Beta Testers Wanted On Updated Lyen Controller Program @370A

amberwolf said:
Tom_D in another thread asked a very good question:
:arrow: Can you just change and flash one parameter value, or do you have to populate them all ? I don't want to upset the rest of defaults values.

No you can't leave a blank field or many ones and try to send the data with this software. In my case ince i have program my controller with the CHXS version of this program i never get back with the "default" value for sure!

Good day!
Black Arrow
 
Lyen,

Hopefully this won't be an issue to you:

Guys,

If you want to try programming your EB2XX model with 116 or even the 846 mcu type using one of the parameter designers but with settings which are "only" available in the "limited" access software mentioned on this thread...

You can always open the actual Parameter Designer.exe using a normal HEX EDITOR and manually change some of the entries for LVC ranges (adding your particular requirements), current settings, etc..

There is still a built in maximum current limit within these controllers, I believe it was previously mentioned as being 57A which sounds fair to right... as indicated you can use the next board size up to change the shunt expectations.

Hope this helps!
-Mike
 
mwkeefer said:
You can always open the actual Parameter Designer.exe using a normal HEX EDITOR and manually change some of the entries for LVC ranges (adding your particular requirements), current settings, etc..
-Mike

It's exactly what CHXS did.. and he corrected some error and added more choices in some parameters.

I'm using it and it work really well.

CHXS is a 17 years old guy from here , Quebec city that built his own very powerrfull ebike and began with a monster version of everything, ( 1.1kWh 28s A123 battery pack, 18 fets controller, 5305, etc.. ) for a first ebike.. I can say this guy really impress me!.. I admit he push the limits a bit too fat but he understand what he do... except he REALLY twisted the axel of his 5305 !!!.. and twice times!!!i saw 240A max current on his 18 fets controller .. ann i can say that his shunt is very well calibrated.. its me who did it with accurate kelvin connections.

The parameter designer he just edited and ameliorated is really good.

btw.. the 120 degree phase advance really increase my speed by 7%


Doc
 
Doc,

Yea leave it to the old school and the still in school to break out the hard core HEX EDIT!
Glad the phase advance got you a bump, have you tried to Hex Edit 125% yet?

I figured this hack out while trying to hand populate ASC/ASV files (the damn text data files) as the UI won't allow anything that isn't in it's combo boxes...

That said, the software (all of it that I've seen or used) simply populates combo boxes from resource encoded text (plain and editable in the exe) and then gets the actual .Text value of the CurrentIndex item of the combo box. It's very simple and straight forward and modifying it should be cake for anyone with Notepad skills.

I've found that if a value isn't legit (meaning if you edited for instance 200% into a speed % field) - the controller doesn't accept the flash so you can have a blast trying different hacks seeing what will and won't work.

Hope it helps!
-Mike
 
liveforphysics said:
Miles said:
liveforphysics said:
Software is either open-source, or it can go to hell as far as I'm concerned. :twisted:
Who pays your wages? :mrgreen: :lol:

You got me there... for now... :twisted: :twisted: :twisted:

too funny :lol: and u work in their datacenter!!?? post a nightly edge build of the next win release candidate on github! you'd be more infamous than the guy who lost his iphone.
 
Hi,

I'm really surprise many people prefer to edit the program, then to use the CHXS version wich is already works and do the job nicely with The Lyen and other controllers. :shock:

P.S. Doc i'm ready to ride with you when you want take care.

Black Arrow
 
Do you think it would be possible to turn regen up to 3?

It isn't quite strong enough braking for my tastes!

Open source is a development method for software that harnesses the power of distributed peer review and transparency of process. The promise of open source is better quality, higher reliability, more flexibility, lower cost, and an end to predatory vendor lock-in.

Firefox is open source. Linux is open source. Apache web server (%75 of the internet) is open source.

You can make it about morality and who's work and insight is being appropriated for commercial gain, but the reality is that open source is an efficient way to develop software.
I'm sure keywin and lyen aren't software pros, and could use some help from interested and willing members of the community.
And after all wouldn't an easy to use software with tons of cool features sell more controllers with programmable usb ports?

The truth is that as battery prices come down and the price of gas rises ebikes are going to become mainstream very quickly, and either an open source standard for controllers will come about OR we will have a dozen companies all offering their lame attempts at creating proprietary software for their controllers. Think itunes and the ipod. The PC and microsoft. Glade air fresheners and their overpriced refills.

