What's new
Van's Air Force

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

How many fuel filters do you have/need?

sirlegin

sirlegin

Well Known Member
We have decided that removing the gascolator from the engine side of the firewall may stop our engine surging in high temps at low rpm's due to the fuel boiling/pre heating here. That leaves a fuel filter at the boost pump and a fuel filter at the spider valve.
This is a 9A with a Mattituck 0-320 with dual lightspeed ignition, fuel injection (Formerly FADEC) and a Catto 3 blade prop.
How many fuel filters are required?
As of right now we get no junk from the wing sumps or water. Hangared.
 
Fuel Filters

You should consult the fuel injection manufacturers and boost pump manufacturers for recommendations on fuel filtration.

The Precision/Bendix injection system manuals recommend 32 micron filtration. The finger screen in the servo is 64 micron I believe and is a "last chance" filter, not intended as a main filter.

My boost pump manufacturer says upstream filtration is required, I have 70 micron filters.

If you're really sure the location of the firewall mounted gascolator is causing the engine surging problem, have you considered running a blast tube to and surrounding it with a cover similar to the one Van's sells for the engine mounted fuel pump? Also have you tried running a blast tube to the engine mounted fuel pump?
 
I have a filter on each tank just ahead of the selector valve.
In the unlikely event that a filter becomes restricted, I can switch tanks and avoid an engine stoppage.
 
One!

One is all you need, and it should have the mesh size required by the fuel system designer.
Good decision on removing your gascolator, should help smooth out your
idle operations.
You have probably seen this in another thread.
FWIW I also have 2 filters one for each tank


Old 07-29-2013, 01:31 PM
Don at Airflow Don at Airflow is offline

Join Date: Jun 2006
Posts: 216
Default Gascolator
This is a letter we wrote to a customer about gascolator use in fuel injected engines.

To fully understand the situation you must first understand why a gascolator was used in the first place. Back when carburetors were the only source of fuel metering on aircraft, water was a detriment to the operation of the engine. Of course water won’t burn but the main reason was that water being heaver than fuel would sink to the bottom of the float bowl. Water also has a higher surface tension than fuel. Since the main jets are located in the bottom of the float bowl and the metering head (the suction created by the venturi to suck the fuel out of the bowl) is low on a carburetor, the water can actually block the flow of fuel through the main jets due to surface tension, thus starving the engine of fuel.

Enter the fuel injection system. There’s no float bowl, the fuel is under pressure (20-30 PSI). So in this case even if there’s some water in the system the fuel control will flow the liquid what ever it is. Granted the engine cannot burn water but there will be no interruption of the delivery of fuel to the engine.

One thing that is in the carburetors favor is that with the float bowl, if any foamy fuel or fuel vapor is in the system, the float bowl will vent off the vapor and the main jets are only exposed to liquid fuel. Correct fuel metering in this case is mostly not effected. Unless the engine driven fuel pump or boost pump is vapor locked the engine will get fuel. In this respect, carburetors are less affected by hot operation. More on that subject later.

Now look at the aircraft fuel system. In an RV aircraft the fuel tanks are the lowest point in the fuel system. Water will be in the sumps of the tanks. If water is found during a sumping operation then further investigation should be made concerning the entire aircraft fuel system condition.

1. Installing a gascolator in any other place other than the wing roots would result in the gascolator being higher than the lowest point in the fuel system. Not the place to catch water.
2. Since the RV aircraft can do some aerobatic maneuvers, rolling the aircraft upside down would dump any water into the fuel system if any were held in the gascolator.
3. Since the fuel injection system does not have a float bowl, fuel vapor is a problem and can cause poor fuel metering. Avgas boils at around 130 degrees F at sea level. The boiling point gets lower as altitude increases and also with a decrease in fuel pressure on the suction side of the fuel pump (flow losses in the fuel system). Since the gascolator is typically installed on the firewall, it is subject to a high heat environment. The volume of fuel in the gascolator does not change very fast at low power therefore the fuel in the gascolator picks up heat, which can lead to fuel vapor issues and possible vapor lock of the engine driven fuel pump.
4. Installation of the gascolator on the fire wall though not recommended on this installation can be accomplished by the following:
a. Make sure the gascolator can withstand 30 PSI, as it will be pressurized when the boost pump is on.
b. Install a blast shield over the gascolator and provide blast air to keep the gascolator cool.
c. Understand that having a volume of fuel on the firewall (gascolator) may result in rough engine operation and poor idle, under hot conditions.
d. With fuel injected installations we want to minimize the volume of fuel in the engine compartment as much as possible. This applies to hose routing and components that increase the volume of fuel that can be heated.


