What's new
Van's Air Force

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

N829MS - SDSEFI system flys

M McGraw

Well Known Member
Friend
Yesterday I took to the air for a first flight. It runs smooth and stable. I now have 4.5 hours on it. I will add the gear fairings in the next few weeks. My wife took takeoff and landing videos.
oJL.jpg


First takeoff. https://youtu.be/zg9WrSeh3sA
First Landing. https://youtu.be/2kz161vYcy8
 
Last edited:
Congratulations

Nicely done Marvin. Great looking 14!
Beautiful field you have there as well.
 
Looks great! Nice strip too.

Please report the tuning and other particulars on the SDS system when you get a chance.
 
Tuning

Looks great! Nice strip too.

Please report the tuning and other particulars on the SDS system when you get a chance.

I have my ignition set up to simulate a magneto currently. Once the rings are seated I plan to play with the ignition timing and compare that to the baseline I'm getting now.

For the fuel map, I have 500 - 1500RPM set at an AFR between 12 and 12.5 then 1600 - 2700RPM between 11.5 and 12. I only have 4.5 hours on the engine so I do not yet have much experience with the system, but I do like what I've seen so far!
 
I think I have the ignition side pretty well figured out, so once you are comfortable feel free to give me a shout. As for the fuel side, I'm some months away from flying, so I'm going to be leaning on you guys for guidance when the time comes.

I'm guessing you are using the integrated fuel flow function? If so, how's that working out?
 
Ignition

I think I have the ignition side pretty well figured out, so once you are comfortable feel free to give me a shout. As for the fuel side, I'm some months away from flying, so I'm going to be leaning on you guys for guidance when the time comes.

I'm guessing you are using the integrated fuel flow function? If so, how's that working out?

I saved a copy of your ignition info from a previous post and plan to follow your lead. I do intend to publish all my number for an IO-390 after the tweaking is done. I will run them by Ross and Barry first so I don't lead anybody off the path. If you don't have SDSDASH get it, it is great for looking at the numbers after a flight.
 
I saved a copy of your ignition info from a previous post and plan to follow your lead. I do intend to publish all my number for an IO-390 after the tweaking is done. I will run them by Ross and Barry first so I don't lead anybody off the path. If you don't have SDSDASH get it, it is great for looking at the numbers after a flight.

For those not familiar, SDSDASH is the optional PC data logging feature we offer. It's been very helpful to assist customers with tuning or running issues. You can email the file to us for advice. We can quickly spot many problems and give you feedback on how to fix them. Great diagnostic tool.

Dave Anders has used it a lot in combination with Savvy during his test flying. We've improved the software many times now, adding features based on customer feedback. You can download the latest version any time after you've purchased this initially. We have some future ideas to improve ease of use next year when we get time to work on it in depth.

log7_zpsxjtdgqea.jpg
 
Last edited:
Congrats Marvin! Great to see that bird in the air.

I still have those wing stands you gave me, and will be putting my wings in them hopefully by the end of the year :)

If you are ever down around the KCHA/KFGU/KRZR area let me know.
 
Phase 1 Update

Hello,
I now have 11 hours on the airplane and I am more impressed with the Vans product than I expected. I am 5'11" and I have the rudder pedals in the middle location with the seat back in the most forward location and a 1" booster cushion under my Classic Aero seats. That position gives me very good visibility and my legs are quite comfortable. The aircraft is truly roomy. Stalls are basically benign both clean and full flaps. No wing drop just a little buffett and altitude loss. To recover simply reduce back pressure to break the stall then add power as needed to regain altitude. The landing flare is predictable and the gear are solid and consistent. It is surprisingly easy to land and stable during the rollout.

The controls are well harmonized without the pitch sensitivity I have experienced in many EAB aircraft. My first two flights concentrated on watching engine gauges and general control feel. I did need to adjust the propeller low pitch stop and the left aileron, but small changes like that are to be expected.

The third flight confirmed the adjustments previously made and I recorded the amp draw of each item I could turn on and off to verify my load shed ability since I have a totally electric aircraft including ignition and fuel injection.

