What's new
Van's Air Force

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

Baffle Mod

DanH

Legacy Member
Mentor
Lycoming cylinders have a classic baffle problem. I don't know if anyone else has taken this particular approach to solving it (lots of ways to skin a cat), but maybe the pictures will help the new guys understand the issue.

Here's the root of the problem. The intake side of the head has no fin depth in the area indicated by the pencil. but standard baffles for the left front and right rear cylinders place a plate directly against this area. There is plenty of air at "A", but there is no way for air to circulate down to area "B"

Intake%20Fin%20Depth.jpg


Details:

Fins.jpg


Here's the standard baffle (right rear) against the head. Flow to the lower fins and baffle wrap is blocked by the zero-depth fin area identified by the green tape. The standard "cure" is to place a washer or some other spacer between the head and the baffle, but that opens a gap along the entire length of the baffle plate.....in particular the area outlined in a black rectangle above. Any air sneaking past the area in the rectangle is pure leakage.

Intake%20Fin%20No%20Clearance.jpg


It is quite easy to hammerform a small duct of sorts in the baffle plate. The "duct" is nothing more than a fancy dent knocked in the sheet.

Intake%20Fin%20Hammered%20Duct.jpg


The result is a passage to bypass the no-depth fin area (arrow). Now air can flow down the back of the head and pass in between the lower fins where you have a baffle wrap. The reminder of the baffle plate is not spaced away from the head and cylinder, reducing undesired leakage. Any air not passing between fins is pure drag; it didn't do any cooling work.

Intake%20Fin%20Clearance.jpg


Not all builders have metal forming hammers, or may not want to try beating on an expensive baffle part as their first metal forming experience. I'll do the left front baffle using an alternate method and get it up here soon.

Have fun.

----->>>>POSTSCRIPT: See posts 24 and 25
 
Last edited:
Great idea Dan. For sure, the clearance is an issue, as I lost 25 degrees just by adding a washer to the back of that baffle. This looks like a much more elegant solution. I'll give that a try on my next one.

Thanks for the tip!!

Chris
 
Great Idea

I used a couple of larger spacers. I really like the idea of hammering a dent to make it work. How great to get to use a hammer in a productive, non destructive manner:)
 
Here's the alternate method. Lots of ways to skin this cat.

Cut a hole in the baffle, carefully deburr it, then make a spacer (0.065" to 0.080" thick) and a cover plate:

Intake%20Fin%20Builtup%20Duct%20Parts.jpg


When assembled, you have a duct to bypass the no-depth fin area:

Intake%20Fin%20Builtup%20Duct.jpg


There is a structural brace (a length of angle) above the inlet ramp and across the face of the cylinder. Space it forward a little so it doesn't block the duct.

Here's the other side:

Intake%20Fin%20Builtup%20Duct%20Back.jpg


The 90 degree flange on the cover plate is important.

Dimple and countersink as necessary, rivet flush on the cylinder side.
 
Last edited:
Great eye...do you lay awake at night thinking of these great ideas/fixes? I love your posts!! Even when the flames are flying!!
 
Another method, that I have used many times, is to rivet a piece of 1/8" flat bar to the outside of the area that Dan is modifying. The flat bar goes in the location where the mounting hole to the cylinder is. It is about a couple of inches long. This is a simple modification, solves the cooling problem, and strengthens the baffle at the attach point.
 
Thank you everyone for your kind comments.

It is an air-cooled engine worth more than my first house, so really good baffles are worth thought and careful fabrication.

BTW, I can't say if a 1/16" or an 1/8" or any other particular dimension is "correct" for this duct. Can't even say if the left front and right rear cylinders should have the same duct size; different pressures in their respective plenum areas. I am sure an install which doesn't allow any air to a whole section of the cylinder head finning is bad, and likewise, all leaks are bad.
 
