Thread Tools
Jun 16, 2016, 11:01 AM
Closed Account
Thread OP
Discussion

NGH9, Evo, OS GT or GGT, Saito FG series???? What for?


A magnet in the propdriver, a sensor tierapped to the front bearing, and 5:1 fuel....

OS max 46 FXi, on gasoline 5:1 mix, no further modifications (2 min 52 sec)


Still getting there on the needles, but the idle is endless (it did idle like that for 10 minutes before I fired up the camara....)

Edit: Checked the engine internally after maybe 30 minutes of idle and an occasional shot of full throttle.
Fair amount of black in the exhaust, but whatever deposits there were still on the piston, they washed right off.
Liner looked OK (no peeling) and compression when cold is very good. When warm, less than cold, but still very acceptable, and provided it is primed right (which I still need to figure out), it starts both warm and cold with one or two flicks.
No kickback whatsoever, despite the ignition being at full advance when below 1000 RPM
Last edited by Brutus1967; Oct 30, 2016 at 01:37 PM. Reason: added a fourstroke to the title....
Sign up now
to remove ads between posts
Jun 16, 2016, 02:38 PM
Registered User
earlwb's Avatar
It is looking good. Nice job. I like the keep it simple approach too. Did you need to change the carburetor out or modify the carb?
Latest blog entry: yes I still fly airplanes too
Jun 16, 2016, 02:55 PM
Closed Account
Thread OP
Nope... nothing done to the engine at all, except for the small hole drilled in the propdriver.

I intend to lower compression a bit though (maybe) and I have a regulated (non-pumped) carb for an RCGF10 on order for just in case....
Looks like I am not going to need it though.

No mufflerpressure or anything...

EDIT: ah, yes, I did change one thing: the Original OS remote needle has a plastic housing, and I was not sure if that would be gasoline resistant, so I changed that for an all metal ASP remote needle unit.... but that is hardly a modification, more a material change....
Last edited by Brutus1967; Jun 16, 2016 at 03:02 PM.
Jun 16, 2016, 04:46 PM
Suspended Account
...
Last edited by John_M_; Nov 01, 2016 at 02:38 AM.
Jun 16, 2016, 05:21 PM
Closed Account
Thread OP
No, it has auto-advance (or more correct, auto-retardation) but that retardation only kicks in above 1000 RPM or thereabout....

If you set the trigger at 28 degrees BTDC, it will fire at that angle from 0 till 1000 RPM (for the purpose of being able to set the ignition angle), at 1000 RPM it suddenly falls back to appr 8 deg BTDC, above 1000 RPM, it linear gets back to 28 deg over the range of 1000 to 4000 RPM. Above 4000 RPM it stays 28 deg.

In the vid it is running slightly below 2000 RPM, so it should fire at approx 14~16 degrees BTDC or so (provided trigger is at 28 deg, I set it by means of running behaviour, not with a graduated disc)

Above 15000 RPM, the response time of the electronics get noticeable, and it slowly falls off again but that is stricktly the whole chain of hall-sensor, processor and coil not being any faster... From 15 to 18K it seems to fall back about 1 degree.

It does not run harsh, the wing is still off and the pushrods rattle a bit because of that....

It was only to see if gasoline will give acceptable results without all kinds of regulators, membranes, pumps etc etc, maybe one day I'll put an ASP 52 or 60 on it.... but not now....
Last edited by Brutus1967; Jun 16, 2016 at 05:55 PM.
Jun 16, 2016, 09:09 PM
Suspended Account
...
Last edited by John_M_; Nov 01, 2016 at 02:38 AM.
Jun 17, 2016, 03:54 AM
Closed Account
Thread OP
No, I stated full advance when flipstarting.... which is a different thing.

I think a reliable idle below 2000 RPM is actually QUITE respectable for a totally unmodified .46 2-stroke.... The plane is manageable with 2500 allready....
Jun 17, 2016, 08:14 AM
Suspended Account
...
Last edited by John_M_; Nov 01, 2016 at 02:39 AM.
Jun 17, 2016, 08:19 AM
Closed Account
Thread OP
Flipstarting is a single revolution, with only ONE passage of the magnet.

