Endpoint calibration + governor mode causing throttle hold not to work?

[kirby_]

I have an CC ICE 50. I went through the throttle end calibration as follows

Setup

• ESC has fixed-points and governor mode enabled w/ 3 values chosen
• TX uses -100 to +100 scale
• TX has a 3-way switch I use for IU mode regardless of collective stick position. IU1 sets -40 Throttle (that's 30%), IU2 sets +40 (70%), IU3 sets +100 (100%)
• TH button overrides all IU modes and sets Thr channel to -100 no matter what

How I calibrated.

• Set both the high and low endpoints on TX to +/-60.
• Set IU3, turn TH button off. Confirm TX is sending what it thinks is +100 on the Thr channel
• Power on the ESC, run the high endpoint on the TX out towards +100 until the ESC beeps at me. Go a couple of clicks out past that point.
• After another second the ESC gives me another series of notes. I don't recognize it. If I pause and wait, I get the same series of notes every second or two.
• I not hit my TH button, confirm the TX is sending -100 on the Thr channel
• Begin running the low endpoint out towards -100, At some point the ESC beeps at me again.
• I know believe the ESC is calibrated.

After this, I test (main and tail blades off). So I unplug the battery verify I'm in TH and plug that battery back in. The ESC arms. I come out of TH, the motor starts to spool up, so I immediately hit the TH button again. The motor does not turn off! I stop the motor by unplugging the battery.

I ran thru the above procedure twice, having someone help me hold and unplug the battery. Same results each time. Any idea what is going wrong?

Before this, I had not done the end-point calibration but I had TH working. The TX programming has not changed. The reason I was calibrating was after looking at the ESC logs I saw that when my TX was sending it's 100%, the ESC was hitting only the 2nd head speed, I suspected I didn't have the endpoints done correctly and that the ESC was never seeing "100%" on the Thr channel. But now, I can't fly it because I can't get turn the motor off!

I can't check CastleLink right now, so I thought I would ask here. Also, I would really like a way to do the calibration and test the results without having the motor hooked up, but I don't see how that could work, I wouldn't here the ESC beeps, and can't confirm the results after rebooting, right?

[Gyropilot]

What are your Tx endpoint numbers?

It'd be helpful if you could download and post the configuration settings from CastleLink when you have the time so everyone could see the exact setup you've programmed.

[Skidpad]

Two things:

1: When setting endpoints, you must have a linear throttle curve - you can't be in gov mode and running "flat" curves at the time - enable gov mode afterwards.

2: Also, when setting the low endpoint, you don't run it towards - 100, you run it towards +100 just like the high endpoint. It seems counterintuitive, but if that's what you did (and the above isn't a typo from you), then your endpoints aren't properly calibrated, and you need to repeat the process and go towards +100 with it.

If you fix those two mistakes, you should be able to calibrate your radio to your esc

One more thing - the esc itself doesn't beep - it sends a pulse to the motor, and it is actually the "speaker" doing the beeping; therefore, you must have it connected to hear any beeps during the endpoint calibration process.

What heli are you working with? You should either remove the blades (all four), or slide the motor away from the main gear and secure it during this time to be safe.

[gmiller0]

When setting endpoints, does it matter if the "set rpm" box is checked in the ESC? Is that ok as long as your TC in your radio is 0,25,50,75,100 when you do the calibration and then after its calibrated change the radio back to flat TC ?????

[Xrayted]

Originally Posted by gmiller0

When setting endpoints, does it matter if the "set rpm" box is checked in the ESC? Is that ok as long as your TC in your radio is 0,25,50,75,100 when you do the calibration and then after its calibrated change the radio back to flat TC ?????

Yes, this is correct. Calibration can be done in gov mode or fixed end points as long as the TC is set correctly.

[gmiller0]

THANKS MAN!!!!

[kirby_]

Ok I solved the problem. The short version is that I had to go much lower than the calibrated min endpoint in order to get ESC turn off the motor.

