanks.

<<Two parallel coolers are equivalent in heat transfer to a single cooler with twice the tubes>>

Good point. I suspect the single large cooler would be an easier install and weigh less.

Ok, back to the thread premise; 8432R vs 10599R. We've learned a lot (well, I have), but one interesting question remains.

Lycoming states engine oil flow volume is 7 to 9 gallons per minute. At 9 GPM (about 65 lbs) an 8432R would be off its chart. If you extend the chart, the pressure drop at 65 lbs oil mass would be 20-23 psi.

So, what determines the difference between a 7 GPM engine and a 9 GPM engine?

Remember, it's a positive displacement pump. In theory (ignoring leakage and loss), volume only varies with RPM. So is the difference in volume (7 vs 9) merely two different pump gear widths, or is there something else? And if the difference is pump size, which engine has which?
 
Prop?

So, what determines the difference between a 7 GPM engine and a 9 GPM engine?

Remember, it's a positive displacement pump. In theory (ignoring leakage and loss), volume only varies with RPM. So is the difference in volume (7 vs 9) merely two different pump gear widths, or is there something else? And if the difference is pump size, which engine has which?

Is it true that people with fixed pitch props have more trouble with high oil temp due to lower rpm on take-off and climb out?

Post #146 at the link below, last pic shows the door I control air flow with.

http://www.vansairforce.com/community/showthread.php?t=1744&page=15
 
Last edited:
Excellent discussion - trying to understand this better myself.

Is the 7-9 gallons per minute through the engine or through the oil pump? I'm guessing it is through the pump. 7-9 approx. corresponds to the rpm ratio 2100 to 2700rpm - supports the conjecture.

Taking the pump rotor dimensions it probably wouldn't be difficult to calculate the pump volume to confirm.

So far my take on the preceeding discussion is as follows:

If 100% of the oil pump output passes through the cooler then the pressure drop will be as per the chart ~23psi. There does not seem to be any issue with this flow rate/pressure drop for this cooler - correct?

If we assume there is sufficient back pressure in the engine and the relief valve operates then the oil cooler outlet pressure will be determined by the pressure relief valve. Add a small delta for pressure drop between the cooler and the relief valve. Lets say this is 80psi.

The pressure at the cooler input would be 80+23psi. Provided this does not exceed the rated pressure of the cooler then all should be well.

If the engine - ie parts beyond the pressure relief valve - does not sustain a back pressure greater than the relief pressure, it will close causing the full oil volume to pass through the engine itself. Say this is 50psi at the valve. The oil cooler inlet pressure would then be 50+23psi.

In terms of oil flow through the engine (what does not get dumped by the pressure relief valve) I would expect that this would vary significantly dependent on engine age etc etc.

Cooler heat transfer capacity would need to exceed the thermal energy dumped into the oil from the engine in order for it to continue to keep temperatures under control. The cooler capacity and the amount of energy being dumped into the oil are both highly variable.

Doug
 
Last edited:
<<Is it true that people with fixed pitch props have more trouble with high oil temp due to lower rpm on take-off and climb out?>>

Hmmm...that's an eyebrow raiser. Theory suggests it should be true, based on what we've learned so far.

For sure oil mass flow would be less (positive displacement pump at lower RPM), so the heat transport mechanism (sump to cooler) would have less capacity.

I'd estimate the engine's thermal input at pretty close to the same as a CS prop setup. There would be fewer firing events per unit of time (lower RPM again), but those events would be at effective early timing (peak cylinder pressure closer to TDC), driving heat load up.
 
Is the 7-9 gallons per minute through the engine or through the oil pump? I'm guessing it is through the pump. 7-9 approx. corresponds to the rpm ratio 2100 to 2700 rpm - supports the conjecture.

Doug

So if you are running hot and want to get the oil temp down perhaps the best way on a CSP equipped plane is to keep RPM up but pull the power (throttle) back. Less heat would be generated but the volume of oil going through the oil cooler would be maximized, maximizing the heat removal. 2500/22 instead of 2200/22 for instance.

I had been reducing RPMs and Throttle thinking fewer ignition events and lower power would maximize cooling but perhaps lower oil flow was reducing the effectiveness of that technique. It was reducing temperature just not as much as I expected. Sounds like it's time for a flight test to compare.
 
So if you are running hot and want to get the oil temp down perhaps the best way on a CSP equipped plane is to keep RPM up but pull the power (throttle) back. Less heat would be generated but the volume of oil going through the oil cooler would be maximized, maximizing the heat removal. 2500/22 instead of 2200/22 for instance.

I would agree but, if you pull the power back, don't we lose the full throttle fuel / air mixture circuit in the carb. Does fuel injection have the same enrichening circuit? Now we are affecting CHT's and EGT's. I'd think just increasing engine RPM might be the way to go... but you'll be making more horsepower and more heat?
 
Recap

So it would seem that when sizing an oil cooler:

We want the vernatherm to run about ½ way open?

If it is closed then there is no margin for engine bearing wear and associated loss of oil pressure as the engine collects hours. Bearing life could be reduced. Well, on second thought we can adjust the oil pressure setting independently on these engines.

If it runs open too far, then we risk over pressure and rupture of the oil cooler or oil filter canister during cold weather or if a blockage in an oil passage occurs. This may be negated or partially negated by using an oil filter with a bypass valve? I’m unclear if using one of these would help with oil cooler over-pressure protection.

But with the vernatherm running partially open we are bypassing hot oil back to the engine, so we need to account for more heat load depending on vernatherm opening.

