Non Bypass Oil Filter Oil Starvation

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

proppastie

Well-Known Member
Joined
Feb 19, 2012
Messages
5,677
Location
NJ
Friend cut apart a short Tempest filter and there was no spring for an oil bypass......My old canister filter which takes the CFO 100 filter element also does not have any bypass provisions. I am making the comparison to the Champion Oil filters that do have a spring against the element for a bypass if the filter fails or is too clogged.

So the question is do we know of any aircraft that have had oil starvation because of a clogged oil filter?......
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
7,230
Tempest has a short, wide spring steel plate that does the job. See it there between the filter can and the element?

1659231892946.png

It's like that to keep the can length to a minimum while still having adequate filter area.
 

proppastie

Well-Known Member
Joined
Feb 19, 2012
Messages
5,677
Location
NJ
We Tried pushing on that part ......it did not move/spring down leastwise not like the champion sprng pressure
 

speedracer

Well-Known Member
Joined
Feb 4, 2020
Messages
360
I use NAPA oil filters on my plane. They have the bypass spring and they're only five bucks.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
17,355
Location
Memphis, TN
I have seen a guy put a car filter on a 172 without the bypass. 50 wt oil and after about 30 seconds, BOOM! What a mess.
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
7,230
I have seen a guy put a car filter on a 172 without the bypass. 50 wt oil and after about 30 seconds, BOOM! What a mess.
Aircraft filters have a thicker can to take the higher pressures, too. Lycs can run up to 100 PSI. I don't know of any auto that does that.
 

Toobuilder

Well-Known Member
Supporting Member
Joined
Jan 19, 2010
Messages
5,707
Location
Mojave, Ca
Most of the "premium" auto filters meet or exceed the burst pressure specs of the aircraft versions. NAPA/Wicks (same manufacturer) and K&N are among them. And yes, there are plenty of auto applications that can exceed 100 PSI.
 

proppastie

Well-Known Member
Joined
Feb 19, 2012
Messages
5,677
Location
NJ
Has any one seen a clogged filter without a bypass that ruined an engine?......thinking about my old canister filter that does not have a bypass
1659280538531.png
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,170
Location
Saline Michigan
A 2.5" diameter filter at 10 psi drop is about 50 pounds, and it only needs a tiny deflection to bypass plenty of oil, lets say 0.020". Did it really get set up to measure 50 pounds and 0.020"? That is a 2500 pound per inch spring.

Coil springs we can squeeze with our fingers are 200-300 pounds per inch, is bypassing at less than 1 psi, or just about all the time. Why bother with a filter at all if everything is going around the filtering element?


Tempest filters have been running fine in our 2200 hour O360-A4M for the last 500 of its 2200 hours.

Now for some other filter info. Oil filters in cars/trucks are intended to pick up the dirt and crud built into the engine plus the debris from first run, breaking in the engine, and then until the first filter change. Once we have broken in an engine, the debris being freed in 50 hours should never do more than mildly load up the filter. It the engine is loading up the filter in our usual use, its remaining life is short indeed. Yeah, you might like to get all of the minutes it has, but how are you going to know the engine only has a few minutes left so you can head for a safe field? All this concern over the bypass contingincy seems to me to allow the lawyers for engine and filter makers to say "we gave them all the time we could".

Billski

PS - Look at posts 17, 20, 25, and 27 by Dan. He covers this all way better than I and covers it correctly.
 
