What's new
Van's Air Force

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

High lower cowl pressure... Baffle leaks or not enough exit area?

agent4573

Well Known Member
Tidying up the final adjustments after phase 1 and still running into some higher temps than I would like. We've gotten the cht's below 400 in cruise but it's only on mild days (75 on the ground, 35-45 at altitude). We would like a little more buffer for when it gets hotter. Any suggestions welcome.

Taken at 5580 pressure alt:
At 158 kias - 16" total, 8.3" delta
At 140 kias - 12.3" total, 6.7" delta
At 120 kias - 9.4" total, 5" delta

Taken at 8240 pressure alt:
At 157 kias - 14.7" total, 7.9" delta
At 130 kias - 10.1" total, 5.5" delta
At 100 kias - 6.1" total, 3.25" delta

These show that the lower cowl has between 3 and 8" of pressure over the outside static pressure. We've verified by taking direct lower measurements instead of using the difference, and they confirm up to 8" of water on the lower cowl. Still need to get a temp guage I can put up there.

Sam James cowl with an IO375. Oil temp is 180-190 in cruise. Need to verify the timing still, but since it's electronic I need a night engine run in order to see the timing light flash. It should have been set at the builder but I want to verify. All 4 cylinders are within 5-10 degrees of each other. Will trim the air dams later to get the spread down, but for now would like to tackle finding 30 more degrees of cooling.
 
Paul,

Congratulations -- from the looks of Flightaware data you've got a speedy airplane on your hands :)

Check the front outboard and inside corners of the rings @ top & bottom cowl for any gaps (leaks). I keep finding these gaps on standard cowls and they're worth ~10°F CHT at least (more at higher IAS).

Does your ignition system spit out advance information? (Lightspeed Plasma II, III -- FlightDataSystems. PMag -- eicad, EIcommander, etc.)? You could check this on your EFIS/EIS...

Brian
(Bay Area Native -- missing the weather, and the food...)
 
The ignition displays it's programmed advance, but it's only truly firing at that angle if the magnet position is set correctly. There's a jig for installing the pickup magnets, but since it was done at Aerosport, I'd just like to verify everything is where it says it is since I haven't seen it firsthand.
 
Paul, what are you using as a static source for your measurements? And could you describe the probe used for the plenum pressure?
 
Last edited:
Your numbers suggest not enough exit air out of the lower cowl.

Steve.

Steve, I think it appears that way because there is too much air bypassing the fins on the engine/cooler. The exit area of the James is quite adequate on the 7, when using a radius down to the firewall/lower skin exit area.

I would suggest sealing the lower intersection baffles, they are real leakers. Wrapping the barrels up even with the crank center line, and ensuring no other bypass leaks. The upper chamber should increase a little, delta increase some more, and the exit will drop.

What oil cooler is being used?
 
Last edited:
Upper port is flush mounted to the outside of the baffle where a 1/4" screw normally goes. In theory, since it's flush mounted to a wall, it should be picking up (nearly) static pressure only. Lower port is on the gear leg, end of tube sealed and small holes drilled through the side of the tube, very similar to a piccolo tube except plastic.

Absolutes were measured relative to cabin pressure. At the conditions, we opened the alt static valve and measured a 200ft rise in indicated altitude.eveything was done at 29.92 to keep it simple. This is equal to 0.2mmhg, or just over 0.1in h2o. Since my lower cowl pressure is reading almost 2 orders of magnitude higher than that, I'm ok ignoring the 0.1 inches for now until I find the larger issue.

I am curious about 2 large gaps I know of, and I'm not sure what's supposed to be done about them. Between the front cylinders exhaust port and the rear cylinders intake tube there a 1/4" gap between the head fins. This continues outward above the primer port as well. The gap is too big for rtv alone, so how are people sealing this area? This seems like a massive area for air to exit without passing over fins, just never saw anyone talking about it before.

2006X oil cooler, mounted to the engine mount. We had a large leak on that seal and just redid it. It dropped the oil temps 20 degrees and helped with the cht's. All the measurements posted above are from after fixing that leak though. I do not currently have the round flow accelerator fabricated or installed. I've seen a few posts about them, but got the idea they were more for reducing drag than helping with cooling.

Also, all temps were done 50 degrees rich of peak.
 

