mauryr

Can't hear very well from inside the model, and the GSU is not visible in the video.
Preheat fuel is the fuel that goes into the engine from the main injectors during preheat phase. The sticks are bathed in the flame from the burner, which is controlled by the pump pw only I think.
This all happens during the preheat phase, not the ignition phase.
Ignition is simply some fuel going through the burner to increase the engine temperature by a couple of degrees. When the ECU reads an increase, it will consider the flame lit and switch to preheat phase.


Am I hearing right - you have absolutely no flame between #0:53 and like, #1:08?
Does preheat start at #1.30?

Flame should start right away at 0:53, when you start hearing the ticking sound of the burner valve. If it doesn't, then something else is wrong and the fuel pools up. Until that's resolved, any further tweak is pointless, because your engine is already flooded before the next phases.

How long and what gauge are the cables you are using for your burner power supply? What battery are you using?

Pull Up Now!

My Feedback: (3)

Mauryr, all good questions. I'll try to answer the best I can. First, there is a temperature rise happening almost right away during the 15 seconds you cited starting at 0:53. Then at some point it goes into preheat, but I'm not sure if it's exactly at 1:30. I'll have to get some lighting and a better video camera than my phone to synchronize the terminal with the sound. The wires are all the same guage from the battery to the ECU, and it's the same gauge as what came with the ECU. It all looks quite heavy. I cut the excess off the wires to keep them short. The battery is a really good question. They are 3S LIFE 2500mah, 9.9V nominal. One point is they are only 5C batteries. So a 5C battery can deliver 5x2500 or 12.5 amps. When I start the turbine, I've watched the led battery voltage indicator and it usually only drops a couple tenths of a volt. This is an indication there really isn't any current starvation going on. But I think this is a good exercise in thoroughness at least. The switch is one of the MPI heavy duty switches. They are physically huge.

mauryr

12.5A is not enough. The burner plus the starter motor draw about 10+ amps, so if the maximum your battery can provide is 12.5 odds are that could be the issue.

If you listen closely to the engine, you can hear a "woosh" sort of sound when the flame ignites. It must happen right after the ticking sound of the pulsing solenoid starts. If it doesn't, either the line is not primed, or the igniter is too cold. I cannot tell on your video, but it does look like it takes at least a couple seconds for the flame to ignite, and that would cause fuel pooling in the engine.

Pull Up Now!

My Feedback: (3)

Hi, I didn't want to write anything more until I had time to absorb some of these suggestions, and do some testing. Mauryr is correct, the whoosh of ignition is late. So to try to sort out, I first hooked up a different battery. This new battery is the original one I'd intended to install in the plane, but it was too big. It's 9.9V nominal, a LiFe, and it has 20-30C rating. But it's a great test battery for the currrent subject, which is: is the installed 5C battery not delivering enough current to heat the heater and run the pump. Thus leading to pooled fuel and late ignition.

I suspected that the 5C battery, though not optimum, is delivering enough current but this needs to be proven. My evidence of this is that the voltage is not sagging during startup. If there was current starvation, that would happen. Anyway, last night I hooked up the 30c battery and nothing changed. There was the same delay before the whoosh sound of ignition, and the same big yellow flame. So my installed 5C battery may not be optimum. You could even say it should be changed, but it's not the root cause.

Next, I went to Auto+1 instead of Auto+2. Nothing changed, so I put that back. Next, I moved fuel preheat from 30% to 32%. Boom, immediate ignition. Well, it didn't go boom. That me talking. There was about a 75% reduction in the duration of the yellow flame, and it wasn't nearly as long. A big improvement, but not a total fix. I'm not sure there is any value in bumping the preheat any higher because I'm already now getting immediate ignition. Am I correct in saying that? If true, I'm not sure what to try next except maybe to bump pump pwr k down a bit now.

