Has anyone played with "BLOCK TIME" on Infineon Controllers?

icecube57

10 MW
Joined
Apr 25, 2008
Messages
3,072
Location
Austell GA
The last time I was in my controller to program it i changed my block time to 10 in the programming software. No im able to peak alot higher and alot longer than I normally would. But here is the catch. I was at 60v and could barely get 50-60A on accelleration. I had peaks in the 80s sometimes. I went to 72v now i can draw 60-70A continuous and the last peak I saw was 110A. This is a 45A 12fet controller. I know we have torque and amperage junkies but im surprised no one has played around with that setting.
 
icecube57 said:
The last time I was in my controller to program it i changed my block time to 10 in the programming software. No im able to peak alot higher and alot longer than I normally would. But here is the catch. I was at 60v and could barely get 50-60A on accelleration. I had peaks in the 80s sometimes. I went to 72v now i can draw 60-70A continuous and the last peak I saw was 110A. This is a 45A 12fet controller. I know we have torque and amperage junkies but im surprised no one has played around with that setting.

You're going to kill your Thundersky's pretty quick that way.
 
Trust me its accidental. The controller is dead anyway. I have to replace 2 mosfets. But the thundeskys are rated at 2C and can pulse 10C. I run at 2-3C. Im looking to getting a 30-35A controller to take it easy on my motor and batts.
 
I've played with it extensively...

The purpose in the eyes of eBike designers is quite simple... Although some hub motors are really torquey, most just arent... to that end, you need the most torque when launching from a dead stop and so...

An engineer decides what wattage and determines the motor's torque output based on 1355 / No Load KV * Amps... Then depending on the requirements, for instance a max load weight of 300 lbs the engineer needs to find a way to get that mass (the 300lbs of you bike batteries and motor) moving along quickly... though accelleration takes more power, it only needs that power for a short time... hence Block time is supposed to be the delay in seconds before the limiting kicks in and brings you back to the programmed primary current.

Another bit you may or may not have realized is the purpose of Phase Current and why it is different than Primary or Battery Current...

At lower speeds the controller can multiply the current at the cost of voltage which is only needed (the higher voltage) once the motor is trying to gain speed.... That's why people with 9C usually recommend 2.5 X the Primary Current for this setting and it's the current your phase windings are slammed with on startup... the block time allows for a momentary surge of Primary Current to enable the output which is current amplified (the FETs do this) which in a 26" DD 9x7 Rear (Loaded kV of 10.10 @ 48v - load rating 100KG / 220lbs) - these motors don't have enough torque at 45A to get you moving and thus reduce duty cycle and power handling of the FETs, phases and every other component... then when you approach speed... you will see your current taper down because "An object in motion stays in motion unless acted upon by an outside force".

By comparison - I have found that using Phase Current of lower multipliers of Primary Current like 1.5 X will result in higher top end speeds because the voltage sag is less in the current amplification process.

If your using geared hubs... these are the most fun, here I begin at 1:1 with block time at about 5 (I do run 69A limit on a 9FET shunt soldered and reprogrammed infineon with stock fets and caps) - I can't keep the nose down and my top speed reaches 30+ mph where as when I had the Phase Current at 2.5 the max speed was 27mph... Next I increased the multiplier until my top speed on a flat without wind dropped measurably... This gives me the absolute best combo of slamming accelleration, higher top speeds and better efficiency since the motor spends less time at low inefficient speeds.

Hope this helps a bit!
-Mike
 
24. BlockTime(s)- Current limit overshoot - higher values will let you overshoot more. Current limit Damping. Values vary from 1-10

http://endless-sphere.com/forums/viewtopic.php?f=16&t=14836

GREAT INFO mwkeefer (Mike) !!! :D

Your analysis is AWESOME! Would you mind elaborating? I love learning new stuff! :D
I think you are the first to really "peg" this and your "tweaks" sound REALLY COOL!

Tao Chie!
 
Yes Mike,

That's the best post I've read in a while, and a real reason to finally connect my controllers to play with on the computer. I wonder how the Block Time is set on my 2 Methods 100v100a controllers? Now I just need to build powerful enough bikes that my appropriate block time is 3 seconds. :twisted:

Is there also a program item for the ramp up time for current? My stock 15fet Infineon board controllers are unbearably slow in their ramp up time for current. Of course maybe that's why they've proven indestructible despite shortening the shunt by more than 1/2 and overvolting to 74v from the designed 60v.

