DIY-EFI

[WesRagle]

Hi Pete,

Wes,
The mounts you found are just the sort of thing that could be employed. It is up to you what approach you take; I have seen several solutions that work fine.

Each solution has its own set of issues, the GPAS manifolds have bosses already moulded into them in about the right place, however they may not be big enough for the injector carriers you have identified and may need the bosses building up to take the threaded mount.

Some things to be considered:
• Holding the injector in place.
• Holding the fuel rail in place on the injector.
• Injector size. (see below).
• Injector wiring/plug clearance
• Fuel line connection to fuel rail.

When putting my system together choosing the correct injector was very tedious because we needed to select an injector that firstly delivered the right amount of fuel in its optimum operational range. We found the CHD 610 which is rated at 25 lb/hr best after trying a range of different sizes. This meant I had to make different fuel rails for each injector tested.

Keep in mind the penetration of the injector nozzle is a major consideration when choosing or building the mount.

Thanks a bunch for your post. You pretty much stated all of the things that are bugging me.

The pile of parts is growing and I now have a set of AeroConversions intakes:



I pulled one of the injectors from the throttle body and and the boss on the GPAS log is just a tad to small to accept the seal provided with the injector. That's why the machinist I talked to suggested adding material.



Of course that leads to the next question. If I were to locate an injector with a standard O-ring bottom seal would that be small enough that the manifold could be machined without adding material? And if I did that would the the injector end up with the proper penetration and "aimed" at the right place (which I assume is the top of the intake valve).

Not much better when considering the "Red Logs". I start looking and I find any number of bosses. I assume I would have to use the correct one for the injector I pick. Different ones, different penetration? And picking an injector is a pretty big chore in and of itself. Different body styles, different spray patterns, ...

All of this to say I'm having a heck of a time gaining any traction. This may sound like a bit of a cop out but ... Knowing that you and Jeff are both working on installation details, and since ya'll intend to share your design/findings, I think I would be wasting my time trying to do the same thing. I'm 100% confident either of you guys will come up with a better solution than I could.

So, I'm going to run off and spend my time considering a VW specific ECU. It will be quite some time before I need the physicals to make progress. I'll be standing by watching for installation details as they emerge.

I do have a couple of questions that I would like answered.

Pete: Do you have a VE and Ignition table/map that you have used on a VW conversion and are willing to share?

And for you guys that are using the AeroInjector as a throttle body. Have you, or do you intend to, forgo the throttle position sensor? I ask the question because I had intended to use the TPS to cross check the MAP sensor.

EDIT1:
I re-read the thread and Pete already answered the question.

There is no need for a position sensor because the whole system works from inlet manifold pressure, throttle open, more air and ECU adds more fuel and visa versa. Switching form EFI to normal operation in flight is no worse than a Sonex burp.

EDIT2:

One factor to be considered in my system was that I use Semi-sequential injection because the Aerovee does not have a distributor, there is no way of knowing which of the cylinders in each bank is going to fire next. Not a problem you have to deal with so you could go Full Sequential.

I don't have a distributor. I have a Dyna S electronic ignition. Unless I'm confused it works just like the secondary ignition on the AeroVee. For full sequential I would have to have a cam position sensor. I don't intend to pursue that at this time. I intend to pursue semi-sequential. Batch in the event of a secondary ignition failure.

Thanks for Your Time,

Wes

[WesRagle]

Folks,

Chris, the esteemed owner/moderator of this forum, has added a section under "Engines And Firewall Forward" specifically for the discussion of Fuel Injection. That should help us keep fuel injection subject matter somewhat organized. So please, keep the discussion going.

A shameless plug: For those who enjoy this group, please don't forget the Donate button located at the upper right of this page.

Thanks,

Wes

[phenry]

OK, This is the beginning of how I have set up my EFI.

This is how I have done it and anything you take away from this is at your own risk.
Most of the brains behind this system is the work of Chris Bishoff and I have only refined the details over time.

The idea was to create a EFI system that could be easily switched back to normally aspirated operation allowing for a fully redundant system that was safe as possible, very uncomplicated, cheap and could deal with the inherent limitations of the Aerovee engine (low available electrical power and no distributor).

My particular problem was that I could not, no matter what I did get the Aerovee/Aeroinjector to idle properly and many times had the engine quit on landing (very embarrassing).

Fuel Plumbing:

Fuel Plumbing.jpg (89.14 KiB) Viewed 5947 times

Gascolator.jpg (62.92 KiB) Viewed 6074 times

Distrution Manifold #1.jpg (21.74 KiB) Viewed 6074 times

Distrution Manifold.jpg (23.54 KiB) Viewed 6074 times

Pump-Regulator.jpg (27.63 KiB) Viewed 6074 times

Installation.jpg (62.13 KiB) Viewed 6065 times

Manifold Perssure Feed.jpg (51.97 KiB) Viewed 6065 times

Modifold Pressure Tapping #2.jpg (48.64 KiB) Viewed 6063 times

Note how the MAP sensor is mounted on the Pump/Pressure Regulator assembly. This made it easy to connect the regulator manifold pressure compensation line and MAP sensor in one place.