Last edited:

proppastie

Well-Known Member
Joined
Feb 19, 2012
Messages
5,677
Location
NJ
Coil springs we can squeeze with our fingers are 200-300 pounds per inch, is bypassing at less than 1 psi, or just about all the time.
now you got me worried about using the Champion Oil filters .
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,170
Location
Saline Michigan
now you got me worried about using the Champion Oil filters .
I hope not... I believe Champion filters have a stouter springs than what I described. I will check out both types when I am at the maintenance shop tomorrow for wheel bearing service on the wife's airplane. After the dental hygenist appointment. Before the new clothes washer install. Busy day...
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
17,355
Location
Memphis, TN
One of Tempest first products started when they got all the filter tooling from Champion for free. Champion stopped making the filters in house and was throwing it away. Tempest did put their spin on it. That’s also why here is a -1 for Champion or -2 on the tempest so the FAA would know who’s were who’s. The early Tempests used straight Champion numbers and the FAA let them until they had some early manufacturing problems, long ago solved now.
 

proppastie

Well-Known Member
Joined
Feb 19, 2012
Messages
5,677
Location
NJ
A 2.5" diameter filter
area against pleats higher?......say 4" deep x 48" long pleat area......but what is the pressure drop across the pleats?......understanding this stuff can get interesting
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,170
Location
Saline Michigan
Look close at the cut down filter in the photo. It has been a while since I even looked at something like this, so my earlier analysis was incomplete... Housing made of can and end cap, spring, seals, filter element containing filter, end cap, and inlet cap. The spring pushes the filter element against the seal on the housing end cap and that push is along the centerline of the filter. You have a difference in pressure (deltaP) across the filter element due to the filter material being a (small) restriction on flow.

Now let's do an analysis of the forces on the spring. deltaP pushes on the filter element in all directions, but the sum of the forces in the axial direction are all the spring must hold against in normal operation. The area of the filter element end cap is Pi*r^2 or PI()*d^2/4. Thing I did not have in before is that the filter element inlet cap has a hole in it where it connects to the inlet. So the area of the inlet cap is is PI()*(Do^2-Di^2)/4.

When D is 2.5", A of the filter element end cap is 4.91 in^2. For the inlet cap, a 2.5" Do and 1" Di, area of that cap is 4.12 in^2. Pressure on the inlet cap is opposite direction then pressure on the end cap, so the net force on the filter element is deltaP times (4.91-4.12) = deltaP*0.785 in^2. A lot lower than by my earlier description. Anyway, if the spring is designed to open the seal at deltaP of 10 psi (pulled that number out of the air, but it has to be less than pump pressure or it could never open) then the spring has to hold 10lb/in^2*0.785in^2 = 7.85 pounds. And when deltaP across the filter paper is less, the spring holds the filter against the seal, and all oil goes through the filter paper...

Now if the spring makes 7.85 pounds as installed, and it moves 1/8" at install, that is a 62.8 lb/in spring. Yeah, you can compress that with your fingers. But the other scheme is that diaphragm spring, and they generally have higher rates. Lets say it is 150 lb/in (another educated guess, but still a guess), so that only gets deflected 0.052" at install. and they both open the seal at 7.85 pounds and 10 psi.


So, no reason to fear either one if they did the job on designing and maintaining their tolerances in manufacture and assembly. If they are not maintaining their tolerances, all bets are off... Sure they can design with a lower rate spring and make plant quality control less critical, but low rate springs usually cost more at the same loads than higher rate springs. Tolerance costs vs part costs, and we can bet they are working to the lower total cost that reliably makes a reliable product.

Billski

PS - Look at posts 17, 20, 25, and 27 by Dan. He covers this all way better than I and covers it correctly.
 
Last edited:

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
7,230
Look close at the cut down filter in the photo. It has been a while since I even looked at something like this, so my earlier analysis was incomplete... Housing made of can and end cap, spring, seals, filter element containing filter, end cap, and inlet cap. The spring pushes the filter element against the seal on the housing end cap and that push is along the centerline of the filter. You have a difference in pressure (deltaP) across the filter element due to the filter material being a (small) restriction on flow.

Now let's do an analysis of the forces on the spring. deltaP pushes on the filter element in all directions, but the sum of the forces in the axial direction are all the spring must hold against in normal operation. The area of the filter element end cap is Pi*r^2 or PI()*d^2/4. Thing I did not have in before is that the filter element inlet cap has a hole in it where it connects to the inlet. So the area of the inlet cap is is PI()*(Do^2-Di^2)/4.

