Jumper T8SG - First TX stock running on Deviation

vlad_vy wrote: If you use fully charged 2S Lipo, you need replace faulty transmitter. It will be second known case with T8SG that can't save settings.

howdy,

I have used a fully charged 2s *saw people with that issue*

I have been able to "copy to" and that seems to allow the save.
it is as if the save doesnt exist or requires a different key press.

I may end up doing that. I will test to see if I can make it work through copy to,

could I format the drive rather than just delete files? then replace files?

cheers

@glenby - It sure seems like the power-down save is not working. If you want to try to troubleshoot that, you could check the following. When you turn power switch off, do the tx power leds remain on for 1-2 seconds? If so, the required power stretch circuit seems to be working. The trigger for the power-down save is the voltage on the bottom pin of the power switch going from ground to +battery voltage. With switch off you can check for + voltage on the bottom switch pin, and also on the main board. The three pin mini plug in the middle of the main board (labelled on/off) brings the switch pins down to the board with the same orientation.

Also be aware the "copy to" workaround can be used to copy on top of (replace) existing model files. And the limit is not 30, but 255 model files. You just have to add modelxx.ini files to as large a number as you want, they just need to be contiguous with no break in the numbers.

@FDR - My own experience has been the following. I was experimenting with the possibility of low voltage operation with the goal of using a single Li-Ion battery. I found that I could set up switching regulators that ensure that the 5v and 3.3v busses on the main board were always maintained even at low battery voltage. I found that the transmitter would operate as expected during a single turn-on flying session - i.e. you can change mixer parameters and model configuration, but at turn-off all changes made during the flying session were lost (went back to the original configuration at previous turn-on), if the battery voltage was below 3.8v. I solved the issue by modifying the hardware voltage divider to make sure the power-down save trigger voltage was high enough even at low battery voltage. The behavior that Glenby sees looks very similar and could be explained if something was preventing the power-down save. I never tried the "copy to" as a workaround but would not be surprised that it could work, since the operating model configuration changes do seem to hold fine during a single power turn-on. Anyway, the real solution is to get the power-down save to work.

Yes, I ran into the batt_critical warning when going to low voltages, and had to reduce it to 2800. But I found that even with that reduced to a low value, the stock hardware fails to initiate the power-down save below a battery voltage of about 3.75v. To go below that you need to modify the voltage divider.

I wasn't specifically aware switching models saved the previous model. But it makes sense to have only one active model at a time which still requires saving.

@glenby - You should try to just switch models instead of a copy to a new location. That might be a cleaner workaround to save your active model without proliferating the number of models in your list.

FDR wrote: Do I understand right, that saving doesn't work only at shutdown, but works in other cases, like copying or switching model?

So far in quick testing, it does NOT save on:
- shutdown
- changing models
- after transmitter config changes.
It DOES save when
- you select "done" for the model name (model changes are not included in this).

If I attempt to load a different model, deviation indicates current the model is unsaved.

Mixer settings and model settings can be saved through a "copy to"

Transmitter settings do not save (to change from mode 2 to mode 1, I had to edit the ini file manually).

Battery Power.
When I shutdown - there is a green and red led lit for a few seconds after turning the unit off.

So the power is being maintained for the power-down save. The save should be initiated by the bottom pin of the power switch going to + voltage, but you would have to open the case to get to that and the on/off main board socket.

I have always seen the model is unsaved warning when loading a new model after editing the current model - so that is normal. At that point I always cycle power to make sure it is saved before risking losing my changes.

ramunas wrote: So the power is being maintained for the power-down save. The save should be initiated by the bottom pin of the power switch going to + voltage, but you would have to open the case to get to that and the on/off main board socket.

I have always seen the model is unsaved warning when loading a new model after editing the current model - so that is normal. At that point I always cycle power to make sure it is saved before risking losing my changes.

hi and thanks all for posting.

I have tried empty models and even resetting a model and neither works.
so right now I am charging a larger 2s battery to see if there is sag or something silly.

in the tx.ini file, what setting could I change to lower the voltage?

cheers all

my old devo10 was a dream with deviation. sigh

@glenby - With the stock transmitter you do not have a choice of going to lower voltages - that requires replacement of the stock voltage regulators, and purely experimental at this point. What is important is that your battery be well above 7.4v. There is not much drain (about 150ma) and any reasonable 2S LiPo will do fine. I may have confused you with discussion of my low voltage experiments. The bottom line was that I ran into symptoms very similar to what you are seeing when I used very low voltages which prevented the triggering of the power-down save. In your case you have plenty of battery voltage so that is not the issue. But any problem in the continuity of the connections from the bottom pin of the power switch to the mcu pin that initiates the power-down save will have the same symptoms. So that is the reason for checking that the + battery voltage is there on the on the bottom pin of the power switch, and that it makes it to bottom pin of the main board Off/On socket, etc. But if you are not comfortable opening the transmitter to chase down the fault in the signal path, don't do it.

ramunas wrote: @glenby - With the stock transmitter you do not have a choice of going to lower voltages - that requires replacement of the stock voltage regulators, and purely experimental at this point. What is important is that your battery be well above 7.4v. There is not much drain (about 150ma) and any reasonable 2S LiPo will do fine. I may have confused you with discussion of my low voltage experiments. The bottom line was that I ran into symptoms very similar to what you are seeing when I used very low voltages which prevented the triggering of the power-down save. In your case you have plenty of battery voltage so that is not the issue. But any problem in the continuity of the connections from the bottom pin of the power switch to the mcu pin that initiates the power-down save will have the same symptoms. So that is the reason for checking that the + battery voltage is there on the on the bottom pin of the power switch, and that it makes it to bottom pin of the main board Off/On socket, etc. But if you are not comfortable opening the transmitter to chase down the fault in the signal path, don't do it.

Howdy,
I will have a look in the next day or so. a bit brain dead after work and the eyes are shot.

with the "copy to" I am able to setup a model and save mixer settings too.
so I think I can work with it (although it is a bit clunky).

I need to read more on mixer settings to enable the 3 position switches for flight modes in drone but that is a small thing I can do later.
right now, I can fly 2 drones with ahfds-2a.
That puts me in front of the turnigy evolution in that I can have more than 5 models.

cheers and thanks

dc59: Are you still reading this thread? And do you have any updates for us about the upcoming T8SG Plus with hall sensor gimbals?

Brainstorm wrote: dc59: Are you still reading this thread? And do you have any updates for us about the upcoming T8SG Plus with hall sensor gimbals?

And the Devo series Gimbals too please

@yets & Brainstorm,

" If " everything is OK , T8SG plus and DEVO hall gimble sample will be shipped to Deviation developers before end of Jan. ,
but I don't know when you can buy in on market ? it's depend on Jumper's policy , T8SG sent samples to developers and start
to sell at same time , but as you know there were many bugs need to fix .........

No problem, good to hear as it went very quiet about them. I'm far happier having developers test them before they come on the market. I'd be happy to test for Devo 7E

dc59: 'Many bugs to fix'? I have a T8SG version 2 (with accessible USB port) and there are no bugs. I use my 7e .ini files. The only thing is I have to rename the switches BECAUSE THERE ARE MORE SWITCHES. The only way to fix that 'bug' would be to remove switches. Don't fix what ain't broke.

Hall effect gimbles? Would you put 'go faster' stripes on your Rolls Royce?