Glowgeek

I started to do pressure testing the 220 with check valves this morning but took them back out in one last attempt to get the engine running right with muffler pressure only. 3 hours of adjusting and readjusting the needles and fuel curve until I got the flattest fuel curve possible and.........although not as bad as before, same thing, with the fuel ch tied to the throttle channel output; coughing out on throttle up. The best transition was with the mixture set rich enough to drop some rpm i.e. too rich. There is no way the engine will transition consistanly with the curve points set any leaner. Still, a 2 second throttle slowdown is required to keep the engine from lean coughing on transition and it will still cough out if the rpm is left in one place long enough for the tank pressure to equalize (drop). This is why I want more tank pressure,to make the solenoid deliver fuel with more authority when opened. I can see now why Chris's pump works so well, not that I'm going there.

So I'm back where I started, no surprise really. I just wanted to catch up on the project because I've been away from it too long. I will try the fuel channel tied to the throttle stick next, leading the throttle opening slightly with fuel delivery.

Glowgeek

Question for you all.

Currently I'm using a clear "bottle style" tank sold for use with gassers. It has some flexibility.

Would a stiffer non flexible tank help?

Cat 1

My Feedback: (1)

Lonnie, Not sure what your curve looked like before but here is an idea - Just an Idea... What if you set the needles to have a low pulse width on the bottom and a long one on the top - the opposite of your flat curve - My single jet setups seem to tend this way (prior to E-pressure) - I'm wondering if the engine would transition better as the solenoid is doing more work and the carb metering less. If your curve is flat the carb tuning is doing the majority of the work - the electronic piece is less responsive in my mind.. I know the theory is a little "rough" but I would ty to see if it made a difference.

1967brutus

I have never thought about that, but I have to say I am using virtually exclusively DuBro tanks (for no other reason than that they are easiest availlable to me), and they are pretty stiff.
But I would not expect that to be a factor, muffler pressure is about 0,05~0,07 bar at peak.

Can you post a vid of how your engine is running? Pics tell more than a 1000 words, so videos tell even more than that.

Glowgeek

Exactly my thinking, but in reverse. I thought that perhaps with my curve looking more like a linear ramp, low pulse width at idle and high at wot, that I was asking the solenoid to do TOO much. To prove this out I messaged the needle settings to create 6 different curve profiles yesterday, that's why my tuning session took so long. In the end I discovered that the flatter the curve, the better, but only very slightly better and still not good enough. I feel something else is going on because, based on that experiment, the solenoid appears to do a pretty good job of keeping up with rapid changes in fuel demand even with a steep curve.

My 180 and 220 exhibit identical behaviour, they need very rich mixture settings to transition without leaning out i.e. requiring to be loaded up with excess fuel at any rpm above idle. With a 2 second slow up/slow down on the throttle channel I can get smooth transition if I quickly throttle up and down. If I pause at any rpm for more than 1/2 second transition problems begin. The longer I pause at any given rpm the worse it transitions. The fuel tank is deflating during these pauses, losing pressure and it's ability to provide excess fuel pressure for transition.

Next test will be with a rigid tank. If that doesn't fix the issue then I'll try Bert's trick of leading the throttle channel with the fuel channel by manipulating slow up/slow down rates.

Glowgeek

Sorry Bert, I'm not set up with fast internet here, just a wireless hotspot with marginal cell phone reception. It routinely takes 30 seconds just to refresh this page.

1967brutus

Well.... If you shoot a vid on your smartphone, and upload it to YT in the night, it does not matter how long it takes, no?

Because I realy have a hard time figuring out what the issue is you're struggling with.
The way you describe it is nothing like my own engines behave.

Cat 1

My Feedback: (1)

I think this might be part of the solution... I haven't had a setup that didn't need this until the e-pressure test but I think it does the same thing by leading with pressure during acceleration. On most setups I can hear the engine - "shut off" during decel as the fuel comes back before the slowed down throttle opening. The sound is instantly noticeable as a different sound than most other glow or gasser setups.

Glowgeek

With OpenTX the channel speed can be adjusted for up speed and down speed individually.

1967brutus

Now this is weird...

I put the tubing on to keep the crud away from the teststand-radiobox... Look what happens (tach)...


I certainly did not expect that...

Raleighcopter

a little more scavenging happening with the tubing on or is it just breathing better?

Glowgeek

Odd, it acts as if it leaned out a bit. Higher rpm but won't hold peak.

1967brutus

I am inclined to think it breathes better... Most likely, since a Wankel operates at a kinda-sorta fourstroke principle, I think the added inertia in the enclosed column of exhaust gas pulls out more spent gas from the combustion chamber, so scavenging efficiency increases.

It actually does hold peak. Not visible in the video, but it does hold that higher peak...
When I noticed the effect, I did not have my camera at hand, and it was peaking out at 12950-ish without, and 13440 with tubing, repeatable and it held RPM for at least 10 seconds (I don't want to hold it WOT too long, given the poor cooling on the teststand). This is a difference of approx 10% in power output, from 0,46 to 0,52 hp.
When I went to get my camera, I left the engine to idle while getting the camera. Probably it pooled a bit of oil, because it peaked at 12500-ish without tubing, but I did not touch the settings.

