What's new
Van's Air Force

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

Looking for best EFI system for lycoming 540

The poster made claims that semi-sequential injection is better. I also would like to understand the rationale for that claim, as I also question it. Also would like to know what he means by it, as the industry only refers to batch or sequential injector firing; Never heard of semi-sequential prior to this post. Nothing wrong with calling out a vendor to back up his claims, advertiser or not. I personally appreciate those that ask for backup details on claims of improvement.

From my experience, proper sequential injector firing is not possible without a cam position sensor.

Larry

My understanding is semi-sequential is similar to wasted spark but for injectors. So instead of firing all [4,6] injectors at the same time you fire pairs twice per revolution. At least that's how it's used in the MegaSquirt world. I don't know how it's used in the EFII32 system.

See this from Toyota's TCCS docs:

Simultaneous = Batch
[2,3] Groups = Semi-Sequential
Independent = Sequential

injector_timing.png
 
My understanding is semi-sequential is similar to wasted spark but for injectors. So instead of firing all [4,6] injectors at the same time you fire pairs twice per revolution. At least that's how it's used in the MegaSquirt world. I don't know how it's used in the EFII32 system.

See this from Toyota's TCCS docs:

Simultaneous = Batch
[2,3] Groups = Semi-Sequential
Independent = Sequential

injector_timing.png


I did a megasquit install on a 6 cyl porsche. It was batch fire with two discrete banks. I still consider all of these approaches as batch, as firing an injector, other than when the valve is open, all pretty much nets the same result, leaving out wall wetting and other minor differences, regardless of how far from valve opening the shot is made. That is my understanding, but am open to being schooled by those more knowledgable.

Note that the toyota approach essentially uses a cam pos sensor (signal picked up on the dist vs the crank) to fire the groups on the intake stroke, so possibly some benefit there. The EFII system does not have a cam pos sensor, so can't do this. They have no way to differentiate the power and intake strokes, as you can't put a cam sensor on the lyc and the mag runs 1:1 unlike an auto dist at 1:2. TOyota says that semi-seq requires knowing the intake stroke, yet Robert still claims semi-seq without it. Curious how he is doing it. Possibly he has a different definition.

Larry
 
Last edited:
Batch fire doesn't care about where the valve or piston is at all, any more than that is a concern on Bendix style mechanical injection which is continuous flow.

When the valve opens, fuel and air mixture is drawn into the cylinder.

Of note, even on "sequential" injection (I prefer the term timed), at high rpm and high throttle, there isn't enough time to inject all the fuel required when the valve is open so even sequential systems inject some fuel when the valve is closed at high power and rpm. Sequential is useful for low rpm, part throttle conditions in cars with emission standards to meet. Not much of those conditions in aircraft (mostly at or near WOT and no emissions).

We've triggered injectors in ones, pairs, threes and six at once over the years depending on application. They all ran just fine from idle to full power. Injection events are typically at half crank frequency (cam frequency).
 
Marketing

Or perhaps it's simply time to get answers to the questions? Or do we just close down threads one after another to avoid have a fact-based discussion in favor of marketing?

I certainly see a lot of “marketing”...

...and that’s all I have to say about that...

Now back to flying! Done with Phase 1!

Thank you Robert, the EFII system has been flawless...
 
Last edited:
Cam Sensor?

All the talk about sequential (can't do on a Lyc. without a cam sensor)...

Ross,

Great information in your post, as always!

I am, however, curious why you would need a cam sensor on an engine with fixed cam timing to use sequential injection? On Lycoming's, since the cam is directly driven at 1/2 crankshaft speed, I would think that you could directly derive the cam timing from a crankshaft trigger wheel such and Dan Horton is using for his EDIS ignition system. Of course it's all a moot point, really, because as you pointed out, there really isn't an advantage to sequential injection on our low speed, steady state (and large induction system volume) Lyconental engines...

EDIT: I may have just answered my own question, because I realize there's that pesky issue of determining which cycle each cylinder is on...

Skylor
 
Last edited:
Ross,

Great information in your post, as always!

I am, however, curious why you would need a cam sensor on an engine with fixed cam timing to use sequential injection? On Lycoming's, since the cam is directly driven at 1/2 crankshaft speed, I would think that you could directly derive the cam timing from a crankshaft trigger wheel such and Dan Horton is using for his EDIS ignition system. Of course it's all a moot point, really, because as you pointed out, there really isn't an advantage to sequential injection on our low speed, steady state (and large induction system volume) Lyconental engines...

