Instant Start 18 fet Infineon Boards are here...

[methods]

Dont worry - those are just peaks.
The current you need to worry about is the one you see about 1 second into a hard acceleration.
Probably be around 100A

My new shunt is set to 0.2200 and it is a real bastard!
I can actually feel the bike hitting the phase current limit... about 3 seconds into a hard acceleration I get a second boost of power like turbo lag.
At first I thought there was some sort of problem then I realized that it really is the phase current.
I actually kind of like it... If I time it right I can come up on someone at full throttle and just as I am passing them I get a boost of acceleration

But... My motor overheats quickly with those sorts of stunts.
Really wish I had a 20" wheel!

Get us some videos!

-methods

 

[Arlo1]

OK so I come up with a simple way to calibrate the shunt. I just simply used my multi meter on 10 am scale hooked it in series with the main ground then slowly and carefully run the wheel up free spining to top speed (not letting it pull more then 10 amps because my multi meter is only rated to that) and held it wide open and watched the ah on the CA and on the meter and compared the two then I figured out the percent difference and re programed the CA then tested it again. It took me 3 tries and about 2 minutes and I ended up at 3110 on the hi mode in the CA Now I can watch my amps on the go and so far with the front still on the ground under acceleration I hit ~120 amps and 3.6 amps at wide open free spin and a peak of 190!

 

[Arlo1]

OK so I hooked up my ebrake switch and it will cut the throtle and the CA gets brighter when I pull it, but it wont slow the bike down.
As well where do you buy a good ebrake? I build one with a micro switch but its not as good as I want.
I belive I have BK to gnd jumped then regen to grond through the switch. And brake level 2 in the software.

 

[Mike1]

Arlo1, have you done the R12 mod to increase regen voltage? If not then you will only get a braking effect if your pack is below 75V.

 

[Arlo1]

Arlo1, have you done the R12 mod to increase regen voltage? If not then you will only get a braking effect if your pack is below 75V.

OK no I have not yeat I did plan to but I was not sure on that. So is r12 the tiny little flat resistor just over from the voltage regulator where it is labled with two others and a line pointing to them?

 

[methods]

You need to do the regen mod and solder BK to ground to enable regen.
go to my new thread - it is explained... but I recommend against regen.
Even if you do the mod I suggest it is only good to 90V - that means you wont have regen till the second half of your pack
DO NOT put regen up to 99V unless you are prepared to replace those fets if something goes wrong.
Uncharted territory.


As far as your current numbers - that number you came up with for your shunt is high I think.
Are you saying that with your ammeter inline you read exactly 10A on the CA and on your meter?
It is bouncing all over hell right?
There is an averaging function in the CA so you need to hold the current steady for a few seconds to get a true reading.

Your freewheel current is too high and your peak amps are too low.
Your freewheel current should settle to less than 2.5A otherwise something is wrong - either with the motor wiring or the shunt value.
I suspect the shunt value since you are getting good performance.

Do you have an iCharger? You can calibrate your shunt with that.

-methods

 

[Arlo1]

Methods what I did wast put my multi meter inline with the ground wire from my battery pack then I slowly crept up to full speed and compared it to the CA full speed freewheel was 3.6 amps. Now before I did this my CA said my freewheel amps were 4.2 with a shunt value set to 2600 my full throtle freewheel speed is 82 km/h. I have my v brakes a little tight but I don't think they are rubbing. I don't have an icharger. But I do have another shunt if needed (the old trick with 8awg wire and a multi meter prope at each end) I feel I am pretty close. I have my settings the same as you except I used 120deg instead of campatible. I am going to open the case and install some wires for the 3 position switch and the resistor mod and I will hook up the usb and just leave it in there and do a hack like you so I can program it without pulling it apart.

 

[methods]

Ok.
You may want to investigate though because 3.5A is too high for a freewheel current on that motor.

-methods

 

[Arlo1]

Ok.
You may want to investigate though because 3.5A is too high for a freewheel current on that motor.