[WesRagle]

Hey Pete,

Very cool!

This is how I have done it and anything you take away from this is at your own risk.

I take full responsibility for anything I do in experimental aviation.

Note how the MAP sensor is mounted on the Pump/Pressure Regulator assembly. This made it easy to connect the regulator manifold pressure compensation line and MAP sensor in one place.

Got it.

You guys have fully convinced me that my idea of implementing electronic TBI as a backup to the primary ECU is a bad idea. Reversion to a fully manual system is the correct answer. Even in the certified world if you want to reduce the design assurance level by using a redundant system you will likely be faced with having to design a "dissimilar" system for backup to eliminate common failure modes. Fully manual is about as dissimilar as you can get. It also eliminates electrical dependency. It also simplifies the ECU design. And for most folks, it reduces the cost of a retrofit. And, simple is good!

As time permits, would you please give part numbers for the regulator, pump, etc.

While considering a ground up ECU I put together a list of analog inputs I would need to monitor.

MAP
IAT
CHT (Engine Temp)
MIXTURE POT
BUS VOLTAGE
FUEL RAIL PRESSURE

Question: Other than MAP, which of these signals do you use?

Again, thanks for your time.

Wes

[lakespookie]

I have questions
1. how are you guys handling the fuel pump interface?
2. do you have a bypass valve for the pump?
3. if not how are you reducing the pressure seen at the aerocarb assume a runaway pump?
4. if you are just going to pull the circuit breaker in that case how are you addressing the fuel feed restriction introduced by the fuel pump?

[phenry]

Fuel tank head pressure only fuel is available to both the EFI pump and the Aeroinjector via the gascolator at all times.
When running EFI the Aeroinjector mixture is closed, no flow to Carb.
When not running EFI, pump is de-energized along with ECU and mixture is open, flow to Carb.

1. how are you guys handling the fuel pump interface?
The pump inlet (in my case) is plumbed directly into the gascolcator. This is only at fuel tank head pressure. When the pump is stopped no fuel is pumped.
2. do you have a bypass valve for the pump?
Because of the arrangement described above, there is no need for a bypass valve.
3. if not how are you reducing the pressure seen at the Aeroinjector assume a runaway pump?
The Aeroinjector is supplied via a separate line directly from the gascolator with tank head pressure only. (see previous drawing)
4. if you are just going to pull the circuit breaker in that case how are you addressing the fuel feed restriction introduced by the fuel pump?
See above.



Hope this helps

[phenry]

Wes,

Again this is how I am doing it. You are on your own to add whatever you want.

A number of items are monitored in the usual manner CHT's etc. However in keeping with the KIS principal all I need to run my EFI is:
Crank position sensor.
MAP sensor

In addition we use a pot wired to the disused (Micro squirt) water temperature input to manually tweak the mixture (not used much once the tune is sorted).

You could add an oxygen (O2) sensor but they hate Avgas, are expensive, need an exhaust modification and don't last long. Once you have it tuned it is set and forget.

[WesRagle]

Pete,

I have no experience with this so all I "know" is what I've read. I'm going to try and read between the lines.

A number of items are monitored in the usual manner CHT's etc.

I had intended to use a single CHT probe in lieu of Water Temp. to implement "Warm Up Enrichment". Warm up enrichment is probably not needed and maybe not even desired. If you want to richen, just twist the knob.

In addition we use a pot wired to the disused (Micro squirt) water temperature input to manually tweak the mixture (not used much once the tune is sorted).

I'm guessing that you have substituted a pot for the water temperature probe and are using Micro squirt's warm up enrichment function to fine tune mixture.?.?

You could add an oxygen (O2) sensor but they hate Avgas, are expensive, need an exhaust modification and don't last long. Once you have it tuned it is set and forget.

I agree that an O2 sensor would run counter the requirement for reliability.

I assume the fuel pressure regulators are very reliable and there is no need to constantly monitor their output.?.?

As for Inlet air temperature, I assume the Micro squirt is using a default value to calculate air density. Do you happen to know what that default value is?

Thanks,

Wes

[phenry]

My EFI system is moving from experiment to refined/final version.
I thought you may like to see the latest and last version of the intake manifolds I have created for my EFI system.

Injector retaining/positioning plate.

Positioned on Aerovee intake.

Plate welded to intake with Hex Cap Head screws pressed in from the rear and ceramic coated.
The cap heads are used to retain the fuel rails.

Hole arrangement with injector seal in place.

Injector installed.

View from port side showing injector outlet.

Kind of like the Ferrari red (should go much faster).

Peter Henry
Waiex #149

[WesRagle]

Hey Peter,

That's awesome!

Earlier you said

The pump I am using draws 2.5 amps and can deliver up to 45 Lt/Hr.

What pump are you using?

Thanks,

Wes