And lets face it anyways; by only letting those who buy from you use the software you're creating a pirate market. This means people will download it with out visiting your site. With out checking out your inventory of bad ass stuff! I spent hours reading and learning on the ebike.ca site, and I know I am buying my clyte motor from them. Be a lighting ride for the community and the community will feed you.
 
mwkeefer said:
Doc,

...have you tried to Hex Edit 125% yet?
...That said, the software (all of it that I've seen or used) simply populates combo boxes from resource encoded text (plain and editable in the exe) and then gets the actual .Text value of the CurrentIndex item of the combo box. It's very simple and straight forward and modifying it should be cake for anyone with Notepad skills.

...-Mike

Don't try this with Notepad :) It's actually very tedious and frustrating to edit these exes. It took me a long time to wade through the code, pick the right section, and count my characaters as I edited to be sure I didn't over- or under-run the length.

@Amberwolf: Some entries I can edit easily and others I can't because what we are doing with the HEX Editor is replacing existing values, not adding or reducing; the length of the string is fixed, so if there's a 0, 1, or 2 in the field then I only have ONE space-width to edit... and if I change that space to any other length then the offset addresses will not be correct and the application will not work (become corrupted).

will_newton said:
...Where the hell are you people? Half of you are evil genius engineers and none of you can do better than this software?!?!?!

I have heard the call, but I am here to tell you that the application was written using an older version of Visual Basic and cannot be reversed except into Assembly. Fear not: I am teaching myself Assembly for just this purpose, but it is on my dime and I have a full plate of challenges.

That being said - I am willing to experiment with these params. I could see value in restructuring the settings like Speed X % to include 125, 130 (whatever) at the sacrifice of lower settings. Personally I'd like to know what happens if we reset the EBS Limit Voltage from 55, 60, 75 to 60, 75, 90. Will it work? I'll find out this week when I hook up Lyen's Mini Monster 8)

~KF
 
Kingfish said:
I have heard the call, but I am here to tell you that the application was written using an older version of Visual Basic and cannot be reversed except into Assembly. Fear not: I am teaching myself Assembly for just this purpose, but it is on my dime and I have a full plate of challenges.
I knew someone somewhere was playing with this. I just wanted to light a fire under some fannies. :D

this CHXS version mentioned earlier....is this a better or more update version of the software? I want to make sure I have the best and/or most complete versions available in manual thread here:
http://endless-sphere.com/forums/viewtopic.php?f=16&t=14836

I've got several versions listed and honestly do not follow the ESC threads, but I will make updates and changes as they are reported to me. If anyone reading this has a second to go behind me and re-read the Manual thread and let me know of any necessary updates. I'll take care of it.
 
will_newton said:
this CHXS version mentioned earlier....is this a better or more update version of the software?.

Is a better or more update version of the software can be used for Lyen controllers and other versions.

But don't use it with a small shunt between 250 to 320 u Ohms like Methods 18 fets 100 Volt 100 amps controller, with this controller it gives you something like 250 amps and over for the "rated current" and about 600 amps and over for "the phase current" in reality, twisted axel or burning motor and controller may occurred with this version of the program.

Good day!
Black Arrow
 
will_newton said:
...
this CHXS version mentioned earlier....is this a better or more update version of the software? I want to make sure I have the best and/or most complete versions available in manual thread here:
http://endless-sphere.com/forums/viewtopic.php?f=16&t=14836
...

@will_newton, I think my contribution is the last set of binaries in the download list, so I will work with those :)

General note to all: I am happy to change the descriptions to what you think is fitting so long as there is space within the string-length to do so; we might need to get creative.

That given, please allow me to display exhibits and delineate:

View attachment 1

PD_116_EnglishV2.png

Biggest difference between the two is the added features of the 116. Otherwise they should be nearly identical.

I suggest we decide to either A) modify the current exes to accommodate all options, or B) craft Power-User versions that encompass those adjustments. Maybe Option-B is best for both the greater and the lesser fish in our pond, yes? :)

I would like to park the conversation about re-write because I’m not there yet and I won’t be until I can sort out the handshaking with the device (hint hint: mwkeefer, are you listening? :) ) Once we understand how this is accomplished then we can start from scratch and rewrite the whole business anew without stepping on anyone’s delicate toes. I can’t fathom that the protocol is hugely different from other similar Infineon products where the SDK is publically available. Small caveat though is that I cannot start this until after my summer holiday... which begins next week (I’m in a mad dash to upgrade P0 since I can’t take P1; long story…).

