Understanding er9x / Ersky9x

[MikeB]

Should be OK, the oXs voltage is mapped to the "Fasv" telemetry value. You just need to use that as the scaler source, sort what the actual number you get is.

For 13.3, you will probably find 89/17 is close enough!

Mike.

[jhsa]

Thanks.. I got to that number.. 89.
0.03V difference to my multimeter.

Now I'm trying to sort the scaler out because the A1 (scaler as you suggested) now has 2 decimals, and the scalers don't read them. So I guess I now have to calculate the multiplier and divisor again for the rage between 9.20 and 10.20V

João

[jhsa]

Mike, I'm now using the scalers as you suggested. It looks like it is a bit smoother but I think it is starting to slightly cut well above the upper limit of 10.20V.
Also (and this happened with the A1 too) it looks like is kinda compressing the throttle and all of it is available on the first 2/3 of the throttle stick movement. I have set a replace mix with the normal throttle and When I compare by switching on and off, in the normal throttle setting the motor spins slower for the same stick position, even if the 10.2V aren't reached yet. And this one I don't understand.

João

[jhsa]

I am now also using a curve on Ch8 and I think I managed to make it start decreasing at the right place now. The main problem is if i really apply a lot of throttle trying to make the voltage sink , big oscillations will start allowing the motor to ramp up too much and making the voltage sink a lot (to dangerous levels) again afterwards. Maybe I should also monitor the throttle stick? How could I do that? My current settings are:

SC1 A1= offset 0 *89 /17
SC2 SC1 offset -970 *221 /11

Ch8 100% SC2 Delay (d: 0.6 u:0.0) Slow (d:2.5 u: 3.0) Curve c9

Curve 9 (-100, -80, -58, -32, -4, 29, 60, 84, 100)

Ch3 + -50% THR Fix offset -50 (ADD Mix)
* -100% Ch8 fix offset 100 (multiply mix)
+ 100% THR
R + 100% THR Switch !AIL (This is for enabling/disabling if necessary)

At the top of the range, around 10.1 and 10.2 V, it does tend to balance and not oscillate so much. I can see it by bypassing the relevant mixes with a switch. When I disable it the RPM immediately rises to where it should be without the limiting, and the voltage of course drops under the 10.1V. So, the system is working more or less ok at the top of the range.The problem is when I try to increase the throttle, which I wouldn't do in normal conditions.. But I would like the motor to decrease RPM and stay there unless I pull back on the stick..

Thank you..

João

[jhsa]

Attached is a log I did a little while ago.. the last couple logs in the file were done using the settings I posted above.

João

[jhsa]

Question, What am I missing here? I'm confused

Take this mixes:

100% Thr
* 100% HALF Switch (GEA) Fix Offset (-100%)

When the sw is OFF it reverses the direction of the channel.

My question is, why is it doing that? The mix should do NOTHING when the switch is OFF right? Or as I said, am I missing something?

João

[ReSt]

In this case, obviously the switch only affects the input (HALF) and not the result of the complete mix line.

I agree, it should do nothing when the switch is off.

Reinhard

[jhsa]

yeah, it confused me this one.. Mike?


João

[MikeB]

HALF and FULL have always been different to everything else. Rather than disable the mix you get the 'other' value.
HALF: switch OFF 0%, switch ON 100%
FULL: switch OFF -100%, switch ON +100%.

Since I've recently added all switches as possible inputs, these are not needed in the same way, but I didn't want to change them as existing mixes may still be using them as they are.

Mike.

[jhsa]

Mike, I think one day we will need a version with all the changes that would have to be done, and can't be done, because of the fear of upsetting other mixes There are many things that can't be done now because of upsetting people's models.. As far as I understand (and please correct me if I'm wrong), that is probably preventing you from doing some things you would like to do, in the way you would like to do.. Stuff like organizing the eeprom structure better for example.