John
 
Knuckles,

Thank you - I was worried I would glaze people over and tried to keep it short and sweet. You are one of us whom at this point I consider an expert with most everything you have touched... great mods and insights and much better documentation skills than I possess. Really I have 11... sorry, just checked 18 versions of my 9 month behind schedule "Infineon Controllers Unleashed" manual I promised everyone way back... I have about 50% of the content complete in much more detail than I could find with reference links so people can follow definitions and examples out on the web relating to each part of the book... But still, its not finished. For as many questions I have answered about infineons in PM and other threads (I attribute 50% of other threads to users not spending time searching for answers first) I could have finished the entire freaking manual and since my work load from the past 2 weeks has finally died down (PM me if you want details, not for public consumption) and my new meanwells nor gary's new 4.0.8 charge controllers, my #25 drive sprockets in 3/8, 8mm and 10mm Bore and teeth from 10 to 60 for the rear sprocket, my front 20" geareds aren't here yet and... well other than some software I need to finish, I suppose it's time to crack down and restructure the book... inserting the videos and graphics of the various mods and so on.... I need to build up some controllers this weekend anyway, I'm selling 2 700C 9x7 kits and though I'm really giving them away already I know the hardest part for most people is opening the case and adding a programming port so install a weather proof phone jack in the end of the controller and will provide the 4 conductor phone cable to 7 pin 6S balance plug Male... then user can adapt to their USB-TTL adapter and will be able to flash... yep the phone jacks have rubber weather stopper when not in use.

Guess I haven't had time to look in on this thread in a while but I always see people asking about Block Time and the generic recomendation about Phase current being set at 2.5 x primary current and it's just not always the right answer and since I spent literally months testing all the abilities of the basic 846 infineon 9FET I think I have resolved 99% of how it functions internally and most of the tricks (sw) that can be done why not elaborate a bit.

Here is the real secret, the hush hush engineers don't want you to know... are you ready?

Although Block Time, Current Limits and Speed % all effect performance and if used / tuned properly can not only get you more accelleration, top end and efficiency but in many cases you can also ensure the life of your motor... but that's not the real purpose of Block Time and Phase Current multiplication......

The brutal and sad truth... a secret that retail vendors, online vendors, distributors, dealers, brokerage agents (frieght forwarders), middlemen and 98% (conservative estimate) of end consumers are going to be pissed with me for revealing the sad truth here in a public forum, or in the case of customers, dealerships (LBS), etc they will be pissed at their vendors, distributors and the manufacturers but I just don't care.

** Please note this before reading further

<-- WARNING: My mind is burnt from all the buzzing around today (300+ miles, a trip to Downtube.com retail then warehouse for a new nova folder and some spare parts, a trip to ebikekit to pickup a 9FS black folder I loaned them for the GMA appearance, converted / automated my fathers house with x10 because he is vision impaired and oh yea got him a new 3GS iPhone and constructed a complete (*Yes, 100% all screen resources/graphics) theme for not only Winterboard but also for every system screen since the iPhone 3GS 3.1.3 accessibility is messed up... sorry I digress) -->


Now you see I may be a bit off my game but I'll give it my best shot:


So have you noticed on a stock controller (just basic stock settings for DD motor 2.5x phase current 100% throttle) that you get a good bit of launch but then it piddles out within 2-3 seconds... still accellerating but not GTO or 442 slamed accelleration? Like your still accellerating just not as rapidly, it's because you aren't!

Here is what I believe happens after careful testing, data collection and just general common sense applied to collected data - where I openly may have somthing fundamental wrong I will prefix with *:

[Example - Block Time, Primary and Phase Currents]
Model: RH205
Wind: 9.7
Loaded kV: 10.10kv - Actually specified as 485 RPM @ 48v (never did get nominal or max or lvc back from 9C)
Maximum Efficiency: 82%
*** These are rough and calculate with known formulas but could be +- 5 to 10%
No Load kV: 11.28 RPM per Volt (10.10 * 1.18) - account for motor efficiency which is 82% so increase loaded kV by 1.18 to get the no load speed.
KT: 120.1241134751773049645390070922 inch ounces of Torque per Amp
ft lbs: 0.6256464243498817966903073286052 torque per Amp

1.) Lets begin with the 1:1 or equal phase and primary currents, best case scenario would be 22A - the torque would be: 13.764221335697399527186761229314 ft lbs / 2642.7304964539007092198581560284 in oz

2.) Now lets assume BT is still at 1 but Phase Current is 2.5 x Primary Current 22A or Phase Current = 55A per phase, generating: 6606.826241134751773049645390071 in oz /
34.410553339243498817966903073286 ft lbs torque - better but wait there is more...