EDIT: I may have just answered my own question, because I realize there's that pesky issue of determining which cycle each cylinder is on...

Skylor

Yes, your edit is correct.

We've been asked many dozens of times over the years how injectors are timed to valve opening- short answer, they aren't and it doesn't matter how the injector drive wires are hooked up, except in the case of systems with fuel trim where we do have to have the right transistor driving the right injector so that you trim the correct injector/ cylinder with the programmer.

Dave Anders did a lot of flight testing on induction tubes lengths (still ongoing BTW) and found the fuel trim indispensable for determining if he was getting more airflow or less. He really wanted to time injector spray to valve open time but there is no way to do this without a cam sensor and again, you'd need very large injectors to get the squirt done in the available time the valve is open at high power.
 
...you'd need very large injectors to get the squirt done in the available time the valve is open at high power.

So Ross, when and how long does an SDS system open its injectors? Pick whatever RPM and A/F you like. Let's see the result...here's the base diagram.
-
 

Attachments

  • Injection Illustration.jpg
    Injection Illustration.jpg
    28.9 KB · Views: 117
So Ross, when and how long does an SDS system open its injectors? Pick whatever RPM and A/F you like. Let's see the result...here's the base diagram.
-

It's going to depend on fuel rail pressure and delta across the injector face - but since Ralph Inkster has been kind enough to publicly share his EM-5 map, we can use his for an example. Mine largely dovetails.
 
Dan,

I'll try to find time tonight to look at some in-flight logging and grab a couple screenshots showing the injector pulse widths at different MAP/RPMs for you. Not sure I logged much in the gauge screen, mostly would have been in the graphical mode which displays duty cycle.

As far as when the injector pulse occurs after the trigger pulse occurs, I'd say I don't know the answer to that as it's not something we have a reason to care about as outlined above. It's unimportant in a batch fire system like SDS. I think you have a couple SDS ECUs there. If you have a dual trace scope, you can check that for your own interest.
 
Ok, here are couple shots of the data log gauge screen, first at 2300 rpm and second at 2700 rpm. No O2 sensor hooked up so AFR is invalid here. On most Lycoming maps, we ramp up the rpm fuel values (richening AFRs) at 2600-2700 rpm for cooling at full power and in the initial climb. That's why the injector duty cycle is so much higher here with only 400 more rpm, same MAP. Users can program values as they wish of course to suit their preferences and mission.
 

Attachments

  • dat1.jpg
    dat1.jpg
    174 KB · Views: 159
  • dat2.jpg
    dat2.jpg
    196 KB · Views: 141
Ok, here are couple shots of the data log gauge screen, first at 2300 rpm and second at 2700 rpm.

Ok, the given values say the all the injectors open together for 50% of a 720 degree crank rotation at 2252 RPM. At 2672 RPM, the injectors are open 72% of a 720 degree crank rotation.

As noted, with batch injector operation it doesn't much matter exactly when the injectors open. Are you triggering from #1 TDC? Here I've arbitrarily started the open periods about 90 degrees before #1 TDC, but it's just a random pick for illustration.
-
 

Attachments

  • Injection Illustration 2252 50.jpg
    Injection Illustration 2252 50.jpg
    40.1 KB · Views: 113
  • Injection Illustration 2671 72.jpg
    Injection Illustration 2671 72.jpg
    40 KB · Views: 113
Last edited:
Code to generate the injector pulse is executed 120 microseconds after the falling edge of the crank sensor pulse occurs and then the injector windings take around 0.4 to 0.7 milliseconds to saturate and start opening (dependent on type, voltage and fuel pressure).

The trigger pulse on Lycomings occurs around 90 deg BTDC, depending on type of Hall Effect sensor used.
 
Last edited:
Code to generate the injector pulse is executed 120 microseconds after the falling edge of the crank sensor pulse occurs and then the injector windings take around 0.4 to 0.7 milliseconds to saturate and start opening (dependent on type, voltage and fuel pressure).

The trigger pulse on Lycomings occurs around 90 deg BTDC, depending on type of Hall Effect sensor used.