mtnflyer14

[TABLE]
[TR]
[TD="class: xl65, width: 193"]
Parameter
[/TD]
[TD="class: xl65, width: 153"]
Value
[/TD]
[/TR]
[TR]
[TD="class: xl64"]Pump Start Point[/TD]
[TD="class: xl66"]Auto+2[/TD]
[/TR]
[TR]
[TD="class: xl64"]Pump Start Ramp[/TD]
[TD="class: xl66"]10[/TD]
[/TR]
[TR]
[TD="class: xl64"]Glow power[/TD]
[TD="class: xl66"]7.4v[/TD]
[/TR]
[TR]
[TD="class: xl64"]Low battery[/TD]
[TD="class: xl66"]7.0v[/TD]
[/TR]
[TR]
[TD="class: xl64"]Starter Power Ignition[/TD]
[TD="class: xl66"]70[/TD]
[/TR]
[TR]
[TD="class: xl64"]5tarter power fuel ramp[/TD]
[TD="class: xl66"]130[/TD]
[/TR]
[TR]
[TD="class: xl64"]100% Starter power[/TD]
[TD="class: xl67"]15,000[/TD]
[/TR]
[TR]
[TD="class: xl64"]RPM off starter[/TD]
[TD="class: xl67"]19,000[/TD]
[/TR]
[TR]
[TD="class: xl64"]RPM to reconnect[/TD]
[TD="class: xl67"]18,000[/TD]
[/TR]
[TR]
[TD="class: xl64"]RPM ignition K[/TD]
[TD="class: xl66"]3000[/TD]
[/TR]
[TR]
[TD="class: xl64"]Pump PW ignition k[/TD]
[TD="class: xl66"]20[/TD]
[/TR]
[TR]
[TD="class: xl64"]Engine min flow[/TD]
[TD="class: xl68"]80%[/TD]
[/TR]
[TR]
[TD="class: xl64"]EGT end preheat[/TD]
[TD="class: xl66"]72c[/TD]
[/TR]
[TR]
[TD="class: xl64"]RPM preheat[/TD]
[TD="class: xl67"]5,000[/TD]
[/TR]
[TR]
[TD="class: xl64"]RPM fuel ramp[/TD]
[TD="class: xl67"]10,000[/TD]
[/TR]
[TR]
[TD="class: xl64"]Preheat fuel[/TD]
[TD="class: xl68"]23%[/TD]
[/TR]
[TR]
[TD="class: xl64"]Ignition time out[/TD]
[TD="class: xl66"]30s[/TD]
[/TR]
[TR]
[TD="class: xl64"]Start mode[/TD]
[TD="class: xl66"]Auto Kero[/TD]
[/TR]
[/TABLE]
I think I gave you these settings from my K-80F on another forum last week, but they work for me here in Phoenix. Not sure how to compare the yellow flame I get to what you are describing, but I don't think it is an issue.
Regards,
Gus

mtnflyer14

Pump Start Point: AUTO+2
Pump Start Ramp: 010
Glow Plug Power: 7.4v
Low Batt. Volts: 6.6v
Starter Power at Ignition: 070
Starter Power at Fuel Ramp: 130
RPM 100% Starter Power: 15000
RPM OFF Starter: 19000
RPM to Reconnect Starter: 18000
RPM Ignition K: 2000
Pump Pw Ignit. K: 022
Engine Min Flow: 080%
EGT End Preheat: 72 C
RPM Preheat K: 5000
RPM Fuel Ramp K: 10000
Preheat Fuel: 23%
Ignition Timeout: 30s
Start Mode: AUTO-KERO

Just for reference, here are the values a friend of mine uses on the K-100G in Florida and he is happy with the results.
Regards,
Gus

mauryr

Ok, so now you know what's causing the flame (the delayed ignition and consequent flooding through the igniter fuel port).
You said if your pump value is below 30, the fuel won't ignite. Is this really true? I have some doubts.
If the pump "moves" the fuel, and the plug is hot enough, the fuel WILL ignite. If you mean it ignites but the engine doesn't start, that's a whole different story.
So as I suggested before, try to lower the fuel pump back to like, 17~23. If too much fuel is pumped onto the plug, it will cool it down and it won't ignite or ignite late. Since you now know it's not a battery related issue, then this is the next thing that could cause your problem. One terribly stupid mistake I made when I was thinking low pump values wouldn't get my fuel to ignite was.... forgetting to open the manual shutoff.... seems pretty obvious the shutoff should be open to have ignition eh?

Get the fuel to ignite right away, every time, consistently. Then we can continue the 'debugging' process and figure out if anything else is not set correctly.
That is, find the real lowest value your pump will move the fuel and get a consistent ignition (woosh sound, don't mind temperature increases or expect a successful start yet).

Pull Up Now!

My Feedback: (3)

Yes, and I apologize for the multiple threads in which I've requested help. It's been my experience one thread doesn't always get the exposure needed to find the experts. And, yes, I resigned myself that if I couldn't find a solution to the yellow flame I would just live with it. Barry even said it's no big deal. But I still wanted to work the issue since I've seen lots of turbines start without it. Who knows if they were kero or diesel though. Read down, though, as some success has been obtained!

Pull Up Now!

My Feedback: (3)

Had a success in last night's testing! Read on.....

"pump value below 30, the fuel won't ignite". Reviewing the chronology of this, using the original setting for diesel from the manual, yep, the engine wouldn't light or start. The DT (data terminal) read "ignition fail". It wasn't until I finally reached pump pw k of 28 that the fuel would light. But the engine flamed out. When I reached 30, the engine started and ran, but it had the yellow plume and the lengthy time to ignition. At Barry's suggestion, I lowered the preheat fuel as much as 4 points with no effect. At your suggestion, I increased preheat fuel a couple points and the delayed ignition went away. Immediate whoosh. That was progress indeed, but the yellow flame was still there.