Hope this helps

Don
Click to expand...
 
One Andair gascolator/filter for each tank (in the wing root), and one 40 micron after the engine driven fuel pump immediately prior to the fuel servo as directed by Prescision (Silver Hawk).

Bevan
injected 0-360
not flying yet.
 
N427EF said:
One is all you need, and it should have the mesh size required by the fuel system designer.
Click to expand...

Probably too much information, but be careful with this statement. Each filter protects something. As previously mentioned, the 70 micron filter prior to the boost pump protects the pumps moving parts. The filter (32 micro in this example) ahead of the servo protects the fuel system parts and orificies. There are probably edge filters for errant particles in the system too.

A finer filter typically means higher pressure drop, so they belong after the pumps and only required (minimal) filtration prior to the pumps. This yields a system with the highest flow and temperature capability.

Thanks for the requote from Don. Excellent, specific information.
 
I asked this question of both EFII and Precision and got the same answer: The pre-fuel pump filter is the only one needed in this scenario: EFII Boost pump, included pre-pump filter and Precision Silverhawk injection.

Maybe this is because injectors in Mechanical FI systems have much larger nozzle holes than with electronic FI.
 
2 in Wing roots and one from AFP before boost pump

I had been battling intermitten clogged injectors and followed Danny King's advice and put an Edelbrock 40 u filter in each wing root. I also rerouted the fuel line on the left wing a bit to better accomodate the Andair fuel selector - which I LOVE.
Since adding the filters in the wing roots about 50 hours ago I have not had a clogged injector. I wish I had followed Danny's advice when he first offered it about 3 years ago. I might have fewer gray hairs!
I opened them up during the oil change a month ago and found a little fuzz in one filter and several black flakes in the other. Either one could have caused a clog.
I have a lot of faith in Don and his product but it would seem that with their 125 u filter protecting the pump there still needs to be a filter to protect the injectors. I know there is a filter in the servo but I have never found anything in that filter.
Time will tell if the filters will take care of the clogged injector problem but right now it looks good.
 
Bubblehead said:
I had been battling intermitten clogged injectors and followed Danny King's advice and put an Edelbrock 40 u filter in each wing root.
Click to expand...

Does the Edelbrock filter have 37 degree fittings or are the fittings replacable? I could not tell from their web page. I would like to put these in my wing roots also.
 
Same functional placement and op logic as Mel

One for each tank prior to selector valve. No problem with fuel flow 1000+ hours.

Bob Axsom
 
Cannot speak for Precision, but...

Looking at micron size vs. hole size one can see that a 125-micron rated filter will pass a .005? particle. This of course is pretty big for filtration in the fuel regulator of an Airflow Performance fuel injection system where typical clearances between the ball and seat can be less than .005?, but we have to be careful here that we do not put any restriction on the inlet side of the fuel pumps that could create vapor problems with flow or pressure drops in the system. In the inlet of the FM-series is a non-relieving 70-micron filter, which will pass a .003 diameter particle. In the Airflow Performance flow divider is an additional 50-micron filter. This will pass a .002 particle. Since the clearances of the flow divider spool valve to bushing are less than .0001, a .002 particle is not an issue for sticking the valve. The smallest injector nozzle restrictor we manufacture is .015? diameter. So with 50-microns as the final filtration in the system there should be no issues with clogged injector nozzles, unless the contamination is in the nozzle lines themselves.

Don
 
Taildrgr said:
Does the Edelbrock filter have 37 degree fittings or are the fittings replacable? I could not tell from their web page. I would like to put these in my wing roots also.
Click to expand...

The ones I used were Edelbrock 8130s and they had the std aviation AN-6 on both ends and have a 40 micron filter.