Attachments

  • PXL_20220425_001225032.jpg
    PXL_20220425_001225032.jpg
    402.6 KB · Views: 400
  • PXL_20220425_001237575.jpg
    PXL_20220425_001237575.jpg
    418.7 KB · Views: 370
  • PXL_20220425_001313336.jpg
    PXL_20220425_001313336.jpg
    464.6 KB · Views: 375
  • PXL_20220425_002605417.jpg
    PXL_20220425_002605417.jpg
    348.5 KB · Views: 398
Last edited:
Tidying up the final adjustments after phase 1 and still running into some higher temps than I would like. We've gotten the cht's below 400 in cruise but it's only on mild days (75 on the ground, 35-45 at altitude). We would like a little more buffer for when it gets hotter. Any suggestions welcome.

How well is the airbox cowl inlet to airbox sealing? A poor seal there pushes air into the lower cowl... with result of less delta P between upper and lower cowl.

Seems to be a common area for a poor seal....
 
2006X oil cooler, mounted to the engine mount. We had a large leak on that seal and just redid it. It dropped the oil temps 20 degrees and helped with the cht's. All the measurements posted above are from after fixing that leak though.

If the oil cooler is mounted to the engine mount, how well does your seal work when the engine thrust causes the engine to shift (a bit) forward? The Lord mounts will give a bit.

If not baffle mounted, then typical (and mine) have a FW mounted cooler fed with Scat tubing, so no issue with that flexing.

And, if your oil temps are down 20 degrees, do you need to mask off some of that oil cooler? If you did, that'll push more air into the cylinders, and reduce the lower cowl pressure.
 
Ok, let's accept the numbers as accurate. You have enough deltaP at all airspeeds. Congratulate yourself, because being left with plenty of lower cowl residual pressure after the baffle drop is a great place to be.

Read what Bill said again. Your task going forward is to increase heat transfer...the quantity of heat being carried away by the existing airflow. The goal is to increase the temperature of the exit air. Make evey single molecule flow close to the hot parts. A temperature probe in the cowl exit, shielded from radiant heating, tells a lot.

I'm on the road right now. I'll take a closer look at your data when I get home.
 
Be anal about sealing any leaks.

Here is some test data on my homemade plenum, James cowl, M1B stock RV7 with the exit fairing at firewall, and 4-into-exhaust pipe. I wrapped every square inch of fin that was covered by the stock baffles, plus on the barrel fins up the 1.5 inch above the crank plane. My stock intersection baffle between the cylinders on each side were huge leakers. They will fool you - my 10 friend said his were sealed well, and he is stickler, but when illuminated it showed it was not. So if it can happen to him . . .beware and careful there.

My static reference was to the static port. I have lots of data from PH I but this is one for comparison to DanH's standard test plan. And, is after I opened up my to-of-barrel wrap gap and added some flow entry straighteners. There are chutes for the front of #1 and back of #3 where the fin length goes to zero at the crank plane.

The oil cooler is stock little Niagara, but has shutters. Test data was the same with shutters open or closed.

I might call this installation extremely sealed, but not extremely restricted.

The 1/4" gap you showed between the heads, I still have them.

IO375 - -is is that a tapered fin barrel? Do you have picts of how you attached the baffles showing under the plenum?

Edit01: I looked at your website - you are using SDS Electronic ignition - to check it you could set mix, cruise, WOT, 2500 rpm, then after oil temp and speed is stabilized, just retard the timing 0.5 deg at a time and wait 2 min and watch the KIAS, do this until the speed drops. Speed reacts pretty fast to power change, CHT takes a few more minutes. A test cell would be 10 min, so just do the timing and stop when you see the speed drop a knot. Then wait to plot the CHT. This will ensure the timing is "correct" and is/not part of the issue. I think it is not the primary cause and until you can show .5 in-h2o in the lower cowl I would not look at timing at all.

WHL Plenum Test Data 2.png
 
Last edited:
The 1/4" gap you showed between the heads, I still have them.

IO375 - -is is that a tapered fin barrel? Do you have picts of how you attached the baffles showing under the plenum?

I'll take some pictures tonight. I downloaded the ECI taper fin baffle template and made my own set that sits pretty well. I RTV'd the edges so even where there is a little gap between the fin and metal, the RTV stops any air from escaping before it supposed to.

Ok, let's accept the numbers as accurate. You have enough deltaP at all airspeeds. Congratulate yourself, because being left with plenty of lower cowl residual pressure after the baffle drop is a great place to be.

