VAF Forums

VAF Forums (
-   RV General Discussion/News (
-   -   Hot #3 CHT - Solved! (

Toobuilder 12-08-2012 09:43 PM

Hot #3 CHT - Solved!
So in the course of changing out my injector lines, I decided to attack my #3 cylinder temperature issue. This cylinder has always been my "problem child" on climb and also in cruise. It typically runs hotter than my coolest cylinder (#4) by a wide margin (usually about 70 - 80 degrees in the summer) in cruise. Always. Anyway, I decided to deconstruct the rear baffle and see what I could do to direct some air through the fins. Then it struck me... With the Vans supplied baffles, there is no way air can get from the top of the cylinder to the bottom on the intake side of the head because the fins terminate at the equator of the cylinder. The wrap around baffle on the bottom has essentially no supply of air. So not only is there no airflow around the bottom, the top fins are simply radiating heat into stagnant air. So I figured we needed to get some air moving past the top and bottom fins to see if the temps would come down. I fabricated a plenum to get the air moving in an organized fashion and attached it to the existing baffle.

Long story short, I finished it up this afternoon and was blessed with good flying weather. I decided to climb to my normal cruise altitude of 7500 feet to see if the mod did anything. I tried an agressive climb to stress the cooling and was greeted with the typical #3 cylinder behavior of leading the numbers. Great! All that work for no gain. But as I pushed over into cruise, the temps started to fall. And wouldn't you know it, the problem child kept falling - eventually stabilizing at a temp only a few degrees warmer than my coolest. So this one fairly simple mod brought my hottest cylinder all the way down to the second from coldest. Now my hottest is #2 - which also features the same airflow blockage by virtue of the Vans baffles. Time to work on that one and get it into line with the others.

Greg Arehart 12-08-2012 10:16 PM

Good Deal!


Mike S 12-08-2012 10:18 PM

This problem has been discussed here quite a lot, usual fix is a washer between the baffle and cyl.

Dan H. did a mod similar to yours, but nowhere near as large----actually he did a couple using different methods..

As long as it is working, all is good :D

Toobuilder 12-08-2012 10:26 PM

The washer trick can't move enough air, IMHO. A washer also does not allow any control over where the air is going. It simply makes it "less" stagnant back there.

DanH 12-09-2012 08:32 AM

Mike's approach has a subtle difference compared to the short ducts I outlined for bypassing the no-fin-depth head section. In Mike's scheme air must first pass between the fins on the upper intake quadrant in order to follow the bypass duct to the lower fins. Between fins is always superior to willy-nilly flow in the general vicinity of the fins. I suspect if you measured the air temperature at the baffle exit below the cylinder it might be a bit warmer than with the short duct under identical conditions, meaning it picked up more heat. That is the goal.

Previous discussion here:

Mike's approach on the right. My own #3 runs abut 25F warmer than the others. The little hammered duct might be on the small side, plus it gives up some heat transfer from those upper fins. I've just been too lazy to fool with it.

Toobuilder 12-09-2012 09:12 AM

Thanks Dan for the illustration. And yes, pulling air through the top fins was one of the objectives. Though it was "fiddly" to make and fit to a flying airplane, it was very effective. On yesterday's flight, I would have expected to see that cylinder 50 degrees warmer than it was.

I am more than a little surprised how often you see baffling like this. I know it is up to the airframe designer to come up with the cooling system, but damming up the airflow on #3 (and #2, for that matter) seems like a significant oversight. Seems like the Z bend in the baffle should be lower on the cylinder so the bottom bend is not adjacent to the "zero fin depth" region. Following the plans, the lower shroud serves no purpose... In fact, you would likely be better off removing it so all the fins could at least radiate.

Also, should have checked to see if this had been discussed on the boards before... I see that you have thoroughly illustrated the issue in the other thread (at least I got to bask in the glow of my own genius for a day...). :)

Toobuilder 12-09-2012 01:47 PM

Just got back from a longer flight and it looks like #3 and 4 are matched within a degree or two. 7500 feet, 51 OAT and 50 LOP I saw 3 and 4 stabilize at 299 while 1 and 2 showed 310 and 318 respectively. Pretty sure that once #2 gets some airflow to the bottom fins it will come into line as well.

RV10inOz 12-09-2012 03:20 PM

Good result!

greghale 12-10-2012 04:14 PM

hot #3 CHT
I was recently down at Gami injectors to have my fuel flows evened up. They include a baffle kit for the turbo bonanzas that includes your mod for the rear cylinders. They found directing the air around the back cylinders of the engine, dropped the CHT down. I plan on doing the same thing while I am installing cowl flaps on my RV10 -- winter project...

Toobuilder 12-16-2012 05:27 PM

OK, decided to take a crack at getting some air around my #2 cylinder ( now my hottest, as reported). Since I have the inlet mounted servo snorkel, I don't have much room to put a large plenum like on #3. So my plan was to go extra wide to make up for the very narrow "air inlet" which as you can see is simply 4 of the existing rivet holes enlarged to 1/2 inch. Capturing the air is a simple wedge shaped .6x4x2.5 inch plenum riveted to the front side of the existing baffle. A simple window was cut to allow an exit for the collected air. Also, I completely removed the "baffle dam" in front of the head fins since there is no way for the air to go "down". At least the in rushing air will have a better chance of carrying off heat this way.

Flight test today revealed that the mod works well, as this is now my coolest cylinder by 10 degrees. I would have expected to see 320 - 315, but it was down to 290. So overall I'm pretty happy with seeing all 4 cylinders under 300 in cruise. Heck, I might need to start necking down the air inlets if this keeps up!

