8FG range check almost a disaster

[stevehof]

One of my 8FG converts was range checking his nearly new Trex 550 after doing Mr. Mel's buzzer retrofit. I was helping him do the range check so this is a first hand account. The transmitter was in 'condition' normal. Throttle curve in normal was a flat 30% to support the Castle ICE ESC governor set up. He does not have 'auto bailout' enabled so his throttle endpoints are around 85 each top and bottom.

He walked out until he just lost control, maybe around 100 feet. Within a few seconds of the swash not responding anymore the ESC armed AND the blades started to spool up. I can't fathom why the ESC would arm at 30% throttle? Anyhow, we tried this again with the ship in throttle hold and the ESC did NOT arm. There is no zero throttle on this ship other than throttle hold and the ESC did not arm when in throttle hold during the range check.

The only thing I found out of the ordinary is that his fail safe was at -126 even though his low throttle endpoint was only around -85. In my 8FG my fail safe numbers mirror my low throttle end points. We did this whole drill twice to make sure it wasn't a fluke or some stick or switch position problem. What the heck are we missing here?

[Phaedrus]

I would confirm that the throttle was set to failsafe and that the value is -100. It sounds like the failsafe is not set correctly. It sounds more like it is set to "hold" (which holds last goo command) since it spools up in normal and not in throttle hold. Try that and see what happens.

[stevehof]

Originally Posted by Phaedrus

I would confirm that the throttle was set to failsafe and that the value is -100. It sounds like the failsafe is not set correctly. It sounds more like it is set to "hold" (which holds last goo command) since it spools up in normal and not in throttle hold. Try that and see what happens.

Throttle failsafe seems to be set correctly. It is a bit lower than I would have expected at -126 when the low throttle travel is only set to 85. However, it does work. That's how my friend shut down the ship when the blades spooled up. He just turned off the transmitter.

[Phaedrus]

OK, now I am lost. So if he shuts off the TX the ESC goes to zero throttle due to the failsafe, correct?

But is he is doing a range test and gets out of range the swash goes to failsafe but the throttle spools up (except in hold)??

The whole idea of the range test is to see what happens when you lose signal from the TX. I guess I am very confused now. What is the motor doing while doing the range test? Is it spooled up or not? If it is it should turn off when he gets out of range. If it is not spooled up what condition is he in? You mentioned he was in normal with a 30% throttle setting, if that is the case why isn't the motor turning during the test?

Like I sad, I'm lost now as to what is actually going on. Sorry.

[stevehof]

The transmitter was in 'condition' normal. Throttle curve in normal was a flat 30% to support the Castle ICE ESC governor set up.

It was range checked in normal mode, which would ordinarily be a set RPM. However, since the throttle is not at zero in his version 'normal' mode the ESC should not arm. This should have been a safe way to range check since we would NOT expect the ESC to arm at 30% throttle. At some point just when the signal was weak or whatever, before failsafe became active, the ESC sensed zero and armed and then, I suppose due to the 30% flat throttle line, the ship started to spool up. It obviously was NOT in fail safe at that time. When he turned off the transmitter after we saw the motor start up, the motor stopped as is should with his zero throttle fail safe setting.

[Phaedrus]

OK, I think I see now. There is about a 1 second delay between the loss of signal and the start of failsafe. During that time it holds last good command, which it is seeing as 30% throttle. Some ESCs will see that as a valid command and arm even if the throttle stick is at full low, since it will arm, then soft start and spool up to the 30%.

I suggest doing range checks in hold with that set up and ESC. There is nothing wrong, it is all a matter of the set up and how the ESC and radio are talking to each other.

On my 18MZ my "normal" (where the radio first starts) has a zero throttle curve. Then I have a hover condition (Idle up 1) which has a normal looking curve ( 0 to 90), and idle up 2 with a 90/80/90 V-curve in it. I do not use a governor right now.

[stevehof]

Originally Posted by Phaedrus

OK, I think I see now. There is about a 1 second delay between the loss of signal and the start of failsafe. During that time it holds last good command, which it is seeing as 30% throttle. Some ESCs will see that as a valid command and arm even if the throttle stick is at full low, since it will arm, then soft start and spool up to the 30%.

I suggest doing range checks in hold with that set up and ESC. There is nothing wrong, it is all a matter of the set up and how the ESC and radio are talking to each other.

On my 18MZ my "normal" (where the radio first starts) has a zero throttle curve. Then I have a hover condition (Idle up 1) which has a normal looking curve ( 0 to 90), and idle up 2 with a 90/80/90 V-curve in it. I do not use a governor right now.

Not too sure I entirely agree that there is 'nothing wrong'. We should have known this information before range checking our E powered ships. This is a safety issue. The manual needs to be updated to reflect the realities of this system with E powered ships.

I tried a range check on my own ship today. I am also using a Castle Creations ESC set to GOV and set rpm. Mine did the same thing as my friend's ship.