Read what Bill said again. Your task going forward is to increase heat transfer...the quantity of heat being carried away by the existing airflow. The goal is to increase the temperature of the exit air. Make evey single molecule flow close to the hot parts. A temperature probe in the cowl exit, shielded from radiant heating, tells a lot.

I'm on the road right now. I'll take a closer look at your data when I get home.

Which temp sensors have you used in the past? If there's a decent setup I can purchase instead of hobbling something together I would prefer that route.
If the oil cooler is mounted to the engine mount, how well does your seal work when the engine thrust causes the engine to shift (a bit) forward? The Lord mounts will give a bit.

If not baffle mounted, then typical (and mine) have a FW mounted cooler fed with Scat tubing, so no issue with that flexing.

And, if your oil temps are down 20 degrees, do you need to mask off some of that oil cooler? If you did, that'll push more air into the cylinders, and reduce the lower cowl pressure.

We put an electronic shutter on the oil cooler opening, so we can control airflow from the cockpit. The oil temp used to be ~210 in flight, now its ~190. We can close the door down a bit and let the oil run at 210 and it does get me a little bit better CHT on cyl 3, but nothing drastic, maybe 5 degrees drop in CHT for a 20 degree rise in oil temp. I will try to get a video camera setup to see if there's a gap forming under power on the oil cooler seal. I'm not particularly happy with the engine mount setup, but I also have no room on the firewall for the cooler since I had to mount all the SDS sensors there.

How well is the airbox cowl inlet to airbox sealing? A poor seal there pushes air into the lower cowl... with result of less delta P between upper and lower cowl.

Seems to be a common area for a poor seal....

The sam james uses a piece of neoprene between the inlet and airbox with hose clamps. I assume there's no leakage there, but I'll take a second look to make sure.

I'm going to do an experiment later this week as well and remove the air dams from the inlet. Everyone has said they fly with them cut down or removed, but my cylinders are fairly balanced and I have the full height dams still. I'd like to see how things react with them removed just to get an idea, but it doesn't seem like the plenum area is really the problem, more with sealing and exit flow. There's also posts that highlight a large exit area difference between the injected james cowl and the carb'd james cowl. I didn't open up the exit at all during cowl fabrication and have the smaller exit area cowl with a 2 pipe exhaust. I'll see if I can fab up the quarter round flow accelerator this week and get it installed before next weekend. I'm waiting on new prop governor cables from Mcfarland that won't be in until Wednesday, so I have some time until then.
 
Hang in there !!

"The sam james uses a piece of neoprene between the inlet and airbox with hose clamps. I assume there's no leakage there, but I'll take a second look to make sure."

I assume this as well, but have not tested - I don't use clamps just the stretch tension of the neoprene. My axial gap is only 3/16".

"There's also posts that highlight a large exit area difference between the injected james cowl and the carb'd james cowl. I didn't open up the exit at all during cowl fabrication and have the smaller exit area cowl with a 2 pipe exhaust."

I may be one of a few that had the 6/7 - and James O & IO cowls on the floor at the same time. The 6/7- James IO all have the same exit profile. It is the "O" that drops an additional inch. The outlines were traced and all the areas were measured. Just keep sealing, that will bring down the lower cowl pressure.

One area we have not addressed and that is the gap behind the spinner. Dan sealed his, I did too (as do many rockets) , but it is only an issue if there is excess lower cowl pressure. This applies to Vans and James cowl.
 
I was looking at sealing the spinner gap. I have a 2.5" prop extension so it gives me an easy surface to seal to. I may look into that more. Any chance I can get a copy of that excel sheet you posted? Sent you a pm with my email. I drop this off for paint in about 2 weeks, would like to know if I have to trim the lower cowl before I paint it, but it's not the end of the world if I have to trim it afterwards.

EDIT: Also, we checked the timing last night. Its set to 25 degrees advance and is right on the money. I haven't started doing the lean-of-peak fuel and ignition curves yet, so for now everything is being measured at 25 degrees advance.
 
Last edited:
Intersection baffle?

My stock intersection baffle between the cylinders on each side were huge leakers.

Are you talking about the factory parts that mount under the center of the cylinders, suspended from an S hook?

If so, what did you do to keep the leakage down?

sam_8589_bd695e1490734544.jpg
 
I'm going to do an experiment later this week as well and remove the air dams from the inlet. Everyone has said they fly with them cut down or removed, but my cylinders are fairly balanced and I have the full height dams still.