Excellent. That means the open point in the diagrams is pretty darn close, certainly close enough for this illustration.

Ok, I've attached a similar diagram for mechanical constant flow, i.e. the Bendix style injection common on our Lycomings (Bendix, AFP, Precision, AvStar).

For those on a learning curve here, mechanical constant flow operates as its name describes, flowing fuel all the time into every intake port. The system varies flow rate to vary the volume delivered. An EFI has roughly the same flow rate any time the injectors are open, and varies the time the injectors stay open in order to vary the volume delivered.

Now, can we get Mr. Paisley or a System 32 user to diagram semi-sequential, and offer a technical explanation of why it was chosen?
-
 

Attachments

  • Injection Illustration Constant flow.jpg
    Injection Illustration Constant flow.jpg
    42.4 KB · Views: 113
Now, can we get Mr. Paisley or a System 32 user to diagram semi-sequential, and offer a technical explanation of why it was chosen
-
Come on now, don't be Bashful...
-
 

Attachments

  • Profile_-_Bashful.png
    Profile_-_Bashful.png
    629.9 KB · Views: 85
Carl,


The biggest challenge of mechanical injection is achieving fuel balance between the cylinders - there is no positive control of the fuel distribution like there is with electronic injection. You can try to balance mechanical injection at one operating point with precision injectors. But the balance will fall apart at every other operating point. And it's really bad at low spider pressures such as during engine starting (hello hot start).

Robert
EFII

My experience with a Bendix RSA-5 and balanced injectors is that this statement is not true. I have essentially zero GAMI spread over a reasonable range of RPM and MAP. The balance does not "fall apart" at all. I can run 100F LOP smoothly at anything between 2300 and 2600 RPM, WOT throttle anywhere from 6,000 to 12,500 ft. I have one electronic ignition and one mag.

During engine start, who the h#*l cares how well the injectors are balanced? Once you get it lit, you can trim mixture to get a reasonable idle. YES, hot starting is a PITA and one of the strongest reasons for EFI. And, yes, sure, the EFI engine likely idles smoother. Not so much that I really care.
 
During engine start, who the h#*l cares how well the injectors are balanced? Once you get it lit, you can trim mixture to get a reasonable idle. YES, hot starting is a PITA and one of the strongest reasons for EFI. And, yes, sure, the EFI engine likely idles smoother. Not so much that I really care.

Well, with a purge valve even the hot restart issue largely goes away. I think the biggest reason for EFI is you could darn near **** in the tank and it would run.

Some people are hellbent on running crappy fuel. I'm not one of those people, and I think you can safely run some mogas with mech-inj if you mix or only use it as cruise, but for sure EFI is better in this regard.
 
I assume Matt is referring to vapor pressure, not price or quality. Any EFI has a clear advantage over constant flow when the fuel has a high vapor pressure (i.e. mogas), as the fuel is held at relatively high line pressure all the way to the injector.

Thread drift. Let's ease back toward "best EFI", as this isn't about fuel choice.
 
I am curious how and to what degree a purge valve helps hot starting on a mechanical FI. My hypothesis has always been that fuel boiling occurs in the fine stainless steel 'spider' lines that run from the fuel divider to the injectors, where they pass just above hot cylinders. The vaporizing fuel forces fuel through the injectors and into the cylinder intake chambers, so the engine is essentially 'flooded'.

If I were to put a purge valve and a return line at the fuel divider, I could flush out hot fuel up to that point, but it would still leave the individual injector lines full of hot fuel/vapor that floods the engine.

Is my hypothesis incorrect?
 
I am curious how and to what degree a purge valve helps hot starting on a mechanical FI. My hypothesis has always been that fuel boiling occurs in the fine stainless steel 'spider' lines that run from the fuel divider to the injectors, where they pass just above hot cylinders. The vaporizing fuel forces fuel through the injectors and into the cylinder intake chambers, so the engine is essentially 'flooded'.

If I were to put a purge valve and a return line at the fuel divider, I could flush out hot fuel up to that point, but it would still leave the individual injector lines full of hot fuel/vapor that floods the engine.

Is my hypothesis incorrect?
It seems correct to me. Thankfully the volume in there is very low so if there is cool fuel just before the spider the starting is a bit easier - at least that's the theory as I understand it.
 