I'll explain in excruciating detail so maybe it'll help others:

My TX is a FrSky Taranis running a current version of OpenTX. In OpenTX, each channel is scaled from -100 to +100, with 0 being the midpoint. The "servos" screen on the TX is where I do endpoint adjustments. This servo screen uses +100 in "high" slot to mean "max channel value", and -100 in the "low" slot to mean "min channel value". It's possible to do extended limits, so I can run either the low or high endpoints on each channel from 0 to 125. For this ESC, I didn't need to do that, it calbrated under 100 on both ends for me.

Other gear details (I doubt this is relevant): Gaui X3, Castle Creations ICE 50, Spirit FBL. I have the throttle cable from the ESC plugged into the FBL, then the RX connected to the FBL via an S.Bus cable. I double checked that don't have governor mode on in the Spirit FBL, so it *should* just be repeating the throttle channel that it gets from the RX. Maybe the Spirit FBL is messing with things?

First things first, I pulled out the head, main shaft, and main gear. I was previously doing this with no blades but otherwise a functional heli. That's not really really safe and since I planned on doing this a few times, I wanted to have the motor just spinning its pinion and not moving anything else. It was quick and easy on the X3. I'll default to this from now on. Trying to unplug the battery near spinning main/tail blade grips isn't fun.

So, my original problem was that when following the calibration proceedure, when I power cycled I could get the ESC armed in TH mode, but as soon as I came out, going back to TH didn't turn off the motor.

I changed two things at once (not very scientific, sorry) to fix this:
1) Changed my throttle channel setup on my TX to go to a linear -100 to +100 curve, which I tied to the left slider control (LS in OpenTX) on the TX. When I'm calibrating stuff, I like collective and Throttle separate. OpenTX makes this sort of thing easy.

2) Changed the ESC from governor mode to fixed endpoints. I verified that I do not have the auto-rotate option enabled in the ESC.

I then re-read the CC doc on calibration and watched the video in the sticky in this forum. It matched what I was doing, so I calibrated again. To do this, I started the low and high endpoints at -30 and +30 repsectively, started at high throttle, etc. My endpoints ended up at around -70 for low and about +94 for high. On each end, this was two button clicks past the point where the ESC chimed at me.

The result was the same as my original post. Once the ESC told me it registered high point then the low point, I kept the throttle control low and cycled power. The ESC armed. I moved the throttle control up slightly and the motor started to spin up. I quickly moved the throttle control back down, and the motor remained on, but clearly at a low RPM. I played with the throttle control a little bit to confirm it was working. The motor reacted appropriately, except "low" on my throttle control (and re-verified as what my TX thought was -100 output) didn't turn the motor back off.

I unplugged that battery and did the calibration again changing nothing else. Same result, ending with the same/similar endpoint settings. This time, I had the idea of leaving the throttle stick low with the motor running and then adjusting the low endpoint even further past the calibrated point. I kept going until the motor stopped. The ESC calibration left me at about, -70 and +94 in the servos screen. The motor finally shut off when I hit about -70.5 on the low end. Each "click" on my TX is a 0.1 incrment, so this was quite a few clicks away from the calibrated low point. I know the videos say "a couple of clicks past the point where the ESC makes a noise", and I did that in both cases, but it appears my setup required even more clicks on the low endpoint.

I did some testing with the linear throttle curve and TH, it worked great. I went back to governor mode on the ESC and flat line throttle values for my Idle Up flight modes and re-tested. Worked great!

I suspect that changing the ESC to fixed-endpoint mode isn't required. The ESC calibrated the same way (same/similar endpoints calibrated) with both modes.

I also suspect that I didn't actually have to change my TXs throttle setup to have a linear throttle curve either. I'm not sure, but I believe the ESC just needs to see "max", calibrate, see "min", then calibrate that. I don't think it needs to see a full smooth transition from +100 to -100 on the channel. (UPDATE: I confirmed this. I redid calibration with with governor mode on the esc and only +100 and -100 values for the throttle nothing in between)

However, doing both of these thing did make diagnosing the low point problem easier, because it was obvious that the motor RPM was going down as I was moving the low servo endpoint until it finally shut off.

