Yamaha YG4i (Apex) Data Collection

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
Ok fam, we have a baseline vibration spectrum on our converted YG4i - SkyTrax PSRU combination after refurbishment, overhaul, conversion and dyno runs. The conversion was nowhere near as complex as we thought, and the engine ran like a dream on the dyno. The plant is now preserved. When we drop it in our aircraft when we complete it, acquiring as much data as possible on the engine is a big priority of ours when the time comes.

Here are the areas where we will gather data:

PSRU Output / Propeller (Lat-Long-Vert)
PSRU Input / Crankshaft Bearing 5 (Lat-Vert)
Crankshaft Bearing 1 (Lat-Vert)
Exhaust Cam (Lat-Vert)
Intake Cam Mid Length (Lat-Vert)
Oil Pump (Lat)
Cooling Pump (Lat)
Engine Crankcase (Lat-Long-Vert)
Propeller Position
Crank Position
Throttle Position
Intake Air Temperature
Intake Air Pressure
Intake Air Flow Rate
Fuel Pressure
Fuel Temperature
Fuel Flow Rate
Oil Pressure
Oil Temperature
Exhaust Gas Temperature (1-2-3-4)
Coolant Temperature In
Coolant Temperature Out
Heat Exchanger Intake Plenum Pressure
Heat Exchanger Air Flow Rate
Heat Exchanger Intake Temperature
RPM Engine
RPM Propeller
PSRU Temp
PSRU Pressure
Crankcase Pressure
Engine Chip Detect
PSRU Chip Detect

Total Powertrain Only Channels: 44

Obviously, all air data, structural, etc, will be integrated as well in our total data package

We are still determining engine-specific test regimes as appropriate for the aircraft, and will likely change as we refine the test plan.

My intent is to release all our data, both in the raw and reduced forms.

All this said, if there are any specific regimes or data points you want us to get, let us know, but just an FYI, we are not planning to be complete/ready for flight test until Spring 2022, provided construction goes as we hope it will; we have alot of white space built in.
 
Last edited:

rv7charlie

Well-Known Member
Joined
Nov 17, 2014
Messages
1,540
Location
Jackson
Reduction drive type, decoupler type, ratio. Drive lube temps vs load. Drive magnetic chip detector monitoring. Prop specs. Aircraft type/speeds.
 

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
G load, yaw, and pitch rates. Gyro and inertial loads on propeller shaft and bearings during maneuvering can be important.
Excellent points! We will be gathering alot more data; the engine parameters I have mentioned are only part of the whole instrumentation package for the aircraft for our program. We are up to 180 channels total, to include air data, structural, etc. Believe it or not, one of the Ospreys I test can collect over 1,500, all telemetered. At one point some years ago, it was over 4,000!

Reduction drive type, decoupler type, ratio. Drive lube temps vs load. Drive magnetic chip detector monitoring. Prop specs. Aircraft type/speeds.
We will definately have information on the exact setup for proper context. For what it is worth, this will be a pusher setup on a Sensenich 3BO ground adjustable, so we will include the prop pitch data as well.

The dyno runs had the vibe channels and basic parameters monitored; our goal was to baseline and see what vibe characteristics & brake horsepower were looking like among other characteristics. Before installation, we will test it on a stand with the flight propeller and all instrumentation for a more precise baseline for calibration purposes. Then repeat in aircraft prior to taxi test phase and beyond. We enlisted the assistance of a couple very R1-savvy guys. They are the SMEs for this plant, and have been beyond helpful.

Thanks, gents!!!

Perhaps if some certain chap sought an Audi SME for a certain project, then maybe he would not be in the pickle he is in....but I digress!
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,941
Location
Saline Michigan
This will be an impressive data set. As an experienced power train torsional vibe guy, I would say that it looks like you are trying to capture radial and axial position of crank, cams, PSRU in and out shafts. True? Is this to track torsional vibe and show which elements are vibrating together? Does that work OK for torsional vibration and torsional resonance? It would seem to me that it would only be really good if the torsional vibe corresponded with positional vibe...

My experience, now becoming stale (retired five years), is to use target wheel (toothed wheels and mag pickups or printed wheels and a laser tach) on each spot of interest, then run a Rotec Munich system to record passage of each tooth. Worked great, but it required the very expensive Rotec signal processing, timing, and recording setup. Then their software, post processing, FFT, etc, to find how much torsional vibe we had at what frequencies, what what was vibrating with what. 2005-2015, that was the way to do it.