A Washer Really Does Work

My 7A has a AFP FI Superior O-360, with a James cowl and plenum. #3 cylinder always ran about 20 degrees hotter than #4. This was not typically a problem until climbing out of Phoenix in the summer, when #3 would get up to 420F or so (per the ACS-2002 readout). I saw Chris' post back in August 2007 and thought it was impossible until I tried it today -- a standard washer really does drop the temperature by over 20F! Seeing is believing! Thanks!
 
Wonder if this applies to the parallel valve engine?

I was looking at my engine today and thought I'd check the clearance between baffles and fin area. Heck, I can see day light all the way down to the bottom baffles on the front and aft sides of the cylinders. There seems to be exposed fin area all the way through except maybe between 2 fins. I even poked a thin rod down to confirm it.

Could it be what your images show closed fin area of the angle valve cylinders? It does not seem to be the case with parallel valve engines.

Could also be I am missing something here....
 
David, mine is an angle valve. I don't remember what a parallel valve looks like on the intake side. It has been ages since I've seen one naked, and well, I'm 53 <g>

Anybody have a close-up picture?
 
Another method, that I have used many times, is to rivet a piece of 1/8" flat bar to the outside of the area that Dan is modifying. The flat bar goes in the location where the mounting hole to the cylinder is. It is about a couple of inches long. This is a simple modification, solves the cooling problem, and strengthens the baffle at the attach point.

Tom, I can't visualize your method. Any pictures?
Roger Evenson, 7A finishing
 
Bumping an old thread.

Doug posted two photos of a new ECI engine on the front page today, and I noticed something interesting about the ECI cylinder head.

Looks like ECI tried to address the "no fin depth" problem on the intake side of the head. Three fin gaps are still fully blocked, but the majority seem to have depth enough for some flow to the lower fin set.
 
Last edited:
Thanks Dan & Doug

Hi Gents,

I'm experiencing high cht's on climbout on my James cowled -4. I remember seeing this thread and resolved to look it up and make some changes. I see Doug is down on donations so I just made one today. This forum has helped me out so many times ... I really appreciate everyone's input.

Thanks to all for sharing your info and Doug for keeping this resource alive.

Cheers,
 
Just out of curiosity, Dan, couldn't the same effect be obtained by using a couple of spacers as others have done but seal the unwanted space with a bead of high-temp RTV?

(I never knew that this was an issue, and since all of my CHTs are good and within about 10-15 degrees of each other, I don't think I'll mess with it.)

[EDIT: I didn't read the whole thread. I have ECI cylinders. Maybe that's why it hasn't been an issue for me. Duh.]
 
Dan and All,

I am mocking up my baffles now. I have hammer formed a duct that is probably about 3/32" deep behind the aforementioned area.

It really isn't too visible in this pic because the bend in the baffle somewhat obscures it. Nevertheless... I am wondering if I should be concerned with the gap I have parallel with the cylinder extending outboard or if I should simply install shims and call it done.

Would that also aid in cooling this problematic area???

I welcome opinions!

IMG_0522.JPG


:) CJ
 
.. I am wondering if I should be concerned with the gap I have parallel with the cylinder extending outboard or if I should simply install shims and call it done.

Sorry for the delay John...missed your post.

In general I'd say promote flow where there are fins present, and minimize gaps that have no fins.

The finless area on the intake side of the cylinder head starts just inboard of the injector and extends outboard to the valve cover. Mine has a bead of black high-temp RTV there, but a small gap (1/16") can't hurt and may even be beneficial.
 
I'll add fly it and see how it works. Having seen this thread before I did my baffles, I made a form and created a nice big clearance behind #3. Turns out it's my coolest-running cylinder. NUmber 2 on the other hand, which has the same fin issue except facing forward, tends to get hot on climbout but then drops to mid-range in cruise. On that cylinder I simply bent/tweaked the baffle a bit, rather than doing a nice job forming a bubble in the baffle. Mostly because I originally ran the Van's air dams on the front two cylinders; turned out they didn't like them and I removed them.

