HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Apr 27, 2012, 12:06 AM
OlliW
Joined Sep 2009
1,969 Posts
Quote:
the board is actually double sided
it's a nice layout nevertheless. I did some layouting with eagle for some projects, and it's easy to spend ages on that...

Quote:
Regarding the bootloader, my chip is '88PA so I've flashed it with the provided BL_4RobbeProgger_m88pa.hex. I have tried both, the diode serial adapter and FTDL USB to serial adapter with different speeds and power-on sequences, also direct RS232 port and USB to RS232 cable converter. I've had no response.
most strange. Another typical mistake is to forget to burn all fuses, but I guess this you certainly did right. Most strange.
BTW: no, there won't be that much revisions... LOL

As regards Chinese weights, since the Sports versions the TRex tail holders have them on it, or at least something which are supposed to be the Chinese weights (looking quickly at HK, the clones seem to tend to not have them).

Olli
OlliW is online now Find More Posts by OlliW
Reply With Quote
Sign up now
to remove ads between posts
Old Apr 29, 2012, 06:46 PM
Registered User
United States, CA, Saratoga
Joined Feb 2010
179 Posts
[QUOTE=OlliW;21447588]
A questions to anyone who experienced the rudder-only-no-gyro problem:
Did it occur (so far) ONLY after a programming step via the USb adapter and the Windows program, or did it even happened at start up also with the last programming a couple of flights back?

OlliW,
I *only* experienced the rudder-only-no-gyro problem immediately after a USB programming cycle. The cure was always to wait with power off. Once it worked, it worked solid until the next programming cycle. Actually, I run into this pretty often, but I've gotten used to dealing with it now.
closedsink is offline Find More Posts by closedsink
Reply With Quote
Old May 02, 2012, 02:12 PM
Registered User
Czech Republic, Olomouc
Joined Sep 2011
126 Posts
Olli,
I have the same experience. I occured just after programming and only in rare cases.
It never happened in field.
By the way FW16 works well on fixed plane on all 3 axis. I only had to lower gain to get rid of wagging in high speed flights and to lower gyro strenght in maneuvers.
ablack is offline Find More Posts by ablack
Reply With Quote
Old May 03, 2012, 02:43 AM
Registered User
Australia, NSW, Sydney
Joined Mar 2011
341 Posts
Glad to have found Olli again. Would love to update the documentation, some questions are not yet answered at the german board.

Currently I am fighting with two GA250 and the latest 0.16 firmware on two HK-250 helicopters. Both are wagging the tail violently and I have almost given up on it.

The first 250 was flying great, but still with some wagging in sharp right turns. Unfortunately I had to replace the cyclic servos and after that the nightmare started. Thought it must be a mechanical issue, so I started to fix my second HK-250 with a reflashes GA-250 gyro. But the issue is the same.

Gain is somewhere between 1% to 20%. It just doesn't matter. In rate mode everything works ok as there is no wag and the heli flies well. No visible vibration. But sometimes in rate, the way of the rudder is limited to lets say 15% of the usual way. On some occasions rudder is using the same amount of way like in HH. When I try to lift off in Rate, I am not able to compensate the spin due to that.

I am also confused about the debounce parameters. On a smaller heli are lower or higher values likely to use? I am currently at 20/25, can this be part of my crazy tail wagging problem? While testing I had with one particular setup the problem that CCW a quick stop counter clockwise was crisp, but clockwise always would end in a wag of death.

My impression is that the applied inner logic is probably not ideal for smaller helicopter. It seemed to be better with 0.15.

Holger
HolgiB is offline Find More Posts by HolgiB
Reply With Quote
Old May 03, 2012, 04:41 AM
OlliW
Joined Sep 2009
1,969 Posts
@closedsink, ablack:
THANKS for reporting back your experience. Would be helpful if more user would share their experience. Anyhow, its relieving to know that at least we three didn't had issues with it in operation. (I think however that the next release will nevertheless include an alert feature for safety).

@ablack:
great that your fixed wing applications works! Maybe you like to share more details about your plane, setup, parameters etc (had the change in the controller structure been important? etc. pp)

@Holger:
Sad to hear that after your initial success and satisfaction you encounter problems now.