A big warning would say, that by using the new version people would have to program all their models almost from scratch. For me that is fine and I wouldn't think twice about moving to the new version, but that's me

João

[MikeB]

Very likely. Back in April 2012 Erazz and I were starting on er9xV2, where we were going to change the EEPROM structure and change several things, like having two full timers and changing the stick mode mapping. I have actually sorted these anyway, but I think a V2 version, which should make it clear there are many significant changes will be the way to go.
I might be able to do an EEPROM conversion routine for eepe.

Mike.

[jhsa]

Yeah, I'm really looking forward to testing something like that.. I can imagine the flash savings would also be enormous..

João

[dvogonen]

You might want to reconsider that idea.

As you know OpenTX 2 included eeprom upgrade code both in the TX and in Companion. Almost everything was converted, but some features simply could not carry over. One example is that some voice alerts no longer worked after the conversion. These relatively minor issues caused a lot of support questions and complaints.

My view is that if automatic data conversion can not be done with 100% accuracy, it should not be done at all. 99% doesn't cut it since that last percent is a guaranteed error and introducing any kind of errors is not acceptable in TX setups.

Sent using Tapatalk 2, the superior forum browser.

[jhsa]

as I understand this will be a parallel version of the firmware and people can stay on V1 until all is working..
Also in openTX, once you move forward you can't roll back to the previous version? That doesn't give people the chance of going back to a working system? or am I wrong?
If no automatic conversion can't be done, it is fine as far as I'm concerned. Having to program all my models again is a very little price to pay for all the benefits that would come from such an upgrade.

But I'm sure Mike could make the conversion work.. I guess it is not the first time. er9x already had an eeprom upgrade before if I well remember..

João

[MikeB]

I have had reservations regarding the EEPROM upgrading, which is one reason I haven't changed it yet.

Mike.

[jhsa]

Of course, but another parallel version (at least in the beginning), the V2, where people expect having to program their models again, that could be done, I think? If we install it, we know that our models won't probably work unless we re-program them
As I said before, as far as I'm concerned, that would be a small price to pay for a much better implemented firmware.

João

[dvogonen]

... Also in openTX, once you move forward you can't roll back to the previous version? That doesn't give people the chance of going back to a working system? or am I wrong?

You are right. There is no way to use OpenTX 2 eeproms on OpenTX 1 radios. Neither is there a way to convert a v2 eeprom to v1.
The best idea is to keep a backup copy of the old eeprom, in case you would like to roll back.
Sent using Tapatalk 2, the superior forum browser.

[jhsa]

Any of you heli guys could explain something to me please??
As some of you know me and Nigel are updating the er9x manual. There's a heli template called "Gyro Setup"
and creates a mix like this:

CH6 50% P2 !GEA
+ -50% P2 GEA

This doesn't look very good to me but I don't know much about helis. Could someone please explain what is needed and if this mix is really correct? And if not please post one that could be used as template
Thank you
João

[Daedalus66]

As far as I can see the results of this pair of mixes are:

With the GEA switch OFF, the first half of P2 travel takes the output from 50 to 100. The rest of P2 travel leaves it at 100.
With the GEA switch ON, the first half of P2 travel leaves the output at 100. The rest of P2 travel takes the output from 100 to 50.

I have a hard time relating this to what I remember about setting gain on a heading hold gyro, but it's been several years since I messed with a GY401 using a DX7.

Just for one thing, only the top 25% of total gain range is affected.

Nigel

[Kilrah]

I think the idea is to have the entire pot range give 0 to +100 when the GEA switch is in one position, and 0 to -100 when it's in the other so that one can set the gain, and flip the switch to go from heading-hold to normal mode or vice versa. But then the template would be wrong as an offset would be missing. Maybe the template got broken when the offset system was changed?

[KAL]

That's the idea behind it.

But unfortunately many of the gyros need different gains for the two modes.
So switching between them needs some adjusting.

Solution: use P1 for -100% to 0% / P2 for 0% to 100%.
Then both gyro modes can be adjusted separately
(Or both pots over the whole throw (-100% to 100%),
then the user can decide to use e.g. two different rate modes
but maybe that leads too far ...).

There is another possibility for those who have a rotary encoder.
Use 2 GVARs as source for the 2 gyro modes and control them
via rotary encoder (in the newer versions you can select via a switch
wether the rotenc alters the relating GVAR [at least with m128 and m2561]).

