[nomocom]

As I work through the design of the fuel system on my 3, I'm struggling with whether to go with two electrics or one electric and the mechanical. Unfortunately it seems like a real grey zone on mogas and the fuel pump. There is anecdotal evidence the mechanical fuel pumps are OK with ethanol blends (Vanguard squadron and others) but fuel pump builders are saying, not approved. They also are not specifically saying what elastomers are in use, which I find a bit odd, if in fact they are using a material that doesn't have good compatibility, it seemingly would offer confirmation of their position by just stating what the material is.

David Carr did a great thread taking a pump apart here...

http://www.vansairforce.com/community/showthread.php?t=73721&highlight=mass+spectrometer

David mentioned the thought of putting the soft parts into a mass spec for the purposes of identifying the molecular structure and composition. If those are in hand, one could presumably look at a standard compatibility chart and get a better idea of what sort of performance to expect from the soft parts in a fuel pump.

I find this to be an especially intriguing idea, as the manufacturers are not interested in increasing their liability, so I expect them to remain cagey on the topic.

Anyone with some mass spec experience willing to comment?
Questions that come to mind.
1) With a mass spec result for a diaphragm in hand, is there a library of materials to match against? Or would one have more testing to do.
2) Is this something the lab would do, or do they just give you the raw results, and leave much up to interpretation?

Comments?

 

[Greg Arehart]

Stan,

The real problem with the mass spec approach is that the materials are likely made of many compounds, so a simple answer is unlikely to fall out of an analysis. Yes, there is a library of materials out there (but not all compounds have a unique mass), and one could probably match the major components of the diaphragm fairly easily and at least get a better sense of whether the diaphragm is likely to be a problem. The other fly in this ointment is that the diaphragm materials have likely changed from time to time and engine to engine, thus unless you analyze your specific diaphragm (or know that yours is the same as one that is analyzed), you cannot be certain that the materials in every diaphragm are compatible.

I have a couple mass specs in my lab, but they are not of the kind that could do this experiment.

Greg

 

[zav6a]

Forget materials

The problem with a mechanical pump and mogas is heat. With our low fuel flows, the fuel is heated to the oil temp/block temp in the fuel pump. Mogas vapor pressure is elevated at high temps and it will boil. It is a given. Perhaps with a carb it will cool and settle down again in the bowl. In an injected engine, the vapor in the lines leads to very unsteady ops. IE, it runs very poorly. Tried it several times. Read Frank's posts.

Even if the material is right, the heat will get you.

Not aware of a plane with a mechanical fuel pump with an auto gas STC.

 

[N427EF]

See Vanguard report

Maybe mogas does not work for you but it works for a whole lot of other people
ethanol or not.
We all know Frankh has a dual electric fuel pump set up that works well for him.
A system with a mechanical fuel pump needs very little modification to keep
vapor lock from occurring.
Now if someone can tell me why I should not believe what one of the members of the Vanguard squadron said, please speak up.


http://www.vansairforce.com/community/showthread.php?t=58334&highlight=vanguard+ethanol+stan

 

[nomocom]

Stan,

The real problem with the mass spec approach is that the materials are likely made of many compounds, so a simple answer is unlikely to fall out of an analysis. Yes, there is a library of materials out there (but not all compounds have a unique mass), and one could probably match the major components of the diaphragm fairly easily and at least get a better sense of whether the diaphragm is likely to be a problem. The other fly in this ointment is that the diaphragm materials have likely changed from time to time and engine to engine, thus unless you analyze your specific diaphragm (or know that yours is the same as one that is analyzed), you cannot be certain that the materials in every diaphragm are compatible.

I have a couple mass specs in my lab, but they are not of the kind that could do this experiment.

Greg

Yup, the diaphragm is not a single material. Looking at it from the edge, there is an elastomer and fabric layers. Thinking about the why, elastomers may provide a pretty good barrier, but they don't have the needed strength for all the cycling- so the fabric is embedded. I'd think the fabric should be tolerant as well. I recall from compatibility studies that the elastomers to varying degrees are permeable, meaning the fabric isn't necessarily protected.

Linking a pump in hand to tested materials could be messy if not done methodically. To start with, we know there are several versions of the high pressure. David showed pics of a black upper (crankcase side) and a black lower diaphragm. The unit I'm inspecting, an AC 40296 has a black upper and an orange lower. Any other colors out there for diaphragms? Brand, diaphragm color, and a 30 day swell test would go along way toward suggesting whether or not it was the same material as a tested pump. The lower orange diaphragm on mine had a convenient tab of 2-3 mm by at least 5 mm. It was easy to remove with a scissors, and it's removal didn't impact pump function.

The mass spec's in your lab won't work. I read a little about the mass spec- I had forgotten about the vaporization. Doesn't seem very promising- I'd think elastomers would burn and decompose rather than give a clean vaporization. I wonder if any mass spec would work?

NMR, no vaporization needed with that tool.

The only thing I used in college chem that I'm still occasionally using is HPLC and GC's. I'm doubtfull one could find a good enough solvent to dissolve the elastomer well enough to shoot it into a HPLC- after all, resisting dissolution is a big part of being an effective seal. They aren't going to want to dissovle. Back to NMR?

This reminds me of a food processer problem I became aware of years ago. Foreign material in the food. I recall they sent foreign material to the lab for identification, so they could better determine what part of the food process was the offender. I'll explore that with some contacts.

 

[nomocom]

The problem with a mechanical pump and mogas is heat. Read Frank's posts.

Even if the material is right, the heat will get you.

Not aware of a plane with a mechanical fuel pump with an auto gas STC.

Thanks for bringing up the issue of heat, Duane. There was a period of time were I ran quite a few Hodges tests on the local fuel- avgas and mogas. I never saw any mogas that really stood out. The readings were within a half lb of one another. I agree heat is a key design problem and would add, whether or not you are a mogas user or not, the fuel system should be built to minimize the addition of heat to the fuel.

Lacking actual heat gain data on fuel going through the fuel pump, one is left to speculate. Having looked at David's pump teardown and recently done a partial one myself, there are some barriers built in to slow heat transfer. There is a gasket at the case/pump interface and two fairly thick elastomeric diaphrams between the heat source and the fuel. The fuel pump to some degree is hollow, so that also would provide some chocking of the heat pathway. There is radiation from the engine and other warm parts that no doubt contributes, but does anyone really know how significant this is? It would be really interesting to have a temp probe at the firewall and another before the servo or carb. It would be a great tool to demonstrate the efffectiveness of insulation efforts.

As far as FrankH's system, yes I've studied it. I consider it an elegant solution, both hydraulically and from the human factors angle. But... for my bird, I'm not quite ready to give up the mechanical, at least at this stage of my investigation. I'd still like better info on the soft parts. I'd like to see some consensus on what one could expect.

Mogas and mechanical pump STC, they are not uncommon. The low wing Piper Cherokees with the low compression Lycomings have STC's available. There may be others.
http://www.autofuelstc.com/autofuelstc/pa/Piper_PA-28.html

 

[Greg Arehart]

Stan,

A quadrupole mass spec could probably work - my mass specs are set up for stable isotope measurements on gases, a very specific application.

If I were going to the dual electric pumps, I would certainly want a backup battery somewhere in the system, considering the low-wing aspect of our airplanes preclude any gravity feed. Especially true for FI engines that require high pressure.

Greg