I am curious how the data recording and processing has progressed. With a four stroke four cylinder at 6000 rpm, that is firing at 200 Hz. To resolve a resonance driven by firing frequency, you would need high resolution samples of all shaft rotational positions at a minimum of about 1600 Hz. If you found you had higher order issues, you would need stIll higher sampling rates. What is your data sampling rate on those channels? What types of hardware are you using for processing and recording this mountain of data, then allowing you to post-process to find what is moving with what, and how badly?

Billski
 

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
This will be an impressive data set. As an experienced power train torsional vibe guy, I would say that it looks like you are trying to capture radial and axial position of crank, cams, PSRU in and out shafts. True? Is this to track torsional vibe and show which elements are vibrating together?
Affirm!!

What is your data sampling rate on those channels? What types of hardware are you using for processing and recording this mountain of data, then allowing you to post-process to find what is moving with what, and how badly?
2000 Hz, and all our engine vibe equipment is Honeywell (Chadwick Helmuth) "borrowed" from our instrumentation engineering folks. In fact, they're using some other proprietary equipment and software as well with us. I won't lie; I am not a vibe guy (I know what numbers are bad on the Osprey when I do a routine vibe diagnostic test, but that's about it), but I work alongside a fellow Bell drivetrain guy who is, and he is sifting through what we have now, so we can characterize what is what.

We know it only gets more complex from here but the Yamaha conversion community deserves it, and it only has the effect of increasing flight safety by way of helping establish a practical & effective maintenance schedule, and powertrain improvement engineering.

Also, at the end of the day, this design of ours is a higher risk planform, so getting a wealth of data is essential, given our desired end state with it.

Our entire flight test instrumentation system is all wrapped up into a software application called IADS (by Symvionics Curtiss-Wright). Same as what is used on the Osprey and across the industry.

It might be a little overkill, but when you have the equipment and a team that knows how to use it, why not? Just don't tell my boss what we are doing!
 

Victor Bravo

Well-Known Member
HBA Supporter
Joined
Jul 30, 2014
Messages
8,920
Location
KWHP, Los Angeles CA, USA
You are very likely doing a tremendous service to the homebuilt community, by tackling an engine test/development project that very very few people are capable of doing.

My initial reaction to the new Yamaha engine was negative (oh, that's nice... another motorcycle engine conversion... ho hum), but apparently the engine does indeed have a lot going for it, and is working in quite a few airplanes. So fair is fair, I have to admit I was more skeptical than I should have been, and I have to accept that it may become a viable thing for a lot of builders.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,941
Location
Saline Michigan
So you guys are doing linear vibe only? No direct rotational vibe? I would ask about and see if a Rotec Munich or similar torsional vibe tool set is available for some speed sweeps for idle to max rpm. This is the major tool for this sort of work in IC engine driven products, like cars, motorcycles, gensets, and the like. You can do this on ground test. I would suggest one target wheel on the crank, another on the PSRU input, another on the PSRU output, and laser tach reading each of them during the speed sweeps. This will give you deflection across the isolator (Crank vs PSRU Input Shaft), vibe internal to the PSRU (Input Shaft vs Output Shaft) and total Crank vs Prop. It will also enable you to correlate torsional vibe to your measured linear outputs so that you can later on interpret the linear stuff.

The Crank vs PSRU Input is particularly important as you get to see how large the rotational cycle is across the isolator. Ideally the resonance is at least an 1-1/4 octaves lower than firing at idle, and so has small cyclic amplitude in the operating range. This also allows you to do your speed vs order vs amplitude plots and identify resonances and significant orders of engine or prop rotation.

I continue to be curious as to if you get rotational vibe from linear measurements on shaft position. Do you simply watch the movement of the shaft (which has to be pretty small and noisy as it is constrained by its bearings and then by elastic deformation of the bearing mounts) or do you install an eccentric and resolve its position to get torsional vibe?

Billski
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
14,654
Location
Memphis, TN
I know what a four channel unit costs, and I have an idea on what the box your borrowing costs, but what is more impressive is having all the sensors and being able to mount them all.
 

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
No direct rotational vibe? I would ask about and see if a Rotec Munich or similar torsional vibe tool set is available for some speed sweeps for idle to max rpm
We are going after direct rotational vibes also. Maybe I misworded/misunderstood; one of my colleagues (a Rolls Royce chap here) is familiar with the Rotec. I have them on point instrumenting what they think needs imstrumented as appropriate. We will be doing speed sweeps also.

What is truly crazy is that they have shown me what a fully instrumented test engine looks like for a fully Mil-Spec/FAA certification process; makes this Yamaha look like it just has a couple wires on it!
 
Last edited:

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
You are very likely doing a tremendous service to the homebuilt community, by tackling an engine test/development project that very very few people are capable of doing.