When D is 2.5", A of the filter element end cap is 4.91 in^2. For the inlet cap, a 2.5" Do and 1" Di, area of that cap is 4.12 in^2. Pressure on the inlet cap is opposite direction then pressure on the end cap, so the net force on the filter element is deltaP times (4.91-4.12) = deltaP*0.785 in^2. A lot lower than by my earlier description. Anyway, if the spring is designed to open the seal at deltaP of 10 psi (pulled that number out of the air, but it has to be less than pump pressure or it could never open) then the spring has to hold 10lb/in^2*0.785in^2 = 7.85 pounds. And when deltaP across the filter paper is less, the spring holds the filter against the seal, and all oil goes through the filter paper...

Now if the spring makes 7.85 pounds as installed, and it moves 1/8" at install, that is a 62.8 lb/in spring. Yeah, you can compress that with your fingers. But the other scheme is that diaphragm spring, and they generally have higher rates. Lets say it is 150 lb/in (another educated guess, but still a guess), so that only gets deflected 0.052" at install. and they both open the seal at 7.85 pounds and 10 psi.

So, no reason to fear either one if they did the job on designing and maintaining their tolerances in manufacture and assembly. If they are not maintaining their tolerances, all bets are off... Sure they can design with a lower rate spring and make plant quality control less critical, but low rate springs usually cost more at the same loads than higher rate springs. Tolerance costs vs part costs, and we can bet they are working to the lower total cost that reliably makes a reliable product.

Billski
The larger area opposite the inlet/outlet cap against the area at the inlet/outlet implies, to me, that the oil pressure against clogged media would force the filter more tightly against the cap, not open it for relief.

I spent years in the air brake component remanufacturing industry and did numerous analyses of the internal pressures of compensating valves, self-releasing control valves, regulators and actuators. This pressure relief spring arrangement puzzles me some. What am I overlooking?
 

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,170
Location
Saline Michigan
I urge anyone who doubts my analysis to visit their favored fluid mechanics text. Equal static pressure everywhere inside a vessel, unequal areas in opposite directions, net force has to exist away from the opening which the spring resists... Now for an example:


Note that the water rocket has to be held on the pump or fill device. Forget to engage the retention and the rocket slides off the pump with the first couple strokes of the pump handle... Launch is achieved when the rocket is pressurized and retention is released. Initial movement is just net force away from the opening. Once the nozzle clears the pump, we get rocket effect due to accelerating ejecta mass and conservation of momentum. If net force were toward the opening, we would need something to nudge the nozzle off the pump. The gadgets do not have such devices, just retention.

Billski

PS - Look at posts 17, 20, 25, and 27 by Dan. He covers this all way better than I and covers it correctly.
 
Last edited:

AdrianS

Well-Known Member
Joined
Jul 5, 2014
Messages
837
Location
Australia
As I understand it, the bypass is for cold starting if the oil is very viscous, but I have no idea what's a realistic pressure drop through the filter.
Assuming 50 weight oil at -5 degrees C compared to 110 C, that's a fair viscosity difference.
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
7,230
I urge anyone who doubts my analysis to visit their favored fluid mechanics text. Equal static pressure everywhere inside a vessel, unequal areas in opposite directions, net force has to exist away from the opening which the spring resists... Now for an example:


Note that the water rocket has to be held on the pump or fill device. Forget to engage the retention and the rocket slides off the pump with the first couple strokes of the pump handle... Launch is achieved when the rocket is pressurized and retention is released. Initial movement is just net force away from the opening. Once the nozzle clears the pump, we get rocket effect due to accelerating ejecta mass and conservation of momentum. If net force were toward the opening, we would need something to nudge the nozzle off the pump. The gadgets do not have such devices, just retention.

Billski
Yup. But the pressure in the rocket is inside it. The pressure on the filter element is outside it. The flow is into the outer area of the can, through the filter, and out the center.
 
Top