A neighbor has a (not a Van's) with a parallel valve Lycoming O-360 that was factory-new in the late 70's. Still has the original cylinders. They look more like the ECI cylinders than the others we see with no fins in the vicinity of the exhaust valve.
 
Lars,

Wilco!

That is why this is an experiment! I will give it a go.

I hope it runs as cool as yours!

Thanks both!

:) CJ
 
NUmber 2 on the other hand, which has the same fin issue except facing forward, tends to get hot on climbout but then drops to mid-range in cruise. On that cylinder I simply bent/tweaked the baffle a bit, rather than doing a nice job forming a bubble in the baffle.

Hey Lars, any chance you could post a photo of how you "bent/tweaked the baffle a bit" on that #2 cylinder? I'd love to find an easy fix to getting the temps on this one down. Thanks.
 
Yep, size matters. In retrospect I should have installed the larger bypass duct on #3 (right rear) and the small hammerformed duct on #2 (left front).

Had the airplane down for a new cooler, so I adopted the idea of a large built-up bypass duct on #3, as presented by Mike Robinson. Didn't go quite as large. This one provides about 1/4" of space to pass air around the pinched no-fin area on the back of #3:

IMG_1233.jpg


The result appears to be a reduction of about 10F. It brings #3 into reasonable accord with 1 and 4. Still have a 24F spread with #2, but I think correcting that (if it really matters) will require reducing flow past that cylinder.

Here's a shot taken yesterday, preliminary testing, in a cruise condition which many would call abusive...roughly 30-50 degrees rich of peak. The cowl exit door is closed so outlet area is around 30 sq in, with an OAT over 60F:

Abuse%20Cruise%20Cooling.jpg


Now here's what I found really interesting. OSH is in three weeks, meaning we all get to drag it in from Ripon at 90 knots. With a new oil cooler and a baffle tweak I just had to check slow flight cooling performance with the cowl exit door open. It was about 80F OAT down low, maybe a little more. I think this will work just fine:

Ripon%20Approach%20Cooling.jpg


Total CHT spread is now 9 degrees, and #3 is no longer the warmest cylinder.
 
Last edited:
Had the airplane down for a new cooler, so I adopted the idea of a large built-up bypass duct on #3, as presented by Mike Robinson. Didn't go quite as large. This one provides about 1/4" of space to pass air around the pinched no-fin area on the back of #3:
Thanks for the experiment Dan. To see if I understand you correctly...

For those of use who have used a washer under the baffle connected to #3 to try and get air there, what we really need is not the 1/16" that the washer provides, but something more along the lines of 1/4"? Obviously 4 washers are probably not going to work here, but was your modification essentially trying to accomplish the same thing?

Also, hard to tell from the picture, but what did you use to seal the sides of the extra piece you riveted to the back of the baffle. It doesn't look like aluminum.

Thanks for the help, as I'm trying to do everything possible to get my #3 down at least 20F.
 
Last edited:
Just another data point from a 1999 Lycoming factory cylinder -

Cylinder-fin-flash_%28Small%29.jpg


Not only is there a "no slot area" as has been mentioned previously, but the adjacent fins are somewhat blocked by flashing from poor molds.

If a cylinder has a poor casting blocking airflow then the "add a washer" technique might be quite beneficial rather than the targeted "slot" technique.
 
Steve,

The new duct you see riveted into place has bent aluminum ends. They just have a lot of Permatex Ultra Black (same as Loctite 598) silicone smeared on them to seal the edge.

Back in post #10 of this thread I professed to not know the exact size needed for these ducts. I still don't, not really. The 1/4" I used here seems to be pretty close. However, there are variables like the fin differences on angle and parallel heads, differences in baffle fit and sealing, differences in inlet shape, well, you get the idea. Heck, look at the CHT values with my cowl exit door open and closed at different speeds and altitudes...with the same baffle!

The "problem" with placing a washer under the baffle at the AN3 bolt is that it spaces the tin away from the cylinder where there are no fins. That seems kinda useless, but you could squirt some silicone in there (black rectangle, post #1) to block the wasted flow....and clearly it has worked fine for lots of builders to date.
 