Since the gyro seems to work perfectly and as expected in rate mode (I at least read your comment such), it's hard to understand what the reasons for your wagging issues could be. The v0.16 controller is totally identical to the v0.15 controller if the Debounce parameters are set to 30 (or CCW=30 and CW=0). The only thin I can imagine is that something "trivial" such as an incorrect parameter (e.g. due to a bug in the codes) or indeed a little bug confuses the hh mode in your particular case. Could you please do this: Connect to AvrConfig, read in your settings, set Diagnose=10 in the Diagnose window, and write the settings. Now power up your heli and toggle your Gain switch a couple of times after the gyro (and reciever) has initialized. Power down, and repeat this a couple of times, e.g. three times. Could you then send the hexdump of the EEPROM as read out with AVRootloader as well as the output in the text output field in AvrConfig after Read (see closedsink's post #105 for what I mean)? This should allow me to double check everything for exactly your parameter setting.

As I said, as everything seems to be fine in rate mode, it's really hard to see that there should be a fundamental reason in the case of 250er in the hh mode. Until proven beyond any doubt otherwise I don't go with the inner logic argument

Quote:
But sometimes in rate, the way of the rudder is limited to lets say 15% of the usual way. On some occasions rudder is using the same amount of way like in HH. When I try to lift off in Rate, I am not able to compensate the spin due to that.
Quote:
some wagging in sharp right turns
We did briefly discuss this already. This sounds very special, and I would honestly rather expect the reason to be in the heli (e.g. cyclic ring problem, could be purely mechanical or swash servo related, etc.). Anyhow, my feeling here is that only a video showing the issue in detail or any further such info would allow a far-distance diagnose. E.g., I do not have any idea what a sharp right turn is actually to you.

By the way, this
Quote:
I had to replace the cyclic servos and after that the nightmare started
is obviously an alerting statement and would rather point to anything else than gyro fundamentals.

Cheers, Olli
OlliW is online now Find More Posts by OlliW
Reply With Quote
Old May 03, 2012, 09:28 AM
Registered User
Australia, NSW, Sydney
Joined Mar 2011
341 Posts
Thanks for your response Olli, I will do the suggested and provide you the requested data on the weekend.

In the meantime I upgraded to better cyclic servos, changed my swash to the Align one with zero play and today I just replaced the main shaft bearings. Also feathering shaft/main shaft itself. I upgraded to a KDS N300 servo (2.2KG, 0.07ms - DS95 clone), tried three different types of blades and in between used all different parameters - but always first the standard which worked initially for me so well.

Its really strange, I spool up the heli on the ground in rate and as soon as I switch into HH, the tail vibrates very fast and violently.

Today after changing all the mentioned and still no succes, I switched the debounce to 15/15, reduced PID I to 2.5 and HH gain from 0.50 to 0.40. No difference. I increased the I deadband from initially 1 to 5. Only difference now is, that the vibration is a little slower. Vibration means fast swining/wagging for about 5mm into each direction.

A counter clockwise turn on the ground stops the wagging for a few seconds. With clockwise turns I will always get into a long (5 - 10cm), violent never ending wag.

For me it feels like my gyro runs like on 100% gain and I can't reduce it. So the data I am collecting from it might be the key... But I am facing the same issue on my other HK-250, using a simple HK-SCM9-6 on the tail. Guess sorted this out on one of them, both helis will be happy back in the air.

Quote:
By the way, this
Quote:
I had to replace the cyclic servos and after that the nightmare started
is obviously an alerting statement and would rather point to anything else than gyro fundamentals.
Its an understandable response from you. Three HXT900 were killed by a shortened wire, the insulation came off on one but luckily it happend on the bench. Everything else survived but after the replacement of the servos, I noticed a tiny tail vibration. But the heli was flying well besides that... unfortunately this "vibration" killed my DS420. It just got too hot in this over attempt of compensation.

So maybe its related to a new gyro/servo combination... the DS420 was just a lucky choice? And again, it might be tempting to explain the issue by "the GA250 is damaged by the servo incident", but same issue occurs to me on the other 250 heli, on a different GA250.
HolgiB is offline Find More Posts by HolgiB
Last edited by HolgiB; May 03, 2012 at 09:45 AM.
Reply With Quote
Old May 03, 2012, 12:32 PM
OlliW
Joined Sep 2009
1,969 Posts
I think the data may help a lot.