If the cylinders are balanced, don't remove or cut the dams in front of #1 and #3. all that'll happen is you will cool the front cylinders more and the back ones less.

I HAVE FULL SIZE dams. There. Not EVERYONE says to cut them!
 
If the cylinders are balanced, don't remove or cut the dams in front of #1 and #3. all that'll happen is you will cool the front cylinders more and the back ones less.

I HAVE FULL SIZE dams. There. Not EVERYONE says to cut them!

I made mine removable since everyone suggested they would need mods. Once I pull the plenum it's 3 screws per side and they come in and out. I'll pull them just to see the affect, but it won't be hard to reinstall to my current setup. It's great to hear there are some people running them fullsize though and I'm not the single weird setup.

Attached are the pics of the taper cylinder baffling. It's hard to get a clean shot of em. Also, I made and installed the exhaust flow reattacher/accelerator thingy. It's hard to tell from the angle, but it only protrudes about 1/16" below the bottom skin, and wraps around to between 1/4-1/2" from the firewall.
 

Attachments

  • PXL_20220426_034818680.jpg
    PXL_20220426_034818680.jpg
    334.7 KB · Views: 372
  • PXL_20220426_034803413.jpg
    PXL_20220426_034803413.jpg
    418.7 KB · Views: 404
  • PXL_20220426_034256797.jpg
    PXL_20220426_034256797.jpg
    310.3 KB · Views: 409
Last edited:
Which temp sensors have you used in the past? If there's a decent setup I can purchase instead of hobbling something together I would prefer that route.

Easy as pie. Buy a few LM34AH-ND sensors. Solder one to the end of a three wire cable. I use three wire shielded just to have the shield braid serve as armor. Make it long enough to move anywhere in the engine compartment and reach the cockpit.

Output is in mV. All you need is a digital voltmeter. If the display says 2.5V, temperature is 250F. Just move the decimal point two places.

Genuine Texas Instruments sensors look pricy these days, but Ebay offers lots of cheaper copies.

If measuring exit temperature, work hard to ensure the sensor sees only air temperature, not radiant or conducted energy. That means good radiant shielding which re-radiates at a low level, and does not conduct via the mounting. The mount in the photos is glass/epoxy covered in fiberfrax under foil tape.
.
 

Attachments

  • LM34 Diagram.jpg
    LM34 Diagram.jpg
    12.6 KB · Views: 264
  • IMG_0202.jpg
    IMG_0202.jpg
    100.3 KB · Views: 337
Did round 3 on baffle sealing today. Used my yet to be installed led strip lights to fill the plenum. Found a few small pinholes around the inter cylinder baffles, and a large gap on the #4 taper fin cylinder baffles. Sealed them all up with RTV. Laid up a glass angle for the temp gauge mount. I ordered a bulb type mechanical water temp gauge. Plan is to use glass/epoxy mount with fiberfrax and aluminum tape and mount the bulb in an adel clamp held in place on a long stainless bolt. It seemed slightly easier than what dan recommended because I don't have to find a 5 volt source and it has a built in gauge. I'll "calibrate" it in boiling water to see how off it is before the install. Should have more flight test data Thursday evening.
 

Attachments

  • PXL_20220427_023718592.jpg
    PXL_20220427_023718592.jpg
    231.1 KB · Views: 326
  • PXL_20220427_031614896.jpg
    PXL_20220427_031614896.jpg
    361.4 KB · Views: 250
Paul,

I see you have the James plenum as well. The center section that molds over the engine case has been reported (by several) to lift in flight and leak. One might take some plastic and seal the upper fins and pressurize that plenum with a backpack blower to see if that is happening. or . . .

Maybe a some tape with good adhesion, or a thread(s) be used across that area as a tell-tale to see if it is lifting. If it is, then some structural mods would be in order.

I have a plan to do that on mine to see how chaotic the flow is from my baffle mounted oil cooler. Someday . . .
 
It seemed slightly easier than what dan recommended because I don't have to find a 5 volt source and it has a built in gauge.

Bulb sender should work, but for the record, the LM34 works with anything from 5 to 20 volts, and the meter is just a handheld VOM, something we all have in the toolbox. It's nice to be able to position it anywhere with a long wire.
.
 