By now, I have an idea of what some of the other ECU parameters do, and can eliminate many of them. One helpful thing, I Googled "RC turbine CAD designs" and hit "images". I got a lot of cutaway views of turbines. The following is my suspicion, and is subject to error! But it's what I think right now based on the diagrams. There seems to be two valves that control two different ways fuel enters the engine. One for ignition, the other for running. I think the initial ignition tube is just that, a simple tube. I can't find a specific cad diagram of Kingtech, so I can't really verify that, so it's just a guess. It squirts a little wet fuel to light the beast at first. Just exactly as you, Mauryr, said, the run injectors sit inside another tube(a tube within a tube) and these outer ones act as preheaters. These two valve are both pulsed by the ECU at various rates depending on what the sensors and firmware are calling for. The preheat ECU parameter must control how early in the ignition process fuel is squirted thru these double wall tubes. This fuel is preheated and finely atomized, so it ignites immediately with a whoosh. The airflow within the preheater tube is from outside the combustion cage to inside, so the atomized fuel is carried along that airflow into the combustion chamber. I think Maury your suggestion to increase this parameter is what got rid of the delay we originally thought was the battery. You really seem to know a lot about these things. I did learn the pump pw k affects both ignition and preheat equally, it's the pulse width modulation for each valve individually that makes them perform in the different ways as needed.

Bumping up my pump pwr k to 30 to get the engine started was really a brute force method. I think I got lucky on that one. So to FINALLY get to the point, I again did what Mauryr suggested and pushed the pump pw k back DOWN to 22, but I left the preheat fuel at 34%. So the only difference between the settings now, and originally when the engine wouldn't start is preheat fuel is up 4% from where it was. The engine started beautifully, immediate whoosh, no delay, and best of all no yellow flame at all. I think the only thing wrong at all, to begin with, was I need more preheat fuel. This has been a real education, thanks for all the help everyone.

Here's a video of the flameless start from last night. http://youtu.be/jk36B****lo

OK, WHAT GIVES HERE. THE VIDEO IS ON YOUTUBE BUT THE LINK WON'T VALIDATE!!!! Is Youtube now preventing us from copying links to our own videos???
Just to to Youtube and search for this and you see it. "flameless start Kingtech k80"

mauryr

That looked like a really good start.
The final check I would do is to stand behind the model and look up the exhaust pipe through the preheat and switchover phase to see if there's any flame still (you can't see anything because you have a long thrust pipe, but maybe the flame still builds up a little and you can fine tune it down do nothing. If you do your setup well, the engine will have NO flame outside the exhaust cone at all. The flame usually likes to creep up a little during preheat starting small and slowly building up and then burns longer during switchover and fades away. At this point, if you see any flame just dial down the pump further more by one or two points and if preheat slows down too much, raise preheat fuel a little to compensate and there you go.

Happy you found your issues and got them sorted

About the youtube link - probably the "bad words" filter on rcu doesn't like the letters Czz and filters them out from your link.

P.S. stand at about 1 meter from the exhaust and make small adjustments, wouldn't want you to get burned by a hot start
And remember to move in front of the turbine as soon as preheat ends, you don't want to be on or behind the plane of rotation of the turbine while it runs.

mauryr

If you are interesting into understanding more of the kerostart process you can look up the fadec manual here:
http://www.espiell.com/FAD10.pdf

There are a few small mistakes and misnomers in it, but I'll leave the fun of figuring out where they are to you, otherwise where's the challenge?!!

mtnflyer14

So the settings you ended up with to produce a quick, flame free start are like what I am using except you have Pump pw ignition at 22 (mine is at 20) and preheat fuel at 34% (I'm using 23)? I may try the higher preheat fuel if that works for you.
Regards,
Gus

Pull Up Now!

My Feedback: (3)

Ha, I've had enough challenges with this new baby. Now looking forward to maiden. I've installed triple control surface rates, and I'll put in CROW next. The maiden should be soon. Tonight, I'll look up the pipe to see if the yellow flame is gone, or just hidden. I will also button up the canopy for a final static thrust test.

I've read the Xicoy manual a few times, but thanks for the link. I reviewed it again. Each time I read it, what they mean becomes a bit clearer. It finally dawned on me that during ignition AND preheat phases, the burner valve does not pulse width modulate with any intelligence. It's fixed width. So not flooding the engine is PURELY dependent on a good pump pw k setting. On the other hand, the main injectors (in their role as preheater) are PWM valves. I think that easily explains why one would need to ramp UP the preheat after having set the pump pw k where it needs to be to avoid pooling.

Pull Up Now!

My Feedback: (3)

Good luck, Gus. Let us know how that works for you.

mauryr

It also states fairly clearly the ignition phase only depends on the pump pw, because the main fuel port valve is closed and only starts pulsating in the preheat phase.