Flights 4, 5, and 6 were used to calibrate the fuel flow and other engine sensors. Toobuilder previously posted a question on the SDSEFI fuel flow calculation. I do not have a flow meter in my system, I use the software inside the SDSEFI. For my IO-390 I turned the fuel pressure up to 49PSI from the factory standard 40PSI. This was done because to adjust air fuel ratio (AFR) at 100RPM increments you input a value between 0 and 255 at each increment which tells the injectors how long to stay open. At 40PSI I ran close to 255 at 2600 and 2700RPM. At 49PSI the numbers are lower and my average injector duty cycle is well below 60% (closer to 40% under most power settings). An obvious question would be why 49PSI and not a nice round 50? Well my Garmin sensor kit included a Kavlico P4055-50G sensor that is a 50PSI sensor and the numbers go blank at exactly 50PSI:). I need to see if there is such a thing as a Kavlico P4055-60G? In the Garmin G3X system, for the fuel flow measurements I chose "Fuelscan 201B" and for the return flow I chose "none". With those selections I needed to increase the K factor about 17.25%; old K = 29500 new K = 34587. The K factor can be adjusted in either the SDS computer or the Garmin system to the same result, I chose to adjust the Garmin. I may tweak the K value slightly, but currently I am very happy with the fuel used versus fuel pumped at fill up time. BTW, increasing the fuel pressure may have increased the need to adjust the K factor?

A quick side note, we are very fortunate in the EAB world to have access to systems like Garmin, Dynon, GRT, etc... I have features I simply could not afford in my previous certified aircraft. These are many of the same features I enjoy at work, but that aircraft cost more than a lifetime of income, okay 20 or 30 lifetimes :). Also the electronic ignition and fuel injection make starting my Lycoming as simple as pushing the start button. Imagine that, it starts as easily as my car which cost far less, no hot start or cold start procedures, just push the button :). It runs VERY smooth and I have not yet balanced the prop!

Flight 7 served no purpose because weather interrupted my plan. I was testing the SDS fuel trim option. One thing I noticed is the Vans baffle system for the RV-14 is very impressive. I trimmed the fuel (about 1 to 2%) to match the EGT's and noticed the CHT's were amazingly even with lots of room to spare for my planned future cowl flap system. All my past experience with Lycoming was to find peak either on the Alcor gauge or by finding the "rough" spot then increasing mixture for best power. I tried to simulate that experience to my great surprise. As I leaned, the engine remained exactly as smooth as before. The EGT, CHT, and IAS changed then suddenly dropped yet the engine never gave ANY hint of roughness. I think I just experienced Lean of Peak and it was anti-climatic :). I would have liked to continue learning, but weather intervened.

At this point, I am satisfied the engine sensors are good so my future flights will concentrate on calibration of the pitot-static system to see what speeds I can get. I really see no point in publishing speed numbers until I know they are accurate. Once I have accurate speed numbers I will move on to Vx, Vy, glide speed etc.. and I will work on my cowl flap system to see if it actually increases speed or merely CHT's.....
 
Last edited:
Outstanding Marvin. I know alot of builders, and a few vendors are very interested in your flight tests!
Tom
 
Congratulations Marvin!

I will be anticipating the walk around of your beautiful plane at Oshkosh.

I also plan to fly my 14A up there too. But I have a lot of little things to tweek to make my plane just the way I want it.

Fantastic job!

Steve Brown
 
OSH

Congratulations Marvin!

I will be anticipating the walk around of your beautiful plane at Oshkosh.

I also plan to fly my 14A up there too. But I have a lot of little things to tweek to make my plane just the way I want it.

Fantastic job!

Steve Brown

Sadly,
I will miss Air Venture this year, it conflicts with my parents 60th anniversary party😊 I would like to see you plane. I go to Port Columbus a few times a year. Which airport do you use?
 
Phase 1 complete!

Today I completed Phase 1. I put the airplane in the hangar at exactly 40.0 hours. Completed the logbook signoffs and washed the plane. I patiently waited for my wife to get home from work. It seemed only fair that she would be the first passenger. If only she had let me do a little of the flying. :)
oJA.jpg
 
Last edited:
Update

Hello Rob,
My last update focused on testing at fixed timing (20BTDC). I have done significant testing on variable timing both ROP and LOP. I need a few more months to make sure my data is repeatable.