(I hope it's understandable 'cause I don't know how to express ...
If needed/whished I can create and post a eepm file during weekend)

Klaus

[KAL]

Here the announced model file.

It contains 4 solutions for using a gyro that is off in mid position (0%) and rate mode / heading hold
in the extreme positions (-100%/+100%) depending on the gyro.

In this examples gyro mode is controlled by the 3 pos switch.

- Channel 4 shows the simpliest solution.

• ID0: with P3 you can set an output value between -100% and 0%
• ID1: output = 0%
• ID2: with P3 you can set an output value between 0% and 100%

- Same game with channel 8, but values can be set separately with P1 (ID0/negative) and P2 (ID2/positive).

• ID0: with P1 you can set an output value between -100% and 0%
• ID1: output = 0%
• ID2: with P2 you can set an output value between 0% and 100%

Advantage: you cannot mix gyro modes by mistake

- Channel 7 is a variant of channel 8 where output can be set between -100% and 100% by each pot

• ID0: with P1 you can set an output value between -100% and 100%
• ID1: output = 0%
• ID2: with P2 you can set an output value between -100% and 100%

Advantage: you can set 2 different positions of the same gyro mode

- My favourite solution (so you have a rot enc) is shown on channel 5.
In principle the same as channel 7 but value is set by the rotary encoder
(uses GVAR1, GVAR2 and curve c1 / I don't know if this is possible with m64).

Klaus

[jhsa]

Klaus, thanks for that..

Mike, what do you think of all this?? I think these examples might cover most of the situations, But we have to pick one up as template?

But I was here thinking.. Don't the heli guys normally use the 3 position switch for something else? therefore the person that wrote the template used a 2 position switch instead?

João

[Daedalus66]

I don't think the template should tie up the 3-position switch, which may be needed for other purposes. And I don't see why a 0 position is needed (if it is, use another 2-position). So I would favor a template that uses say ELE to switch between two pots, one for -100 to 0 (rate), the other for 0 to +100 (HH).

[MikeB]

For those builds that have templates, it would be possible to have two gyro templates, possibly the channel 4 and channel 8 ones.
The existing template uses the GEA switch, so, perhaps combining the ideas, keep with the GEA switch, and have two templates, one uses P2 for both sides (as now), and the other uses P1 and P2, one for each side.
Also correct the template(s) to allow for the "Fix Offset" option that defaults to ON.

Mike.

[jhsa]

Also correct the template(s) to allow for the "Fix Offset" option that defaults to ON.

Mike.

Definitely

[KAL]

... Don't the heli guys normally use the 3 position switch for something else? ...

The heli guys mostly use the 3pos switch to select between "idle up" and 2 stunt modes.
For me one stunt mode is enough at the moment .

I have chosen the 3pos switch because I have a gyro that has to be set to mid position
to reset the aimed heading in heading hold mode.
But that can also be done by a second switch if needed.

With my 9XR pro in mind an idea arose: what about an option to tie two 2pos switches
together to create a 2nd 3pos switch while dropping one 2pos switch?
But this can also be done without changing the firmware if one knows how to handle that .

Klaus

[MikeB]

The 9XR-PRO has options to convert some 2-pos switches to 3-pos switches quite easily, all supported in the firmware and hardware.

Mike.

[jhsa]

You can replace a 2 position switch with a 3 position on the pro, the firmware already supports it.. I did and my ELE switch is a 3 pos now.. but this is for er9x not ersky9x. Therefore there will be only one 3 position switch.. Or am I misunderstanding something, which wouldn't surprise me..

João

EDIT: oops, Mike was faster

[KAL]

I did this mods to my pro (now three 3pos switches).

I meant the 9X. I had the idea to drop one of the 2pos to install a 3pos switch.
Let's say e.g. drop "Rud" and "Ail" switch and connect their wires to a 3pos switch.
A little awkward to program compared to sIDx but manageable .

The icing on the cake would be if er9x could support this .
But on the other hand there are things like e.g. save/restore to SDcard that seem
to be more important .

Klaus