What you describe is most puzzling. It seems rather systematic, hence I wouldn't think that it is related to a new gyro/servo combination. However, because it's at a sudden that systematic it's my general feeling, and also my experience, that if something did work before and suddenly doesn't work anymore, even if replaced by identical parts, then it's most likely not this something which is at the heart of the problem. For instance, I doubt that a DS420 would die upon whatever attempts to compensate... I would bet that it died because of another reason... and see a finite probability that this other reason is the cause for many of the issues. Don't get me wrong, I am not saying that the issue can't be with the GA250, there are really many possibilities for little bugs in the softwares (you certainly remember the servo dir bug)(it could be e.g. as simple that the newest version of AvrConfig doesn't set PID rate or PID I properly and so on and so forth), I just see a significant danger that one is focusing on the symptoms but not the cause.

By the way, I have not responded to your Debounce question. In some simplified way the Debounce can be seen as an estimate of how fast the tail can respond (this speed does depend on the servo speed but also all other factors such as the rpm of the tail blades, size of the tail blades, inertia of heli etc pp). 30 means that the tail is ultra fast. 1 means that the tail is ultra slow. I can't tell for a 250, but for a 450 a typical value for debounce seems to be something like 10-12. I would not think however that the tail of a 250 is fundamentally faster/slower than for a 450.

For your wag issue, since it doesn't seem to depend on Debounce, which actually does make sense to me, I think you just can ignore the Debounce (and De/Accelartion) parameters, i.e., have them set to 30.

What does actually happen if you stay in hh mode but set PID I = 0, and for PID Rate the same value as you use in rate mode? With PID I = 0 (and the Debounce/Acceleration to 30) the gyro should essentially behave identically in both rate and hh mode. Form what you say I suspect that this is not what you would find, but it would be an interesting test.

Another point which you have not explicitly stated, if you revert back to v0.15, not only for the firmware but also ALL AvrConfig files, your original "good" behavior comes back or you still get this strange wagging thing?
OlliW is online now Find More Posts by OlliW
Reply With Quote
Old May 04, 2012, 09:13 AM
Registered User
Joined Mar 2006
28 Posts
Hi Holger,

I have been flying version 0.16 and it works perfectly for me. The symptoms you describe are almost identical to issues I had wit a friends heli. Almost gave up on it but we eventually tracked it down to a receiver problem - must have been getting noise on gain channel. Friends heli worked fine in rate mode but violent oscillations in HH. Maybe try fixing the gain in firmware if you have not tried this. Also try adding capacitor to 5V on receiver and/or ferrite on controls to gyro.

I have seen some servos noise interfere with other channels on receivers. For example on one of my 450s I have corona digital servos on cyclic and I had to add a capacitor between 5V and ground on my FRSKY receiver to fix problem. I have also seen problems with Corona receivers. May give you some clues and new noise from new servos may be your issue.

Cheers
Bruce
BrucePage is offline Find More Posts by BrucePage
Reply With Quote
Old May 04, 2012, 10:12 AM
Registered User
Australia, NSW, Sydney
Joined Mar 2011
341 Posts
Hi Bruce,

what you mention doesn't sound far fetched. I remember that I had issues with HXT900 on clylic and a Corona Servo on the tail together with the GA250. Everytime I used rudder, all cyclic servos were shaking. Capacitor certainly worth a try, too bad that I didn't order one from Hobbyking but one from an old board might do the job.

First thing I will try tomorrow is fixing the gain with the software. Thats something I used earlier we I did the first tests. When that doesn't help I will do what Olli suggested and provide some data.
HolgiB is offline Find More Posts by HolgiB
Reply With Quote
Old May 04, 2012, 12:09 PM
OlliW
Joined Sep 2009
1,969 Posts
the capacitor(s) should be as low ESR as possible otherwise it won't help much, not sure any one from any board would do that, and you should plug it with leads as short as possible directly into an unused reciever port (e.g. the binding port).

the problems might then also be related with a too weak BEC, or to the servos

(I don't think that this is then a particular GA250 problem, since IMHO it is as well designed as it's possible, and it's maybe not its fault that it is operated in a tricky environment).