My initial reaction to the new Yamaha engine was negative (oh, that's nice... another motorcycle engine conversion... ho hum), but apparently the engine does indeed have a lot going for it, and is working in quite a few airplanes. So fair is fair, I have to admit I was more skeptical than I should have been, and I have to accept that it may become a viable thing for a lot of builders.
Agreed, and thank you!! When the relative simplicity of conversion, low cost, high performance, reliability and all the advantages it appears to have over the Rotax became known, I was quick to ask "whats the catch?", hence our data collection efforts. This test equipment we are using has a price tag that is way more than what I could possibly budget out of pocket for, but it was available for use.

If this engine is truly a viable alternative to the 912, or if its a 500 hour or less bullet or whatever, this testing will help find the answers.

It also helps that we surrounded ourselves with good SMEs, ie Yamaha engine tuners/builders, vibration and instrumentation gurus, etc. If it was just my two partners and I, doing this kind of engine R&D would have become a science project!

We are about to get Edge Performance in the loop also; they have done some great work on their apex projects...turbo-ing them past 300 bhp!

Proper characterization is so important. We want to get a darn good idea how long parts last in this engine, and what should be improved.
 

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
I continue to be curious as to if you get rotational vibe from linear measurements on shaft position. Do you simply watch the movement of the shaft (which has to be pretty small and noisy as it is constrained by its bearings and then by elastic deformation of the bearing mounts) or do you install an eccentric and resolve its position to get torsional vibe?
I'll double check, but we are using an eccentric. My two vibe gurus are enjoying this so far, as they say it is "so much more exciting & complex than a turbine!". Now I know why they're nicknamed Sheldon & Leonard!
 

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
I know what a four channel unit costs, and I have an idea on what the box your borrowing costs, but what is more impressive is having all the sensors and being able to mount them all.
I'll have to get a pic of our instrumentation engineering shop at V-22 flight test sometime. I've held signal conditioners, strain gauges and sliprings (among other items) that cost more than what most of us make a year!

I do have to say though... they have some low cost solutions; they made our static cones from parts sourced at Lowes!
 

daveklingler

Well-Known Member
Joined
Jan 22, 2013
Messages
128
Location
Albuquerque
Brian, do you have a good idea what the final weight and dimensions of your installation are? I've been wondering about the overall envelope and CG.
 

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
Brian, do you have a good idea what the final weight and dimensions of your installation are? I've been wondering about the overall envelope and CG.
In the dyno config, which included the stock intake and exhaust out of the sled (which will change), we are looking at the following:

Height: 18.25
Overall Width: 26.10 (Engine only is 14.00)
Length with SkyTrax PSRU (Minus Prop + Hub): 26.20
Weight (as tested, all equipmemt installed): 158.75

Obviously, some dimensions will change based on final exhaust and intake design, etc. We are still defining mass properties.
 

fly2kads

Well-Known Member
HBA Supporter
Joined
Jan 2, 2010
Messages
1,797
Location
Justin, TX
I spent a few years working at National Instruments, in a non-technical role. That was my first exposure to data acquisition systems. The things our customers did blew me away. I am still fascinated by this stuff, so keep posting!
 

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
I spent a few years working at National Instruments, in a non-technical role. That was my first exposure to data acquisition systems. The things our customers did blew me away. I am still fascinated by this stuff, so keep posting!
Will do. Like I mentioned up above, we will publish everything we got. I am no vibe expert by any stretch; just a dumb stick wiggler, but yeah, our instrumentation and vibe & acoustics guys are something else! They've even given me predictions on anticipated cabin noise levels! Geeks, but good dudes!
 

Marc W

Well-Known Member
Joined
Mar 31, 2017
Messages
672
Location
Colorado
Speaking of noise levels, I will be most interested in the muffler setup you use if you are willing to share that. I understand that one problem with these engines is the earsplitting exhaust noise. An effective lightweight muffler system would be good to have.
 

cblink.007

Well-Known Member
Joined
Jul 7, 2014
Messages
632
Location
Maryland, USA
Speaking of noise levels, I will be most interested in the muffler setup you use if you are willing to share that. I understand that one problem with these engines is the earsplitting exhaust noise. An effective lightweight muffler system would be good to have.
Right now we are using what was on the sled for initial post-build testing, which is an MBRP Powersports Trail Muffler and stock manifold and pipes. The exhaust system will change, as it is optimized for the original transverse installation in the sled...and pretty long. We have a couple ideas on the new layout.

All this said, it wasn't all that loud on the bench...even when we took her up to 10,000 rpm!
 
Top