SP140 First Flight Debrief and Throttle Mod Discussion

[skivvie (Loren Cox)]

After MANY many years of wanting and EPPG, and many months of delays due to shipping damage and parts shortages… I finally got to first flight my SP140 2 days ago with my Sirocco 3 24. I’m putting this in the Throttle Controller category on the forum because that’s the main issue I had…

Overall… I love it. The ease of setup and lack of having to deal with gas and oil is as awesome as expected. Plug and go is the way. Made my gas buddies jealous when they were having starting issues on a Moster.

Noise: Not as quiet as expected when on throttle, but not surprising. I’ll have to remember the ear protection going forward. The silence when off throttle is golden.

Comfort: I’m light so I’m a bit pressed forward after setting the hang points as far back as possible and adding webbing behind the seat back to add a bit of spacing. Overall no worse than other motors I’ve flown in terms of arm angle. Harness is great otherwise. Hopefully some redesigned arms will help us light folks with a better hang angle, but with the thrust line as high as it is, I didn’t find the hang angle bad at all once I was flying despite it being around 17 degrees on the test hang. Getting out of the harness and forward for landing was slightly more difficult than I’m used to just due to hang angle, but once my foot touched it was no problem standing it up. Having the weight so high on your back makes ground handling much nicer, friend who tried it said the same.

THROTTLE LEVER: This is my main issue so far and what my thoughts focused on most of the pre-flight and flight. I used CHILL mode the whole time, so keep that in mind… but most of this applies regardless of mode.

The lever arm offers almost no feeling or resistance feedback and is too easy to press. While wearing thin gloves and grabbing risers and toggles, I bumped the throttle multiple times… My buddy flew it after me and did the same. I can’t imagine it with thick winter gloves, you wouldn’t be able to feel anything. In the future I won’t be arming until wing is over head to prevent accidental spin ups until I sort a better lever arm solution.

It is also difficult to find a brake toggle position that allows you to hold the brakes while not bumping the lever because the pressure is so light and lever is long. Again, with thicker gloves this will suck more. Open to suggestions here but I tried 1 finger, 2 fingers, all 4 fingers through the toggle and all seemed to bump the lever too easily while gripping. The obvious solution would be brake toggle on hand first then throttle, but this is also obviously a bad idea. I’ve used the Polini throttle for years and it’s by far my favorite, so I’m not new to this configuration of throttle. It offered good resistance and feel while not being in the way, possibly because the grip was much thinner. I’m sure I’ll get used to this with more flights and tweaking.

SOLUTION ideas:

A. Make the throttle lever half as long. This would get the top of it out of the way of the brake toggle making it harder to accidentally hit, so you can hold toggle with top to fingers and trigger throttle with bottom two. I know someone else on the forum did this already. I’ll print another lever and try it out.

B. Add a spring over the plunger. I’ve played with this and it helps a ton with resistance feeling on the bench. But it will also shorten the throw so you might not get full throttle … I have to test it plugged into motor to check. Hopefully I’ll have time to model up a modified lever with spring cup and shorter length soon. This could be fixed in software if spring prevents full throw of plunger. Maybe a clothespin or carabiner style spring would work better… will have to look at it closer.

THROTTLE CURVE: Wonky in chill mode. The lack of resistance feel on the lever doesn’t help, but the throttle curve could use some tweaking. Again I only tried Chill mode so this is limited to that mode. But, I was struggling to find the sweet spot to foot drag as it was just too sensitive. I would LOVE a software solution that would allow us to tweak a visual curve and save custom modes to the throttle. How amazing would that be to be able to tune the curve to your liking. I’ll give the SPORT mode a go next time and see if that helps as I imagine it’s way different.

Overall I’m in love. It’s so close to perfect given the available battery tech. A few tweaks on the throttle stuff and I will have nothing to complain about. Thank you to everyone who put time in developing this thing, it’s a work of art as far as I’m concerned.

-Loren

[skduncan560 (Scott)]

Awesome flight report. Thanks for sharing! It’s quite a different machine than anything else.

@skivvie

I would LOVE a software solution that would allow us to tweak a visual curve and save custom modes to the throttle. How amazing would that be to be able to tune the curve to your liking.

1000% yes to this. This would be awesome!

Also, which firmware were you using? 5.6 smoothed out the response in chill mode and I liked that change quite a lot, but would still love some sort of customization like you describe.

As far as accidentally triggering throttle on the ground, I only arm once I’m holding risers and brakes and ready to go and hold my middle and ring finger behind the throttle lever so it can’t move until the glider’s up and I’m actually ready for power. Haven’t tried with thick gloves yet though. This seems to work for me, but YMMV…

[skivvie (Loren Cox)]

@skduncan560

Also, which firmware were you using? 5.6 smoothed out the response in chill mode

5.6, the latest. I hadn’t flown the earlier ones so nothing to compare to. Overall the curve was nice but I just found a zone right around my ground skimming speed at the trimmer setting I was on that was super touchy and hard to lock in.

@skduncan560

hold my middle and ring finger behind the throttle lever so it can’t move until the glider’s up and I’m actually ready for power

Ohhh… nice tip. I’ll have to try that. Good idea.

[MikeD (Mike)]

I flew the older FW 5.3 last week for the first time and the throttle was really touchy to say the least. I just updated to FW 5.6 and could tell right off it’s going to be so much smoother, test flight tomorrow.