-methods

Yeh what I am going to do is back the brakes way off and test again. And look at a few things. This is a motor stevo used to have and he changed the bearings in it he might have used something with more resistance. One amp seems to be pretty easy to waste eg. if the brakes drag ect. I also got the regen working and man from a free spin it stops fast! So tonight I will try them in a real worl test. I am not going to tell you where I got the resistance to on R12 At leaste not yeat :lol:

 

[methods]

Tell you what Dr. Danger...

If you set your regen where I think you have (someplace stupidly close to 100V) then I may sponsor you.
I have been looking for someone to do some testing on that.
If you smoke out your fets I will replace them at the playa price of $1.50 each. That sound good?
If you dont smoke them then I want to know what god-damn value you used

Dont forget that you are not just spoofing the regen with that R12 mod - that value spoofs the ECU's entire perception of the input voltage.

Also dont forget that the mosfets and capacitors are only rated for 100V.
Just because the DC battery voltage is only registering as 98V (or some other number below 100V) does not mean that the dynamic voltages and currents are not bouncing off the walls in there.

-methods

 

[Arlo1]

Tell you what Dr. Danger...

If you set your regen where I think you have (someplace stupidly close to 100V) then I may sponsor you.
I have been looking for someone to do some testing on that.
If you smoke out your fets I will replace them at the playa price of $1.50 each. That sound good?
If you dont smoke them then I want to know what god-damn value you used

Dont forget that you are not just spoofing the regen with that R12 mod - that value spoofs the ECU's entire perception of the input voltage.

Also dont forget that the mosfets and capacitors are only rated for 100V.
Just because the DC battery voltage is only registering as 98V (or some other number below 100V) does not mean that the dynamic voltages and currents are not bouncing off the walls in there.

-methods

You got a deal! I will try to test it tonight on a half pack then a full pack. I am more worried about shorting the fets and causing the rear wheel to lock so I will start at slow speeds. Then after all of this I will let you know all my data.

Does anyone know where to buy ebrakes hall or a switch?

 

[methods]

You can just use any old momentary switch
I suggest keeping the regen off of your main brakes
Problem is that you can not "ease" into it.
If you tap your brakes while leaning into a turn it throws you off balance

Leave your brakes separate
Set up a little thumb button for the ebraks.

That is the advice I give to people running big power.
For the lower power bikes ebrakes are fine.

My kelly used to kick on the 2KW ebrakes so hard that I would almost go over the bars
That was with it turned down to only 5% !! I did not want to push more than 20A back into the batteries

The next big-dog will be the guy who figures out how to get the regen to hit harder on these controllers....
Not sure what metric it is using... I guess just the R12 voltage and the shunt?

-methods

 

[Arlo1]

Thats why i am wondering about the hall brakes. I am wondering if it is suposed to be a 0-5 volt verance to gradualy control the brakes. I still have a couple good automotive throtle position sensors I can wire one in and try it as well.

Edit#1 Test one. so far so good I started with a battery at 90v and it is 24s 1p turnigy so LVC will come around easy for me (other pack is on its way)
I started testing in incraments from 10km/h and up incase it was to much. I think 10k is to slow to feel it. But as I went faster it hit harder. I toped out at about 65km/h and tried it and it was good it is not that bad having it wired to your brake lever like I do but I don't recomend it. I will be using a seprate switch instead. It is cool doing regen up hills ect but around a corner it would be to much if it is slipery out. I think I can also fell my axle moving in the dropouts when it is used it twists just a little so I will have to fab up another bracket to go on the outside of the dropout kinda like methods did. Oh and I had 9.7% for that ride and peak amps at 30 so not bad.

 

[Arlo1]

So test #2 I started with a 99.1v pack and had no regen at first but with my 1p pack it only takes a couple blasts to get it to start. I actual got it to regen the pack back up to just over 97v and then the regen shut off a couple of times. This time I got a regen of 8.5% and max -amps or -30.8 I tried to watch the amps on the go and -15 seemed to be the most it would do continous. My motor smells a little hot but its also the brakes lol I have sure burnt alot of my rear v brakes off in the last week. I tested it with a lazer temp gun and through one of my vent holes the hottest I could see was 163 deg f with an ambient temp of about 8 degC. I also checked freewheel amps and with the brakes adjusted back a bit it is down to 2.9 amps. So my r12 value when mesureing r12 is 900ohms seems to get regen to about 97volts and I would say that is perfact so a unballanced pack should not be damaged. I am not sure if the caps or anything are changing my mesurement of r12 or not but to get that I used a 3000 something ohm resistor in parallel with r12 to get it down. I will try to test more tomorow on a longer trip around town. For now I just though my quick 100v charger in my backpack!