I am curious how and to what degree a purge valve helps hot starting on a mechanical FI. My hypothesis has always been that fuel boiling occurs in the fine stainless steel 'spider' lines that run from the fuel divider to the injectors, where they pass just above hot cylinders. The vaporizing fuel forces fuel through the injectors and into the cylinder intake chambers, so the engine is essentially 'flooded'.

If I were to put a purge valve and a return line at the fuel divider, I could flush out hot fuel up to that point, but it would still leave the individual injector lines full of hot fuel/vapor that floods the engine.

Is my hypothesis incorrect?

Correct, BUT

Yes, the fuel in the SS lines is boiling. However, after shutdown, you engine compartment quickly gets to around 180-200* After several minutes, all of your fuel back to the firewall will be around 180*. Therefore, even after getting it started, it will be rough for a while until cooler fuel gets to the injectors. The purge valve won't do a lot for the initial start, as your already flooded from the boiling fuel in the injector lines, but will help a great deal in not having a rough engine for 2 minutes after start.

Larry
 
Last edited:
Correct, BUT

Yes, the fuel in the SS lines is boiling. However, after shutdown, you engine compartment quickly gets to around 180-200* After several minutes, all of your fuel back to the firewall will be around 180*. Therefore, even after getting it started, it will be rough for a while until cooler fuel gets to the injectors. The purge valve won't do a lot for the initial start, as your already flooded from the boiling fuel in the injector lines, but will help a great deal in not having a rough engine for 2 minutes after start.

Larry

Indeed. On my EFI system I can reach down and feel the return fuel line from the cockpit. A hot soaked engine will see that section of fuel line rise to a temperature thats almost too hot to touch, then quickly cool to the tank temperature in about 10 seconds. That is a lot of fuel considering the EFI is bypassing all of it at 40+GPH. The same volume of fuel would take a deadhead Bendix system several minutes to approach the same level of cooling (if ever). A purge valve gets cold fuel to the flow divider very quickly and that's what gets you the benefit.
 
Correct, BUT

Yes, the fuel in the SS lines is boiling. However, after shutdown, you engine compartment quickly gets to around 180-200* After several minutes, all of your fuel back to the firewall will be around 180*. Therefore, even after getting it started, it will be rough for a while until cooler fuel gets to the injectors. The purge valve won't do a lot for the initial start, as your already flooded from the boiling fuel in the injector lines, but will help a great deal in not having a rough engine for 2 minutes after start.

Larry

Thanks, this is helpful. I rarely get enough heat soak to have a prolonged rough running. Usually a quick run-up clears it. Once in a while when it is really hot, it coughs and barks until take-off power is applied. But the key is the purge system won't prevent the initial flooding. So the same hot-start procedure would be used: WOT, mixture idle c/o until it fires, then the quick two-step of push the red knob in and pull the black knob out.
 
I’ve found that running up the engine to 1200 rpm and then shutting the purge valve helps on subsequent hot starts. Maybe it helps burn off more of the fuel in the injector lines?
 
I'm in the process of closing on a HR2 with no purge valve(yet?). Does anyone kill the engine via fuel selector to cutoff? Any chance the would help until a valve or EI is installed...?
 
I think the hard starting of mechanical FI is way overblown. Proper technique makes it a non issue, even with a pair of magnetos. Look at the competition Pitts guys from the 70's... None of them had a starter! All hand propped, all magnetos. Unfortunately, "proper technique" is very airplane specific. The type of fuel pump used, the number of 90's on the suction side, gascolator or not, ect. All define a system that needs to be learned by experience, not rote process from a forum. Learn the systems "needs" and never have an anxious moment at the gas pumps.

EFI simply takes the skill out of the equation. You just push a button and its running.
 
EFI simply takes the skill out of the equation. You just push a button and its running.

...which is a significant plus for either currently available PWM FI system, as compared to constant flow FI.

Which brings us back to the OP's question.

I've attached a semi-sequential injection diagram below, compared to batch injection. If installed on identical engines, either must inject the same fuel quantity for similar power, so the difference here is merely when the injectors are opened. It's probably a good representation of the System 32. If not, I'm sure we would all appreciate a correction. One wildcard, for example, might be injector flow rate, which would change the pulse width.

Take a hard look. Is there an advantage to one or the other, and if so, why?
.
 