3.) If durring #2 you were to go to 90% throttle (return of 4.5v), Block Time would kick in - Although I spent maybe 10 minutes on confirming (to my satisfaction) that the setting was the number of seconds which
you can exceed the current limit for.... I do not believe it has anything to do with how much current to allow to slip through (ie: 1=XAmps, 2=XAmps, etc) , just how long to hold off on before imposing the proper
current limit. My process behind this conclusion/observation is that when I set Block Time to 2 and did a trainer with full resistance full throttle test: The current limit was set for 32A (otherwise stock) and it
pushed right up to 60 some AMPS which was when I let go of the disc... then I tested using Block Time of 4 (Infineon will cutout in 5 seconds in stalled / semi stalled condition), same everything in programming
32A limit and all - The result... same limit to begin with but 2 extra seconds of "run away" and I peaked at 70A.

4.) Although I cant get the factory settings (won't release them) or a factory ASC file... I don't know what they normally use for BT out of the box, I stick with 4 seconds now which works perfect for me.

Now Really the reason for all this is because most of the "vendor" batts won't handle more than 20-25A...

Oh well, truth is out...

BTW: Give this a shot... open Parameter Designer in Hex Edit and modify the section where block time entries are located to give higher block times, if it's in the GUI the controller will accept it, this works for LVC and Tolerance, Speed % (like 125%, etc) and several other things you may want to tweak.

I'll get working on finishing the book and release it as an Alpha hopefully friday night saturday morning...

That said, I have a call to arms guys...

I have Infineon EB806, EB809, EB812 - 846 in 6 FET (P75N75), 9FET (P75N75) and a single 12 FET (P75N75) -

I am still looking for the following for review, documentation and testing:

EB206, EB209, EB212 - 6 FET with 116 Processors. I have none of these and would be willing to either send a 9FET in exchange (brand new), test and return or purchase... I also need 2 more 6 FET Infineon EB806 controllers also for testing as Ive just toasted my last 2 boards.

If you have any of these you want to trade, sell or loan... please PM me. I would like to include details on them in my book if possible (maybe it will have to be Volume #2)

Thanks in advance!

-Mike
 
John in CR said:
Yes Mike,

That's the best post I've read in a while, and a real reason to finally connect my controllers to play with on the computer. I wonder how the Block Time is set on my 2 Methods 100v100a controllers? Now I just need to build powerful enough bikes that my appropriate block time is 3 seconds. :twisted:

Is there also a program item for the ramp up time for current? My stock 15fet Infineon board controllers are unbearably slow in their ramp up time for current. Of course maybe that's why they've proven indestructible despite shortening the shunt by more than 1/2 and overvolting to 74v from the designed 60v.

John

John,

If I recall from your posts you run geared hubs right? I will be happy to assist you in proper configuration and walk you through the steps with your motors because I think you have a type I don't (and I have most) if its okay with you - I will include the proper configuration and why for yoru particular hub(s) once we resolve it.

Unless methy has changed his SOP or you asked for special somthing... it is set to 1 (Methods can you confirm that please?)

I don't understand how you mean powerful enough bikes that block time of 3 is appropriate, I know your smart.... I've read many many of your posts and your little ride with our buddie (where I remember the geared info from) - Could you elaborate, depending on other settings that may be appropriate or it could provide lower efficiency becasue it may not be the right target to shoot for... again I am happy to help via PM, email, skype or phone... I need to see if CR is on my vonage free list but if so, I can even call you to discuss things...

What model is stamped on the PCB of your 12 FET infineons (revision code too) and the date (backwards) - finally... what MCU (846, 886, 116, whatever the new one is I like to calll 747) are you running? I ask because unless you have the incorrect settings for phase current and speed percentage (I guess based on the question of BT you don't know your settings right now) the only other answer is slow start which totally defeats the purpose of everything but no worries we can figure it out and likely it's just not programmed right for your needs - though I hear your rides are even more awesome than mine and I still want to see the John in CR's quest for 300mph while keeping the wheels on the ground (detachable jetison wheels?)