 

[ZapPat]

Sounds like you're having fun discovering the pleasures of powerfull ebikes, Arlo!

For such a monster bike I would go with two torque arms, specially when using regen at the same time. Personnaly I've not had any dropout problems even when using *lots* of regen (it's still the only brake on my 2wd), but with two torque arms per wheel. After a fresh build I also check the main axle bolts before or after every ride, since there is often some settling in of the axle assembly during the first few rides.

You can lower your ebrake power level with the software's "EBS level" box - put it at "1:" for less braking than at "2:". I've never tried "0:" though, because I assume it would turn regen brakes off... or it might just be the weakest setting too? At any rate. lowering the EBS power level will help your controller deal with the regen... EDIT - What setting did you have it at for your tests?

As for the fears of regen stress, the most important thing to avoid is using lame high internal resistance batteries with regen. This is important to your controller as this means it will be seeing higher voltages during regen, since any previous "voltage sag" your batteries have during driving turns into "voltage peaks" during regen. This is potentially hard on the caps and FETs in your controller, specialy if you are operating at voltages close to component maximums. Also, the lame batteries themselves might not like unregulated regen if the current peaks are too high, but ones with OK C ratings will work well. In the same line of thought as this, also use heavy and short wires between the controller and battery pack. Thin wires would add resistance which would have a similar effect as having a lame battery pack, and long wires also add inductance which will make your caps work hard and create bigger voltage spikes in your controller.

The next big-dog will be the guy who figures out how to get the regen to hit harder on these controllers....
Not sure what metric it is using... I guess just the R12 voltage and the shunt?

I don't follow you here, Methods, are you saying you want it to brake *more* than it does now?? Better understanding the current sensing circuit would be the first step towards any regen current mods I guess. I see a couple resistors and caps on the PCB before the signal goes to the MCU. Maybe a positive voltage shift (for the regen current) and some filtering by the looks of it...

 

[Arlo1]

I have mine programed with ebs level 2 But I like it there! The regen is very powerfull at higher speeds but thats cool becasue you need to stop faster at those speeds. It progresivly decreases till about 10km/h where there is no regen. It is like the tourque of acceleration but in compleat reverse!

 

[ZapPat]

I have mine programed with ebs level 2 But I like it there! The regen is very powerfull at higher speeds but thats cool becasue you need to stop faster at those speeds. It progresivly decreases till about 10km/h where there is no regen. It is like the tourque of acceleration but in complreat reverse!

Groovy!

One really cool thing about the ebrake power slowly droping with wheel speed - This gives you something of an anti-blocking braking mecanism! I've gone down many really steep wood lot roads at fair speeds with my 2wd bike, and the surface was not nice with lots of loose gravel and water ruts and I felt pretty darn secure just on these ebrakes. Call me crazy if you like, but at least I have a redundant system in case on would ever fail... :lol: I still haven't installed Method's backup brake system suggestion, a 2"X3" on a pivot to the back tire, but am considering it.

 

[Arlo1]

Yeh it is hard to grasp the concept of not having a variable brake like we normaly use but I just ordered a ebrake to keep trying it. I have mine hooked to the rear brake lever and it is cool but I want something less geto. I may put a momentary switch in series with the ebrake so I have to push it as well but I am charging up right now to go ride around town to try it out and see if I want to get used to it. As for the axle..... well from all our crashes and my last burnout I blew the tire and we ground the end of the axle so when I pulled the nut off it wrecked it and then striped when I got everything back together so I used some washers to get by. But I oreded a new axle and nuts as well. My dropouts have been welded but I will be adding a pinch bolt on the end so in order to spread the dropouts apart it will have to brake the bolt. I will take pics when I am done. I do agree it feels alot like abs!