Attachments

  • 72%.jpg
    72%.jpg
    74.5 KB · Views: 166
Thanks, this is helpful. I rarely get enough heat soak to have a prolonged rough running. Usually a quick run-up clears it. Once in a while when it is really hot, it coughs and barks until take-off power is applied. But the key is the purge system won't prevent the initial flooding. So the same hot-start procedure would be used: WOT, mixture idle c/o until it fires, then the quick two-step of push the red knob in and pull the black knob out.

I also sometimes give a .5 second shot with the boost pump before starting with that technique. It floods it a bit more, but get fuel into the SS lines so that it is less likely to quit after it catches.
 
Thanks, this is helpful. I rarely get enough heat soak to have a prolonged rough running. Usually a quick run-up clears it. Once in a while when it is really hot, it coughs and barks until take-off power is applied. But the key is the purge system won't prevent the initial flooding. So the same hot-start procedure would be used: WOT, mixture idle c/o until it fires, then the quick two-step of push the red knob in and pull the black knob out.

I installed a fuel return line on my system, similar to the purge valve. It goes from the servo back to the tank, through a valve and small orifice. It is mostly for mogas, which will often be very rough after a hot soak. It gets rid of a lot of the hot fuel before starting.

The 100LL seems to do pretty well after the heat soak, but mogas is much worse, with it's lower boiling point.
 
...which is a significant plus for either currently available PWM FI system, as compared to constant flow FI.

Which brings us back to the OP's question.

I've attached a semi-sequential injection diagram below, compared to batch injection. If installed on identical engines, either must inject the same fuel quantity for similar power, so the difference here is merely when the injectors are opened. It's probably a good representation of the System 32. If not, I'm sure we would all appreciate a correction. One wildcard, for example, might be injector flow rate, which would change the pulse width.

Take a hard look. Is there an advantage to one or the other, and if so, why?
.

Good work bringing this back on topic.

Sounds like you are trying to decipher what Robert means by, "System32 provides semi-sequential fuel delivery timed to intake valve operation."

I suspect that all that means is that he has a dedicated injector channel for each cyl so he uses the crank position sensor to determine where he thinks the cyl is at in the cycle and injects fuel at a specific time as opposed to once per cycle for all cylinders.

Whether that makes a lick of difference really boils down to two things in my mind:

1. As discussed earlier, how would he be able to determine if we are on the intake or exhaust stroke. Without this information he is blind, so everything else is moot.

2. Even if he did know exactly where the piston is in it's stroke, does the fuel atomize better if it's sprayed on the back of a open or closed valve? In our application, I suspect it matters little.

On his comparison page he mentions a lot of marketing stuff that doesn't really matter, but he also mentions a few things that are nice.

1. Internal data logging to SD card.
2. Software is field updatable.
3. Provisioning head configures both ECUs at the same time.

He also talks about EMI protection and voltage input range, but either of these systems is susceptible to poor aircraft electrical system design.

As for mechanical vs EFI. I think that the real advantage of EFI is in the starting (especially hot starting) of the engine. Mechanical or electronic injector is more or less doing the same thing, especially when the electronic injector is in the same hole as the mechanical one, which is my understanding from both setups.

Ross, when you finally do start on EM-6 a few things that I think you should make better:

1. The three things above.
2. Internal battery fail over like the CPI-2.
3. Enough injector channels for an 8cyl.
4. Wifi access point that allows engine monitoring from any wifi/browser.

The ESP32 microcontrollers are pretty easy to work with, fast, robust and have built in wifi/bluetooth. If you were especially clever, you could provide a link in the web interface to upload all of the config parameters, engine sensors, and system logs directly to you for troubleshooting.

Also, I know you are in love with your website, but many of us aren't. Please consider using a real web framework, fonts, javascript, and other modern web tech that doesn't look like it was written in notepad in 1997.

Matt
 
I installed a fuel return line on my system, similar to the purge valve. It goes from the servo back to the tank, through a valve and small orifice. It is mostly for mogas, which will often be very rough after a hot soak. It gets rid of a lot of the hot fuel before starting.

The 100LL seems to do pretty well after the heat soak, but mogas is much worse, with it's lower boiling point.

I'm planning on something similar. A tee, with a small orifice that routes around 6-8GPH back to the tank to give a place for vapor to go and to have constant fuel flow to cool the mech pump.