Ahhh... I suppose I will finally officially endorse (remember I do not work for e-bikekit and though this is stable for me, it may not be for everyone) that I run (on a geared hub) a normal 9FET infineon with the P75N75 fets on 18S (75.6) and with a soldered shunt offset calibrated to limit at 69/70A - this runs great for me and has for months... I can't say everyone would have this experience but it works for me!

I hope those controllers aren't slow start but if they are... you can always offer them to someone with bafang or other weaker plastic geared motors cause teh soft start will save their motors... it reduces ramp up speed so the smashing of the gear train when the motor catches up to wheel speed doesn't cause impact. that was the idea anyway, not implemented right - I have my own implementation for geared motors which monitors wheel RPM + BLDC RPM (hall sensors) to determine if motor needs to spool up to avoid damage... it's just an addon like a CA without a UI... configure from your PC right now - all this is in prototype stage (as it it works, for me, abour 90% of the time).

I have to close... exhausted and lots of bikes to build tommorow, chargers to build, wheels to build / respoke, etc... I will get to the book and publish it even if its just what I have so far as my explaintions include observational data, parameter logging and video / pictorial how to for modding up Infineons...

I need to mention, Thud has just done a very keen job as he always does!

-Mike
 
I thought it was pretty obvious that stock controller settings were optomized for less warranty returns on batteries and controllers. If you had a good battery, wouldn't it be a bit harsh on these cheap kit controllers to pull those kind of amps for longer? Or is there enough headspace built in for that, as there is in the motors themselves.
 
Mike,

I run 22lb direct drive hubs used on some small Chinese motorcycles and scooters. They come stock with a 15fet Infineon that is called 60v30a. I've be running 74v nominal (81v off the charger) and I shorted the shunt wire by over half. No issues in over a year of 15-20 miles/day avg, 115lb bike + 250lb me + often 100lbs of kids. I typically cruise at 35-40mph, but there are some good stretches of smooth road that I regularly go 50, and I've had it up to a wind assisted 60mph. There's no delay in start like I've heard Methods describe soft start, but it doesn't hit hard, and feels to me like a long current ramp up time that I'm familiar with on Curtis controllers.

If we could somehow use a switch to make controllers do like this one for normal use of for slippery conditions, but switch to "power mode" when we want performance, that would be great. My high Kv motors seem hard to drive, because they got my Methods controller too hot to touch for more than a few seconds.

I'm flush with controllers, so if you want me to send you one of these stock 15fet Infineons, just pm me your address. If their performance can be unlocked a bit, this motor/controller combo is the cheapest and most durable route to high performance.

John
 
Thanks for the very interesting insights Mike. Not sure what the original BT setting was on my Lyen eb212 but I've always set it at 1 since I've programmed it, so I'll try some other values now I know what it's affect is. Thanks.
Rob
 
dogman said:
I thought it was pretty obvious that stock controller settings were optomized for less warranty returns on batteries and controllers. If you had a good battery, wouldn't it be a bit harsh on these cheap kit controllers to pull those kind of amps for longer? Or is there enough headspace built in for that, as there is in the motors themselves.

I think they're just tuned to not burn up motors. Maybe some attempt at optimization is attempted for specific motor/controller combos, but I doubt it. A good example is the big hubs I have. They're run in wheels ranging from 10" scooter rims to 17" motorcycle rims with controllers that are set up identically, which tells me there's been no real optimization. My guess is that it's more like whoever orders the controllers only supplies input about options desired like current voltage limits, along with stuff like regen and how the brake cutoff works.

It sounds like the stuff Mike is doing could lead us to some real matching of these controllers to the motors we run with them, with far better results than stock. Kudos to Mike for the effort he putting in and so generously sharing with us.

John
 
dogman said:
If you had a good battery, wouldn't it be a bit harsh on these cheap kit controllers to pull those kind of amps for longer?
I wouldn't call the XieChang controllers cheap. Inexpensive is a better term.
I use original IR mosfets in all my controllers. In the end that is the heart of PWM anyway.
Just a few mods and these units perform excellent and can last forever.