I powered up in range check mode and in normal condition, which has a 30% flat throttle to support the ESC governor. I walked out about the same distance as we did with my friend's ship and sure enough about the time we lost control of the swash, the ESC armed and soon after the ship spooled up. I had the blades off so I walked another 10-15 feet and the ship went into fail safe and the throttle shut down. The blades were spinning a lot longer than one second.

It is now obvious that between the time that the link is actually lost and the receiver goes fail safe there is a period when the signal is messed up enough to send a false zero throttle to the ESC that armed the ESC and then the signal went back to the 30% flat throttle curve and started the motor.

The bottom line is that it is NOT SAFE to range check a Futaba receiver unless the throttle channel is at zero. This may seem like a no brainer after the fact but most everyone at our field who flies electric with a Castle gov walks to the line in idle up and just arms ESC when everything is ready to go by going into throttle hold. So, doing the range check in a mode that did not have a zero throttle didn't seem like a dangerous thing to do. In fact, we tried the same experiment on a DX8 and even when the throttle was set to zero, the ESC did not arm in range check mode.

[Phaedrus]

I'll post the same answer here that I did over on the other forum:

Your issue is not with the receiver, it is with the ESC and how it acts in the second between loss of signal and the start of failsafe. As I told you on the other forum the Futaba receivers will go into hold for ~1 second before failsafe activates. You would not want failsafe to happen at the slightest loss of signal. So the Futaba RX waits that 1 second and then says "Yep, I've lost contact with the TX. Go Failsafe".

But you are correct in the sense that you should never range check any radio with anything other than a zero throttle setting unless you know exactly what your ESC is going to do. Your failsafe is working exactly as it is designed to do. It is the ESC that is not playing nice. With power planes/helis you never do a power on range check unless the aircraft is restrained, why is an electric any different??

[stevehof]

Originally Posted by Phaedrus

I'll post the same answer here that I did over on the other forum:

Your issue is not with the receiver, it is with the ESC and how it acts in the second between loss of signal and the start of failsafe. As I told you on the other forum the Futaba receivers will go into hold for ~1 second before failsafe activates.

Sorry, just not buying your explanation at all. The head was turning for a good 4 to 5 seconds before failsafe kicked in. It nearly got to full rpm. 30% throttle would have been the 'last known' throttle input as you suggested in an earlier post. Castle ESC's will NOT arm at anything other than what they see as zero throttle and you have to calibrate low throttle before you can even use the ESC. Somehow the signal was messed up during the transition between proper linkup and failsafe. Futaba's failsafe system needs a tune up in this area.

[Phaedrus]

Looks like we'll carry on this discussion in stereo.

That's fine. I explained how the system works and the cause of your issue. That's all I can do.

BTW - I just attempted to replicate your situation with three different aircraft, my 550 with an Align ESC, my pattern plane with a Jeti Spin 99 ESC, and another 62" (5s) plane with a Hacker Opto-Pro 70 ESC. I set a flat 30% curve, none armed with the TX on. Then I walked away in range check mode until I lost signal and went into failsafe. None of them spooled up when the signal was lost. Go figure. I took the blades and/or props off in all 3 aircraft. The receivers were a R617FS, R6203BB and a R7008SB.

[stevehof]

Originally Posted by Phaedrus

Looks like we'll carry on this discussion in stereo.

That's fine. I explained how the system works and the cause of your issue. That's all I can do.

BTW - I just attempted to replicate your situation with three different aircraft, my 550 with an Align ESC, my pattern plane with a Jeti Spin 99 ESC, and another 62" (5s) plane with a Hacker Opto-Pro 70 ESC. I set a flat 30% curve, none armed with the TX on. Then I walked away in range check mode until I lost signal and went into failsafe. None of them spooled up when the signal was lost. Go figure. I took the blades and/or props off in all 3 aircraft. The receivers were a R617FS, R6203BB and a R7008SB.

Good for you. I'm glad your system and ESC's performed properly. The ESC's you use are not what most guys are using. Castle is probably the 'everyman's' ESC these days. The guy with the DX8 who also had a successful and uneventful trial range check today was using a Castle ICE 75.

I can assure you that my R6203SB and my friend's R6208SB did not play well with the Castle ESC during range check. If you want to lay this all on the ESC, be my guest. All I know is that the Castle ESC will not arm until it sees zero throttle. Some time during the transition between proper link and fail safe it saw zero throttle which should not have happened and the head ran for far longer than the 1 second you keep mentioning. As far as I'm concerned your explanation is still lacking substantiation. Just because the receiver works with other brands of ESCs does not mean the signal sequence and timing will not inadvertently arm some ESCs other than Castle ESCs.

Considering Murphy's Law, my friend should have been in throttle hold. However, the way he conducted his fail safe check should have been fine based on the knowledge that he had at the time.