If you are now using the Gain plug but not the internal rate gain values as you did before, couldn't it then simply be that your transmitter/reciever sends a too high gain signal? (didn't you told me once that your transmitter system is known to has some gain adjustment issues?)


PS: good comment, Bruce, thanks.
OlliW is online now Find More Posts by OlliW
Reply With Quote
Old May 04, 2012, 10:52 PM
Registered User
Australia, NSW, Sydney
Joined Mar 2011
341 Posts
Yes Olli, in the beginnig (for the other, I was the very first tester) I suspected a problem and just to exclude that as a possible reason for issues, we switched to setting the gain from the AvrConfig. The issue is only know to arise together with a GY401b. Later I switched it back to the gain from the TX and that worked well.

However, today I tried to set the gain to 0.05 in the AvrConfig and to permanent HH. It works! No wagging and the tail holds good. Now I can dare to change the other paremeters.

Maybe it would be interesting (if you got time) to provide you the data as suggested earlier. Others with the T9X might run into the same issue and with the data the GA250 is able to collect, we should be able to find out the reason for the issue.

Well that also explains, why I am having the same trouble with the other 250. Its because the TX/RX combo which is the same for both helis.
HolgiB is offline Find More Posts by HolgiB
Reply With Quote
Old May 05, 2012, 04:31 AM
OlliW
Joined Sep 2009
1,969 Posts
I am glad that things improve, was probably a frustrating time the last weeks.

BTW: You probably use "Gain Input Function = not used" at the moment. You may wish to try "Gain Input Function = toggle mode". This mode allows you to switch between rate and heading hold mode via your gain switch at the transmitter, but the actual gains are those set via the parameters "Rate Gain" and "PID Gain".

Even if things improved, I am not yet totally convinced that you got to the bottom of the problem. I mean, at the point where you switched to using the transmitter gain you should have seen immediatly the wags showing up, what you tell however reads to me as if history had been such: you first used intrenal gain with your first 250+GA250 which works, you then changed to transmitter gain without complains, you then had this servo incident from which on your problems started, which now can be resolved by switching back to using internal gain... this to me sounds as if using internal gain just helps avoiding a yet undetected problem.

If it's an electrical problem, along the lines Bruce suggested, I would think they should show up already at the ground, i.e. on your workbench. That is, on the workbench how does the tail servo behave by just moving the transmitter sticks around (with the gyro in rate vs heading hold, and the gain used from the transmitter, i.e., the problematic setting)? What happens on the same conditions if you push the tail around?

I am looking forward to inspect your data and try out myself your setting.
OlliW is online now Find More Posts by OlliW
Reply With Quote
Old May 05, 2012, 12:12 PM
Registered User
Australia, NSW, Sydney
Joined Mar 2011
341 Posts
Quote:
Originally Posted by OlliW View Post
I am glad that things improve, was probably a frustrating time the last weeks.

BTW: You probably use "Gain Input Function = not used" at the moment. You may wish to try "Gain Input Function = toggle mode". This mode allows you to switch between rate and heading hold mode via your gain switch at the transmitter, but the actual gains are those set via the parameters "Rate Gain" and "PID Gain".
Thanks for mentioning that, I just changed it! And it was indeed frustrating. After having on two heli this kind of issue and very attempt of mechanical improvement did not improve the tail at all...

Quote:
Even if things improved, I am not yet totally convinced that you got to the bottom of the problem. I mean, at the point where you switched to using the transmitter gain you should have seen immediatly the wags showing up, what you tell however reads to me as if history had been such: you first used intrenal gain with your first 250+GA250 which works, you then changed to transmitter gain without complains, you then had this servo incident from which on your problems started, which now can be resolved by switching back to using internal gain... this to me sounds as if using internal gain just helps avoiding a yet undetected problem.
Yes, you recounted it precisely and guess you are right. But when I look back, when I tested this GY520 gyro, I had this crazy tail wag issue as well. Doesn't matter hat type of heli - 250 or 450, with both the same issue. Others here were somewhat statisfied with the GY520, but for me it was completely unusable. So its probably the transmitter and the way it handles the gain. The RC doesn't seem to matter - I am using the samll 6 and large 8 channel version and on both the GY520 failed big time. However worth investigating, I might be not the only one.