Thanks to everyone for the advice here.

[Gladius]

I believe that most TX do full point increments, So, if yours beeped at -70, the advice to "go 2 clicks past where it beeps" would result in a low setting of -72, not -70.2.

[kirby_]

Oh! That's really useful to know. I wondered about that so I tried to figure out what was happening on the Futaba in the calibration video on this forum, but I couldn't make it out clearly. Ok. I'm going to recalibrate tonight, go two full points past each ESC beep.

Thanks again!

[btheli]

I had a problem a few months ago where my CC ESC governor would not go into IU2 after it got warmed up...IU1 was as far as it would go. I worked with Castle tech support and we finally figured out that I had only gone 0.2 past the max and min. Once I went a full 2 points (NOT clicks on the Taranis) past max and min, everything started working great.

I'm also running the Spirit FBL. I could not get the governor working to my satisfaction, so I just a few nights ago I disabled it and tried to get the Castle gov hooked back up. Like you, I ran into difficulties calibrating the endpoints. I discovered that running the throttle signal through the FBL somehow changed the mapping between the Tx and the ESC. I unplugged the throttle from the FBL and plugged it back into my Rx and the ESC calibration worked as it had before.

The main problem I had with the Spirit governor was getting the autorotation bailout to work. If you get that to work, I would be very interested in hearing how you did it.

[kirby_]

I remember you saying in the Spirit forum you had trouble with the Spirt Gov. I'm not using it on this heli (my 2nd Spirit) because I bought the X3 used, it came with a CC ICE 50 that doesn't have an RPM sensor, and I haven't thought about adding a sensoryet. On my first heli, with a Phoenix Edge and Spirit FBL, I was happy with the Spirit gov.

I haven't noticed any Thr signal problems I can pin on the Spirit, it was just this 1.0 vs 0.1 endpoint adjustment thing. I will eventually try the auto-rotate CC setting and let you know how that works out.

[btheli]

The problem you described in post 1 and 7 re calibrating the ESC and the motor still running is what I was talking about. If you took the throttle out of the Spirit and put it back into the Rx, your calibration would work as expected.

[kirby_]

Originally Posted by btheli

The problem you described in post 1 and 7 re calibrating the ESC and the motor still running is what I was talking about. If you took the throttle out of the Spirit and put it back into the Rx, your calibration would work as expected.

Oh! I didn't connect those dots. Ok, so Thr cable RX->ESC would result in 2 actual clicks (0.2 change in Servos screen on OpenTX) needed for calib, but Thr cable Spirit -> ESC results in needing 2.0 change in Servos screen on OpenTX.

[Gladius]

No, you should to go 2 full points past the beeps regardless of where the cable is plugged in.

I believe that what btheli was trying to say is that FBL systems don't always pass a full throttle and zero throttle signal to the ESC, and that interferes with the programming/calibration. So, by connecting the ESC to your receiver, you're removing that issue.

Btheli, after you do the calibration, should you reconnect the ESC to the FBL?

[btheli]

Yes, 2 full points whether plugged into the Rx or the FBL.

Yes, I was saying that the FBL somehow changed the throttle signal when it was just passing it through (so it's not just passing it through). I did not spend much time studying it, and I probably could have calibrated it, but I went all the way down to -100 on the low end and never got the ESC to arm. I could have enabled the extended limits in the Taranis, but I decided to just move the throttle back over to the Rx and that fixed it.

I have an SBus cable connecting my Rx to the FBL, so no, I did not need to reconnect anything from the ESC to the FBL.

And kirby_, I meant to mention this earlier in response to your comment in post 1:

After another second the ESC gives me another series of notes. I don't recognize it. If I pause and wait, I get the same series of notes every second or two.

When you send +100 to the Castle ESC and hold it there, the repeated chirping it's doing is telling you it just cleared the logs. This is a slick way of doing it without having to plug in the Castlelink.