Not only is there a "no slot area" as has been mentioned previously, but the adjacent fins are somewhat blocked by flashing from poor molds.
If a cylinder has a poor casting blocking airflow then the "add a washer" technique might be quite beneficial rather than the targeted "slot" technique.

Either approach will bypass the flashing and shallow fin depth.
 
Last edited:
Dan, I take it you are not addressing the upper fins with your latest mod?

If not, it might be worth while to add something to draw the air through the upper fins on the way to your duct just to see if it is of any measurable benefit.
I recently duplicated "my" baffle mod on a buddy's 200 HP RV-7 to good effect. Would be nice to know if the upper portion of my duct is doing anything
 
Great! It will be nice to see the results.

One thing I've never understood is the lengths the automotive, fan cooled engines (VW, Porsche, Corvair) go to to push air all the way around the cylinders with very elaborate baffling while the aircraft engines simply go only half way. With the much more demanding duty cycle of the aircraft engine, it would seem they would benefit the most. And for all practical purposes, the cooling scheme is the same - i.e. high pressure on one side, lower pressure on the other. The car drives a fan to deliver this pressure differential, the airplane uses forward motion. Why are the baffles different?
 
Am going to duplicate your bypass duct Dan. Thank you for the good photos.
The washer has given me much improvement. I'm fascinated with how much cooling improvement is gained by just paying attention to the baffling details. Thanks to all for the info.
 
Parallel Valve - M1B

Here are some picts of my parallel valve. It appears to have some (small) depth for airflow. Do you think the either bump, or alternate method, or nothing needs to be applied to avoid temp imbalance?

Great thread BTW, very timely bump for me as the baffles are coming out of the box now.

IMG_0718.JPG


IMG_0717.JPG
 
One thing I've never understood is the lengths the automotive, fan cooled engines (VW, Porsche, Corvair) go to to push air all the way around the cylinders with very elaborate baffling while the aircraft engines simply go only half way.

The best (i.e. most recent) data for air cooled engine design is probably found with the SAE. I'm not a member and not curious enough to spend non-member money for SAE books.

Heat transfer from metal to air is governed by temperature delta. The air entering a full wrap baffle is at ambient temperature, so heat transfer from the hot metal is high. Further around the cylinder the air is now much warmer, and heat transfer is lower. The result is less and less cooling as the air proceeds further around the cylinder.

There's an early NACA work illustrating the issue. See Table 1 and Fig 2:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20090014186_2009013404.pdf

It is interesting to note that our stock baffle scheme is something like Fig 2a; small exit radius, close entrance width, no exit duct...the worst configuration tested in terms of uneven cooling and energy loss.

To be fair the deformation of the cylinder from perfectly round to some unknown shape may be minimal. I don't know. It should also be noted that there is a difference between cooling cylinders and cooling cylinder heads. Obviously the fin area is shifted toward the exhaust side, and aluminum has a far higher ability to transfer heat by conduction as compared the steel barrels.

Like you I'm tempted to try full wraps. I have hesitated because of economics, not lack of curiosity; IO390 cylinders are very expensive. I've mostly concentrated on extreme sealing and a bit of airflow control to the regions left lacking. I do have partial wraps on the barrels between cylinders, thus each cylinder has symmetrical airflow.

Some of the EZ bunch have done radical cooling experiments. You might look there.
 
Here are some picts of my parallel valve. It appears to have some (small) depth for airflow. Do you think the either bump, or alternate method, or nothing needs to be applied to avoid temp imbalance?

Again, nobody really knows exactly how much clearance is optimum. Field experience suggests that it's more than Lycoming has provided. Would I build in some additional room for airflow to the neglected bottom fins? Yes indeed.

Keep in mind that you want to keep the air between the fins all the way to the exit on the bottom of the cylinder. It's common to see sloppy baffle wrap fit.

