XPD: open-source keywin e-bike lab replacement

Get all your technical information about electric bikes here.
User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Max regen voltage ?

Post by NeilP » Apr 21 2013 1:39pm

Just been setting up a 6 FET today and wanted to put the regen voltage up to 84 volts, but the max is either 77.7 or 80 volts.

I can't work out yet why some of the saved profiles I have allow 77.7 and one allows 80.1

Anyone got any clues ?
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

User avatar
Kingfish   10 GW

10 GW
Posts: 4064
Joined: Feb 03 2010 11:23am
Location: Redmond, WA-USA, Earth, Sol, Orion–Cygnus Arm, Milky Way. Age: > yesterday < tomorrow
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by Kingfish » Apr 21 2013 1:59pm

If your controller is the type that can be modified, I used the "R12" mod to gain range. This is well-documented with a small search :)

Here you go: R12 mod for high voltage regen

Best o' luck, KF
* My 2WD Garden Wall
* Kinaye MotorSports
* Primary ride: 2WD Disc 9C 2806-equiv / Dual Lyen 12FET / 20S7P LiPo.
* Epics: Going to California: 2011 8)
* 50-mph, 101, 10k-Club. 12,527 miles-to-date, 7037 as 2WD.

It is by caffeine alone I set my mind in motion.
It is by the beans of Java that thoughts acquire speed.
The hands acquire shakes, the shakes become a warning.
It is by caffeine alone I set my mind in motion.

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » Apr 21 2013 2:35pm

Sorry, I should have been more clear..Yes, I know about the R12 mod..done my 18 FET...just not thinking to mention it..I was more interested why one file I have allows 80.1 as its setting max, where the rest allow 77.7

But I now see the reason...EB3xx board setting allows 80.1, EB2xx board settings allows only 77.7

I just bought 4 6 mosfet controllers, and need to mod them anyway..so will add the R12 to the list

Lyen is now adding a second shunt to the 6 MOSFET units..and then you set them up with the board type as a 12 EFT rather than a 6 FET...he did give the reason as
ED LYEN wrote:When the resistance on the shunt reduced, which help the FETs output higher current at the same power level.
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

User avatar
methods   10 GW

10 GW
Posts: 4506
Joined: Aug 08 2008 12:08pm
Location: Santa Cruz CA
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by methods » Apr 24 2013 12:57pm

This is a Methods Technology 24 Fet controller specific version - folks have been asking for it. There are only two files: Setup and Full Power (what else would you need :o )

Methods Controller Specific Programming Version

Reference the dummy guide if you get stuck: super simple guide

Setup is what you want to load the controller with any time you are screwing around... even if you short the outputs everything should be fine.
Full Power is what you should load if you want to smash your face.

Please remember:

The entered DC current is 1/2 the programmed DC current (i.e. 75A means 150A)
The entered AC current is 1/2 the programmed AC current (i.e. 150A means 300A)
The entered voltages are a ratio resulting in a max regen voltage of ~99V, so if in doubt leave the voltage settings alone
You must NEVER, EVER, EVER EVAR.... change the current overshoot window to anything other than 0 seconds! On this controller it will only result in a completely un-ride-able bike and if you make the tiniest mistake it will result in the controller eating itself alive. (not joking....)

All you have to do is get the software working
Plug in the programming cable
Tell the software to program
Then *hold down* the button on the programmer (it is NO and when you hold it down it powers the chip and flashes - so keep it held down for a second)
Programming should happen in less than a second
You do not want the controller powered while you do this - just turn it off if it is on the bike.

Be smart - keep the rear wheel off the ground the first time you turn it on after any firmware changes.... 100V 150A is not a forgiving setting....

-methods
Increasing battery voltage and controller current limit will result in a non linear experience

User avatar
methods   10 GW

10 GW
Posts: 4506
Joined: Aug 08 2008 12:08pm
Location: Santa Cruz CA
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by methods » Apr 24 2013 2:26pm

NeilP wrote:
ED LYEN wrote:When the resistance on the shunt reduced, which help the FETs output higher current at the same power level.
What he is saying is that with only one shunt wire (or however were in the original design) the shunt can not handle the power load. Power across the shunt is equal to:

Pshunt = I^2*Rshunt

So if the calculations works out to some power level even near where the shunt wires are rated they will heat up... as they heat up the resistance will go up... as the resistance goes up the controller will read a higher and higher incorrect current reading... so for instance when it is putting out 20A it will think it is putting out 25A...

Vshunt = I * (Rshunt + Rtemprise)

So adding the second shunt will improve reliability, improve CA AH calculations, improve power output (i.e keep it steady instead of hot and heavy off the line and then soggy once hot), etc.

It is a good move.... and it will make modding the controller a lot easier :mrgreen: No more need to solder the shunt - just program as 6 fet and turn *down* the current in software.