This set-up is not the most stable (lot of play in the throttle rod, and the carb is a pretty crappy design for something made by OS...)

cmulder

Looks like you created a primitive tuned pipe.
See if by making it longer or shorter you can find a "sweet spot" for the peak rpm you are looking for.

John_M_

That long straight pipe acts as a header, increased scavenging... what are you using for CDi.

1967brutus

No intention to "tune", but at the very least I get the feeling this "thing" that goes for a muffler is to say the least, extremely sub-optimal.
Since the engine is going to be used in a pusher config, but the muffler cannot be turned 180 degrees, it would end up transverse to the airflow, resulting in a lot of drag and turbulence.
Thinking about something better... But not now...

John_M_

Without the pipe attached, you can hear the occasional miss fire, and when you put the pipe back on it picked up rpm and cleaned up... It almost runs as if it were running on glow plug ign.

There's no overlap persay between the induction / exhaust ports on a rotary engine, but the increase in scavenging on exhaust from the length of straight pipe clears any residual spent gases, and in turn improves induction fill.

Certainly would work well as a pusher... see what you can come up with in terms of a custom pipe of sorts with a small expansion chamber off the exhaust port, and extend the pipe in a gentle return aft.

1967brutus

I am not sure, but I do think, that occasional misfire has other reasons, that might very well are affected by scavenging efficiency, but not caused by it.
As it is, this engine differs slightly from full size Wankels in that it has a separate recess for the glow plug, that is connected to the combustion chamber by a tiny hole. There is not, like in conventional glow engines, a 1/4"-32 threaded hole straight through to the burnroom. This is done to offer the apex-seals an as uninterrupted passage as mechanically possible. Under normal circumstances (meaning, the engine being used as a glow engine) the cavity of the glowplug that holds the filament, connects directly to the tiny passage between the plug recess.

However, when a sparkplug is used,the plug-recess becomes a kinda-sorta precombustion room, but this space is only "scavenged" by compression and the blow-out after ignition.
So a succesful ignition depends on there being an ignitable mixture in that precombustion room. I have the impression that a succesful ignition is more or less a matter of "luck" for lack of a better word.

I know that full size Wankels require special sparkplugs with recessed electrodes for exactly this reason, but there are no "Wankel sparkplugs" in this size.
In fact, for this project, I had to slightly modify a sparkplug, shortening the mass electrode otherwise it would close off the tiny passage between recess and combustion chamber, and I had to select plug washers for thickness to prevent the sparkplug bottoming out. I could not simply cut away the entire mass electrode, otherwise the spark would most likely jump to the metal of the engine and erode away the engine locally.
As it is, I can see the clean spot on the mass electrode, so at least I have reasons to believe that the spark is jumping at the correct place.

cmulder

Don't have a wankel so working from memmory here
if you have acces to tube bending tools would a tube bend so it wraps around the engine not be a option?
And if your club is not too strict about sound just leave it a open ended tube ?

John_M_

You mentioned that before regarding the plug chamber... it reminds me of those anti-oil foul spark plug adapters used on high mileage piston engines that burn oil... basically creates a small chamber to shield the spark plug from that tarry oil residue condensing on the plug electrodes... works with some reservation, a sorta fix all... the full sized rotary's use a ground electrodeless / rim fire spark plug... Interesting design engine, but I never thought much of them... although the rotary has proven to be worthy in the racing environment... but I'm a diehard / old school piston engine lover going back a few generations.

John_M_

Take a look at these spark plugs.... top plug in the list 1/4-32 "side fire" ... Only unknown, whether the rcxel plug cap will fit or not... I don't think it has any other unique properties to offer other than the ground electrodeless design.

http://www.miniaturesparkplugs.com/

TheEdge

My Feedback: (788)

Happy Birthday Bert

1967brutus

Dang, someone remembered!

Thanks!

Not the nicest of days, down with the flu and thus more or less in self isolation (not so much because of the 'rona, I just absolutely did not feel like organising any form of social gathering),

Messed about a bit more with the Wankel, and to my amazement, I saw another 100 RPM. Not stable but also not as an outlier "freak-peak", it was fairly repeatable.
So peak RPM today was 13550.
Given that 2 days ago it still topped out at 12950, that is an increment of near 14% from 0,46 hp to 0,53 hp (acc Pé Reivers, Lets not focus on the absolute values, whatever they are, the increase is about 14~15%).
I get the feeling, this engine was not yet even properly broken in, OR there is an awful lot of old castor crud washing out.

John_M_

Is that right ! ... well happy birthday... sorry to hear you're feeling under the weather... I haven't seen the flu in ages... eat plenty of root veggies, onions / garlic, parsnips, and citrus fruits, good immune system boosters.

Anyway, best wishes... hope you feel better soon.

1967brutus

Thanks... Allready coming back from the near-dead...