Here the air just leaks out the side (small arrow) rather than proceeding all the way to the base of the head. The factory's positioning of the rubber anti-vibration comb didn't help at all:

Wrap%20Leak%202.JPG


Sometimes your engine shop contributes to the stew. Here they helpfully added a blob of orange silicone so the center baffle plate would not wear against the fin tips. As a result air flow to the preferred exit is compromised (big arrow), encouraging it to instead flow out the open side (circled arrows):

Wrap%20Leak%201.JPG
 
Last edited:
Again, nobody really knows exactly how much clearance is optimum. Field experience suggests that it's more than Lycoming has provided. Would I build in some additional room for airflow to the neglected bottom fins? Yes indeed.

Keep in mind that you want to keep the air between the fins all the way to the exit on the bottom of the cylinder. It's common to see sloppy baffle wrap fit.


Sometimes your engine shop contributes to the stew. Here they helpfully added a blob of orange silicone so the center baffle plate would not wear against the fin tips. As a result air flow to the preferred exit is compromised (big arrow), encouraging it to instead flow out the open side (circled arrows).

9bhf79.jpg

I've seen an engine shop simply use a piece of this stuff -

http://www.aircraftspruce.com/catalog/appages/ams3320siliconebaffleseal.php

bonded to the baffle to both prevent leaks at the edges and prevent wear. Spruce says it's good to 550F so contact with the cylinder should be OK.
 
I've seen an engine shop simply use a piece of (silicone baffle seal) bonded to the baffle to both prevent leaks at the edges and prevent wear. Spruce says it's good to 550F so contact with the cylinder should be OK.

Yep. The lower sections of my aluminum baffles are fully lined, so I don't depend on the aluminum to keep air between fins. The aluminum wraps are not much more than a backup to ensure the fiberglass/silicone wraps stay in place. Best I can tell none has shown any sign of debonding.

This is Loctite 598 (aka Permatex Ultra Black) rolled into ordinary 9 oz plain weave fiberglass between plastic sheets, then cut to size with scissors and stuck to the fins while wet.

P1250001.JPG


P1250003.JPG
 
Last edited:
Yep. The lower sections of my aluminum baffles are fully lined, so I don't depend on the aluminum to keep air between fins. The aluminum wraps are not much more than a backup to ensure the fiberglass/silicone wraps stay in place. Best I can tell none has shown any sign of debonding.

image clip

Wow, great minds... I used baffle material similarly between the baffle and fins. It is effectively self "bonded" in there to the fins from the heat. Of course mine aren't nearly as pretty as this powder puff :).
 
...There's an early NACA work illustrating the issue. See Table 1 and Fig 2:...

Good stuff Dan. Fig 2(i) appears to have the best design from a TLAR perspective, and the the numbers seem to agree. This would be fairly easy to accomplish around the cylinder barrels, but the heads are going to require some thought. In any case, I think that there is a LOT to be gained without much additional work.

The "problem" I'm running into however, is that every improvement in cooling "power" also starts begging for a variable geometry cowl outlet. As I stated before, my most recent flight started out at 120 degrees on taxi, and was a still very warm 71 degrees at 8500 - yet I had two cylinders under 295 in cruise. Thats a bit too cool for best efficiency.
 
As I stated before, my most recent flight started out at 120 degrees on taxi, and was a still very warm 71 degrees at 8500 - yet I had two cylinders under 295 in cruise. Thats a bit too cool for best efficiency.

Michael, what numbers are you wanting to see there for best efficiency??
 
Well, it's a heat difference machine, so hotter is better as long as the materials can survive. And I understand that the engine is quite happy at 350-375. So that's my target.

Of course, you don't want a "false" temp reading by choking off portions of the head - you want the right airflow at all points. So that's what is driving the quest for good overall airflow through the system, then choke down the whole thing to regulate the temps. If that makes any sense.
 