On the throttle lever softness, maybe try taking a length of music wire (5") bend in the middle 90 degrees. Then place it under the lever from the hinge side and against the throttle body, the bend shouldn’t hinder the plunger. I use this type of spring resistance on RC DLG sailplanes and spoiler setups.

[skivvie (Loren Cox)]

@MikeD

On the throttle lever softness, maybe try taking a length of music wire (5") bend in the middle 90 degrees. Then place it under the lever from the hinge side and against the throttle body, the bend shouldn’t hinder the plunger.

Hrm. Not sure I’m visualizing that right. Could you draw a quick diagram?

I modeled up a shorter lever and printed a test. I’m making another version now that will have a nipple that will hold a 9mm x 22mm spring I happened to have in my junk spring bag. Since the spring offsets the max throw preventing the lever from pressing the plunger all the way in when it is fully compressed, the nipple should push the plunger in the extra 2-3mm. It’s printing now.

[skivvie (Loren Cox)]

Hrm. Yeah that seems to work very nicely. Maybe I’ll give this a go. Anyone see any reason this is a terrible idea that I’m missing? (Yes I accounted for the fact that the nipple at the start of the throw in full open position will press a bit on the plunger… but not enough so I still had to shave off a mm of material on the bottom outside of the lever where it contacts the throttle body so it can open fully and not press any throttle. When I first tried it and armed the throttle it was ever so slightly pressed so the motor started spinning immediately … no worries no prop on*)

*this brings up safety point… the throttle needs to check if the plunger is pressed at any amount upon arming, and if it is, disable throttle until it is let off fully and reset to zero. That way no surprise high throttle events on arming should the throttle be pressed by something, jammed toggle or whatever, and the pilot not be aware of it. @Pdwhite @glydrfreak @zjwhitehead

[zjwhitehead (Zach Whitehead)]

@skivvie

*this brings up safety point… the throttle needs to check if the plunger is pressed at any amount upon arming, and if it is, disable throttle until it is let off fully and reset to zero. That way no surprise high throttle events on arming should the throttle be pressed by something, jammed toggle or whatever, and the pilot not be aware of it.

Yep. This has been the case since day 1. It wont arm if you are on the pot

Each pot is slightly different from the factory so I suspect if you’re able to do it at like 2% power then its simply a calibration difference.

[glydrfreak (Braedin Butler)]

@skivvie

the throttle needs to check if the plunger is pressed at any amount upon arming, and if it is, disable throttle until it is let off fully and reset to zero. That way no surprise high throttle events on arming should the throttle be pressed by something, jammed toggle or whatever, and the pilot not be aware of it.

Here’s the code that handles that:

#define POT_SAFE_LEVEL 0.05 * 4096  // 5% or less

// Returns true if the throttle/pot is below the safe threshold
bool throttleSafe() {
  pot.update();
  if (pot.getValue() < POT_SAFE_LEVEL) {
    return true;
  }
  return false;
}

case AceButton::kEventDoubleClicked:
    if (armed) {
      disarmSystem();
    } else if (throttleSafe()) {
      armSystem();
    } else {
      handleArmFail();
    }

5% is the threshold on the throttle for arming. If you try to arm it with more than 5% throttle, the arming process will fail. The reason why 5% was chosen and not 0% or 0.5% is because there’s a tiny bit of noise while reading the potentiometer and arming could occasionally fail. I’m sure this 5% threshold can be decreased but I’m doubtful that we would find success taking it below 2%. Maybe I’ll take some time to figure out exactly how much noise there is while reading the potentiometer and I’ll try to fine tune this threshold. Maybe the ESC endpoints and throttle endpoints can be tweaked so there will be no overlap with this threshold. The other issue I foresee is that every potentiometer will come with a different tolerance and this threshold may need some margin to cover that.

[glydrfreak (Braedin Butler)]

ha. Zach beat me to it.

[skivvie (Loren Cox)]

Nice! I failed to test it at a higher throttle percentage to see if this safety existed. Very glad ya’ll were on top of that one from the start.

@glydrfreak

Maybe I’ll take some time to figure out exactly how much noise there is while reading the potentiometer and I’ll try to fine tune this threshold. Maybe the ESC endpoints and throttle endpoints can be tweaked so there will be no overlap with this threshold. The other issue I foresee is that every potentiometer will come with a different tolerance and this threshold may need some margin to cover that.

Got it. Happy to help test. Ideally it would be anything over zero that would tigger handleArmFail() but yeah I get the noise and calibration issue. Interesting design problem to solve. There could be a first time setup calibration run that gathers min and max values of the pot to get the baseline and adjust this setting from there, with a fallback if the calibration isn’t run.

[eagleuh1 (Jim)]

Glad to see there’s a safety feature in there to prevent a power up when arming the throttle. One less thing to worry about for a clutch like me.

[eagleuh1 (Jim)]

Clutz like me! ■■■■ spell check

[evan]

The tolerance is understandable but to avoid the issue above (it is possible to start it at a very low power setting) the initial position can also be sampled and used as the reference for zero motor power. That will take into account variation between units, kind of a self calibration.

[evan]

I like it. I was thinking it would be nice to have a “kickdown” type of action on the throttle lever. So that the first 2/3rds of travel (where you would use for launch and cruising) is lightly sprung, and you can easily hold it against the kickdown detent to get a nice steady push while running and managing your wing.
Then for climbing you pull harder to get into the final 1/3rd of travel. It means you are less likely to accidentally put in a blip of full power that could make for a bad launch. This would be an alternative to making the throttle response really slow.