Quote:
If it's an electrical problem, along the lines Bruce suggested, I would think they should show up already at the ground, i.e. on your workbench. That is, on the workbench how does the tail servo behave by just moving the transmitter sticks around (with the gyro in rate vs heading hold, and the gain used from the transmitter, i.e., the problematic setting)? What happens on the same conditions if you push the tail around?

I am looking forward to inspect your data and try out myself your setting.
Tomorrow morning is my usual flight slot. I tried to streamline my setup and once I return I will put everything together. The "tail vibration" seems to remain an issue. The servo is a different one, the KDS N300 and seems to be capable to deal with it. Turning the PID I from 3.5 to 3 reduced it significantly. I will take my laptop with me tomorrow and play with some settings.

Holger
HolgiB is offline Find More Posts by HolgiB
Reply With Quote
Old May 06, 2012, 06:28 AM
Registered User
Australia, NSW, Sydney
Joined Mar 2011
341 Posts
Hi Olli,

I did what you said and switch to diagnose. Here are the data from power on/off the heli 3 times and switching a couple of time between rate and HH. In the TX HH is set to 10 and rate to 65.

Code:
Reading Options... please wait!
x00 x20 Version: x10 x00  -> 16 (x10)
x00 x2f Gain Input Function: x00  -> 0 (x00)
x00 x30 Gyro Mode: x02  -> 2 (x02)
x00 x31 Rate Gain: x05  -> 5 (x05)
x00 x32 PID Gain: x32  -> 50 (x32)
x00 x33 PID I: x19  -> 25 (x19)
x00 x38 Debounce CCW: x19  -> 25 (x19)
x00 x37 Debounce CW: x14  -> 20 (x14)
x00 x27 De-/Acceleration Limit Right: x1e  -> 30 (x1e)
x00 x29 De-/Acceleration Limit Left: x00  -> 0 (x00)
x00 x2c Gyro Orientation: x01  -> 1 (x01)
x00 x50 Servo Type: x02  -> 2 (x02)
x00 x51 Servo Direction: x00  -> 0 (x00)
x00 x52 Servo Mid: x00  -> 0 (x00)
x00 x53 Servo End Right: x46  -> 70 (x46)
x00 x54 Servo End Left: x3a  -> 58 (x3a)
x00 x26 Rudd Dead Band: x0c  -> 12 (x0c)
x00 x24 Rudd Mid Mode: x01  -> 1 (x01)
x00 x25 Rudd Mid: x02  -> 2 (x02)
x00 x2b Rudd Rate: x0a  -> 10 (x0a)
x00 x4a Pitch Mix: x00  -> 0 (x00)
x00 x4b Pitch Zero: x00  -> 0 (x00)
x00 x39 Controller Structure: x00  -> 0 (x00)
x00 x34 I Dead Band: x01  -> 1 (x01)
x00 x35 I Limit: xf4 x01  -> 500 (x1f4)
x00 x4c Gyro Limit: xe0 x01  -> 480 (x1e0)
x00 x4e PID Limit: xe8 x03  -> 1000 (x3e8)
x00 x5e Diag: x06  -> 6 (x06)
x00 x5f dRuddMid: xff xff  -> 65535 (xffff)
x00 x61 dGyroMid: xee x1f  -> 8174 (x1fee)
x00 x63 dppmGainRate: xff xff  -> 65535 (xffff)
x00 x65 dRate Gain: xff  -> 255 (xff)
x00 x66 dppmGainHH: xff xff  -> 65535 (xffff)
x00 x68 dPID Gain: xff  -> 255 (xff)
Options read!
Find attached the content of the eeprom. For the test of course, I reconfigured the gyro to gain from the TX... I neede to change the extension to TXT, otherwise I doesn't let me attach it.

Holger
HolgiB is offline Find More Posts by HolgiB
Reply With Quote
Old May 06, 2012, 03:21 PM
OlliW
Joined Sep 2009
1,969 Posts
Hey Holger

THANKS for the data.