I was concerned like others about having the fuel flash in high temps with a filter on the suction side of the pump. The filters I used are rated for much higher flow rates with small pressure drop. I think the website I bought from had the ratings on it. i will try to find it and post tomorrow.

Many people are using filters on the suction side and some down below 40 microns without problems so I think it will be fine.

Here is a link to Summit Racing for the filters. Jegs also has them and there are some deals to be fuond on Amazon and Ebay. http://www.summitracing.com/parts/edl-8130/recommendedparts

Here are a picture of the installation on the left side.

DSCN0476%202-M.jpg
 
Last edited:
I have to agree with Don here, adding restrictions on the inlet side of a pump is risky business. This is especially true in hot climates where the fuel temp can get quite high from the aircraft just sitting out in the sun makes the situation even worse. Personally I'm more willing to risk a clogged nozzle than a total loss of power.

If you must add a very fine filter than it makes more sense to have it after the engine driven pump so fuel is always under pressure.

I have a 74 micron filter (same physical size as the AFP filter with lots of surface area) after the selector valve, thats as fine a filter as I was willing to go on the inlet side.

I know John (Bubblehead) probably doesn't want to here this but I needed to say it anyway.
 
Last edited:
Walt said:
I know John (Bubblehead) probably doesn't want to here this but I needed to say it anyway.
Click to expand...

Walt, I welcome dissenting opinions as long as they are based on fact and logic. Your opinion meets both criteria and I respect your opinion as well as Don's and others. And please speak up if you think I am headed in a dangerous direction.

My bottom line is I have had several losses of power and most (maybe all) were clogged injectors. The worst 2 (not definitively clogged injectors) were bad enough that I could barely maintain altitude and the engine was shaking violently. Those events make me willing to try something different. Suction side filters of 40 micron or even less have been used by several people here in North Texas and have not caused catestrophic loss of power. That does not prove it will never happen, but the risk is worth it for me.
 
Bubblehead said:
And please speak up if you think I am headed in a dangerous direction.
Click to expand...

Well the truth is I don't know if you are or not. Perhaps someone that has the capacity to do some actual testing in this area can determine if it's dangerous or not.

I do know that lowering the pressure in the fuel line is not a good thing, does adding this filter lower the pressure enough to cause problems... don't know.
 
To those having clogged Injector issues, what sort of filter did your FI system have, and was it cleaned at every 50hr?

The traditional setup AFAIK / am familiar with is 1 filter prior the fuel pump(s), and then the FI "finger screen". I have seen filters, especially after the first few hours with lots of swarf / junk, but getting nowhere near "blocking flow" in that 90% of the filter area was still clear.

I have had loss of power / rough running twice, in each case caused by the FI finger screen being blocked. This was where the owner had late on added a flop tube, and not flushed out the taken correctly :confused: The filter got the big stuff, but the smaller made it to the screen. You cannot "see it", but trying to blow through the FI screen filter indicates the problem.

I've also had one clogged injector, new build, and suspect the particle was there from build. Aircraft / engine flew well, but "something" was wrong with EGTs / leaning, and a bit of analysis led us to the problem. I now run a new engine up to say 1200 RPM, lean it off a bit, and look for even EGT rise.
 
I have the standard AFP 125 micron filter between the selector valve and the boost pump, plus the standard finger screen in the servo and the standard one in the flow divider, all standard parts. When I had clogged injector nozzle problems I would pull the orifice and could see the blockage with a magnifying glass. Most times I pulled the main filter and sometimes I pulled the finger screen in the servo and never found anything more than a small black spec. During one incident I pulled the injector and found crud close to the end of the injector. That was one of the times when it shook like heck and I had to make a precautionary landing. Thanks to some help from a RV-6 builder/driver at that airport I was able to clean it in an ultrasonic cleaner and continue the trip.

I had recurring problems without wing root filters and have had no occurrences since installing them. I will check the airframe log book this weekend to find out how many hours I have flown with them but I think it has been about 60 hours. That is not conclusive proof that the filters are working but every hour without a problem makes me more a believer.