What I can say is the timing curve for ROP is very flat so increasing timing does increase performance, but I?m not certain it is enough of an increase to justify the unknown reduction in detonation margin.

LOP results are trending much better and I do like what I see; however, I really need to make certain my data is repeatable.
 
1st flight

Yesterday I took to the air for a first flight.

Congrats Sir great job and the fun is just starting!
looking forward to seeing the EFII map you post
Lee
 
Be aware that most of the modern SDS setups are using different injectors with different flow rates in different placements and orientation from the rebranded setups. Fuel maps will not be comparable as a result.

Ignition maps won't be comparable either unless on the same engine, CR and fuel (and be sure to verify actual ignition timing and Magnet position). Some builders have found actual ignition timing has been advanced 1-3 degrees when checked from the pre-set values.
 
Update at 175 hours

I have been testing variable timing for quite some time. It appears my aircraft does best at about 9500PA (11,500DA), 65%[email protected]. Below is a comparison of our return trip from Air Venture 2018. I cannot identify a speed advantage of 125ROP over 100ROP. The screenshot below is at 100ROP at 20BTDC. I have tested ROP timing and found that 23BTDC is optimal; however, the speed increase was not worth the unknown detonation margin for me.
2dcfzns.jpg


This second screenshot is three minutes later at approximately 30LOP and 27BTDC, 64%[email protected]. The LOP curve proved very very flat. I tested significantly more advance than 27BTDC, but again my tolerance level for running a $40,000+ engine is conservative. When I go LOP I loose about 10kts; however, advancing the timing recovers a significant portion of that loss while saving 2 or more gallons per hour. The leaner the mixture more advance is required. This is simply a snapshot in time. I have gone as far as 175LOP with the engine running smooth at incredible fuel flows; however, the speed was not so incredible.
25qdh6q.jpg


There is a post from approximately 2 years ago by a Vans pilot (Scott M.) who got 169KTAS at 8.8gph with fixed magnetos. I pay a penalty for increased weight and drag of larger landing gear. My future plan is to compensate with a cowl flap. With CHTs in the low 300s there is ample cooling drag margin to gain what I suspect will be ~5Kts or 0.5gph.
 
Last edited:
Here is the RV14 using stock mags last fall on my way to Arizona.

I have been a long time user of one electronic ignition/ one mag, on my previous planes but went with mags for simplicity on the RV14. Both systems have advantages but the numbers on this comparison look pretty similar to the posted images above. The engine is the stock Van's supplied IO390.



25as1dy.jpg
 
When making changes to my aircraft I always fly beside a buddy before and after the change. We note his power setting for the first flight and then he flies that same setting for the "change" It is very clear if there have been any differences in speed. Using this method you get away from differences in temperature, pressure etc. I used the gps triangle system for many years but the differences in power due to temperature alone made the comparisons difficult.
 
Thanks for posting your timing findings Marvin. The TAS vs. FF seems pretty good with the draggier gear. If it's costing 5 knots then you are burning maybe about .5 to .8 gph less than Tom for the same TAS.

Hopefully you'll find a few knots with the cowl flap. That will be interesting to see too.

175F LOP, wow.
 
Last edited:
I cannot identify a speed advantage of 125ROP over 100ROP.

That aligns with dyno sources.

The screenshot below is at 100ROP at 20BTDC. I have tested ROP timing and found that 23BTDC is optimal; however, the speed increase was not worth the unknown detonation margin for me.

If it helps, other than short experiments, I've been at 23 since the first dyno run.

This second screenshot is three minutes later at approximately 30LOP and 27BTDC, 64%[email protected]. The LOP curve proved very very flat. I tested significantly more advance than 27BTDC, but again my tolerance level for running a $40,000+ engine is conservative.

I was out at 35 BTDC, at 16.5K and 30 LOP for a bit last week. Not desirable. Higher CHT and OT, and 5 knots slower.

I'm beginning to suspect flat advance curves are a characteristic of the angle valve. Looking forward to your data.
 
