Jump to content

Wheelemetrics: Custom Gotway/King Song -app


esaj

Recommended Posts

  • Replies 146
  • Created
  • Last Reply

Great, here is the table for voltage / charge state of the battery:

 if(val < 53.0) {
                        String mEnergyShow = "0%";
                    } else if(val < 54.1) {
                        String mEnergyShow = "10%";
                    } else if(val < 56.2) {
                    } else if(val < 57.7) {
                    } else if(val < 59.1) {
                    } else if(val < 60.5) {
                    } else if(val < 61.7) {
                    } else if(val < 62.8) {
                    } else if(val < 63.9) {
                    } else if(val < 65.0) {
                    } else {
                    }

But seems there is code missing for string of values 20 up to 100%.

Link to comment
Share on other sites

Chinese English Pinyin Dictionary

jiào zhǔn
to calibrate
 

And this code for calibration:

        if(jiaoZhunStart) {
            jiaoZhunCNT = (jiaoZhunCNT + 0x1);
            if((jiaoZhunCNT % 0x6) == 0x1) {
                sendMessage("c");
            } else if((jiaoZhunCNT % 0x6) == 0x4) {
                sendMessage("y");
            } else if(jiaoZhunCNT > 0x11) {
                jiaoZhunStart = false;
                jiaoZhunCNT = 0x0;
            }
        }

Link to comment
Share on other sites

20 hours ago, esaj said:

One thing that came to me to try might be modifying the GotwayProtocolCodec and trying to drop the first 0x04-byte from the MSG_TAG -header array.

Another option is to select the protocol manually in case of none is detected. But this could be more complex. 

Could you build the instance of app with forced GotWay protocol selection for me to test in the evening?

Link to comment
Share on other sites

18 minutes ago, Andr said:

Another option is to select the protocol manually in case of none is detected. But this could be more complex. 

Could you build the instance of app with forced GotWay protocol selection for me to test in the evening?

https://www.dropbox.com/s/6u21bx3nzlalk0w/app-release-force-gotway.apk?dl=0

Didn't test it in any way ;)  

Link to comment
Share on other sites

1 minute ago, Andr said:

Will try to connect tonight.

P.S. I have one MOSFET burnt anyway :) 

20160818_223113~01~01.jpg

Ouch, that's been quite a current spike... the TO-220 -package legs should be capable of around 75A.

Link to comment
Share on other sites

12 minutes ago, Andr said:

I'm thinking of soldering another 6 transistors in parallel for redundancy.

I think that you should take into account that you'll be doubling the gate charge/capacitance too, which will affect the rise/fall times. But it could work just fine, and I'd actually be interested to hear how it goes ;)

It's also possible that something else got taken out along with the mosfet.

Link to comment
Share on other sites

@esaj, yeah, we will see if it was only mosfet or not. Now the board turns on (motor disconnected) and is giving two short beeps constantly. 

Your app could not connect. Only the GotWay app. 

 

19 hours ago, electric_vehicle_lover said:

Can you please tell us in what situation the mosfets burned??

I took a hard ride in the park with ground paths, for 7 kilometers, then hit a root - the wheel shutdown. Then I tried to turn it on what resulted in sparks and smell. After that it turned into a punch-hammer. 

19 hours ago, electric_vehicle_lover said:

Can you please put pictures of the full board?

Here you are!

 

03.jpg

02.jpg

01.jpg

Link to comment
Share on other sites

29 minutes ago, Andr said:

@esaj, yeah, we will see if it was only mosfet or not. Now the board turns on (motor disconnected) and is giving two short beeps constantly. 

Your app could not connect. Only the GotWay app. 

 

I took a hard ride in the park with ground paths, for 7 kilometers, then hit a root - the wheel shutdown. Then I tried to turn it on what resulted in sparks and smell. After that it turned into a punch-hammer. 

Here you are!

  Hide contents

03.jpg

02.jpg

01.jpg

That board is the MicroWorks 30B4: https://github.com/EGG-electric-unicycle/documentation/wiki/MicroWorks-30B4---30km-h-controller-board-with-bluetooth

And is important to see that they had to revert the board for it to work (the dissipator is on outside direction) and this means they could not change the firmware and had instead to rotate the board.

Thanks for the description of the problem. I could not figure that could result in such a burn mosfet.