Attachments

  • 2012-11-10_16-44-54_99.jpg
    2012-11-10_16-44-54_99.jpg
    360.8 KB · Views: 248
Paul,

I see you have the James plenum as well. The center section that molds over the engine case has been reported (by several) to lift in flight and leak. One might take some plastic and seal the upper fins and pressurize that plenum with a backpack blower to see if that is happening. or . . .

Maybe a some tape with good adhesion, or a thread(s) be used across that area as a tell-tale to see if it is lifting. If it is, then some structural mods would be in order.

I have a plan to do that on mine to see how chaotic the flow is from my baffle mounted oil cooler. Someday . . .

One of the tests for this weekend is to seal the edges of the plenum with tape and see if the numbers change significantly. It doesn't show any leaks on the grounds, but yeah, in flight it might be a different story.
 
Mike, since you asked, the photo is marked where it should be touching the fins, and every other gap should be sealed. The areas at the case and under the finless barrel section is the usually the worst. It is hard to reach, but making an extension for the RTV tube in a caulking gun can reach.

Screen Shot 2022-04-27 at 8.10.51 AM.png
 
The other common leak is the rear baffle wall, at the short un-finned section at the base of the cylinder.
 
If I was a betting guy I’d have made a lot of money re finding air leaks when customers have said they have sealed baffles fully:D
In addition to areas that very knowledgeable people like DanH have mentioned there is another method.
Use a flashlight to find unsealed areas around the baffles, you’d be surprised how many leaks are still there after convincing yourself that you got them all. Have a friend help by looking for the light while you’re shining it up from the bottom & back of the cylinders.
Btw - on the Lycs the #3 & #4 aft vertical baffles, where they meet the engine case is an area that is quite often missed, I have seen big gaps there.
 
All data today taken at 6500 indicated, 30.10 alt setting. Difference between cockpit and static port is 0.2" water. Plenum and cowl pressure readings are referenced to cockpit. Mixture was set to 50 rich of peak at WOT and left there for all other data points. 2400 rpm.

75% power, oat 47, 160 kias, 176 true, exit air 180 degrees, chts: 399, 403, 386, 409. Plenum 16 in H2O, cowl 6.9", delta 9". Oil temp 200.

61% power, oat 48 degrees, 140 kias, 153 true, exit air 175, chts: 369, 373, 361, 376. Plenum 12.3", cowl 5.4", delta 6.8".

49% power, oat 47, 120 kias, 133 true, exit air 180, oil 192, chts: 349, 351, 342, 356. Plenum 9.3", cowl 4.1, delta 5.2.

43% power, oat 48, 100 kias, 114 true, exit air 195, oil 194, chts: 346, 347, 344, 356. Plenum 6.7", cowl 3, delta 3.8.

In the pattern and on the ground, the cowl exit aire is 195-200 pretty steady.
 

Attachments

  • PXL_20220430_031129464.jpg
    PXL_20220430_031129464.jpg
    241 KB · Views: 181
Puzzling

The temperature rise from inlet to exit seems high, but at the same time the choked exit indicates that mass flow is also high. Assuming (for the moment) that the temperature measurements are accurately represent the air, then that is a huge increase in heat rejection over my M1B for a baseline.

It certainly could be more than one issue here. If it was simply too much bypass air, then the higher mass flow (indicated by high low cowl pressure) would/should be cooler - i.e. less temperature rise across the engine/exhaust pipes.

177 ktas at 2400 rpm means your engine is pretty strong. My 180 hp IO360 takes 2650-2700 rpm to get that. You have my data and the delta-T across the system is substantially less. This part alone still points to higher than normal heat rejection - - and going back to make absolutely sure that operational timing is correct.

This is my thinking at the moment, but mental processes will keep going. If it changes, I'll report. Something more than leaks or flow past fins is going on here. It could be heat rejection or exit temp measurement error . . . Or both.

edit- or pressure measurement error.
 
Last edited:
The timing light shows the timing is right on on the ground. it got windy this afternoon, but we'll go do another flight tomorrow and start retarding the timing a degree at a time like you said, to make sure we're at the least advance possible to still get full power.

The plane goes to paint next weekend, so I think I'm going to open up the lower cowl just a bit before I send it off. Nothing major, but 1/2" or so off the exit ramp just to give me a few more SQ inches of exit. I'm planning on doing the spinner seal when it comes back from paint, hopefully I can fit that in before Oshkosh.
 
Last edited:
Today's results?