However for now I we have a bit of time to update our current apps and I am happy to oblige on adjustments whilst I have the hood open and am tinkering.

Branching,
  • What labels do you want changed? Use the images above as a reference, then suggest a concise edit and I will see what can be done.
  • What values to a matching feature would you like to see changed? I am hoping that like mwkeefer said that the application is using the TEXT values and not the INDEX values.
Delivery: I will provide drops up to say Friday the 25th; after that I can’t promise as I have to catch a train, with my bike, with a colored faring, without a trailer, over 101 miles/day, possibly charging on the fly, with the latest BMS, through the desert, and thinking of Yoda… :shock: :twisted: :mrgreen:

~Mad Max... er, KF
 
I've been thinking about this a bit laterally, and wondering whether sniffing the bidirectional serial data link to the controller might yield a way to reverse engineer the programme?

What I have in mind is to intercept the two serial data streams (the data transmitted to the controller from the PC and the handshake data from the controller to the PC) and then work out what is actually being transmitted. I'm guessing that the data transferred is just some new variables, not the entire controller code. It seems likely that this data might even be ascii, rather than hex or binary, as the controller can probably parse strings to accept changed variables.

To do this effectively would need a way to read two serial data streams synchronously, which is beyond my limited competence. I guess using a straightforward terminal programme would work, but I don't know how to capture two data streams at once.

Once the raw data being transmitted to the controller is known, then designing a new programmer to change variables should be easy enough.

Jeremy
 
I used to use Frontline Serial spy to do similar captures...but you have to buy it..

I was just a matter of having two serial ports.
We'd tap the TX and RX and send them both to the RX line on the RS232 ports on the PC.

Realterm (freeware) may provide power enough data capture utilities to do this....
 
Jeremy Harris said:
I've been thinking about this a bit laterally, and wondering whether sniffing the bidirectional serial data link to the controller might yield a way to reverse engineer the programme?

What I have in mind is to intercept the two serial data streams (the data transmitted to the controller from the PC and the handshake data from the controller to the PC) and then work out what is actually being transmitted. I'm guessing that the data transferred is just some new variables, not the entire controller code. It seems likely that this data might even be ascii, rather than hex or binary, as the controller can probably parse strings to accept changed variables.

To do this effectively would need a way to read two serial data streams synchronously, which is beyond my limited competence. I guess using a straightforward terminal programme would work, but I don't know how to capture two data streams at once.

Once the raw data being transmitted to the controller is known, then designing a new programmer to change variables should be easy enough.

Jeremy

http://endless-sphere.com/forums/viewtopic.php?&p=133441#p133441
 
Gosh Mike1 :roll:
I recall reading that and mentally filing it in the dusty part of my tiny brain. Thanks for the refresher! I guess my next task then is to create a little serial app and test the flash. RX/TX: Here I come…
 
Good to know that someone's already done this!

I guess the remaining thing to do is painstaking work through the settings, one by one, to decode the full data set transmitted to the controller. From then on its just a matter of creating a nice, user friendly, application that allows the required settings to be entered and then sends the programming string to be sent.

Jeremy
 
Following on Jeremy’s lead,I have a schedule of tasks worked up, and a quick status to report.

EDIT: snip!

I have moved my contribution to a new thread that is more appropriately designed to follow the new development.
Link: Building the new Controller Programming App

Lyen, please accept my apologies; it was not my intention to hijack this thread :oops:
~KF
 
Please can someone here is able to edit the parameter designer to change the preset value o fthe LVC?.. We are alot of ebikes using 24s lipo and some 40s.. so the 66V LVC max value is not suitable... Can someone add LVC choice value around 75 to 85V please?

Doc
 
Doctorbass said:
Please can someone here is able to edit the parameter designer to change the preset value o fthe LVC?.. We are alot of ebikes using 24s lipo and some 40s.. so the 66V LVC max value is not suitable... Can someone add LVC choice value around 75 to 85V please?