AND now that Keywin has truly mastered fabrication (better than XieChang corporate IMHO) ...
Well these are darn good HIGH QUALITY lil bitches! 8)

btw ... over 100V is the way to go. Man I love the performance ... like night and day.

mwkeefer said:
I have about 50% of the content complete in much more detail than I could find with reference links so people can follow definitions and examples out on the web relating to each part of the book... But still, its not finished. ... If you have any of these you want to trade, sell or loan... please PM me. I would like to include details on them in my book if possible (maybe it will have to be Volume #2)
Thanks in advance! -Mike
Mike,

Very cool. But don't sweat the "book". Technology changes so fast.
I am just lucky as Keywin is my "bud" and we "chill" on skype and stuff. :lol:
I get all the best new shit.

Of course I am happy to share. Sharing is Caring! :roll:
I actually do all this for fun. I think it's cool.

I have the latest stuff coming in soon. Same with Steveo.
You need "stuff"? No worries. At Cost sound OK?
Sell off all your old "stuff" on ebay or in the ES market.

Cheers,
Knuckles
 
So you confirm that setting phase current to 1.5X will increase top speed?

Doc
 
dogman said:
I thought it was pretty obvious that stock controller settings were optomized for less warranty returns on batteries and controllers. If you had a good battery, wouldn't it be a bit harsh on these cheap kit controllers to pull those kind of amps for longer? Or is there enough headspace built in for that, as there is in the motors themselves.

In a way I agree, they are Optimized for the cheaper (assume 25A maximum @ 60 seconds rate) SLA, LiFePo, etc - You could say its to limit warranty returns on the batteries but personally I think were saying the same thing from opposite ends of the stick - the design is to protect those low discharge capable batteries from damage.

I must disagree with regard to the controller damages and returns, considering with nothing more than a software flash I can repurpose a 9FET infineon for use at 18S (75v) and 38-40A Primary and 75A Phase Current and that reprogrammed controller will work forever (mine ran all last summer and through the winter and those blizards)... since the fets at WOT see a maximum 66% duty cycle and even the P75NF75 FETs go full on/full off pretty quick (not as fast as IRFB4110s) I don't believe the controller is limited for it's own well being, again it's for the low end BMSes and Batteries... the controller LVC is for it's protection, below that and there is not sufficient voltage on the Gate rail to effect switching and you get a short and a blown out fet.

To answer directly with regard to a decent or high end battery... I'm not Luke (live for physics) or Methods but... I would consider my 15 and 18S 2P pack to be a "GOOD" battery - capable of 200A continuous and 300A for 60 seconds... The infineon I use with my steel geared hub motor on that is only modified with a soldered shunt (not even calibrated), then I set the lowest current primary and phase limit and held the brake and got 40A... I adjusted this until I saw 69A which I think was programmed setting of 14A and I use that for both primary and phase which keeps my voltage nice and stiff at the motor since there is less multiplication... in either case, I run both the 15 and 18S packs on this controller and other than the common sense step of artic silver paste when I reassembled after soldering the shunt and adding the programming port - they are fully stock, caps fets and all... This has been running for about 5 months now and never gets warm... I didn't even beef up the traces on the FETs.

So I can confirm these controllers (all of the infineons I've experienced anyway) have more than ample headroom...

If your going to do with a DD hub motor, the phase lines are the limit really... with a geared the current tapers to nominal very quickly... a direct drive hub will require more sustained current, now add the multiplication factor to it and imagine your bursting 150A per phase from dead stop (2.5x recommended setting) - this is okay because it's only on launch but where your DD would consume 60A ... a geared nominal for the same speed would be less than 1/2 of that... I am assuming a 60A primary current limit and the 2.5x phase current.

I hope that helps clarify rather than make it more complicated and again this is all imho I am not a real engineer = )_

-Mike
 
Knuckles,

Book may be a misnomer, more like the Infineon Technical Information.doc but about 10x the content... I know it all changes so it's not the specific settings I'm trying to explain but rather the principals behind their workings... to me this stuff is like windows, there are 1000 ways to accomplish any given task... there are also classes available for each new version of each new application but I never took one, I find that understanding the inner pinnings and functioning (GDI, User, Kernel, etc) gives you the ability to adapt and still understand even new sw/hw because you understand the concept and functional implementation... in a nutshell it is the essence of hacking right, understanding?

You have provided a great wealth of information to the forums, I am finding though that people regurgitate facts but they don't understand the underlying mechanism / implementation... those are the people who need to go to a new class for each version... I am a firm believer in personal experience being the best teacher and I hope I'm not being too convoluted with this, but I guess it's more of teach a man to fish... thats my intent anyway and until someone comes up with perpetual motion or antigrav I think the method of 3phase commutation won't change much and as I am certain you know... all these little "tweaks" or settings are just logical extensions of the base commutation controller... right?