Link to comment
Share on other sites

3 hours ago, electric_vehicle_lover said:

That board is the MicroWorks 30B4:

Nope, this is older version. Just 30B. I bought it on aliexpress more than a year ago. And as you might have read, I did all of this upgrade to my Airwheel :) 

And I still do not understand why wheelmetrics cannot read what GW app does. :(

Link to comment
Share on other sites

 

Hello Esaj,

On 23/08/2016 at 0:18 PM, Andr said:

Will try to connect tonight.

P.S. I have one MOSFET burnt anyway :) 

20160818_223113~01~01.jpg

Ouch, that's been quite a current spike... the TO-220 -package legs should be capable of around 75A.

>> I'm not agree because it must be to use with a radiator and vented for 75A !! => there's a problem design  !

 

Nicolas

 

Link to comment
Share on other sites

24 minutes ago, Nicolas Kerspern said:

Ouch, that's been quite a current spike... the TO-220 -package legs should be capable of around 75A.

>> I'm not agree because it must be to use with a radiator and vented for 75A !! => there's a problem design  !

@esaj was though talking (if I'm not mistaken) purely about the leg of the TO-220 casing which is designed for currents up to 75A ;) 

Link to comment
Share on other sites

5 hours ago, Nicolas Kerspern said:

 

Hello Esaj,

Ouch, that's been quite a current spike... the TO-220 -package legs should be capable of around 75A.

>> I'm not agree because it must be to use with a radiator and vented for 75A !! => there's a problem design  !

Nicolas

Yes, there shouldn't be that high currents passing through it anyway, but I was referring to the burnt mosfet seemingly having blown an entire leg off. Like HEC said:

4 hours ago, HEC said:

@esaj was though talking (if I'm not mistaken) purely about the leg of the TO-220 casing which is designed for currents up to 75A ;) 

Yes, the "practical limit" for TO-220 casing is 75A, here's an IRF (International Rectifier, a company that produces power mosfets among other stuff) FAQ-note:

http://irf.custhelp.com/app/answers/detail/a_id/373/session/L3RpbWUvMTM5ODg4NjMyOC9zaWQvZio3TGc2VGw%3D

HEXFET® MOSFET package current limits

Answer ID 373  |   Updated 04/08/2009 12:43 AM

Question

IRF3205 has a TO-220AB package. This device is rated at 110 amps. How is it possible that this package type will conduct 110 amps of current. It looks small to us and the 3 leads look too small to handle 110 amps.

Answer

The value on page 1 of the data sheet has a "5" asterix against it. Footnote #5 says package limitation is 75A. 

The 100A figure results from the industry standard way of defining silicon current limit by reference to an infinite heatsink, that is junction temperature is at 25C ambient temperature. Older generations of silicon had conduction properties that were inferior to copper, so with older type devices the silicon determined the conduction limit not the package type. The current generation of low Rds silicon is a more efficient conductor than copper. As a particular die is used in more than one package style we state the silicon limit as a common characteristic, again using the industry standard. Designers can then use the standard thermal equilibrium equation, which includes the Rds value, current through the part and the thermal resistances of the package and any heatsink, to determine the maximum junction temperature that will occur in their application. The actual current limit for a given package depends on what current density and therefore what voltage drop and heating in the leads and frame a designer is willing to accept. Any heat generated in the leadouts and the leadframe will heat the silicon die, increase its Rdson value and thus increase its conduction losses for a given constant current value. Generally 75A is accepted as the practical limit for TO-220.

Emphasis mine.

Link to comment
Share on other sites

On 24.8.2016 at 4:18 PM, electric_vehicle_lover said:

It also don't work with my EGG.

Did you try the EGG app?

I can't see the difference in the board picture from 30B4 to 30B. Do you know?

No, I just use gotway 3.4.51. (Actually, the tuned app that I uploaded here http://electrotransport.ru/ussr/index.php?topic=26454.msg814534#msg814534)

Can you point me directly to the .apk?

I believe 30B4 has improved progrm. Such as store odometer. Am I wrong? Capacitors also looks different.
Here's the link^ http://www.aliexpress.com/item/Factory-price-electric-self-balancing-unicycle-mainboard-one-wheel-scooter-PCBA-controller-DIY-balancing-car-accessories/2034551122.html?spm=2114.13010608.0.73.icchGa

The description of 30B4: 

Model: 30B4

CONNECTING wires method (same as 30B )

upgrade:  chips, programs and fixed alum board. (apply more models of the shells )

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...