Doc

Hey, Doc,
of the various Parameter Designer versions I have seen, the most voltage they go up to is 66.5V, so it looks like that is not enough. But, by formula, the desired LVC would be (voltage you want +2)/2, as others have mentioned. So, a 66.5V input value would really be a cutoff of around 2x this, or around 130V. I have not tested this, however. :wink:

-- Tom-D
 
Tom_D said:
Doctorbass said:
Please can someone here is able to edit the parameter designer to change the preset value o fthe LVC?.. We are alot of ebikes using 24s lipo and some 40s.. so the 66V LVC max value is not suitable... Can someone add LVC choice value around 75 to 85V please?

Doc

Hey, Doc,
of the various Parameter Designer versions I have seen, the most voltage they go up to is 66.5V, so it looks like that is not enough. But, by formula, the desired LVC would be (voltage you want +2)/2, as others have mentioned. So, a 66.5V input value would really be a cutoff of around 2x this, or around 130V. I have not tested this, however. :wink:

-- Tom-D


I dont think so because mine is set to the max availlable ( 66V) and the battery are 100V.. so if that 2X factor would really work, my system would not work ( LVC 2 x 66V = 132V and battery are 100V.. but mine still work.. so LVC is probably really at 66V.. My battery go down to 78V sometime and nothing is cuting.

Doc
 
Doctorbass said:
Tom_D said:
Doctorbass said:
Please can someone here is able to edit the parameter designer to change the preset value o fthe LVC?.. We are alot of ebikes using 24s lipo and some 40s.. so the 66V LVC max value is not suitable... Can someone add LVC choice value around 75 to 85V please?

Doc

Hey, Doc,
of the various Parameter Designer versions I have seen, the most voltage they go up to is 66.5V, so it looks like that is not enough. But, by formula, the desired LVC would be (voltage you want +2)/2, as others have mentioned. So, a 66.5V input value would really be a cutoff of around 2x this, or around 130V. I have not tested this, however. :wink:

-- Tom-D


I dont think so because mine is set to the max availlable ( 66V) and the battery are 100V.. so if that 2X factor would really work, my system would not work ( LVC 2 x 66V = 132V and battery are 100V.. but mine still work.. so LVC is probably really at 66V.. My battery go down to 78V sometime and nothing is cuting.

Doc

Thanks, good to know some real testing. For now, I will use the Cycle Analyst for setting pack LVC.
-- Tom
 
I wana test this.
 
The old way to modify your LVC was simply to put a resistor (raise it anyway) across R12 (the shunt return) which I would assume but haven't confirmed works on the EB2XX series boards from Lyen (but I'll find out soon) - I don't know how it will map but I think the 116 boards (the EB2XX series) use a 1200 ohm resistor for R12 also... someone (can't remember who) said a 3K across this (bottom side) set the max regen to 97v for his 100v 24S pack...

I will be working this out shortly also... but I am sure someone else will have an answer for the HVC options, I don't 'think hexediting the parameter designer software with a proper value would work but I could be wrong :)

-Mike
 
mwkeefer said:
The old way to modify your LVC was simply to put a resistor (raise it anyway) across R12 (the shunt return) which I would assume but haven't confirmed works on the EB2XX series boards from Lyen (but I'll find out soon) - I don't know how it will map but I think the 116 boards (the EB2XX series) use a 1200 ohm resistor for R12 also... someone (can't remember who) said a 3K across this (bottom side) set the max regen to 97v for his 100v 24S pack...

I will be working this out shortly also... but I am sure someone else will have an answer for the HVC options, I don't 'think hexediting the parameter designer software with a proper value would work but I could be wrong :)

-Mike
I think that was me I was the first on the methods 18 fet to get regen to 97 volts. What I did was run another resistor in parellel with the one that was al ready there. I since had to replace it because the oem 1200 ohm resistor failed. I now just simply run a 800 ohm resistor in the cuircit and I always have regen. My pack never ballances ovewr 100 volts and by the time you ride it you will never get regen over what you charged it to. DISCLAMER UNLESS YOU LIVE ON TOP OF A BIG HILL! lol
I am totalty stoked to get my bmx fliping more people off! I guess my 10ah 20c 24 s batteries need to be upgraded for the soon to be higher demand on them!
 
Lyen,

Is there any way we can change it so the 3 speed switch input instead changes 3 different current limit settings, or any other way to achieve at least 2 different current settings via a flip of a switch instead of plugging to a computer? To me that would be far more useful than multiple speed settings.
 
Back
Top