Were today's results with or without the deflectors on the front cylinders?

How many HP is this engine supposedly?
 
Radiant Heat?

Oh, isn't that temperature bulb going to get radiant heat from the exhaust pipe? I think you need to shield it so it can't see the IR radiation.
 
Oh, isn't that temperature bulb going to get radiant heat from the exhaust pipe? I think you need to shield it so it can't see the IR radiation.

It's mounted between 2 reflective and insulated mounts to limit the amount of radiative and conductive heat it's receiving. Not sure if the pic shows it well, but it's fairly well protected. After engine shutdown, it almost immediately drops down below the minimum gauge reading as well, which means it isn't getting a lot of radiative heating from the engine or exhaust since that would be the only way to keep it hot after shutdown.

Were today's results with or without the deflectors on the front cylinders?

How many HP is this engine supposedly?

Today was with the full deflectors still installed. On paper it's 195. In reality it's probably 205.
 
OK, I had an optical illusion when I looked at the temp sensor pic. However... you are also measuring how much the air was warmed by the exhaust pipes / headers.

#4 is the hottest cylinder, so cutting / removing the deflectors on #1 and #2 are just going to steal more air from #4.

Is this a CS prop? how much clearance is there between the spinner and the prop blade roots? Turbulence there isn't helpful. I notice that if you run at 75% power... of a 160HP engine... you've got no issues (which is 58% on your engine). :rolleyes:

Still curious if the oil cooler gasket is actually sealing under engine thrust.

Is the engine fully broken in?
 
Puzzling

The temperature rise from inlet to exit seems high, but at the same time the choked exit indicates that mass flow is also high. Assuming (for the moment) that the temperature measurements are accurately represent the air, then that is a huge increase in heat rejection over my M1B for a baseline.

Don't be fooled by the high exit air temperature. The CHT's are also high and OAT is low, so epsilons are pretty much in the average range for a parallel valve. Given the four conditions above, we have...

(exit temperature - OAT) / (average CHT - OAT) =

0.377 @176 KTAS
0.394
0.439
0.489

You have my data and the delta-T across the system is substantially less.

Careful here too. Bill's installation appears to incorporate very restrictive baffle wraps; the database places the results well outside the norm. I've attached a selection below.

Don't treat the values as set in stone. As much as possible, all were gathered using the same equipment and methods, but still, variations tend to creep in. For example, mixture is uncontrolled...some are full rich, others leaned. I note here that Paul is running 50 ROP, maximizing CHT.

#4 is the hottest cylinder, so cutting / removing the deflectors on #1 and #2 are just going to steal more air from #4.

Not necessarily.
.
 

Attachments

  • ScreenHunter_1142 May. 01 08.37.jpg
    ScreenHunter_1142 May. 01 08.37.jpg
    96.3 KB · Views: 160
  • ScreenHunter_1143 May. 01 08.41.jpg
    ScreenHunter_1143 May. 01 08.41.jpg
    29.4 KB · Views: 133
Previously I said "Let's accept the numbers as accurate". Bad assumption. Calculate the available dynamic pressure based on reported altitude, temperature and TAS, then compare to reported upper plenum pressure. See below. The reported upper plenum pressures cannot be accurate, as it is impossible to have a Cp of 1, i.e. an upper plenum static pressure equal to full available dynamic pressure. It's not a closed end system like a pitot.
.
 

Attachments

  • ScreenHunter_1144 May. 01 09.45.jpg
    ScreenHunter_1144 May. 01 09.45.jpg
    46.7 KB · Views: 145
Last edited:
Oh, 50 ROP. Check out https://www.avweb.com/ownership/the-savvy-aviator-59-egt-cht-and-leaning/

Summary - The advice is to run 100 ROP if you are running ROP.
Curious how much that would help you.

About 10 degrees on cht between 50 and 100 ROP. I choose 50 because it's worst case. I planning on running LOP, but that doesn't help me on 100 degree high DA takeoffs, I still need an efficient cooling system. It is a CS prop, and the gap is set to 1/4".

Dan,
I kept getting more than 100% pressure recovery as well, and was hoping I was doing the math wrong. I'll go fix my plenum tube and install a more reasonable piccolo style tube. The lower style is already a piccolo style though, so I don't think the plenum reading should affect the lower cowl reading. I took both readings relative to the cockpit, and a direct differential reading. I've also moved the lower pickup tube to a different spot closer to the exit of the heads, and it's reading stayed pretty consistent. Also, cp greater than 1 is possible, but it would have to be the world's fastest rv7 in order to get a shockwave into the intake.
 