If it helps, other than short experiments, I've been at 23 since the first dyno run.

Actually, It helps significantly. I don’t like to stray too far from established data without a quality check. Are you running 23BTDC at takeoff power? The SDS system does allow me to run 23BTDC for cruise with a 3 degree retard at takeoff power. That would allow the same detonation margin as set by Lycoming.

I went back through my data and the additional 3 degrees is worth about 2kts at cruise.

On another note, I stopped my LOP advance testing at 31BTDC because the curve increased speed noticeably up to 27BTDC but went flat after that. I could go up to 29BTDC for another knot, but decided to back off since I’m not a Reno racer and every knot is not critical for my type of flying. At 31BTDC I was still well below the point of noticeably increasing CHTs

In a previous post in this same thread I provided my airspeed and fuel flow calibration data. My airspeed is good to within 0.7 knots; therefore, measuring 1 knot increments in a free atmosphere can be difficult.
 
Last edited:
Appreciate all the good info here from Marvin and Dan. We'll be able to build better baseline maps for 390 users. :)
 
?..I was out at 35 BTDC, at 16.5K and 30 LOP for a bit last week. Not desirable. Higher CHT and OT, and 5 knots slower.

I'm beginning to suspect flat advance curves are a characteristic of the angle valve. Looking forward to your data.



...On another note, I stopped my LOP advance testing at 31BTDC because the curve increased speed noticeably up to 27BTDC but went flat after that. I could go up to 29BTDC for another knot, but decided to back off since I?m not a Reno racer and every knot is not critical for my type of flying. At 31BTDC I was still well below the point of noticeably increasing CHTs...

Not sure I'm understanding the data presented here. Seems to be in conflict.

Mr. Horton: Did you jump to 31 degrees straight away, or did you incrementally get there? If the later, why did you continue to advance in the face of dropping speed? Did the speed drop right off your 23 degree setting, or did it climb, peak and decline between 23 and 31?

Mr. McGraw: Your speed increased up to 27, but went "flat"? Are you suggesting the peak speed was found at 29 but fell off after that? One additional knot between 27 and 29 is still a pretty sharp slope in my mind.

Personally, I'd have no issue squeezing that last knot out of it - if the power (speed) is still increasing, detonation is not present. And at altitude, detonation is highly unlikely anyway. Look at Dans experience: 5 whole knots too advanced and no detonation.

Interesting data on the angle valve behavior. Looking forward to my own testing on the RV when I install Ross' system.
 
Not sure I'm understanding the data presented here. Seems to be in conflict.


Mr. McGraw: Your speed increased up to 27, but went "flat"? Are you suggesting the peak speed was found at 29 but fell off after that? One additional knot between 27 and 29 is still a pretty sharp slope in my mind.

Personally, I'd have no issue squeezing that last knot out of it - if the power (speed) is still increasing, detonation is not present. And at altitude, detonation is highly unlikely anyway. Look at Dans experience: 5 whole knots too advanced and no detonation.

Interesting data on the angle valve behavior. Looking forward to my own testing on the RV when I install Ross' system.

Michael,
The two photos are my data with respect to Density Altitude. The red area is ROP at 75% Power. The peak is at 23BTDC. (Ignore that the advance is not linear, that is a sorting error in my spreadsheet. I did infact fly with a linear increments)

The blue areas are at 25LOP and different Density Altitudes. Notice that the April 6th LOP data is significantly faster than the second date. I set the LOP on my G3X at 20BTDC and 25LOP. as I advanced the ignition the LOP would change so I adjusted the FF (fuel flow) to get 25LOP. I suspect that is incorrect so I discarded that data. NOTE: The CHTs trend is the same as the second data set.

In the second photo (3 months later), FF was held constant. You can see the LOP data peaked at 27 or 28BTDC and started down at 31BTDC. Not shown in the picture, but in all cases the CHTs climbed steadily from about 300F to 319F

I was not man enough to continue past 32BTDC. Also, each run took 7 minutes so I got a bit tired.
25z330l.jpg


15n9cf5.jpg
 
Last edited:
...In the second photo (3 months later), FF was held constant. You can see the LOP data peaked at 27 or 28BTDC and started down at 31BTDC. Not shown in the picture, but in both cases the CHTs climbed steadily from about 300F to 319F

I was not man enough to continue past 32BTDC. Also, each run took 7 minutes so I got a bit tired...