How is ANY system going to derive a rotational speed from that, unless you first program the duration (in degrees of crankangle) of the magnet passage so the CPU can compute something? (and I do not know of any ignition system that has the feature of programming that dwell-time)
Whichever ignition you have, only at the second passing of the magnet, it can derive a rotational speed....

You can have the first passage at 1000 RPM or higher, and the first spark will still come on the trigger, only after the second passage, there is retardation.

It is a common known thing for RcExl to usually have a strong kickback on handstarting, and the reason why most people advise the use of an electric starter.

I believe OS solved this kick-back issue by having the spark of the first passage delayed until the next passage (the ignition is Always "one revolution behind") which prevents the kickback, but necessitates the use of an electric starter. Apparently, they put more worth on preventing injury when somebody turns the prop with an armed ignition, and sacrifice handstarts for it....
Last edited by Brutus1967; Jun 17, 2016 at 09:16 AM.
Jun 17, 2016, 01:32 PM
Suspended Account
...
Last edited by John_M_; Nov 01, 2016 at 02:39 AM.
Jun 17, 2016, 02:47 PM
Closed Account
Thread OP
The RcExl sensor is just a switch. its signal does NOT contain any information about fieldstrength, rate of rise, or other, it is just a logic 1 or 0 and has a vertical upramp and downramp.

I have had contact with somebody (you can find him at www.ignitionmaster.com, he is also active on this forum at times) that has hacked the processor of an RcExl and read the embedded software, according to him, there is nothing in the programming to provide for a delayed first spark, or other form of starting help.

Now this is second hand info, and I lack both skills and equipment to verify it first hand, but this is a video of the behaviour of an RcExl unit (mind, the scale is accidentally reversed...)

RCeXL-01 (0 min 34 sec)
Jun 17, 2016, 03:04 PM
Suspended Account
...
Last edited by John_M_; Nov 01, 2016 at 02:39 AM.
Jun 17, 2016, 03:28 PM
Closed Account
Thread OP
No it does not....

I have checked this extensively on a scope. It is a 1 or a 0 and nothing inbetween.
Not "from what I understand", but my own observation from when I was searching for alternatives. I had to first put the Original sensor on the scope, in order to know what kind of signal my alternative sensor needed to provide.
There is no speed dependent wave form, only a blockwave, of which the dutycycle is entirely depending on the relation between hub diameter, and magnet size. RcExl does not supply a magnet (they sell one as optional part, but leave it up to the customer to use that one or provide his own) and neither do they know the hub diameter of the engine the unit is going to be used on. Ergo, RcExl does not know the dutycycle of that blockwave, and thus have no data to work with. The programming seems to indicate that this is true, and so does the video....

Trust me, there is NO retardation, from 0 to 1000 RPM

The video shows a rotating disc with a mark, and strobe light inside it triggered via the RcExl ignition unit. It has all the relevance, it exactly shows what the unit is doing.
You can clearly see the disc slowly start spinning and increasing speed. At certain speed the timing "jumps" (which is the retardation, you can see that from the direction of rotation when the disc initially starts to move) and gradually climb back to its Original angle.
The only thing it lacks is a speed indicator (unfortunately).
Last edited by Brutus1967; Jun 17, 2016 at 03:37 PM.
Jun 17, 2016, 03:35 PM
Suspended Account
...
Last edited by John_M_; Nov 01, 2016 at 02:40 AM.
Jun 17, 2016, 03:46 PM
Closed Account
Thread OP
First of all: I am NOT talking about CH (and never was....), only RcExl because that is what I am working with on this engine, and second of all, I do not really care whether you agree or not, as after all I was the one having had the RcExl on the scope for myself, and developed an alternative sensor for it that allows to detect the crankpin directly through the backplate.....

I happen to KNOW how their advance curve functions...


Quick Reply
Message:

Thread Tools