Last edited:
I kept getting more than 100% pressure recovery as well, and was hoping I was doing the math wrong. I'll go fix my plenum tube and install a more reasonable piccolo style tube.

Not sayin' it must be done a particular way, but the standard setup used for the database has two piccolo tubes in the upper, and two in the lower, each pair plumbed together, and arranged fore and aft so they average across each volume as much as possible.
.
 

Attachments

  • Piccolo Installation.jpg
    Piccolo Installation.jpg
    33.5 KB · Views: 156
  • Piccolo Tube.jpg
    Piccolo Tube.jpg
    58.7 KB · Views: 160
  • Upper1.jpg
    Upper1.jpg
    590.1 KB · Views: 214
I will get the "standard setup" made up so I have comparable numbers. The plane goes to paint on Friday though, so all this will likely have to wait until ~August. I expect to get it back from paint only a few weeks before Oshkosh, so I'm going to table this until then, and I'll revive the thread once I'm back into it. Thanks for all the help so far everyone.

As an extra data point we flew up to Reno yesterday. ~50-75 LOP the hottest cylinder was 373 (9,500 @ 34F). It was averaging 163 true @ 7.4 GPH (58% power) and 168 true at 7.7 GPH (62% power). Still need to play around with the SDS LOP ignition timing to find best timing, but it's relatively happy once I'm up and cruising leaned out. Just gotta figure out the climbs.
 
Last edited:
Hello all, i bring up this thread, because of hight CHT in some conditions.

RV-4
O360 A4A
Catto 3 blade fix prop

Below 70°F OAT and without PAX we can do "everything" with the plane, without getting high CHT, but above 70°F OAT, we have the problem, that CHT is getting to high, if we do Aerobatic or if we climb with high rate. Together with PAX it is much more a problem.

I have read, that one check should be the pressure difference between upper and lower cowl and yesterday i tested it.

If i climb with 85 kt and WOT i have about 2300 rpm and the pressure difference is 0,13 psi
Only PIC, OAT about 55°F and all temps are fine

attachment.php



In cruse condition i had 155 kt IAS, 2500 rpm and a pressure difference about 0,33 psi

attachment.php


What do you think about the pressure difference, is this OK or is this something we have to make better for lower CHT in high OAT conditions?

Ralf
 

Attachments

  • climb.jpg
    climb.jpg
    241.5 KB · Views: 330
  • cruse.jpg
    cruse.jpg
    516.2 KB · Views: 321
Last edited:
What do you think about the pressure difference, is this OK or is this something we have to make better for lower CHT in high OAT conditions?

Ralf, I've attached a few O-360 cooling charts, a cleaned up version of the old hand drawn Lycoming charts. The typical chart only plots a few horsepower curves, and all the values are empirical, based on some unknown "standard" baffles, thus great precision is not possible. However, they are useful.

The red lines describe a climb case. I have assumed 1500 feet MSL and 70F OAT. The intersection of the pressure drop/altitude line and the OAT line is below the 75% power line (135HP@435F). This power level is barely sustainable at low altitude given 3.5" H2O (0.13 psi) at 70F. The situation will get worse with increasing altitude, i.e. as density is reduced.

Can it be made better? There are three possibilities, (1) increase upper cowl pressure, (2) decrease lower cowl pressure, or (3) increase heat transfer.

Upper cowl pressure will be some percentage of available dynamic pressure. You mentioned 85 knots, which I will assume to be true airspeed. At 70F, available dynamic pressure (q) is 4.36" H2O. With good inlets and well sealed baffles, it's possible to convert 85% of available q to upper cowl static pressure. Here that would mean 3.7" H20 (4.36 x 0.85).

Given you measured a 3.5" deltaP across the baffles, we'll assume your lower cowl pressure is just a little (0.2") above freestream pressure. It's not a crazy assumption with a clean, open cowl exit.

It we accept the data and assumptions as stated, it would be very hard to improve these upper and lower cowl pressures. So what to do?

First, increase climb airspeed, to increase available dynamic pressure. Given the same conditions, q is doubled by climbing at 120 KTAS. The deltaP across the baffles would jump to more than 7" H2O. See the green line.