We do this with all our controllers by doubling the shunt power handling ability (8W on the MethTek controller)

-methods
Increasing battery voltage and controller current limit will result in a non linear experience

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » Apr 24 2013 2:36pm

Ah yes, was not considering the shunt wire heating up, makes sense.
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

Ebikeman   1 mW

1 mW
Posts: 16
Joined: Jun 18 2010 7:40pm

Re: XPD: open-source keywin e-bike lab replacement

Post by Ebikeman » May 01 2013 3:51am

I am confused as to what values to program an EB309 Cellman IRFB4110 for Battery current and Phase current.
Last year I read that the EB312 profile is used and setting the current value twice smaller than desired. That was with an earlier version of XPD; I am using the latest version of XPD 0.3.2 which has profiles for the EB309/Cellman.

So do I now enter the actual values I require?

e.g.
Battery current limit 25.5A
Phase current limit 51.4A

The shunt value for my controller is marked as 2.17mohm.

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » May 01 2013 4:11am

As far as I am aware, the profiles are based on the number of shunts fitted, so sounds like that at 2milli ohm you have two shunts, same as a 312 board. I am guessing your controller is modified with an extra shunt.

I think the profiles should maybe renamed to '1 shunt' , '2 shunts' etc
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

Ebikeman   1 mW

1 mW
Posts: 16
Joined: Jun 18 2010 7:40pm

Re: XPD: open-source keywin e-bike lab replacement

Post by Ebikeman » May 01 2013 4:55am

Ok NeilP for your thoughts, but that does not help me as to what values to enter now. Do I enter the required ACTUAL value or half what I require.

And what is the significance of the three different Infineon 3 profiles; in particular the one with a description 12fet_32s compared to the one with description 12 FET for BMC V2.

I am using the GNG Brushless motor 450W 48v on a 39v battery.

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » May 01 2013 8:29am

I did not give any figures, because I do not know the answer , but it is not difficult to work it out. The only way you are going to really find out is to try it and see.

Do you know what you entered with the older version of XPD? and do you still have the paperwork that Paul sent you with the controller? or an e-mail of the settings? I know he sends out if you ask, the settings you need, as screen shots of the settings you should use.

All you can really do if you want to use the latest XPD is try some really low values of half the value you want, then go and ride and see what results you get. If you have a Cycle analyst then you will see the max current values that result. if no CA then you will just have to use the seat of your pants feel of the bike after adjustments compared to how it was before.


I did not have any Profiles named Cellman in the version of 0.3.2 that I downloaded, so i can't try them and see what they produce, sorry.

But this does bring up and important point as to what the different profiles are. Is it as I suggest merely a shunt issue or is it more deep than that?
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

User avatar
gwhy!   1 MW

1 MW
Posts: 1651
Joined: Jun 08 2011 3:47pm
Location: Bristol, Uk

Re: XPD: open-source keywin e-bike lab replacement

Post by gwhy! » May 01 2013 10:23am

the lyen 6 fet controllers may well have 2 shunts for the reasons given above and makes perfect sense but this is also addressing the controllers cutting out problem that some see due to the hardwired phase current limit.

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » May 01 2013 12:23pm

gwhy! wrote: to the hardwired phase current limit.
???
Are you saying the EB3 boards are fixed max phase current?
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

User avatar
methods   10 GW

10 GW
Posts: 4506
Joined: Aug 08 2008 12:08pm
Location: Santa Cruz CA
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by methods » May 01 2013 12:55pm

Ebikeman wrote:Ok NeilP for your thoughts, but that does not help me as to what values to enter now. Do I enter the.....
Anyone who gives you a direct and concrete answer does not know what they are talking about :D Listen bro... these board blanks come from a dozen factories and they are populated by 100 different random Chinese guys and then sold to 200 more guys like me who may or may not change this or that. There is no answer to your question - and even if there were a correct answer it would be tentative (at best) and subject to change at any time.

So... Set your current limit to 20A DC
Set your CA up with the correct shunt info (Make sure it is CORRECT!)
Go load the bike down on a steep hill

Did you see 10A? 20A 40?

This will give you the current multiplication factor.

Now - set the LVC to something like 30V
Hook up your controller to a variable supply (or hook in and out packs as best you can to vary voltage)
Keep turning down the voltage until the controller cuts out

What voltage was it? 30V? some other value?
You can back-calculate any R12 mods by the difference between these two values... or you can just open the controller up and directly measure the R12 value.

This is super DIY territory - dont let the professionalism of this programming software lull you into thinking that this is some how "explored territory". Since they came on the scene several years back the Infioneon type controllers have been in a constant state of flux - changing monthly - and then changing even more as feedback from customers, guys like me, guys like cell man, guys like Lyen, guys like Keywin... feed back what needs changing etc.