What happens on your system when you leave the valve open all of the time? Does that help with running mogas?

Matt
 
Some automotive sequential systems actually started spraying on a closed valve under certain low load conditions. Fuel flashes off a hot valve just fine.

In any case, sequential has to spray on a closed valve as MAP and RPM increases so that becomes moot anyway on aircraft applications where most of the time is spent at those conditions anyway. Only way around this is an immense injector which would compromise idle quality.

We are working on the user re-flashable code now. Plan is to be able to log to and read from an SD card.

We may be able to selectively change values in both ECUs however 4 cylinder backup ones have only minimal sensors to run and use different default values as a result so that doesn't make sense in that case.

The current EM-5/ V32 has considerable hard EMI protection as well as lots of software to protect from this. We don't see many problems in this regard.

The other brand mounts the injectors in the primer ports with the heat drawbacks from cooling air and adjacent exhaust stacks. Ours are mounted on top where Lycoming mounts theirs- in cool air.

No plans for auto battery switchover on the EM-6 at this time. With low voltage warnings on glass panels today, lots of time to switch to backup power if the main alternator takes a dump.

We've never fitted a system to a 720 yet and the market is too small to justify designing for those. With single plug engines like LS V8s, doesn't make much sense to back up fuel but not spark. We already do quite a few ECUs for these engines in aircraft. Trim isn't needed because they have properly designed manifolds already.

No wifi planned. Want to get this to market in a reasonable time frame. It's already 3 years late because we've been so busy. Will use the same family of micros as well, no need for anything more in this market.

I won't be changing the website because other people think we should. We are buried in orders as is. It will remain informational rather than artsy which serves no useful purpose in my view. We're a small company, no time to do this with such an immense site. Lots of older folks (our main market) say they like the present format. Can't please everyone and I'm not going to mess with success.
 
Last edited:
Some automotive sequential systems actually started spraying on a closed valve under certain low load conditions. Fuel flashes off a hot valve just fine.

In any case, sequential has to spray on a closed valve as MAP and RPM increases so that becomes moot anyway on aircraft applications where most of the time is spent at those conditions anyway. Only way around this is an immense injector which would compromise idle quality.

We are working on the user re-flashable code now. Plan is to be able to log to and read from an SD card.

We may be able to selectively change values in both ECUs however 4 cylinder backup ones have only minimal sensors to run and use different default values as a result so that doesn't make sense in that case.

The current EM-5/ V32 has considerable hard EMI protection as well as lots of software to protect from this. We don't see many problems in this regard.

The other brand mounts the injectors in the primer ports with the heat drawbacks from cooling air and adjacent exhaust stacks. Ours are mounted on top where Lycoming mounts theirs- in cool air.

No plans for auto battery switchover on the EM-6 at this time. With low voltage warnings on glass panels today, lots of time to switch to backup power if the main alternator takes a dump.

We've never fitted a system to a 720 yet and the market is too small to justify designing for those. With single plug engines like LS V8s, doesn't make much sense to back up fuel but not spark. We already do quite a few ECUs for these engines in aircraft. Trim isn't needed because they have properly designed manifolds already.

No wifi planned. Want to get this to market in a reasonable time frame. It's already 3 years late because we've been so busy. Will use the same family of micros as well, no need for anything more in this market.

I won't be changing the website because other people think we should. We are buried in orders as is. It will remain informational rather than artsy which serves no useful purpose in my view. We're a small company, no time to do this with such an immense site. Lots of older folks (our main market) say they like the present format. Can't please everyone and I'm not going to mess with success.

I respect that, and I didn't mean to come across as overly critical, the idea was to give you some feedback from a consumer that has more understanding than most.
 
I respect that, and I didn't mean to come across as overly critical, the idea was to give you some feedback from a consumer that has more understanding than most.

No worries, I appreciate your comments and suggestions. We have to prioritize our energies on customer support, work on the new products, production and getting orders out the door. It's a 60 hour week for me these days and same for Barry. We've added some more staff here trying to keep up with all this demand. :)
 
...trying to decipher what Robert means by, "System32 provides semi-sequential fuel delivery timed to intake valve operation."

I doubt it is timed to intake valve operation...at least not all the intake valves. More likely half of them.

I suspect that all that means is that he has a dedicated injector channel...