We may need to account for the differing engine speeds of constant speed vs. fixed pitch propellers during take-off and climb out.

Since oil flow is basically at a fixed rate due to the recip. pump, (we are usually limited to full rpm on take-off due to engine fuel mixture concerns), does it make sense to vary the air flow across the cooler with cowl flaps or ram air or do we just do the same thing by increasing airspeed and let the climb performance suffer?

Would the aforementioned cable operated manual bypass instead of an automatic vernatherm be the way to go? Either that or air control doors add complexity and operator work load and human performance failure modes.

In one of Tony Bingellis books, (Firewall Forward)? I seem to recall pictures and commentary involving an air door hooked to a bellows type thermostat.

Question: Does a propeller governor which is also a boost pump add heat to the oil?

Dan, If I'm screwing up the intent of the thread let me know, please. Are you wanting to focus on just the oil cooler differences?
 
Last edited:
<<So it would seem that when sizing an oil cooler: We want the vernatherm to run about ½ way open?>>

No. Oil cooler sizing assumes the vernatherm is fully closed, ie blocking the bypass port and routing all oil to the cooler circuit.

<<.... we risk over pressure and rupture of the oil cooler or oil filter canister during cold weather or if a blockage in an oil passage occurs. This may be negated or partially negated by using an oil filter with a bypass valve? I’m unclear if using one of these would help with oil cooler over-pressure protection.>>

Can't over-pressure a cooler with the bypass port open. Can't over-pressure a cooler with the bypass port closed either, because the vernatherm also acts as a cooler circuit pressure relief valve. If cooler circuit flow resistance goes above 60 to 90 psi (Lycoming figures), the vernatherm opens by compressing its spring and dumping pressure and flow to the bypass. The coolers themselves are good for several hundred psi.

Both the cooler bypass and the cooler circuit feed the oil filter. In theory a blocked oil filter could overpressure the cooler, but at that point you have bigger problems anyway.

<<If I'm screwing up the intent of the thread let me know, please. Are you wanting to focus on just the oil cooler differences? >>

Keeping a thread perfectly on topic would be an unrealistic expectation given the mechanics of forum participation ;)
 
<<Is the 7-9 gallons per minute through the engine or through the oil pump?
I'm guessing it is through the pump.>>

So am I.

<<7-9 approx. corresponds to the rpm ratio 2100 to 2700rpm - supports the conjecture.>>

Perhaps Lycoming does mean the 7-9 variation is dependent on RPM. I'll drop a note to a rep.

<<So far my take on the preceeding discussion is as follows: If 100% of the oil pump output passes through the cooler then the pressure drop will be as per the chart ~23psi. There does not seem to be any issue with this flow rate/pressure drop for this cooler - correct?>>

That's what we're trying to determine.

<<If we assume there is sufficient back pressure in the engine and the relief valve operates then the oil cooler outlet pressure will be determined by the pressure relief valve. Add a small delta for pressure drop between the cooler and the relief valve. Lets say this is 80psi. The pressure at the cooler input would be 80+23psi. Provided this does not exceed the rated pressure of the cooler then all should be well.>>

Doug, I think that's a very realistic analysis, and quite likely what is actually happening in the field. Surely there are folks with 9 GPM flows running 8432R coolers.
 
Quick notes.

Apparently there are several pump impeller widths, the most common being roughly 3/4 and 1", and width determines if you have a 7 GPM system or a 9 GPM system. Makes sense. I think 18109 series impellers are 3/4". Not sure about 1" part numbers. Maybe we could get some input from our resident engine builders?

The nice lady at Stewart Warner checked with engineering and reports their coolers are proof tested to 400 psi. They don't have a specified maximum working pressure, but the design engineering manager feels 200 psi in service shouldn't be a problem. Good info.
 
Dan,

Looking through the manuals etc and now understnd that most 4 cyl Lycomings have the 0.75" oil pumps, the 6 cyl engines have the 1".

I also found a cross section of the oil pump so I was able to estimate the impeller size. So far as I can tell the impellers each have a swept annulus area of abut 1.166 sq inches. (I used an OD of 1.5" and ID -smallest diameter of the 'teeth' - of 7/8" perhaps someone could confirm these sizes..)

The Oil volume per revolution is the product of: The proportion of the swept volume that the oil occupies (0.5 is my conservative estimate - probably closer to 0.6), the width of the pump (0.75" for 4 cyl Lyc), the annulus area (1.166 sq. in) and the number of impellers (2).

Oil volume approx = 0.5 x 0.75 x 1.166 x 2 = 0.8745 cu.in per rev.

Note 231.9 cu. in per USG if I am not mistaken.

At 2700 rpm the oil volume pumped is about 2700 x 0.8745 / 231.9 or 10.2 USG/min.

So I suspect the 7-9 G./min figures are probably about correct if not conservative.

Hopefully someone can post accurate impeller dimensions.

Doug
 
There is a lot of great information being shared in this thread. I especially like the analytic approach by DanH and Doug.

The question that popped into my mind is can the 1" pump be used on a 4-cylinder Lyc?
 
8432 installation pics?

Not to get too far off topic, but I've pretty much decided to try an 8432 on my 200HP IO360 RV7. I have an 8406 on there now and it's marginal on very hot days (90+).

I've combed this site but can't seen to find a good picture of an 8432 installation. Does anyone have pictures of one? Preferably baffle mounted on an RV7?

Thanks in advance,
 
Back
Top