I couldn't find any bug in my code (which doesn't mean that there isn't one, just that I couldn't find any).

There are some things in your data however I don't really get.
1) In your diagnose setting I can see that your diagnose is 6, and dGyroMid is indeed at a reasonable value, dGyroMid=8174. However, all other values are just default (0xFF=255 for bytes and 0xFFFF=65535 for words). In contrast, in your EEP-File, diagnose is 7, dRuddMid=1522, dGyroMid=8177, dppmGainRate=1478, dRate Gain= 2, dppmGainHH=1797, dPID Gain=120.
EDIT: this is probably because you first used AVRootloader and then AvrConfig, but didn't do the flip thing after AVRootloader. Please do AVRootloader after AvrConfig.

2) In your EEP file many fields which are actually not used and should be 0xFF are not 0xFF. Not sure how this comes about.

If we assume that the EEP values reflect your situation, we can conclude

HH = 10 at TX -> 1797 us ppm signal -> PID Gain = 1.20
Rate = 65 at TX -> 1478 us ppm signal -> Rate Gain= 0.02

So, if the above readings represent your case the two gains for rate and HH are really VERY different... which they certainly shouldn't. Furthermore, a Rate Gain = 0.02 is really redicoulously small... I am surprised that your heli is flying with that... on the other hand, PID Gain = 1.20 sounds reasonable, but could be too large hence your wagging... (with my 450 I am typically in the range of 0.7).

Anyhow, all this is very confusing and I can't tell you how this came about, and what it means, I only can tell you that something is very strange with your readings. PID Gain = 1.20 and Rate Gain= 0.02 doesn't make any sense.

Maybe you can do the following. Open AvrConfig, read your settings, and store them via the menu settings->save. Then open AVRootloader, go to the EEprom tab, read in teh EEprom. Now edit the whole fields such that everything is 0xFF. Now write the EEprom to your GA250. This should essentially reset the EEPROM. Switch on and off your GA250 a couple of times (wait until the initialisation is completed). Read the EEPROM again. There shoud be now many more 0xFF inbetween the readings. Now open AvrConfig, load your previously stored settings and write them to your GA250. Switch on and off your GA250 a couple of times (wait until the initialisation is completed). Now connect to AvrConfig and set Diagnose to 10. Switch on and off your GA250 a couple of times (wait until the initialisation is completed). Switch on your GA250 and wait until the initialisation is completed. Then, and only then flip the gain switch at your transmitter a couple of times. Power down the GA250. Now connect to AvrConfig (not AVRootloader!) and read in your settings. Have a look into the diagnose window, it should show now some values <>0xFF. Save the output of AvrConfig as you did before. Disconnect form AvrConfig. Now, and only now, connect to AVRootloader and get the EEProm readings as before.

You are sure that you use the v0.16 dev file in AvrConfig, and not maybe the old v0.15? (in the first line of AvrConfig you should see GA250-Cp-Gyro-v016).

cheers, Olli

EDIT: I think you simply don't get around figuring out what your relation of the TX gain settings with the gain ppm signal at the reciever is. Go through the settings in 10% steps and read of your results e.g. with a servo (or on the TX monitor).

EDIT II: since with the gain plug active (Gain Input function = set gain & mode) the gyro initialized, the gain signal itself is at least OK and valid.
OlliW is online now Find More Posts by OlliW
Last edited by OlliW; May 07, 2012 at 02:00 AM.
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion A Free firmware project for the GA250 Gyro cruzado Mini Helis 344 Apr 29, 2014 05:19 AM
Discussion GA250 gyro issues rangamatchstic Electric Heli Talk 29 Jan 20, 2012 10:33 PM
Sold Assan MEMS GA250 gyro Jonathan Ott Aircraft - Electric - Helis (FS/W) 0 Jan 12, 2012 08:21 AM
Sold assan ga250 mems gyro $15 shipped ohsnap Aircraft - Electric - Helis (FS/W) 2 Oct 27, 2011 10:57 AM
Found ga250 gyro tcb22185 Aircraft - Electric - Helis (FS/W) 1 Sep 21, 2011 07:44 PM