[HybridCycles]

@Kirby,

I haven't gone past calibrating my endpoints yet on my Edge 160hv + Spirit + Taranis.

I was able to get the high endpoint beep but not the lower one. You had to move the stick/slider to high and low position then adjust the EP's accdngly right?

I can't get the esc to arm, I was able to get it to arm without calibration on airplane/Fixed endpoints though. I tried spooling it up on this mode but noticed that on the Spirit gov screen the RPM reading is not registering(my white wire is on PIT pin on spirit) maybe the Edge doesn't spit out the rpm data on Airplane mode??

So I have 2 issues here first is calibrating the Fixed EP's(I'll try your posted steps) and 2nd would be the rpm readings to appear on the gov settings screen. Hopefully the first solves the 2nd.

I have successfully setup my 450l with the align stock esc + HW sensor +Spirit, Gov works beautifully solved all my tail problems with it!

Hopefully I can get the edge to work too....

Originally Posted by kirby_

Oh! I didn't connect those dots. Ok, so Thr cable RX->ESC would result in 2 actual clicks (0.2 change in Servos screen on OpenTX) needed for calib, but Thr cable Spirit -> ESC results in needing 2.0 change in Servos screen on OpenTX.

[kirby_]

If you're using the Spirit Gov then you're using a Throttle signal generated from the Spirit. So you need to do two calibrations. Post #12 in this thread in the Spirit FBL forum describes the process.

The short version of it ends up being:
1) Calibrate your TX Thr signal to Spirit FBL. Move your TX endpoints until the diagnostics page on Spirit's setup wizard shows 0-100 at min-max throttle
2) Calibrate Spirit's Thr signal to the ESC. Do the normal Castle ESC calibration (stick high, power on, etc), But the "endpoints" you adjust will be only in the Spirit setup software. They call this "throttle" range under their governor settings.

If you have more questions, you'll probably get better answers in the Spirit forum, because calibrating a CC ESC with Spirit governor is more about Spirit than Castle.

[HybridCycles]

Cool,

I thought the endpoints on the TX would have to be moved...
Thanks, I'll check out the post.

Originally Posted by kirby_

If you're using the Spirit Gov then you're using a Throttle signal generated from the Spirit. So you need to do two calibrations. Post #12 in this thread in the Spirit FBL forum describes the process.

The short version of it ends up being:
1) Calibrate your TX Thr signal to Spirit FBL. Move your TX endpoints until the diagnostics page on Spirit's setup wizard shows 0-100 at min-max throttle
2) Calibrate Spirit's Thr signal to the ESC. Do the normal Castle ESC calibration (stick high, power on, etc), But the "endpoints" you adjust will be only in the Spirit setup software. They call this "throttle" range under their governor settings.

If you have more questions, you'll probably get better answers in the Spirit forum, because calibrating a CC ESC with Spirit governor is more about Spirit than Castle.

[kirby_]

Originally Posted by HybridCycles

Cool,

I thought the endpoints on the TX would have to be moved...
Thanks, I'll check out the post.

You do move the TX endpoints, but only in step 1.

(all this assumes linear throttle curve that goes from 0-100 as you move the left stick)

Step #1, move your TX endpoints so the Spirit FBL diagnostics page shows the throttle range at 0 and 100 when your TX throttle stick is at min and max. This has nothing to do with the ESC yet. You could leave the ESC disconnected here if you wanted to.

Step #2. Run the CC calibration. During this step, you will be moving the TX stick, but you'll adjusting the "throttle range" settings in the Spirit software INSTEAD of adjusting your TX endpoints.

Think of it this way. With the Spirit FBL governor enabled, you TX tells your FBL what to do and your FBL tells your ESC what to do. The Spirit FBL is a middleman. The throttle signal from your TX is NOT passed directly on to the ESC. From the point of view of your ESC, the FBL is the thing sending throttle signal so you have to calibrate the ESC to the "endpoints" in the FBL (The Spirit Throttle Range settings).