Having a skypemance with Keywin must be nice = ) (had to sorry) and I won't claim I haven't made a few bucks nor can I deny that eventually and gods willing I will offer the steel geared hubs I've been working on for months now to the public and yes I want to make a profit... however, normally (and you can ask anyone) I do this for fun... heck even half the dealers / vendors / manufacturers I provide some engineering support to, I don't charge them most of the time (again you can ask)... I believe we (and most of us here) are alike in this way... Personally if I can help other manufacturers, dealers and vendors deliver a safer and longer life product... and if they are open to constructive suggestion (not the type who know nothing but think they know everything) it only helps our community and the experience even novices or non technical people get from ebikes.

Thank you for the offer of sharing... I will very likely take you up on it but most of my old stuff ends up just given away or Repurposed... try this sometime, an old motor your not going to use.. a normal powered controller and a rebuild battery pack... find a thrift store bike and just give it away to someone. Ive done this about 4 times now, 3 were over 65 and now can enjoy riding on a nice day again like they used to when they were still a young couple... one I gave to a fellow who had the misfortune of a bad reaction to his medication combined with sugar shock (diabetic) and since he was loopy the cops did a blood test and he just happened to have smoked a joint about a week before... instant DUI even though he wasn't driving (the keys were in the ignition) and he wasn't drinking or under the effects of any illegal drugs... I don't really need to sell off stuff for $$$, not that I'm rich but my "profession" lets me work 4-10 months a year and earn enough to live for 2 years if I am careful...



In either case cost is very generous but like I tell everyone, your time is worth money so make somthing on it... lets take the rest of this conversation PM or email.

-Mike
 
John in CR said:
Mike,

I run 22lb direct drive hubs used on some small Chinese motorcycles and scooters. They come stock with a 15fet Infineon that is called 60v30a. I've be running 74v nominal (81v off the charger) and I shorted the shunt wire by over half. No issues in over a year of 15-20 miles/day avg, 115lb bike + 250lb me + often 100lbs of kids. I typically cruise at 35-40mph, but there are some good stretches of smooth road that I regularly go 50, and I've had it up to a wind assisted 60mph. There's no delay in start like I've heard Methods describe soft start, but it doesn't hit hard, and feels to me like a long current ramp up time that I'm familiar with on Curtis controllers.

If we could somehow use a switch to make controllers do like this one for normal use of for slippery conditions, but switch to "power mode" when we want performance, that would be great. My high Kv motors seem hard to drive, because they got my Methods controller too hot to touch for more than a few seconds.

Surprise, the basic switch is there already! Just reprogram with Speed %3 at perhaps 20%... that doesn't limit your max speed like the Speed Limit loop % does... it limits your current which would provide you a ramp up then flip to S2 for normal riding and even go to S3 if you want and reprog it to 115/120%

In reality... I have a better solution for wet conditions, electronic traction control... it's a hold over from my initial pre-controller ideas months back, but it could be done in analog too fairly simply. In essence all that's needed is RPM monitor of both wheels and the CURRENT flow... then its a simple switch case/select case logic tree, if they are within 2-5% of matching speed... all is right with the world, if the drive wheel is > 2-5% check current, has it dropped ? If so slippage is beginning and a variable shunt regulator can then dial back the throttle return line just enough until the condition ceases... poof, real world traction control and it really does work - I was out in the 26" blizard on a FWD 9C 9x7, passing plow trucks without chains!

Before you ask... I don't have it available for sale, I cant even offer artwork right now but it's not a whole lot more than an Arduino Pro Mini 5v, 2 hall sensors for RPM and a hall based current sensor salvaged from a fried EagleTree logger... If there is enough interest maybe I will work on an analog circuit to accomplish the same thing, digital is the way to go though since wheel size changes with tire type and so on... digital can be reconfigured. Problem is I have no time right now to do this but if someone else wants to undertake it, I would be more than happy to assist, code review, testing, etc...