Second, it may be possible to improve heat transfer. The goal is to heat every air molecule as much as possible by maintaining turbulent flow in contact with hot parts to the greatest extent possible. Usually that means better sealing, notably baffle wraps which keep the air between the fins all the way down around the cylinder head to a point near the spark plug and CHT probe. Most baffle wraps allow a lot of air to escape out the sides well before reaching that point.
.
 

Attachments

  • COOL O-360 Ralf Climb Example.jpg
    COOL O-360 Ralf Climb Example.jpg
    50.4 KB · Views: 87
  • COOL O-360 Revised.jpg
    COOL O-360 Revised.jpg
    44.4 KB · Views: 76
Thank you for your information and the charts.
If i check the charts, the temperatures i have, are not unexpected.

Our alert is set to 435°F.
If the alert comes up, it take some time until the reaction takes place, so that the CHT is climbing to 450°F or little bit higher, before it comes down again.
Would you say this is acceptable for climbing with hight climbrate and aerobatic for some minutes?
I know Lycoming mentioned 500°F as max CHT, but seems like it is better to stay away from that.

We will double check if we can optimize the air flow.
I will share some pictures next time, maybe here are ideas for optimization.
 
If the alert comes up, it take some time until the reaction takes place, so that the CHT is climbing to 450°F or little bit higher, before it comes down again.
Would you say this is acceptable for climbing with high climb rate and aerobatic for some minutes?

Too high for me, but admittedly an opinion without absolute evidence.
 
We attempted to make the "standard setup" for measuring the plenum pressures, and it looks like it just isn't going to happen. There isn't enough space between the plenum and cylinders to fit the required probes. We've learned to live with the temp restrictions we have now, and I'll likely rebuild the plenum sometime next year if I find time.

The temperatures are controllable now (~380) during take-off and climb if we richen up to ~18 GPH on a warm day sea level takeoff. Max power from Aerosport was documented at ~16 GPH, so we're burning an extra 2 GPH in the climb for cooling. Once we level off, we trim it out to 165 knots true LOP @2400 rpm, which normally puts us 50-100 LOP, ~58% power, 385 CHT (highest), and 7.5 GPH fuel flow. With this motor, I think we're leaving 10-15 knots on the table in cruise, since going rich of peak means going WAY rich of peak for cooling. Trimming out around 70% power gets us close to 175 knots true, but at 100 ROP our CHTs are close to 420. To get them under control means running almost 200 ROP and way more fuel flow than I'm willing to pay for.
 
We attempted to make the "standard setup" for measuring the plenum pressures, and it looks like it just isn't going to happen. There isn't enough space between the plenum and cylinders to fit the required probes.

Wow. They typically sit on the pushrod tubes. That is a really low plenum.

Paul, you have a beautiful airplane, so how about some pictures? In particular, the cowl and plenum ;)
 
Doing an oil change this weekend. I'll take a bunch of pics. For the most part the piccolo tubes fit, but the forward 2-3 inches of them gets crushed by the plenum where it turns downward and starts morphing into the round inlet. I may make another set of them and just make them a few inches shorter.
 
Last edited:
.......

Absolutes were measured relative to cabin pressure. At the conditions, we opened the alt static valve and measured a 200ft rise in indicated altitude.eveything was done at 29.92 to keep it simple. This is equal to 0.2mmhg, or just over 0.1in h2o. Since my lower cowl pressure is reading almost 2 orders of magnitude higher than that, I'm ok ignoring the 0.1 inches for now until I find the larger issue. .......

I think most of your angst re high outlet bottom pressure has been produced by your mis-calc due to the the cabin static you are using.

The .1 in H2O error you presume is actually about 3 inches of water error as indicated by the 200 ft altitude difference. Simply, the density ratio of air and water is about 820 so the pressure of a 200 ft column of air will be about equivalent to a water column of 3 inches. Big error. So, reduce your reading by 2 or 3 inches and see if that doesn't make things a little more normal.

The rest of the comments by Dan, Bill and other are pretty good stuff.

Its all about mass and momentum.

Be well,
Ron
 
Last edited:
Use a flashlight to find unsealed areas around the baffles, you’d be surprised how many leaks are still there after convincing yourself that you got them all.

I thought I was done then I read this thread so I got a flashlight and ended up using an entire extra tube of RTV for a lot of very tiny holes everywhere but now it's pitch black in there, thanks !
 
Back
Top