Yep. The lower sections of my aluminum baffles are fully lined, so I don't depend on the aluminum to keep air between fins. The aluminum wraps are not much more than a backup to ensure the fiberglass/silicone wraps stay in place. Best I can tell none has shown any sign of debonding.

This is Loctite 598 (aka Permatex Ultra Black) rolled into ordinary 9 oz plain weave fiberglass between plastic sheets, then cut to size with scissors and stuck to the fins while wet.

2zxur5v.jpg


14bnwn6.jpg

Can we see a pic of the top side?
 
Can we see a pic of the top side?

Not much to see. The wraps stop at the 3 and 9 positions. Wraps on top (say 3 and 9 up to 1 and 11, like the ones Don posted) are kinda low on my experiment list. Next is probably cleaned up cylinder baffle exits and the addition of exit ducts.
 
After seeing the pictures of the Formula One and Dan's fin sealing I think to myself what a great idea and probably worth doing. I then realize that if the aluminum baffle around it doesn't seal to the fin seal then some of the air will be between the seal and the baffle. This would prove to be even less productive on the wrong side of the seal/tape. So I guess I will stick with the old plan rather than having to spend that much effort right now. I have the taper fin ECI cylinders and getting a good seal there will be an effort.

Looked at a Piper last week and could not believe how poorly the baffles were sealed. There was a 3/8" gap all the way around the rear crankcase.
I guess they make up for poor sealing with large inlets and poor speed.
 
I then realize that if the aluminum baffle around it doesn't seal to the fin seal then some of the air will be between the seal and the baffle. This would prove to be even less productive on the wrong side of the seal/tape

Little bead of silicone eliminates that leak.
 
I was thinking a soft dam of some kind but a bead of silicone would make a perfect and easy soft dam. Thanks.
 
Inlet and outlet gap sizing

While thermodynamics is not my primary area of expertise, and I have yet to purchase, let alone baffle, an engine, one thing that occurs to me as I view the baffles pictured that are adhered to the outer ends of the fins is that the width of the gap between the baffles on the inlet and outlet sides of the cylinder may (or may not) be critical.

The direct contact between the baffle and the fins, keeping all the air in contact with the sides of the fins is certainly the most efficient manner to transfer heat at any given airflow. However, the total airflow is an important factor also, and the subject of this thought.

I suspect that there is an optimum gap width between the baffles, measured along the fins, that will result in the optimum airflow along the fins. As the gap is made wider, thus the length of the fins in contact with the baffle being less, the airflow will increase due to the area of the inlet and outlet being larger, and the pressure drop, or friction loss as it's sometimes called, of the air passing through the passages formed by the baffles and fins being less due to the reduced length of said passages. However, the cooling efficiency at a given airflow will be less due to the larger percentage of unbaffled fin length. At the other end of the spectrum, as the gap gets smaller, the airflow will decrease due to the smaller inlet & outlet, and the increased length of the passage between fins and baffle resulting in a larger pressure drop through the passages at a given flow rate. Somewhere between very wide and very narrow gaps, I would expect to find an optimum dimension of gap width on the inlet (top) side and the outlet or bottom side. The optimum might occur with different widths on the inlet and outlet sides, maybe slightly larger on the intake side. Or vice versa, due to the air expanding on the way through.

'Twould make an interesting development project, if it has not already been researched.
 
Last edited:
Bet it's been done

Oldsam,
I'll bet some F-1 driver has done the math. I was also thinking that there must be a perfect radius for the inlet and outlet portions of the cylinder wrap. There was a post by Paul Lipps, as I recall that showed how he set his cylinder wraps up.

BTW, Dan - fantastic thread!!!
 
Well, it's a heat difference machine, so hotter is better as long as the materials can survive. And I understand that the engine is quite happy at 350-375. So that's my target.

Of course, you don't want a "false" temp reading by choking off portions of the head - you want the right airflow at all points. So that's what is driving the quest for good overall airflow through the system, then choke down the whole thing to regulate the temps. If that makes any sense.
Yes, thanks Michael. :)
 
Back
Top