John in CR said:
I'm flush with controllers, so if you want me to send you one of these stock 15fet Infineons, just pm me your address. If their performance can be unlocked a bit, this motor/controller combo is the cheapest and most durable route to high performance.
John

John I appreciate the offer but if it's a 15 fet 846 EB815 it will be the same (from a core perspective) as EB809 and there is nothing to be gained shipping one to me... we should PM chat or email about this, if it's somthing I've not seen yet I may take you up on the offer (I will pay for s&h both ways too) just to document the design... Depending on who really built it (its getting harder and harder to tell) 15 fets indicates 5 fets per phase so I would assume 3 high and 2 low side per phase... if the fets were the P75NF75 it would probably be safe to push upwards of 140A Primary Current Peaks and a Phase Current of 210+ Amps - sadly that would FRY or weld your windings...

What MCU (processor) does yours use, what is the board marked EB8, EB2? whats the date on the board and finally which FETs is it equipped with? I guarantee if the PCB is in good shape you can get amazing performance (too much) out of one = )_

-Mike
 
icecube57 said:
The last time I was in my controller to program it i changed my block time to 10 in the programming software. No im able to peak alot higher and alot longer than I normally would. But here is the catch. I was at 60v and could barely get 50-60A on accelleration. I had peaks in the 80s sometimes. I went to 72v now i can draw 60-70A continuous and the last peak I saw was 110A. This is a 45A 12fet controller. I know we have torque and amperage junkies but im surprised no one has played around with that setting.

Can't believe I didn't answer your post but instead your topic...

When you went to 72v the maximum RPM of the motor went up - assuming were talking a rear hub 9x7 (good enough for example) with a kV of 10.10 loaded...

At 60v your maximum loaded speed on a 26" wheel would be:

60 * 10.10 = 606 RPM / 336 = 1.8035714285714285714285714285714 * 26" = 46.89 mph maximum *theory - which would also be the most efficient speed again in theory
I assume when you say 72 you mean nominal? so we need to assume 83v hot
83 * 10.10 = 838.3 RPM / 336 = 2.4949404761904761904761904761905 * 26" = 64.868452380952380952380952380952 mph max in theory

Even if I have the wrong motor, the example still works...

What is happening, when the motor is run at higher voltage with a higher Potential top speed it will consume more current for longer to get to that speed... your not reaching the max efficiency or top speed for your voltage and so the motor is pulling huge currents trying to reach that nominal state... With the lower voltage, the top speed possible is lower and your getting closer to it faster so you use less current even on launch and for less time before approaching that nominal "sweet spot" where your at speed but using far less current... with the higher voltage / potential speed you are just never reaching the maximum efficiency and thus you consume far more on launch and continue to draw more current while the motor tries like hell to get up to speed.

Actually when I look at your numbers, think about it... higher voltage and speed require more current/torque to maintain and offset wind resistance... below 15mph you fight rolling resistance, above you fight wind resistance = )_

I hope that answers your query more accurately.

-Mike
 
Well I like the fact that the Block Time does allow me to overshoot on current but yes it indeed does dump a crap load of current in the process. Bad for the batts I have mine at 10. I will recuce to 3 or 5 when i open my controller up to change mosfets. I think I will program one of my speed switches around 60-70% to simulate my bike at 60v. It wont truely limit my current draw but it will limit the speed to keep the motor in the butter zone efficency in the high 30s. I know that the motor was solid and had no issues at 60v but at 72v 5 minutes out of the racing gates im smoking. Literally. Your totals were surprising pretty close. My no load is 66mph at 79v. and 56mph and 66v.
 
I've put together my own little reference booklet, solely for my own use, on the various features of these controllers, some of it gleaned from various informed posts on here, some based on things I've found out when building high power 6 FET controllers up from Keywin supplied boards and some that's come from the original XieChang presentations. It would be a great help if there was a way to have a single, locked, repository of all this information somewhere, rather than have it buried in threads all over the place. The major problem that I found when trying to get hold of information was that just about every Infineon/XieChang thread here has lots of misinformation and deviation from the core topic, so it's quite hard work trying to sift out the stuff that is valid.

One thing worth noting about the "block time" setting is that I believe that it disables the secondary peak current limiting function for the duration of the set time, as well as disabling the primary, programmable, current limit. This secondary current limit circuit doesn't have a programmable current, it's set in hardware by fixed component values. As gwhy and I discovered, there are ways to change the settings for this limit by changing component values, but it's a bit dodgy, as it's potentially fairly easy to get peak currents at startup that exceed the FET limits, even allowing for the relatively low duty cycle and modest pulse lengths.