Toobuilder 06-23-2013 04:46 PM

....Sure, but how does it work in the Summer?
OK, its now six months later and I had a chance to really check this mod out today. I left Phoenix at noon with the temp passing through 105. On my normal stair step climb to the west staying clear of the class B sections, the CHT often runs right past 400 on a few cylinders within minutes, but not today. I maintained WOT and best power EGT for the entire 15 minutes it took to get out from under the PHX airspace and I never saw anything hotter than 385. Once to 8500 feet and LOP, the temps settled to 325 -350.

I know that 1 data point does not make a trend, but my first "hot" flight of the season sure went well.

Soon, I will try a climb at VY all the way to altitude and see how it behaves.

alpinelakespilot2000 09-24-2015 04:20 PM

With some help from Michael Robinson (Toobuilder) and Pete Howell, I finally got around to attacking my hot #2 cylinder. It was always the hottest in climb at full throttle, and also a bit hotter than the rest in cruise. I had already tried the washer trick on both #2 and #3, but neither seemed to help that much. Given that lack of success, I pretty much followed Michael's approach this time around...

Cut a bypass hole in my #2 baffle to allow air to get around the lack of fins:

Built a plenum to funnel air into that bypass:

The end result at 65% cruise? Averaged over a few flights before and after the mod...
Before plenum, #2 ran just slightly hotter than #3.
After plenum, #2 runs 25-35F cooler than #3.

Before plenum, #2 ran about 35F hotter than #4 (traditionally my coolest).
After plenum, #2 runs about 5F hotter than #4.
The long and short, it appears that the plenum cooled #2 by about 30F. In addition, though I don't have hard data for climb temps, #2 used to be my limiting CHT by a fair margin. Now #2 and #3 keep pace as my hottest, but it is now easier to climb WOT than before.

Now, I need to attack my #3. Michael's solution there is a bit more complicated to picture, and to fabricate, but definitely is at the top of my "to-do's".

Toobuilder 09-24-2015 06:19 PM

Glad this worked out for you Steve. I have performed this mod to my Rocket and experienced similar results. I have also developed a much simpler plenum for the rear cylinder which is just as effective as the one shown at the start of this thread. I'll send you the drawing and/or publish it here. If you have not started cutting metal yet- wait until you see my new one. It will save you a bunch of time.

SeanB 09-24-2015 07:09 PM

Keep it coming

All this is very interesting. I believe a lot of us would like to see and could possibly benefit from your latest rear plenum idea. Hope you decide to post it on this forum.

Steve, your data is also appreciated.


BillL 09-24-2015 08:18 PM

Here is what I did for the rear head. When I did all this, I had not seen this thread!

Head Side

A plot of areas show this will provide all the air needed for the lower fins.
Edit: 125 hrs and tons of temperature data over many test types say this works. 5.22.20

Toobuilder 09-26-2015 01:18 PM

Bill, that's really clean and a huge improvement over the "washer trick". Mine goes one step further however, with particular emphasis on pulling air through the upper fins. My line drawing is at the bottom, and here are the installed pictures of the new plenum. Notice the upper section which hugs the upper fins.

If you look at the above picture and compare it to Dan's line drawing in post#5 you can see where I got my inspiration for the new design. Thanks Dan!

This plenum sits on the “flat” part of the head, with extra material to make the wrap sections on top and bottom.

Keep in mind that solid lines are “cut” and dashed are “bend”. Also, the bend order that works well is:
1. Sides
2. “back kink”
3. Flanges.

DanH 09-26-2015 01:52 PM


Originally Posted by Toobuilder (Post 1017094)
Bill, that's really clean and a huge improvement over the "washer trick".

All of Bill's work looks like that.


If you look at the above picture and compare it to Dan's line drawing in post#5 you can see where I got my inspiration for the new design.
Your idea. I just do pictures.

Speaking of pictures and ideas...check out these from a certified Grumman Tiger. Yep, that's good 'ole #3. It's not like we invented any of this. It's a well known problem and fix:

Toobuilder 09-28-2015 09:16 AM

Notice that post #16 is updated with my drawing for the "new" plenum.

Greg Arehart 10-27-2015 11:47 AM

I finally got around to shamelessly stealing these ideas for my hot #3. Made a version of Bills modification and it took ~30-40 degrees off my CHTs on the test flight this morning. Now my #3 is just a few degrees cooler than the #1 & #2 and about the same as #4.

Absolutely recommend this modification for anyone with #3 problems - took me about two hours to do it, including removal and reinstallation.


Toobuilder 10-27-2015 11:53 AM

My hope is that Van will eventually come around and "shamelessly steal" this idea.

...And in one fell swoop fix most of the cooling problems these airplanes have.

lakespookie 09-25-2019 11:58 AM

You should submit it to vans or tag one of their accounts to this post.

B Cunningham 12-21-2019 01:10 PM

I wish all of the pictures could be reposted with a host that actually shows them...:mad:

David Carter 05-22-2020 07:39 AM


Originally Posted by B Cunningham (Post 1394370)
I wish all of the pictures could be reposted with a host that actually shows them...:mad:

I agree! Especially since this thread is linked from the Engine Cooling sticky.

gmcjetpilot 05-22-2020 08:50 AM

This is like the washer trick to hold the baffle off the fins to allow more airflow around the back and lower part of the #3 that has been around for about 30 years or more, but like this channel design..... More design... and hey it works. It could be taken further. I think some composite shapes that fit better and guide air better with less leaks could be the ticket...

All times are GMT -6. The time now is 08:07 AM.