Experimentation and successive approximation FTW... and remember... confirm your shunt value with a constant current source if you can. Just because it is written on the controller does not mean that it is correct (coming from a guy who manufactures controllers and writes shunt values on cases)

-methods
Increasing battery voltage and controller current limit will result in a non linear experience

User avatar
gwhy!   1 MW

1 MW
Posts: 1651
Joined: Jun 08 2011 3:47pm
Location: Bristol, Uk

Re: XPD: open-source keywin e-bike lab replacement

Post by gwhy! » May 01 2013 2:30pm

NeilP wrote:
gwhy! wrote: to the hardwired phase current limit.
???
Are you saying the EB3 boards are fixed max phase current?
no, they are not fixed and the phase current can still be adjusted but there is a safety cut out built in to these controllers so if the phase current do not decrease in x amount of time the controller will cut out, the 12 fets 'x' amount of time is longer than the 6 fet .. more details about this problem can be found in this thread http://endless-sphere.com/forums/viewto ... =2&t=16910

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » May 01 2013 2:47pm

That will explain what I was seeing with the Delta terminated motor then....cutting out

Zombiese mentions the 'unusual' phase current limiting (and regen ) in EB3xx boards, in this thread http://endless-sphere.com/forums/viewto ... =2&t=49586
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » Jun 01 2013 1:41am

When you change a profile in XPD, what settings does that change?

Is it only scaling of battery current / current through the shunt or are there effects?
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

User avatar
methods   10 GW

10 GW
Posts: 4506
Joined: Aug 08 2008 12:08pm
Location: Santa Cruz CA
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by methods » Jun 01 2013 2:40pm

Every time I get an email letting me know there was a post to this thread I feel inclined to come back and once again say....

This is kick-ass software and we all owe a great debt to the originator for taking the time to make such a useful tool. One of the most honorable things a person can do with their life is make tools for other people to use. Teaching kids... and making tools... the two things I respect the most.

If you are reading this thread for the first time - use this software. It kicks ass.

Later - if you can - find a way to repay the favor (by making tools for others)

-methods
Increasing battery voltage and controller current limit will result in a non linear experience

User avatar
dan974   1 kW

1 kW
Posts: 414
Joined: Jun 12 2012 1:36am
Location: Reunion Island

Re: XPD: open-source keywin e-bike lab replacement

Post by dan974 » Jun 02 2013 11:07am

Hi.is there any chance that the soft work with Kelly controllers or does it only support the infineon protocol ?

User avatar
methods   10 GW

10 GW
Posts: 4506
Joined: Aug 08 2008 12:08pm
Location: Santa Cruz CA
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by methods » Jun 02 2013 1:17pm

dan974 wrote:Hi.is there any chance that the soft work with Kelly controllers or does it only support the infineon protocol ?
No chance.
FYI Kelly will freely share the programming and flash software you need. Be careful flashing - you can brick. Stick to the programming software - it is pretty powerful. Remember if running thrust throttle that it is scaled to your peak phase output... so if you have resolution issues try lowering your phase current.

-methods
Increasing battery voltage and controller current limit will result in a non linear experience

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » Jun 04 2013 2:07am

No one know?

NeilP wrote:When you change a profile in XPD, what settings does that change?

Is it only scaling of battery current / current through the shunt or are there effects?
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

User avatar
methods   10 GW

10 GW
Posts: 4506
Joined: Aug 08 2008 12:08pm
Location: Santa Cruz CA
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by methods » Jun 04 2013 2:30am

I remember that being answered earlier - I thought it also dealt with R12 - have to read the whole thread I suppose.

-methods
Increasing battery voltage and controller current limit will result in a non linear experience

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » Jun 04 2013 4:05am

Umm yes, i too remembered an R12 section to this thread, but I can't remember it answering it in a way I easily understood. There is a lot on page one...but it is making me glaze over a bit. I may have to sit down in a quiet room and just try and get my head around it.
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

User avatar
methods   10 GW

10 GW
Posts: 4506
Joined: Aug 08 2008 12:08pm
Location: Santa Cruz CA
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by methods » Jun 04 2013 12:00pm

Lol - tell me how it works once you figure it out :mrgreen:
I just figured out enough to make my own profiles. All the voltages and currents are still screwed - I really would like to do what needs doing to make the current values and voltage values correct. Just too low on my priority list.

Speaking of Priority lists... is it just me or does the list just get longer and longer as you get older?

I have like 32 different projects going on in at least 8 different fields. Wives, and kids, and chicken coups, and jeeps to repair and houses to finish building and a business to run and wells to fix and ebikes to build and languages to learn and forums to surf and software to write and relationships to maintain and a shop to put a roof on and... and.... phew.

-methods
Increasing battery voltage and controller current limit will result in a non linear experience

User avatar
NeilP   10 GW

10 GW
Posts: 4967
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: XPD: open-source keywin e-bike lab replacement

Post by NeilP » Jun 04 2013 1:13pm

yep the list gets longer. Tractors, Harley, more and more e-bike projects.. cameras, learning new things
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

iovaykind   10 kW

10 kW
Posts: 572
Joined: May 16 2011 5:02pm
Location: United States

Re: XPD: open-source keywin e-bike lab replacement

Post by iovaykind » Jul 13 2013 7:05am

Has there been any development in regen settings other than 0(none), 4(weak), 8(strong), or 255(unlimited)?

Desperately looking for something between 0 and 4..

Post Reply