Jeremy
 
Any relationship between the 115% speed setting performance( phase advance?) and block time and phase current multiplier ??

BTW.. the 115% speed dont really affect my 5302 on 20" wheel.. BUT REALLY POSITIVELY AFFECT MY 5305 TOP END !!! !!

Doc
 
Jeremy,

You are right on with your assessment of the secondary being disabled for a time with Block Time and since you teased at it, I might as well hit the nail... in software you can get around this secondary limit (or well increase it considerably) by programming for a different board model (IE: 812 for an 806 and so on).

Now I'll mention the caveots, the programming parameters must be adjusted in a fashion similar to when you solder a shunt for more current or indeed the FETs will be TOAST! For everyone who hasn't done this, there are some posts here on ES about it (search) but... it is dangerous (to your controller and possibly your batteries / bms and motor) and YOU DO IT AT YOUR OWN RISK!!!

I agree also on the need for a locked resource... half of what is out there is bad or misleading or at best difficult to understand... that was my fear in explaining the innerds of BT, Primary and Phase current limiting (and why I hadn't mentioned the secondary limiter, most people have difficult enough time understanding the concept of just one)... and it seems now that you, me, methods, Knuckles and a few others have our own "books" written up for our own reference... though I am sure we have much overlap I am also certain we have unique information, ideas or tricks to contribute.

To this end, I propose we form a google docs or source forge project (source forge would be better as I have schematics and firmware to go with some of my enhancements) and I think any of us who are interested and willing to contribute our time and knowledge should form a comitte and be admins on this project... others can join and ask but it would in essence be locked.

We could do this here on ES I suppose but not enough of us are admins and I for one don't actually want to be an ES admin, too much political BS to deal with...

If you (all of you) think a wiki with locked editor rights would be a better method for posting a definitive collection of information.... I would be happy to host it on my cluster... Heck, I can set it up on e-bikemike.com (check for yourself, no pages, no adverts just a file storage area for me) for now but I am more than willing to pay for and register a more appropriate name (as in not my name) that is agreed upon by all.

Even if others don't want to contribute (Jermey I'd love to peek at your "book") I am still going to do this and will include as much valid tech data as possible.

Jeremy - you have perhaps one of the most useful collections of info given your full rewind (multiple right) of several hobby motors and build up on several infineons plus you run on 6S or 5S right - those low voltage mods are awesome.

Doc,

The 115 setting only effects your 5305 top end? I must admit I am not too up on Xtalyte gear... the only motor I ever had I blew in days (my own fault, nothing to do with the motor) on my test bench... could you elaborate a bit on the effect noticed on the 20" vs the 05?

-Mike
 
on my current setup i had to use the 15fet setting to up my amps as i was only seeing 32amps on my turnigy meter even though the settings were maxed for the 18fet then using the 15fet setting i now see 50amps
i am looking forward to your "book" mike
 
wasp said:
on my current setup i had to use the 15fet setting to up my amps as i was only seeing 32amps on my turnigy meter even though the settings were maxed for the 18fet then using the 15fet setting i now see 50amps
i am looking forward to your "book" mike


Let me ask... on the old setup with 15fet settings and 32A on the turnigy... what was the controller (FETS), what were the primary and phase settings you programmed and finally... when you first started off, did you feel a kind of "knock" or jolt shortly after beginning to ride with those settings?

The reason I ask... what you are describing is the controller hitting it's secondary limiter which I normally just call it's internal protection circuit for simplicities sake... you can confirm this by going back to the old setup but begin with primary current set at 30A and Phase at 45A... it should work and the watt meter should display properly... then add 5 amps to primary and 10 to phase and test again... keep testing until you get the results of 32A as before and then roll back a bit....

This will get you to the maximum (without soldering the shunt or my preference adding a second shunt in parallel with the first) your controller can use without reverting to the internal defaults which I think is what happened...

Before I began the double shunt or soldered shunt mods... I used to have a similar problem with a 9FET... I would program for 40A but when I saw 38 would feel an odd hiccup and then the meter would show 30sh (actually I think it reverted to 24A)... if I set the limit for 38 it never happened and when I soldered the shunt and recalculated my current limits... well, 69A and no reversion to the lower default.

Hope it helps!

-Mike
 
Thanks Mwkeeer. I will be reading this alot more later and building a better cooling system for my x5! I have my block timer on 1 in my methods 18 fet!
 
Back
Top