Edit: More tests and and it will regen if pack is below 97volts and it seems to charge back to 98 like there is a bit of and hystarious to it. I do wish I could have the regen with a variable amount of brakeing. I will test the tps (what you guys call a hall sensor) hopefully withing a few days. I have found my amp hour/km has gone from the low 50s to the low 40s just by using the regen and it says regen % 6.5 after todays ride. It was 45feg f out side and after using 300whs my motor is at 170deg smells a little hot but not bad and concidereing it is ripping up and down hills and passing cars lol its doing good.

 

[methods]

It is not your lipo batteries that I am worried about... Hell - I charge mine at 30A on the regular.

It is the mosfets I am worried about...

-methods

P.S. By hit harder I mean Extending form 0-2 on up to 3, 4, 5 etc. My Kelly would put out 40A no problem on low settings.

 

[Arlo1]

I realy think it is good I think It is good my pack has to be down to 97-98 volt think it will be ok I put on 30km today and all is well.
I just installed the 3 speed switch and found freewheel speeds for 60% on x1 to be 42km/h so that will be cop mode. But x2 at 80% and off at 110% are the same as well I am able to see the same free whell paek amps reving it up at 46amps I set it like this
60%
80%
110%

So thats with a 1/2 pack I am charging it up right now so I will go test in the real world in a bit.

Edit: so I went buy the post by BenMore and used pads x1 and x2 but I read I should have used x1 and x3? Man this thing is fast on 60% setting still wheelies actualy eaiser to handle and tops out at 42km/h up hill or on a flat! Speed settings 2&3 are the same but I think it is not working because It must be x3 I was suposed to use.

 

[Mike1]

Arlo, X2 is the correct pad but it activates the speed 3 setting. So you should have programmed it

Speed 1 60%
Speed 2 110% (Default setting with no switches on)
Speed 3 80%

 

[ZapPat]

It is not your lipo batteries that I am worried about... Hell - I charge mine at 30A on the regular.
It is the mosfets I am worried about...

When I was talking about lame high Ri batteries being bad for regen (see post above), I meant that they are hard on your controller components (FETs+caps) as these will let the voltage rise higher the more regen current you are pushing back into the battery. Of course good quality 20C Lipos have very low internal resistance so are very good for regen, as they will protect your controller by sucking in the amps without having their voltage rise very much at all. This is what is at the heart of the "regen is hard on your controller" argument, since regen is not any harder on your controller in any other way exept for this higher voltage caused by the negative current going into your battery.

As for raising the regen braking power, lowering the shunt's value already does this to the same extent as it raises the regular drive current limit. Other than this, it's up to the MCU to decide, so I would think it'll be impossible without reprogramming the thing. I doubt that we could just hack the programming software and add option "3:" to the EBS level setting box, but who knows?

 

[Arlo1]

Thanks mike yeh I did program it that way I just posted wrong.
SO I wonder if it is switching and 80%=110% or if I screwed up and soldierd it in the wrong whole?
I think all my current limiting is now in the batterys I only have 5ah @ 20 c of lipo so when I double up my pack I will be able to push it harder. I used the .250mOhm shunt and re-atached the top shunt and soldered it a bit too. I am at 2500 on the ca for programing.
I don't want the ebrakes to be any stronger on my bike. Zappat
I found they are to strong in most cases so I would like to be able to apply them more gradualy! I also don't want them weaker just don't want them to hit as hard when a car is behind me!

 

[GGoodrum]

I have no idea if this will work, but have you tried connecting a regular throttle to the brake input, instead of just a switch? Most brake inputs simply parallel the throttle input, so what you are doing when the switch is activated is grounding the throttle input. With this regen/brake input maybe partial "throttle" will give partial regen? I think this is how the Kellys work.

-- Gary

 

[Arlo1]

Thanks gary.
SO I did try applying the throttle as I use the brake and it is the same. But I am wondering about a throtle hooked to the brake to gradualy apply it as well. I will try this soon. It would still be nice in a perfact world to have something hooked to the rear brakes. Like a strain gauge that varies in presure the very the rear brakes!