Probably true, but how to know why? There's no operating manual available for download. Apparently permission to read it requires a decoder ring and membership in Orphan Annie's Secret Society. Until we can buy enough Ovaltine, the only clue is numbered injector harness wires.

Too bad we can't get some clarification, and a good explanation of why it's claimed to be better.
 
I doubt it is timed to intake valve operation...at least not all the intake valves. More likely half of them.



Probably true, but how to know why? There's no operating manual available for download. Apparently permission to read it requires a decoder ring and membership in Orphan Annie's Secret Society. Until we can buy enough Ovaltine, the only clue is numbered injector harness wires.

Too bad we can't get some clarification, and a good explanation of why it's claimed to be better.

"Be sure.... to drink..... your Ovaltine!"

"A crummy commercial?... son of a B..........ch!!!"
 
I'm planning on something similar. A tee, with a small orifice that routes around 6-8GPH back to the tank to give a place for vapor to go and to have constant fuel flow to cool the mech pump.

What happens on your system when you leave the valve open all of the time? Does that help with running mogas?

Matt

I did the calculations to limit flow to about 10 GPH through the restrictor. While SOP is to close the valve at take off, it can be left on indefinitely, as the pump produces about 45 GPH of flow, so there is plenty of overhead. Downside, is that it takes a good minute or so to get cool fuel into the lines. Higher flow rates will cool faster, but require more diligence in ensuring the valve is closed under high power regimes.

Larry
 
PIREP From A Happy SDSEFI Customer

I am stepping into an unknown area here. I have an injected 540 but looking to step that up and get better performance and control. I am not an engine guy so looking for input to make my 540 fly the best. Thanks to all that can give feedback

Hi

I thought I'd answer the original question for this thread. I selected the SDSEFI system for my RV-10 IO-540 (CAI + 9:1 compression).

To my mind there are three key elements that must be considered : Support / reliability / function. All are important but I'll deal with support first.

Support

I became aware of SDS when I became part of the Eggenfellner / Subaru engine debacle. When Eggenfellner's support for this engine package disappeared, it was Ross at SDS who stepped up and helped me get the engine working. I throught I bought a plug and play engine but was sadly mistaken.

Ross supplied ECU's to Eggenfellner who specified how they were programmed and locked (to prevent user adjustment). Ross had no responsibility to the final purchaser - this was Eggenfellner's responsibility.

Even so, Ross unlocked my ECU and guided me through adjusting the ECU values and installing sensors so I could get the engine working safely. For all his help he never charged me a dime. His commitment to his customers and the community at large won my greatest respect. I credit him in keeping me safe during my first 50 hours with the Subie engine. Postscript - the engine did die an early death due to cooling issues (unrelated to the ignition / injection system).

It is one thing to support your own product, entirely another to support some else's.

Reliability

Through my involvement with the Subie I became quite aware of the SDS ECUs operated and how robust they were. When I learned that there were nearly 2,000 ECUs in service, there was no doubt that I would install an SDS system in my next RV10. Over the past two years I have logged 190+ hours on my SDS equipped -10. The system has been flawless. It has met and surpassed all my expectations. My third -10 will have the same system.

Functionality

The SDS system does everything I need it to do. In terms of functionality, what I need the ECU to do is run my engine smoothly and efficiently. The SDS system does both. I run LOP whenever appropriate and see about a 9% fuel savings with only a small airspeed loss. I could save more but I like to fast and high. My engine runs very clean.

Okay - there is one other factor - Bling. I like aircraft bling. I installed Ross's new square programming head because it looks pretty on my panel. The old round instrument - not so much.

I have also installed all the CNC anodized engine parts that Ross sells (valve covers etc, etc, etc). They make an engine look quite spiff.

Full Disclosure

For the reasons above I am a big supporter of Ross and SDS. Some of you may have seen his RV-10 videos taken in my -10. My engine pix are also on his site. This I have done as a way of saying thank you. I won't even let him buy gas or discount my purchases. I owe a great debt as he kept me safe with the Subie engine.
 
FlyEFII.com for sure.
These guys have the next generation product after the older EFI SDS.
They provide everything needed for conversion/installation and they install the magnets on your flywheel as part of the installation cost.
In process converting my Lyc O-360 now and have been very impressed with their quality parts, support, and advice.
 
Back
Top