Anyway, the point of this thread was to save someone else from possible injury and let Futaba know that there is a chance for a poor outcome on their fail safe check. So, in that context I'm satisfied at this point and we need not continue this thread. We will have to agree to disagree.

[Phaedrus]

Why you do not recognize this for what it is I do not know. It is not an issue with the radio. It is an issue with the CC ESC. Why not contact them and ask them why their ESC arms when it should not do so?? Since I have shown that at least three other brands do not act this way with a Futaba radio how can it be an issue with the radio and not the ESC??

panzlflyer wrote this over on the other board:

Castle states that ANY other setting other than Zero (0000) in Normal can lead to a startup, its in their document section.
Your normal is not set up correctly for Govenor mode.

Seems to indicate that at least one other person sees where the issue really is.

[stevehof]

My normal mode is NOT the traditional setup for normal. I am just using the 'normal condition' so that I can have all three RPM ranges that are available in the Castle ESC gov mode. It is set up exactly as it should be for the castle governor operation.

I suspect the Zero 0000 throttle comment in the Castle document is referencing a traditional zero to 100 throttle curve and not a governor setup.

[Phaedrus]

I suggest given how you and your friend have your helis set up that you do range checks with the heli in throttle hold.

So far this seems to only happen with a Castle ESC and not at least three other brands. I am not clear how this is the fault of the radio and not the ESC that is the outlier. Plus it only manifests itself if the range test is done with the heli in a normal mode with the governor set and a 30% flat throttle curve set. It does not happen in Hold with a 0% curve set.

[stevehof]

From the other forum

The op states that he had a 30 across the board, so when it failsafes it sees Zero if thats how you set failsafe in the radio, Arms, comes out of failsafe and sees 30 so goes to spool up.
All Zeros in Normal or at least 0 30 30 30 would prevent that if the throttle is at low or keeping it in Hold.

I think this is what is actually happening. His failsafe was set to absolute zero on throttle. So, the receiver goes in and out of failsafe for a while. Depending on the exact distance and other factors, maybe this in and out of failsafe cycle goes on for several seconds or more.

Again, things being what they are during a 'range check' we should have been in throttle hold or whatever condition was all zeros. If we had considered the possibility of the receiver cycling in and out of fail safe, we certainly would have proceeded differently. Live and learn, that's why I posted.

The continued reference to the need to have all zeros in normal doesn't necessarily apply to some 8FG/Castle ESC set ups. I do have auto bail out set on my Castle ESC and I have Idle3 set to all zeros to arm my ESC. If I could rename my conditions in the 8FG this would not be so confusing to those who do not understand the Castle set rpm gov set up.

[Stolla]

Personally I would never do a range check in anything but hold condition with a heli! , you must always expect murphy's law to enter the equation. Especially when using with flybarless systems where thr may go into the fless system it now becomes the job of the system and not rx. Always check failsafe out on the bench without blades by switching tx off in any and all conditions. I'm not sure whether last hold position can be set for thr but I would deactivate that as well, if you get power restored motor will pick up again and you have energy stored in blades to keep it airborne

[ChapaWolf]

When Mr Futaba said and i quote "when the transmitter thinks its lost the signal it goes into a hold" when he said a second it could have been 3 to 5 seconds depending on how fast you were moving away and quite possibly moved your radio at the point of loss of signal and then regained signal before completely losing signal that sounds more like the condition you are talking about. In my opinion it would be foolish to have a flat line 30% throttle curve in normal especially with the cc esc and further why would you want to take a chance on powering up your motor during a range check and even further if it was only 30% what are you complaining about? When do you plan on doing a range check to arm your esc to fly?

[BustedRaptor77]

By any chance have the EPA values been changed since calibrating the ATV fixed endpoints in the CC ESC? Your first post eluded to that you thought something was wrong with the values.

On mine I calibrate the low and high then use a signal that is higher than the low to control the governor. I set throttle hold to be lower than the low value, and activate fail safe as hold. As soon as you change any EPA value the ATV calibration in the ESC needs to be redone.

[stevehof]

Originally Posted by ChapaWolf

When Mr Futaba said and i quote "when the transmitter thinks its lost the signal it goes into a hold" when he said a second it could have been 3 to 5 seconds depending on how fast you were moving away and quite possibly moved your radio at the point of loss of signal and then regained signal before completely losing signal that sounds more like the condition you are talking about. In my opinion it would be foolish to have a flat line 30% throttle curve in normal especially with the cc esc and further why would you want to take a chance on powering up your motor during a range check and even further if it was only 30% what are you complaining about? When do you plan on doing a range check to arm your esc to fly?

OK, this is an OLD thread and I have already plead guilty to making some bad decisions on how I proceeded with this range check two posts back up the line. So, no need to spank me further...

[BustedRaptor77]

Originally Posted by stevehof

OK, this is an OLD thread and I have already plead guilty to making some bad decisions on how I proceeded with this range check two posts back up the line. So, no need to spank me further...

Oh sorry just noticed the date.