OK. Understood. When doing a full sweep (from 20 to 30 degrees, for instance), the EGT is going to move well away from your initial setting. However, once you have the range narrowed down, the difference between your "optimum" ROP (or peak) and your usual LOP isn't as dramatic. In my testing, I don't touch the mixture once the initial LOP is set.

With that said, once you identify the speed peak, there is no reason to be "man enough" to go further with the advance - (I'm not either) - it's only going to go slower and hotter after that. I go to the middle of the range of the peak and call it good.
 
Last edited:
Mike,
Straight to 35. Just happened to be the max on a non-optimized map#2. Flying at 16.5K, so MP was very low.
I gotta get a laptop I can use in an RV8 cockpit so I can make incremental changes. Wouldn't be such a space problem in a side by side. Right now I can only flash an entire map on the ground with the big laptop.
 
Mike,
Straight to 35. Just happened to be the max on a non-optimized map#2...

Copy Dan. Though I have not dabbled with advance much beyond peak speed (not much reason to), what I have done seems to indicate that the drop in speed and increase in temp is pretty sharp the further past optimum you go. It could be that the angle valve is even more prone to this behavior. You might have been fairly close to optimum, but just a touch too far. Will be interesting to see how your engine compares with Marvin's when you get a laptop situated.
 
Marvin, remember there can be minor variations in how well an airplane is leveled during AHRS calibration, plus G3X gives you the ability to enter a pitch offset in flight. I would not use pitch attitude to try to compare performance. IMHO the more important reference is the vertical speed which is zero in the first image and around 250 fpm descent for your airplane to achieve the same numbers...so if you let it stabilize in ALT hold you probably would see the expected reduced cruise speed (due to your larger wheels).
 
Keep at it Marvin, I?m liking and learning with your post updates.

I thought I would have a defenitive timing map to post after a couple test flights since receiving my programmer, but it?s taken me 50 hours (ignition only) of piddling to feel like I?m starting to get the hang of the testing-seeing-recording part of it all. Lots of variables that pop up during testing to skew true results.

I?ve long since given up on the time and money it costs for Test flights, I just try to work it in as I can on flights I?m gonna do cause I?m a fly-in anyway.
 
Hello Andy,
I agree the testing can be time consuming and expensive. I went down a bit of a rabbit hole trying to compare to another aircraft on the rich of peak side, all the while knowing the actual advantage of the variable timing will be on the lean of peak side. When comparing across airframes I cannot control weight, form, trim or cooling drag. I cannot control calibration, repeatability, etc...

I repeat all my data a minimum of three times to see if any should be dicarded. The value of the data is a controlled change against a baseline not the total number. For me the time and expense is more of a learning experience since I?m not trying to feed three kids like you:)
I post the data because people on this board can point to errors I may have overlooked. That is the value I find on this messageboard.

Sandy and I did look for your airplane at OSH several times. We came up short even though I know you were there.

Short OSH for me due to work, had to leave Tuesday am. Glad to catch you for the short visit!

And it?s 4 kiddos......and a horse :D
 
Marvin, remember there can be minor variations in how well an airplane is leveled during AHRS calibration, plus G3X gives you the ability to enter a pitch offset in flight. I would not use pitch attitude to try to compare performance. IMHO the more important reference is the vertical speed which is zero in the first image and around 250 fpm descent for your airplane to achieve the same numbers...so if you let it stabilize in ALT hold you probably would see the expected reduced cruise speed (due to your larger wheels).

Yes, you should use level flight as indicated by zero feet per minute and/or your velocity vector (the little green airplane symbol in the upper picture) on the horizon line. The yellow waterline marks represent the deck angle, and you'd expect that to be a little above the horizon in cruise unless you've programmed an offset so it's on the horizon at typical cruise conditions. And even then it will vary--at light weights and high speeds it may well be below the horizon in level flight.
 
Back
Top