If I clog an injector and cannot dislodge the particle with the boost pump and by going full rich I am out of options. If I get flashing in a filter I have several options. I can:

1) change tanks
2) reduce power thereby reducing flow and the pressure drop and reducing the likelihood of flashing
3) turn on the boost pump which may be able to overcome some of the flashing.

Also, during cool weather or once at altitude the temps are cooler so the chance of flashing is reduced.

It is a classic case of choosing your trade-offs and doing what works for you. Right now wing root filters appear to be doing the job and so far have not caused problems. I live in Texas and flew Sunday afternoon when it was about 102 F on the taxiway and I suspect the hanger was hotter than that. If I did not have a problem with fuel over 100 F I probably will not have a problem.
 
Maybe you've gotten all the crud out of your fuel system. :)

My concern would be that at low altitude and full throttle, flashing/vaporization would be more likely, when things are most critical: Takeoff/climb. At high altitude where it may be less likely, you have options.

Ultimately, these are our aircraft and we choose where we want the safety tradeoffs.
 
This is an interesting and informative thread, by guys with expertise in the field.
As a vendor supplying hose assemblies to alot of you, we are always interested in what problems you are having, and what eventual solutions you come up with.
My .02 cents worth for Don. I have never met the man, although we have communicated through email. Don is the master, and I dont think I'll get too many arguments over that. He has the test benches to provide absolute factual data for his products, and I would bet most others as well. He gives his recommendations based on years of FI experience, and knowing his systems requirements.
I guess the point I'm getting to is this: if you have to run alot of different rated filters to have a reliable fuel system, this would tell me that I would have to look at all components. Everything from the tanks, pickup tubes, hoses/rigid tubes, selector valve, boost pump, mechanical pumps----and so forth. If you are getting debris in the filters, thats a red flag for me to see WHAT the debris is. Maybe its NOT from the fuel, maybe from a rubber lined hose in the system that is deteriorating. How many of you are using a 701 hose as a flop tube, or a pickup tube?
Many of our engines are using the stock recommendations and are extremely reliable. Granted, high heat/humidity creates issues that we need to solve. We are dealing with the experimental world, so everyone is going to be different.
Tom
 
More data points

Walt A. and I had a great conversation about this on Saturday and he suggested trying to quantify the pressure drop. I did some thinking about that in the afternoon and this morning I called the Edelbrock Tech line. They did not have any technical information about pressure drop at different flow rates (aka characteristic curve) for the filter but the guy said they get the question from time to time and the engineers told him the restriction from the filter was less than the restriction from the AN-6 fittings at each end. That is not a lot of help.

I also started thinking about the whole fuel system from the tank to the cylinders like Tom suggested. This spring I changed the fuel pickups in both wings to comply with the "anti-rotation" SB that Vans had. The pickups were the old style slotted fuel line and I installed the new screen type pickups. I need to find my pictures of the old pickup tubes but those slots were not very big and there were not many of them and it seems like that could be a source of pressure drip too. Remember the flashing does not have to happen at the most restrictive place. Pressure drop is cumulative along the entire fuel flow so lowering it at the tank 1" H2O or so can make enough of a difference to cause flashing somewhere downstream.

Here's my bottom line for now, subject to change upon further events or information. I have rebuilt or replaced all the major fuel system components; fuel pickup, new fuel valve (Andair), Don rebuilt the FI servo and spider, I installed new injectors both body and sleeves, and the main fuel filter is clean and installed properly. There are no kinks in fuel lines, no leaks, and the vents are clear. I did everything I could and still had a rough engine problem traced to dirt in an injector. I don't where the dirt came from or why it did not get picked up in the normal filters but I do know it is not fun flying a plane that occasionally sputters and coughs and shakes and loses power.

I've balanced the risks and rewards and know these filters have been flying in multiple airplanes for several years in North Texas where I am based. I only have 20 hours with the filters (and new pickups) so it is too soon to tell, but when I inspected them after 10 hours I found a little black spec in one and a fuzzy fiber in the other. Only time will tell but for now I think this is a reasonable approach.

I have tremendous respect for Don and appreciate his input and wisdom along with the input from Walt and everyone else who have spoken up.
 
You must log in or register to reply here.
Back
Top